Advances in Agronomy, Volume 79

vi

CONTENTS

VI. Applications of 15N Pool Dilution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Factors That Regulate Gross N Fluxes . . . . . . . . . . . . . . . . . . . . . . . . B. Quantifying the Actual Magnitude of N Cycling . . . . . . . . . . . . . . . . . C. C and N Mineralisation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Competition for Substrate Availability . . . . . . . . . . . . . . . . . . . . . . . . E. Soil Disturbance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Activity of Ammonia-oxidisers in Soil . . . . . . . . . . . . . . . . . . . . . . . . G. N Saturation Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII. Future Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

96 96 99 102 104 105 107 108 109 110 110

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS K. S. Dhillon and S. K. Dhillon I. II. III. IV. V. VI.

VII. VIII.

IX.

X.

XI. XII.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Historical Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Properties and Uses of Se . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rocks and Minerals Containing Se . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Formation of Seleniferous Soils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Se Additions to the Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Natural Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Anthropogenic Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Geographical Distribution of Se Toxic Soils . . . . . . . . . . . . . . . . . . . . . . . Se in the Seleniferous Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Seleniferous Soils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Se in Water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Se in Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Se in the Atmosphere . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Effects of Se Toxicity on the Components of the Ecosystem . . . . . . . . . . . A. Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Animals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Selenium Toxicity in Humans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Management of Seleniferous Soils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Soil Mixing/Covering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Soil Washing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Thermal Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Immobilization of Se in the Toxic Environment . . . . . . . . . . . . . . . . . E. Presence of Competitive Ions in Soil Solution . . . . . . . . . . . . . . . . . . . F. Selecting Plants with Low Se Absorption Capacity . . . . . . . . . . . . . . . G. Phytoremediation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . H. Bioremediation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Future Research Needs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

120 121 123 125 127 130 131 131 134 137 138 143 144 150 151 151 152 154 156 157 158 158 159 159 160 160 165 172 173 174

CONTENTS

vii

THE ROLE OF ARBUSCULAR MYCORRHIZAL FUNGI IN SUSTAINABLE CROPPING SYSTEMS L. A. Harrier and C.A. Watson I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. The Role of AM Fungi in Sustainable Farming Systems . . . . . . . . . . . . . . A. Crop Nutrition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Crop Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Crop Water Relations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Plant Reproduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Root Architecture and Longevity . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Soil Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G. Soil Microbial Populations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . H. Plant Populations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. Managing AM Fungi in Sustainable Agriculture . . . . . . . . . . . . . . . . . . . . IV. Direct Impact of Crop and Soil Management Practices on AMF . . . . . . . . A. Rotation Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Intercropping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Crop and Varietal Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Cultivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Nutrient Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Liming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G. Crop Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . H. Grazing Livestock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I. Inoculation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Effects of Different Agricultural Systems on AMF . . . . . . . . . . . . . . . . . . A. Organic Versus Biodynamic Versus Conventional Farming . . . . . . . . . B. Long-term Versus Short-term Effects . . . . . . . . . . . . . . . . . . . . . . . . . VI. Managing AMF in Sustainable Agriculture—Prospects for the Future . . . . A. Economics and Product Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Research and Development Needs . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

AS

186 189 189 193 194 195 196 196 197 198 198 199 199 201 201 202 203 205 205 205 206 206 206 208 209 209 209 210 210

CATCH CROPS AND GREEN MANURES BIOLOGICAL TOOLS IN NITROGEN MANAGEMENT IN TEMPERATE ZONES Kristian Thorup-Kristensen, Jacob Magid and Lars Stoumann Jensen

I. Why Use Catch Crops—How Do We Want Them to Affect the System? . . 228 A. N Effects of Catch Crops, Solving Environmental Problems . . . . . . . . . 230 B. N Effects of Catch Crops in Agriculture . . . . . . . . . . . . . . . . . . . . . . . 231

viii

CONTENTS

II. Uptake of N and Soil Depletion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Growth and Uptake Potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Root Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Relationship Between Soil Depletion and Nitrate Leaching Loss . . . . . D. Leguminous Green Manures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. Catch Crop Effect on N Supply for Subsequent Crops . . . . . . . . . . . . . . . . A. Calculating the N Effect, NEff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. What Determines Pre-emptive Competition . . . . . . . . . . . . . . . . . . . . . C. What Determines N Mineralisation . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Depth Distribution of Inorganic N . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Gaseous Losses of N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Field Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. Other Effects of Catch Crops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Effects on Other Nutrients Than N . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Effects on Soil Microbiological and Faunal Activity . . . . . . . . . . . . . . C. Effects on Soil Physical Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Effects on Soil Water Content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Effects on Pests, Pathogens and Weeds . . . . . . . . . . . . . . . . . . . . . . . . V. Making the Most of Catch Crops in Cropping Sequences and Whole Crop Rotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Placing Catch Crops in the Crop Rotation . . . . . . . . . . . . . . . . . . . . . . B. Establishing Catch Crops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Incorporation Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Choosing Catch Crop Species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Model Simulation of Catch Crops in the Crop Rotation . . . . . . . . . . . . F. Placing Catch Crops at the “policy Scale” . . . . . . . . . . . . . . . . . . . . . . VI. Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

232 233 235 237 240 241 244 244 245 250 253 254 259 260 262 263 263 264 266 267 269 270 273 280 289 291 292

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.): ITS ECOLOGY , AND USE AS A PASTURE LEGUME IN AUSTRALASIA Michael L. Smetham I. Characteristics of the Species Trifolium Subterraneum L . . . . . . . . . . . . . . A. Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Adaptation To Soil Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Physiological Response to the Environment . . . . . . . . . . . . . . . . . . . . . . . A. Vegetative Responses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Reproductive Responses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Influence On Germination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. Management of Subterranean Clover . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. History Of Use . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Choice Of Variety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

304 304 306 307 308 308 309 311 316 316 319

CONTENTS C. Establishment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Seed Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Impact Of Hardseeded On Performance . . . . . . . . . . . . . . . . . . . . . . . F. Hardseed Carryover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. Productivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Herbage Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Herbage Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Use As A Pure Sward Or In Mixture With Grass . . . . . . . . . . . . . . . . . V. Pests and Diseases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Viruses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Foliar Fungal Diseases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Root Fungal Diseases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Insects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Oestrogenicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ix 324 326 329 330 332 332 334 335 337 337 338 339 339 340 341 341

BREEDING HEVEA BRASILIENSIS FOR ENVIRONMENTAL CONSTRAINTS P. M. Priyadarshan I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Growing Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Ideal Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Marginal Areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. Constraints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Geo-climatic Stresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Biotic Stresses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. Hevea Under Marginal Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Immature Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Yield Depression, Patterns, Regimes and Specific Adaptation . . . . . . . V. Breeding Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Polyclonal Seedlings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Recombination Breeding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Increasing Genetic Diversity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Molecular Breeding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

352 356 356 357 360 360 366 369 369 371 376 376 378 382 384 391 392 393

INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 401

QUALITY SOIL MANAGEMENT

39

the water (sodium adsorption ratio—SAR and electrical conductivity—EC), and the amount, timing, leaching fraction and evaporation path of irrigation water at the soil surface affecting salt deposition relative to plant rows. These management factors govern the ability of salt-threatened soil to function more than intrinsic soil properties (Rhoades, 1972, 1998).

IX. FOCUS THE MESSAGE AND PRIORITIZE THE EFFORTS Pimentel (2000) listed three reasons why erosion control has not received the research and mitigation support it deserves, given the magnitude of its threat to humanity. The reasons were erosion’s insidious nature, its slowness relative to human perception, and the public’s lack of regard for the value of soil. He reasons, therefore, that “the soil erosion issue is out-competed by many other more dramatic events requiring public attention.” We submit that there are other reasons why erosion abatement does not receive the research or conservation support it deserves—reasons for which we, the soil science community, share culpability. They are our failure to prioritize, communicate unambiguously, and, as communicators say, “stay on message.” We believe that holistic lumping together sets of problems, focusing on whole-system assessments of what Leopold (1941) termed “sickness,” rather than using reductionist diagnostics, prioritized problem identification and amelioration, is a mistake. The environmental “health” philosophy is hotly debated across the environmental and resource sciences. As stated earlier, we recognize the value of collections of measures to characterize soils or other ecosystems or ecosystem components. It has been a routine approach to the science for decades. But, even health assessments rely on “triage” to prioritize action. In the emergency room, the chest wound takes priority over the blistered foot. In the doctor’s office, cancer is controlled before prescribing an exercise regime for muscle tone. Charities raise funds to battle specific maladies such as muscular dystrophy, heart disease and leukemia, not to defeat “poor health” nor to promote “good health.” Similarly, we believe that environmental stewardship and soil conservation, are better served by staying on message. To be effective, we need to cogently communicate specific prioritized problems. We need to focus attention on research and action toward prioritized, clearly identifiable, important and achievable solutions. We should avoid confusing the pedologically uninformed public and its funding agencies with unspecific concepts that we as scientists do not agree upon. Failing to stay on message, to be specific and to categorically prioritize, risks leaving a poorly informed public the option to overemphasize popular, but less critical issues while underemphasizing more critical but less well recognized or less politically correct issues. If we want to control erosion,

40

R. E. SOJKA, D. R. UPCHURCH AND N. E. BORLAUG

we need to identify erosion as the problem, not poor soil quality. Erosion imparts a specific image and target for conservation activists. Soil quality means different things to different audiences, in different places and even on different days. How soil quality takes on different meaning and policy implications in different settings can be better understood by comparing European and US agroenvironmental perspectives. Potter (1998) explored the basis for different expressions of agro-environmental reform policies in the USA, UK and EU. He postulated that in the US, erosion abatement and production management dominate the influences that have shaped policy, whereas in Europe chemical pollution abatement and concern for cultural integration of agriculture as manifest through landscape management are driving influences. In this context, soil quality assessment as a soil profile contamination-fighting tool is a conceptually discreet approach that fits and serves the European outlook reasonably well. In Europe, continental erosion is less dramatic than in the US and agricultural production is more highly subsidized and seen as essential to strategic and cultural independence, rather than as a spark plug of the economic engine. Institutionalization of soil quality in the US has been far less tentative than in Europe, despite a much more complex US definition and potentially expansive implementation and ramifications. Thus, international gatherings addressing soil quality must translate what each group means by soil quality. A largely production/erosion-driven soil conservation paradigm has shaped the soil quality movement to suit the US focus, whereas a pollution-driven paradigm suits Europe. Does it not make much more sense to address erosion, pollution, etc. in the first place? Karlen et al. (2001) chronicled development of the soil quality concept, listing the scientific disciplines and agencies that contributed to and influenced its direction and principles. However, we note that while calls for indices and institutional frameworks by scientists and government agencies are documented, no public call for the resulting indexing approach and institutionalization is documented. In their discussion of soil quality indexing they state “The expert opinion process functions best when a multidisciplinary team of scientists representing agronomy, ecology, economics, engineering, entomology, pathology, soil science, social science, or any other discipline deemed critical for the assessment being made can be assembled with land owners, operators, and other stakeholders.” This statement points to the complexity of indexing soil quality. However, it also removes any allusion to the base problems of managers that would explain the need for institutionalized indexing as an outcome. Lackey (2001) pointed to the same problem of balancing ecosystem assessment or indexing needs, versus the agenda of scientists or bureaucrats with vested interests in concept development. He stated: Understanding the values and preferences of society is crucial to appropriately implementing concepts of ecosystem health, but obtaining such understanding

QUALITY SOIL MANAGEMENT

41

credibly is difficult. To assert, however, that concepts of ecosystem health are merely scientific constructs is incorrect. As Russow (1995) concludes, “The claim that scientific descriptions in general or measures of ecosystem health in particular are value neutral is simply false.” The likely alternative to public involvement is that the values of scientists and other technocrats will be used as surrogates for societal values and preferences. We think implicit in Lackey’s statement is also the responsibility of public institutions promoting such arbitrary concepts to acknowledge, respond to and affect change based on criticism and dissent from within the scientific community.

X. ADVOCACY VERSUS SCIENCE Karlen et al. (2001) defend advocacy and incorporation of external values in soil quality assessment. They state “. . .all decisions are value-laden and dominated by personal experiences and expectations (Keeney and Raiffa, 1976; Mayhew and Alessi, 1998). Even seemingly objective decisions, such as which grant proposals to fund, are driven by personal social values and preferences (Keeney, 1992). Given that all decisions are biased, who is better qualified to interpret scientific indicators than the scientists who developed them.” We take exception to every tenet in the above quote. We question the validity of value-laden indicators to begin with, and see advocation of them as a compounded problem. Index developers have an obvious conflict of interest regarding interpretation of validity and scientific merit of the index they developed. Those steeped in the debate about value intrusion in sciences (especially applied sciences) are quick to emphasize that this debate is complex and arguments favoring detachment versus involvement cannot be set aside trivially. As Rykiel (2001a) stated “Scientists should be both objective and concerned. However, they bear a special responsibility to make a distinction between scientific statements and the values they associate with those statements.” Elsewhere Rykiel (2001b) notes that “Policy, which is our attempt to implement what ought to be, is based on values, not science.” Whereas he states “The work of science is to understand what is and how what is can lead to what might be. The work of policymakers is to wrestle what is and what might be into what ought to be.” Government decisions and public policy may contain bias, they may be forced to. We contend, however, that the science that serves as the information base for making decisions should strive to be as free as possible of bias and values. The Karlen et al. (2001) quote above is insensitive to the soil quality paradigm’s and institutional infrastructure’s consistent failure to even inform users that there is

42

R. E. SOJKA, D. R. UPCHURCH AND N. E. BORLAUG

scientific debate about the concept, its use, and the interpretation of soil quality indices—a debate which advocates have only acknowledged when forced to do so. That itself is one problem of advocacy within science. Who is better to interpret? In the case of soil quality, the existence of criticism within the science implies that not everyone agrees on whose interpretation is or should be regarded as authoritative. This is particularly noteworthy where interpretation leads to institutionalization and public policy implications or even recommendation of enforced policy (National Research Council, 1993). If nothing else, the existence of credible critics implies that any given interpretation is not unilaterally and doctrinally authoritative and would seem to demand care to at least note and cite counterarguments if not indeed present and consider them in detail. The difficulty of taking an advocative stance, that is, promoting a set of arbitrarily assigned values and policies within the construct of a supposedly empirical analytical index, is that the scientist ceases to be a scientific mediator among information users (managers), other stakeholders, and policy formulators. Instead, the scientist becomes one of the biased factions with vested interests in the outcome. Mackey (1999) explained the danger of scientists as advocates, “. . .the critical role scientists should be playing is that of mediators rather than advocates. The Oxford Dictionary defines an advocate as one who pleads the cause of another, or one who pleads, intercedes, or speaks for another. By practice, an advocate does not take an objective look at a situation and weigh the pros and cons to arrive at a reasoned position. The classical behavior of a lawyer in a court is therefore anathema to a good scientist.” He goes on to say that in contrast to an advocate, a mediator, “. . .uses the skills and knowledge at their disposal to help resolve a situation. The role of a mediator is neither neutral nor weak. On the contrary, it implies there is a concrete goal to be achieved, and that there are feuding advocates who need the wisdom of the mediator to help achieve a satisfactory solution. There are plenty of activists but precious few mediators.” We might even go further than this and argue that mediation is not the right construct either, but that as scientists our role is to reliably and without bias inform the debate. Pouyat (1999) said much the same thing, “if biologists and ecologists wish to be taken seriously in the policymaking process, they must work at being viewed as members of the scientific community rather than as part of the advocacy community.” As Rykiel (2001a) commented on this quote, “policymakers want the truth from scientists, not their personal opinions.” Truly scientific decisions are not biased and they do not depend on personal values or the beliefs of the person making or interpreting the measurements. Deciding if one object is hotter or colder than another does not require personal values, it requires temperature measurement. Deciding how much of chemical A must be added to chemical B for a complete reaction depends on knowledge of the system’s chemistry and physics, not personal values. Deciding if it is reasonable to expect a soil to support nitrogen fixation depends on presence of

QUALITY SOIL MANAGEMENT

43

symbiotic bacteria, not personal values or opinions or even the output of a soil quality model. Deciding if it is likely that a seed will germinate is determined by the critical soil water content for seed germination, the seed response to water, bulk density, temperature, soil chemistry, salinity and aeration, not personal opinion, economics or social values. Deciding on whether a given soil has a 100% rating for crop productivity is entirely dependent on the subjective social, economic, and philosophical parameters and perceptions that affect the choice of and interpretation of cropping system and arbitrarily acceptable or unacceptable system inputs. To argue, as do Karlen et al. (2001), that soil quality assessment is scientific, and in the same breath state it is justifiably as socially biased as grant proposal evaluation, is a contradiction in terms. If the rationale is that soil quality assessment is as humanly biased as grant proposal evaluation, then we argue for abandonment of the soil quality paradigm based on that presumption alone. A scientific index does not hinge on quixotic social, cultural, political, economic, programmatic, or topical values. A scientific index is accurate whether employed by a scientist or a non-scientist, under any circumstance by simple virtue of its adherence to universal scientific principles and physical reality. All decisions are not biased. Decisions depending on personal, cultural, economic, political, programmatic, or topical factors are inherently biased, but they should be forthrightly recognized as non-scientific decisions. And if it is these kinds of decisions that support or determine an index, then the index must be recognized as other than an objective scientific index. Objectivity is the purpose of science. Subjectivity is the purpose of values and belief systems. Science is evidence-driven. Values and belief systems are faith or convictiondriven. Karlen et al. (1997) noted in their conclusions that the concept of soil quality is “emotional and evolving.” We think it is fair to ask why that is so and what the implications of that statement are for a scientific concept, especially one already being institutionalized and suggested as a basis for government policy formulation. What issues have made the soil quality concept emotional? Karlen et al. (1997) attributed the emotionalism to different cultural views of soil. We submit that the reasons for the debate being emotional go far beyond that. Sources of emotion include concerns stemming from scientific, social and cultural value differences, problems and/or perceptions of unfairness and/or exclusivity, regional or taxonomic bias, unobjective data and parameter evaluation, premature institutionalization, concept redundancy, concept ambiguity, devaluing/renaming previously established functional concepts, disagreements about scientific approach, excessive emphasis on organic matter and organically oriented agricultural approaches, and concern for policy implications and intrusion of political correctness into the management philosophy of the soil resource. Acknowledging that the concept is still evolving, while already institutionalizing it, troubles many scientists’ sense of system logic. This concern is

44

R. E. SOJKA, D. R. UPCHURCH AND N. E. BORLAUG

compounded by the fact that many are unconvinced that the new paradigm has fully met the challenge of the scientific method. In 1984, Stephen Gould stated that “Science is all those things which are confirmed to such a degree that it would be unreasonable to withhold one’s provisional consent (Mackay, 1991).” Consensus does not exist in the community of soil scientists regarding soil quality, despite institutionalization. Science strives to eliminate any doubt as to the facts determined. Interpretation of facts, setting goals, and establishing environmental indices are matters of policy or belief systems, with inherent capacity for ambiguity, confusion, disagreement and even hostility (Lackey, 1998a,b; Zeide, 1998a,b; Callicott, 1998). A number of scientists have noted the potential pitfalls of the soil quality concept because of its heavy reliance upon numerous subjective concepts and value judgments (Linser, 1965; Letey et al., 2003; Scho¨nberger and Wiese, 1991; Singer and Ewing, 2000; Singer and Sojka, 2001; Sojka and Upchurch, 1999). Karlen et al. (1997) proposed tying soil quality evaluation to the relational non-absolute environmental philosophy of Aldo Leopold. The logic, ethical consistency, and scientific credibility of Leopold’s “Land Ethic” were critically examined by Zeide (1998a), raising significant questions as to its technical validity and appropriateness as a cornerstone for soil science—a discipline in which Leopold, a forester and game manager, had little actual expertise. Perhaps more importantly, contrary to the premise of Karlen et al. (1997), we do not believe that most soil scientists fail to assign adequate intrinsic value to soil, nor do we believe that they feel any less of a “special relationship with the earth” than “naturalists.” Rather, it is because of the soil science community’s general high regard for the soil resource that assigning “low quality” ratings to broad categories of soil is disturbing to many soil scientists. Referring to communication dilemmas associated with the soil quality lexicon, Karlen et al. (1997) stated: “. . .what would seem to be a relatively simple choice of words, can result in very different messages when delivered to our clients.” Some key words in the soil quality vocabulary bear heavy burdens of multiple meaning. “Quality” can be interpreted as degree of excellence, as in the conformance to a measurable standard; or it can refer to a categorical attribute or characteristic; in the environmental context, it has come to mean freedom from pollution. “Value” can mean financial, spiritual, emotional, cultural, or strategic worth; or it can mean the quantified numerical measure of a statistically analyzable parameter. Doran et al. (1996) also noted communication problems among various interested constituents concerning the term soil health. He noted that the dictionary defines health as the condition of an organism or one of its parts in which it performs its vital functions normally or properly. However, he also pointed out that this was by no means a satisfying definition for all scientists or stewards of the land, stating that the term has factionalized academics,

QUALITY SOIL MANAGEMENT

45

environmentalists, farmers and public land managers, with the end result that “The producers, and therefore society’s management of the soil, are caught in the middle of these opposing views and the communication failures that result.” Nonetheless, Doran et al. (1996) chose to use the term soil health as one completely interchangeable, and even preferable to soil quality. We see this as adding to the confusion of the US soil quality lexicon, which has an institutional definition that reserves the specific term soil quality to indicate the absolute potential of a given soil compared to other soils, and refers to the present state of the soil as its condition or health (Mausbach and Tugel, 1995). If the reader is not convinced that the soil quality paradigm and its various academic and institutionalized definitions and nomenclature have detrimentally impacted the technical lexicon of soil science, we would refer them to Patzel et al. (2000) who attempts to derive the etymology and epistemological basis of the terms soil fertility and soil quality. You may share our surprise at their conclusion that soil quality is a more precise and functionally specific term than soil fertility, stating “the concept of soil fertility has an almost infinite number of definitions.” They further assert “Firstly, the term “soil fertility” cannot be shaped as a technical term of natural sciences. . . Secondly, the term “soil fertility” is considered to be a qualitative dispositional term, which is not completely operationalizable in natural sciences, as its actual value can never be verified.” The paper by Patzel et al. (2000) ignores 100 years of clear communication by soil fertility experts and texts and attempts to convince us the last decade of disagreement about the meaning of soil quality is preferable. The Soil Science Society of America Glossary of Soil Science Terms clearly defines soil fertility as “The relative ability of a soil to supply the nutrients essential to plant growth,” (Soil Science Society of America, 1998). Lamentably Patzel et al. took no account of the mounting specific arguments in the literature (of which they were eminently aware) criticizing the term soil quality for having etymological attributes precisely the opposite of those that they assert. Meanwhile, to date, the literature uses the term soil quality ubiquitously and almost exclusively to mean productivity. This is despite an elaborate and repeated recitation of the mantra that soil quality refers to current status of the soil for a given function, based on the seminal rationale of Larson and Pierce (1991, 1994) and Pierce and Larson (1993) who proposed the term as a means of assessing transient soil status and expressly advocated movement away from a dominant application of the term to be productivity. To date the only application of soil quality has been as a substitute term for the given function of soil productivity. The preference of Doran et al. (1996) for the term soil health as a substitute hardly helps reduce the confusion deepened by Patzel et al. (2000). Most soil biologists and microbiologists, meanwhile, weight the term soil health to connote the diversity and/or activity of the spectrum of soil biota. Whereas, institutionally, transient soil status is defined interchangeably (for any specified use) as soil condition or soil health, even if that use is to support structures,

46

R. E. SOJKA, D. R. UPCHURCH AND N. E. BORLAUG

provide insulation, or any other engineering use, which often prefers soil sterility over biotic activity, or even SOM enrichment. Such inherent ambiguities, while a common aspect of policy debate, have always been regarded as unacceptable in development of scientific vocabularies and tenets. They create the potential of unintended outcomes when use of formulaic interpretations are taken out of the hands of scientists and left to the discretion of end users who could range from farmers to agricultural scientists, legislators to environmentalists, bankers to realtors, or lawyers to government bureaucrats. Many concept users will not have the soil science training or acumen needed to understand the subtleties of the concept, its ambiguities, or its potential pitfalls if improperly interpreted. Management does legitimately utilize scientific input to make decisions, where the objective inputs are coupled with values to determine an outcome. However, if the inputs are already biased by values, the manager will unwittingly be at the mercy of someone else’s judgment. This may be tolerable if the manager understands and agrees with the pre-existing bias. But where most index users are oblivious to the scientific debate or left uninformed of the debate surrounding an index, it makes managers using an index pawns to someone else’s value system, possibly even supporting a paradigm that the manager would otherwise choose to oppose if better informed. In a recent essay, Deichmann (2000) discussed the spectrum of implications and potential consequences for linking science to expedient policy, even if defined by proponents as being for the public good. He stated “. . .the call for politically responsible science, and hence more power for scientists, does not guarantee an ethical stance.” Environmentalists’ attempts in the 1980s to create a “political ecology” as the “guiding science of post-modernism” is a case in point. The intellectual origins of their criticisms of “causal reductionist” science lie in the 1920s when German ecologists, among them Karl Friedrich, proclaimed ecology as a path to “a view of the world, in which everything is related to everything else, everything directly or indirectly affects everything else.” Friedrich expanded this view of biology as a doctrine aimed at serving “the benefit of the people,” which was quickly subverted by the emerging political regime under the “doctrine of blood and soil.” We might add: when values are mixed with science, science can lose control over which values and agendas are ultimately served and whose interpretations will ultimately be empowered or for what motivations. We feel there needs to be clear separation between scientific analysis and formation or enforcement of public policy. Furthermore, between the two must be a scientist-mediator, knowledgeable about specific data, communicating with an information user who is knowledgeable about specific systems needs, limits and potentials. Substituting formulaic indices and simple analytical kits so that the uninformed in science or policy can make oversimplified do-it-yourself decisions seems an imprudent course for agronomic farm management, for environmental management, and for regional or national stewardship of farms, farmers,

QUALITY SOIL MANAGEMENT

47

the environment or the general public. We concur with the sentiment of Butler (2000) that “the role of science is to illuminate political choices, not enforce them.” And we share his concern for movement toward institutionalization of a nonuniversal scientific interpretation when “the most outspoken scientists on the matter tend to be those with interests in seeing the technology progress.” When there is debate, and particularly when there is sharp debate, Butler notes it is important that the institutions of science (not government or policy) should step to the forefront to guarantee balanced investigation and analysis of issues. Among his suggestions is the need to broaden the expertise of advisory committees. In his editorial he speaks of broadening into the community, but we submit it should also mean broadening to insure that opposing scientific views are aired openly and with equanimity. Singing to the choir does not make for harmonious science. He further suggests that scientists and their professional societies will need to be more active in carving out a role as honest brokers who can help clarify the issues and ensure impartial information. We feel these sentiments apply to the formulation of and institutionalization of the soil quality paradigm. Given the polarization among the soil science community regarding the soil quality concept, it is surprising the lack of acknowledgment and consideration of the counter arguments that appear in soil quality philosophical and promotional literature, bordering on what Sommer (2001) termed “bahramdipity” which he defines as roughly the opposite of serendipity. Where serendipity might be termed recognition of lucky discovery, bahramdipity might be thought of as intellectual “denial,” or sublimation. Bahramdipity can include insistence upon the correctness of an interpretation even in the face of facts presented that undermine or disprove it, and an unwillingness to even consider modifications that correct the disparities. We certainly hope that soil science has not adopted the Red Queen’s philosophy that “It’s too late to correct it. When you’ve said a thing that fixes it, and you must take the consequences.” Sojka and Upchurch (1999) included a section discussing “Plausible Ramifications and Unintended Outcomes.” That section dealt with plausible impacts of the soil quality concept and the manner of its advocacy on the soil science profession itself and with conceivable broad public policy implications based upon the published statements of soil quality advocates and institutional literature. Rather than repeat that discussion, which is somewhat removed from the concept and philosophy discussion we have assembled in this chapter and which has been alluded to in several of the preceding sections, we direct the interested reader to the previous publication for full consideration of those issues.

XI. GLOBAL PERSPECTIVE The environmental movement that began in the 1960s brought a unique, and appropriate new view to agricultural production and land management, globally.

48

R. E. SOJKA, D. R. UPCHURCH AND N. E. BORLAUG

However, in recent years, the movement has been captured by elitists, and has evolved more and more toward an anti-science, anti-technology reactionary force. Many of its leaders oppose high-yield crop production technology, including chemical fertilizers, herbicides, insecticides, fungicides, and now genetically modified high-yielding varieties. Critics of modern agricultural technologies should consider the impact on the environment had these technologies not been adopted over the past 40 years. An additional 67 million ha would have been required to provide the same wheat production in India, had farmers continued the use of the low-yielding pre-Green Revolution technology. This is but one example of the positive environmental impact of adoption of highyield production techniques, including both improved varieties and improved soil management. To put this in perspective, on a global scale, world cereal production increased from 650 million tons in 1950 to 1887 million tons in 1998. Using 1950’s technology, it would have required an additional 1150 million ha of cultivated land to produce this yield, while only 650 million ha were actually cultivated. While we agree that farmers should strive to return organic matter to the soil, through appropriate crop rotations, green manure crops, and animal manures, this does not address the full nutrient needs of the crop nor the environmental consequences discussed earlier in the chapter. Only 60% of our current world population can be supported without the use of chemical nitrogenous fertilizer (Smil, 1999a,b). The Sasakawa-Global 2000 (SG2000) program sponsored by the Nippon Foundation is aimed at food crop production technology transfer projects in sub-Saharan Africa. SG2000 and the Ministry of Agriculture jointly developed a package of improved crop production technologies for increasing food crop production. These include: (1) the use of the best available commercial varieties or hybrids, (2) proper land preparation and seeding to achieve good stand establishment, (3) proper application of the appropriate fertilizers and, when needed, crop protection chemicals, (4) timely weed control, and (5) moisture conservation and/or better water use if under irrigation (Borlaug and Dowswell, 2002). Local farmers using these practices have achieved crop yields two to four times higher than is typical with traditional production methods. It is generally agreed that world population will increase from the current 6 billion to around 7.6 billion people by the year 2020. It is likely that the demand for cereals, which accounts for 70% of our food supply, will increase by 40 –50%. The global arable land area potential for further expansion is limited. Therefore, most increases in global food supply must come from agricultural land already in production. It is estimated that 85% of the total growth in food supply must come from increased yield on land currently under cultivation (Pinstrup-Anderson and Pandya-Lorch, 2000). Formidable challenges exist for bringing unexploited, potentially arable, land into agricultural production. The Brazilian Cerrado, or savanna is a good case in point. The central Cerrado, with 175 million ha in one contiguous block, forms the bulk of the savanna lands. The soils of this area are

QUALITY SOIL MANAGEMENT

49

mostly various types of deep loam to clay-loam latosols (oxisols, ultisols), with good physical properties, but highly leached of nutrients. Thanks to targeted soil and agronomic research, today there is an agricultural revolution underway in the Cerrado. A new generation of improved crop varieties are moving on to farmers field’s. Improved crop management systems, built around crop rotations and minimum tillage, that facilitate infiltration and reduce runoff and erosion have been adopted. However, as in all of agriculture, further research is needed to solve specific, identifiable problems. Research is needed to define more exact fertilizer recommendations for various crops grown in the area. Since zero tillage is in widespread use, it is absolutely essential to develop crop rotations to minimize foliar infection with diseases that result from inoculum left in the soil or in crop residue from the previous season. The opening of the Cerrado will help assure an adequate world food supply, assuming wise policies are used to stimulate production. From a global perspective what is discussed in this chapter is only a portion of a broader issue. The current backlash against agricultural science and technology evident in some industrialized countries is difficult to comprehend. The world has the technology, either available or well advanced in the research pipeline, to feed on a sustainable basis a population of 10 billion people. The more pertinent question is whether farmers and ranchers will be permitted access to the continuing stream of new technologies to meet the challenges ahead. We are not short of new theories, we need people who understand land management.

XII. CONCLUSIONS Lal and Pierce (1991) stated, regarding stewardship of the soil resource, that “mismanagement and neglect can ruin the fragile resource and become a threat to human survival.” Evidence continues to mount that the survival of civilizations has been far less related to the “inherent soil quality” of their lands than to their ability to manage the lands. Indeed, some of the longest continually occupied population centers of the world exist in settings and on soils that the Sinclair et al. (1996) model would rate as low quality. While, soil properties certainly played a role, and the management’s success was related to the effect of agriculture on soil properties, the scenario was not the opposite. Namely, soil properties did not determine the success and longevity of the civilizations, independent of the management (Mann, 2000). The needed outputs from soil science to meet the requirements of a sustainable civilization are highly specific and easily identifiable research products and management goals. Their attainment is poorly served by obscuring them in vague quality, condition or health assessments that require deconstruction for interpretation, are not even universally agreed upon by advocates of the approach

50

R. E. SOJKA, D. R. UPCHURCH AND N. E. BORLAUG

and which, still worse, divides the scientific community rather than uniting it to achieve specific goals. Gomes (2000) noted that at present, about 80% of the world’s population lives in the poor, developing world. At current rates, in one lifetime of 75 or 80 years, poor countries will increase by 400% while the rich countries will grow by only about 8%. Today 47% of the global population lives in cities, and with virtually all of current growth happening in urban settings, this will be the last generation of humans to live mostly in rural areas. Despite this growth, the land under cultivation, which doubled from 1900 to 1960, has been nearly constant since then. The miracle of that last sentence deserves some comment. The Green Revolution, which began in the 1960s, marked the triumph of a philosophy that human ingenuity, turned to the management of natural resources and genetic potential (pronounce that agriculture) could end starvation and assure the survival of the species, while simultaneously halting the relentless advance of agriculture onto more and more of nature’s domain. Humans continue to expand their presence on the landscape, but the expansion is predominately for nonagricultural purposes. This was accomplished by focusing agricultural research on improving management to enhance production on existing lands. Along with this, again since about the 1960s, there has been an ever-increasing emphasis on achieving production goals while avoiding environmental degradation. This planetary accomplishment was achieved not by meta-scale analysis and metascale indexing, but by tackling individual discreet obstacles to production and environmental protection, one at a time, specifically, systematically, relentlessly, using the established and proven reductionist approach to science. As a result of the soil science and agronomic research that preceded the current institutionalized soil quality paradigm, each American farmer now feeds 143 people (Burton, 2000), more than a doubling since World War II. In the US, barely 2% of the population is engaged in farming and the average urban family spends less than 7% of its annual income on food (Abelson, 1995). Life expectancies in the US and worldwide have increased markedly over the last few generations because of improved food supplies and quality. Over the same time 25– 30 million km2 (the area of North America or Africa) have been spared from the plow because of improved agronomic and soil management technologies. This one fact is the greatest act of environmental protection achieved in the history of humankind. It has prevented billions of tons of erosion annually, staved off destruction of continental tracks of habitat, and avoided annual introduction of billions of pounds of additional agrochemicals and fertilizers on the lands spared. At the same time, in the last two to three decades, we have greatly reduced dependency on agrochemicals, nutrients, and destructive practices on lands we manage for agriculture. These achievements did not come from devoting the creative engine of agricultural research to renaming existing productivity indices and concepts. They resulted from working to solve

QUALITY SOIL MANAGEMENT

51

pressing and obvious production and environmental problems. These feats point to both the success of the American agricultural research system to provide relevant solutions to the American farmer and to the need to maintain our efforts. The need to remain focused and to solve problems has not changed. In the next generation, global agricultural production must rise more than 2% per year to meet rising population needs (Waggoner, 1994). This assumes we do not expand agriculture into new lands but, instead, provide our needs from existing farmland. By one estimate, more food must be grown in the first generation of the new millennium than was grown in the preceding 10,000 years of farming (Paarlberg, 1994). Patrick Moore, President of Greenspirit, a non-profit organization devoted to environmentalism, and one of the original founders of Greenpeace, has joined a group of world-renowned environmentalist, political leaders, and scientists (including three Nobel Prize winners) in signing the Declaration in Support of Protecting Nature with High-Yield Farming and Forestry. This group has recognized that the greatest protection to our global environment will come from supplying this increased food demand using currently cultivated land. Why have we been so outspoken about our concerns for the soil quality paradigm? As this chapter and the Sojka and Upchurch (1999) editorial should make clear, there are a variety of reasons. They can probably be summarized in a few categories. There are an extensive number of technical problems with the concept. To date, soil quality research has not addressed the important problems pointed out to them by scientists critical of the concept. By its own definition soil quality must function in several ways simultaneously. Many of the functions have contradictory requirements. Optimizing for one requirement can seriously impair others. No effort has been made to attempt integrated assessments. Many individual index components do not objectively weigh the full range of negative and positive impacts on all functions, including the production function. As with any broad index, a negative score must be deconstructed back to original component inputs to guide management. Managers prefer a full range of specific readouts over a red/green warning light. An index scale must have a threshold for reaction. If we continue to implement soil condition indexes that are little more than red/green indicators, how will we determine either the threshold value for reaction or what the reaction should be? How does one practically integrate this concept to cope with simultaneous conflicting functions? Who will determine the threshold and what are the policy implications for triggering an institutionalized regulatory reaction? Such triggering is determined by weighting of inputs to the index. Weighting can target the wrong functions or may represent social, political, cultural, institutional, and economic biases that do not accurately reflect the management problem or needs. We are far better served by having separate soil productivity indices, soil environmental indices and pedobiological indices that do not

52

R. E. SOJKA, D. R. UPCHURCH AND N. E. BORLAUG

confuse the user or run the risk of creating significant negative unintentional outcomes in other functions. There is regional and taxonomic bias in the concept. This is obvious from the models and maps that have been produced by the soil quality infrastructure that fail to account for documented regional agricultural inputs, environmental damages and economic production. An indexing system that devalues the least subsidized and most profitable soils while ignoring environmental problems in highly-rated soils does a disservice to American agriculture, our environment and the health of citizens. Andrews and Moorman (2002) implied that these criticisms are about regionalism and that because soil quality experiments have now been conducted in several states the criticism has been dealt with. They are wrong. The criticism is about the failure of the soil quality paradigm’s indexing approach, which simply becomes obvious when evaluated on a regional scale. It was not our choice to model the nation’s soil quality to a single relative scale. Soil scientists cannot be expected to ignore analysis of the model output given its implications and given its failure to explain production or input requirements. We stress the importance of management, because that is what agriculture and soil science and land stewardship embody. That set of values by its very nature is about erasing regional constraints through management. It de-emphasizes regional setting, initial conditions and indexing and looks to performance and outcome. To us every hectare is important. We eschew the labeling of soils as low quality, with the risk of marginalizing their importance. If the soil quality paradigm is positive potential-based, then we suggest placing less emphasis on identifying and classifying relative soil limitations and returning to emphasis on researching the management requirements to eliminate them. Soil performs a multitude of functions simultaneously. Only management can address the simultaneous needs of soil function. Integrated indexing of simultaneous functions has not been achieved. If it is, again, it will require establishment of response thresholds and index deconstruction to individual inputs for interpretation. What will the index do if the assessed status can only meet the criteria for one function by interfering with another? Ultimately the answer will lie in management to cope with the simultaneous and contradictory needs. It will be a pity if that need cannot be met because research resources were absorbed by indexing efforts instead of finding management solutions. To meet our obligations to the future, we feel we should prioritize the targeting of our research toward known discrete problems, that are clearly identifiable and defensible, and whose solution can be clearly pointed to with easily measured impact and cost benefit analysis. If we expect to be supported in the work toward solution of known important problems, we need to stay on message, not confuse the public by constantly renaming the problems we are attempting to solve, nor watering down their specific individual importance by homogenizing them in ambiguous feel-good vocabularies that mean all things and nothing to the tax paying public.

QUALITY SOIL MANAGEMENT

53

The philosophical basis of the soil quality paradigm would lead one to believe that an entirely new and highly complicated soil assessment construct is needed to identify the planet’s or individual farmers’ most critical problems impacting food and fiber production, economic return and environmental protection. Nothing could be further from the truth. The problems are obvious. The expression of the problems is anything but subtle, and sadly, their impacts on production and the environment are often egregious, not veiled. The key deficiency in natural resource based research is not the ability to locate, identify, and rate problems. The deficiency is our inability to solve them, affordably, promptly, effectively, permanently, sustainably and using approaches, technologies and communication techniques that can be understood by, are practical for and that will be accepted by managers. We do not need investment in developing new yield plateau plots; we need new management to raise the plateaus. The challenge of our generation is to achieve this while simultaneously continuing to improve our protection of the environment. In the not so distant future, a generation or two, when the authors of this chapter are already achieving the status of mere anecdotal asterisks, the Malthusian projections alluded to in the earlier paragraphs will either be realities, or potential problems that were averted. The latter scenario demands that we make significant choices now about how we choose to prioritize our research and invest our research dollars and energy. We submit that choosing to elevate the direct solution of known critical problems is a wiser path than the development of subjective indices for debatable assessment of subtle soil status variations. The next generation of soil scientists, agronomists and environmental stewards and our children and grandchildren will be better served by full stomachs, clean water, clean air and preserved wild lands than by incremental improvement in an arbitrary soil rating.

REFERENCES Abelson, P. H. (1995). International agriculture. Science 268, 11. Akinremi, O. O., McGinn, S. M., and McLean, H. D. J. (1999). Effects of soil temperature and moisture on soil respiration in barley and fallow plots. Can. J. Soil Sci. 79, 513. Albrecht, W. A. (1938). Soils and Men, 1938 Yearbook of Agriculture, House Document 398. In “Loss of organic matter and its restoration”, pp. 347 –360. U.S. Government Printing Office, Washington, DC. Albrecht, W. A. (1975). In “The Albrecht Papers”. (C. Walters, Ed.), Acres USA, Raytown, MO. Alegre, J. C., Pashanasi, B., and Lavelle, P. (1996). Dynamics of soil physical properties in Amazonian agroecosystems inoculated with earthworms. Soil Sci. Soc. Am. J. 60, 1522–1529. Alexander, M. (1971). An Environmental Improvement—Agriculture’s Challenge in the Seventies. In “Agriculture’s responsibility in establishing soil quality criteria”, pp. 66–71. National Academy of Sciences, Washington, DC.

54

R. E. SOJKA, D. R. UPCHURCH AND N. E. BORLAUG

Allan, D. L., Adriano, D. C., Bezdicek, D. F., Cline, R. G., Coleman, D. C., Doran, J. W., Haberern, J., Harris, R. G., Juo, A. S. R., Mausbach, M. J., Peterson, G. A., Schuman, G. E., Singer, M. J., and Karlen, D. L. (1995). SSSA statement on soil quality. Agron. NewsJune, 7. Alvarez, R., Santanatoglia, O. J., Daniel, P. E., and Garcia, R. (1995). Respiration and specific activity of soil microbial biomass under conventional and reduced tillage. Pesq. Agropec. Bras. 30, 701–709. Alvarez, R., Santanatoglia, O. J., and Garcia, R. (1995). Effect of temperature on soil microbial biomass and its metabolic quotient in situ under different tillage systems. Biol. Fert. Soils 19, 227–230. Anderson, J. E. (1991). A conceptual framework for evaluating and quantifying naturalness. Conserv. Biol. 5, 347 –352. Andreasen, C., Streibig, J. C., and Haas, H. (1991). Soil properties affecting the distribution of 37 weed species in Danish fields. Weed Res. 31, 181–187. Andrews, S. S., and Moorman, T. B. (2002). Symposium papers introduction. Agron. J. 94, 1–2. Annin, P. (1999). Down in the dead zone. Newsweek 134, 60–61. Anonymous (1996a). Soil Quality Information Sheet: Soil Quality—Introduction. USDA Natural Resources Conservation Service National Soil Survey Center and Soil Quality Institute, and USDA Agricultural Research Service National Soil Tilth Lab. Anonymous (1996b). Soil Quality Information Sheet: Soil Quality Indicators: Organic matter. USDA Natural Resources Conservation Service National Soil Survey Center and Soil Quality Institute, and USDA Agricultural Research Service National Soil Tilth Lab. Anonymous (1996c). Soil Quality Information Sheet: Soil Quality Indicators: Indicators for Soil Quality Evaluation. USDA Natural Resources Conservation Service National Soil Survey Center and Soil Quality Institute, and USDA Agricultural Research Service National Soil Tilth Lab. Anonymous, (1997). In “A Crop Protection Reference” (13th ed.), pp. 2230. C&P Press, New York. Anonymous (1998a). http://www.statlab.iastate.edu/survey/SQI/. Anonymous, (1998). In “What’s New. . .Soil Quality Test Kit Guide”. USDA Soil Quality Institute, Ames, IA. Anonymous, (1998). In “Willamette Valley Soil Quality Card”. USDA Soil Quality Institute, Corvallis, OR. Arshad, M. A., and Coen, G. M. (1992). Characterization of soil quality: physical and chemical criteria. Am. J. Altern. Agric. 7, 25 –31. Avery, D. T. (1994). Hudson Briefing Paper, No. 162. In “The Organic Farming Threat to People and Wildlife”, 12 pages. The Hudson Institute, Indianapolis, IN. Avery, D. T. (1995). In “Saving the Planet with Pesticides and Plastic: The Environmental Triumph of High-yielding Farming”. Hudson Institute, Indianapolis, IN. Baltruschat, H., and Dehne, H. W. (1988). The occurrence of vesicular–arbuscular mycorrhiza in agroecosystems. I. Influence of nitrogen fertilization and green manure in continuous monoculture and in crop rotation on the inoculum potential of winter wheat. Plant Soil 107, 270– 284. Banks, P. A., Santlemann, P. W., and Tucker, B. B. (1976). Influence of long-term soil fertility treatments on weed species in winter wheat. Agron. J. 68, 825 –827. Barriuso, E., Baer, U., and Calvet, R. (1992). Dissolved organic matter and adsorption–desorption of dimefuron, atrazine and carbetamide by soils. J. Environ. Qual. 21, 359–367. Barriuso, E., Laird, D. A., Koskinen, W. C., and Dowdy, R. H. (1994). Atrazine desorption from smectites. Soil Sci. Soc. Am. J. 58, 1632–1638. Beauchemin, S., Simard, R. R., and Cluis, D. (1998). Forms and concentration of phosphorus in drainage water of twenty-seven tile-drained soils. J. Environ. Qual. 27, 721–728. Benoit, D. L., Derksen, D. A., and Panneton, B. (1992). Innovative approaches to seedbank studies. Weed Sci. 40, 660 –669. Bergamaschi, B. A., Fram, M. S., Kendall, C., Silva, S. R., Aiken, G. R., and Fujii, R. (1999). Org. Geochem. 30, 835–842.

QUALITY SOIL MANAGEMENT

55

Bjorneberg, D. L., Kanwar, R. S., and Melvin, S. W. (1996). Seasonal changes in flow and nitrate-N loss from subsurface drains. Trans. ASAE 39, 961–976. Black, B. D., Harrington, G. W., and Singer, P. C. (1996). Reducing cancer risks by improving organic carbon removal. J. Am. Water Works Assoc. 88, 40–52. Blair, J. M., Allen, M. F., Parmelee, R. W., McCartney, D. A., and Stinner, B. A. (1996). Changes in soil N pools in response to earthworm population manipulations in agroecosystems with different N sources. Soil Biol. Biochem. 29, 361 –367. Blum, W. E. H. (1993). Soil protection concept of the council of Europe and integrated soil research. In “Integrated Soil and Sediment Research: A Basis for Proper Protection”. (H. J. P. Eijsackers and T. Hamers, Eds.), pp. 37 –47. Kluwer Academic Publishers, Dordrecht. Boone, R. D., Nadelhoffer, K. J., Canary, J. D., and Kaye, J. P. (1998). Roots exert a strong influence on the sensitivity of soil respiration. Nature 396, 570– 572. Boorman, G. A., Dellarco, V., Dunnick, J. K., Chapin, R. E., Hunter, S., Hauchman, F., Gardner, H., Cox, M., and Sills, R. C. (1999). Drinking water disinfection byproducts: review and approach to toxicity evaluation. Environ. Health Perspect. 107(Suppl. 1), 207–217. Borken, W., Gru¨ndel, S., and Beese, F. (2000). Potential contribution of Lumbricus terrestris L. to carbon dioxide, methane and nitrous oxide fluxes from a forest soil. Biol. Fert. Soils 32, 142–148. Borlaug, N. E., and Dowswell, C. R. (2002). Prospects for World Agriculture in the 21st Century, In “Proceedings of the Workshop on No-till Farming in South Asia’s Rice –Wheat System: Experiences from the Rice –Wheat Consortium and the USA, February 21, Columbus, OH.”. Bosch, C. (1991). Bodenfruchtbarkeit und Bodenschutz im Konzept der o¨kologischen Landwirtschaft. Berichte u¨ber Landwirtschaft. Neue Folge. 203. Sonderheft. 59–63. Bouma, J. (1997). Soil environmental quality: a European perspective. J. Environ. Qual. 26, 26–31. Bradley, P. M., McMahon, P. B., and Chapelle, F. H. (1995). Effects of carbon and nitrate on denitrification in bottom sediments of an effluent-dominated river. Water Resour. Res. 31, 1063–1068. Brechelt, A. (1987). Wirkung verschiedener organischer Dungemittel auf die effizienz der va-mykorrhiza. J. Agron. Crop Sci. 158, 280– 286. Brechelt, A. (1989). Effect of different organic manures on the efficiency of VA mycorrhiza. Agric. Ecosyst. Environ. 29, 55– 58. Bremer, D. J., Ham, J. M., Owensby, C. E., and Knapp, A. K. (1998). Response of soil respiration to clipping and grazing in a tallgrass prairie. J. Environ. Qual. 27, 1539–1548. Broussard, M., Lavelle, P., and Laurent, J. Y. (1996). Digestion of a Vertisol by the endogeic earthworm Polypheretima elongata, Megascolecidae increases soil phosphate extractability. Eur. J. Soil Biol. 32, 107 –111. Bucks, D. A., Sammis, T. W., and Dickey, G. L. (1990). Irrigation for arid areas. In “Management of Farm Irrigation Systems”. (G. J. Hoffman, T. A. Howell and K. H. Solomon, Eds.), pp. 499–548. ASAE, St Joseph, MI. Bundt, M., Widmer, F., Pesaro, M., Zeyer, J., and Blaser, P. (2001). Preferential flow paths: biological ‘hot spots’ in soils. Soil Biol. Biochem. 33, 729–738. Burkart, M. R., and James, D. E. (1999). Agricultural-nitrogen contributions to hypoxia in the gulf of Mexico. J. Environ. Qual. 28, 850–859. Burton, V. (2000). Research, education, and economics. USDA News 59(8), 5. Butler, D. (2000). The role of science is to illuminate political choices, not enforce them. Nature 403, 6 –7. Callicott, J. B. (1998). A critical examination of “Another look at Leopold’s land ethic”. J. Forest. 96(4), 20 –28. Calow, P. (1995). Ecosystem health—a critical analysis of concepts. NATO ASI Series 1. Global Environmental Change. In “Evaluating and Monitoring Health of Large Scale Ecosystems”. (D. J. Rapport, C. L. Gaudet and P. Calow, Eds.), Vol. 28, pp. 33–41. Springer, Berlin. Calow, P. (2000). Critics of ecosystem health misrepresented. Ecosyst. Health 6, 3– 4.

56

R. E. SOJKA, D. R. UPCHURCH AND N. E. BORLAUG

Cambardella, C. A., and Elliott, E. T. (1993). Carbon and nitrogen distribution in aggregates from cultivated and native grassland soils. Soil Sci. Soc. Am. J. 57, 1071–1076. Cantor, K. P., Lynch, C. F., Hildesheim, M. E., Dosemeci, M. E., Lubin, J., Alavanja, M., and Craun, G. F. (1998). Drinking water source and chlorination byproducts in Iowa. I. Risk of bladder cancer. Epidemiology 9, 21–28. Carter, A., Heinonen, J., and de Vries, J. (1982). Earthworms and water movement. Pedobiologia 23, 395–397. CAST, (1999). In “Interpretive Summary. Gulf of Mexico Hypoxia: Land and Sea Interactions”, (2 pages). Council for Agricultural Science and Technology, Ames, IA. CENR, (2000). In “Integrated Assessment of Hypoxia in the Northern Gulf of Mexico”. National Science and Technology Council Committee of Environment and Natural Resources, Washington, DC. Chiou, C. T., Kile, D. E., Brinton, T. I., Malcolm, R. L., and Leenheer, J. A. (1987). A comparison of water solubility enhancements of organic solutes by aquatic humic materials and commercial humic acids. Environ. Sci. Technol. 21, 1231– 1234. Chotte, J. L., Ladd, J. N., and Amato, M. (1998). Sites of microbial assimilation, and turnover of soluble and particulate 14C-labelled substrates decomposing in a clay soil. Soil Biol. Biochem. 30, 205– 218. Clancy, K. L. (1986). The role of sustainable agriculture in improving the safety and quality of the food supply. Am. J. Altern. Agric. 1, 11 –18. Clark, R. M., Goodrich, J. A., and Deininger, R. A. (1986). Drinking water and cancer mortality. Sci. Total Environ. 53, 153–172. Coghlan, A. (2001). Common Cause. New Scientist. 172, 17. Cohen, S. (1997). Macropores in turfgrass soil columns. Ground Water Monitor Remed. 17(4), 59 –62. Comis, D. (1999). Tracking manure-borne pathogens. Agric. Res. 47, 22. Condron, L. M., Cameron, K. C., Di, H. J., Clough, T. J., Forbes, E. A., McLaren, R. G., and Silva, R. G. (2000). A comparison of soil and environmental quality under organic and conventional farming systems in New Zealand. NZ J. Agric. Res. 43, 443 –466. Cook, N., and Hendershot, W. H. (1996). The problem of establishing ecologically based soil quality criteria: the case of lead. Can. J. Soil Sci. 76, 335–342. Cox, C. (1995). Soil quality: goals for national policy. J. Soil Water Conserv. 50, 223. Culliney, T. W., Pimentel, D., and Pimentel, M. H. (1992). Pesticides and natural toxicants in foods. Agric. Ecosyst. Environ. 41, 297 –320. Curry, J. P. (1976). Some effects of animal manures on earthworms in grassland. Pedobiologia 16, 425–438. David, M. B., Gentry, L. E., Kovacic, D. A., and Smith, K. M. (1997). Nitrogen balance in and export from an agricultural watershed. J. Environ. Qual. 26, 1038–1048. Davidson, E. A., Belk, E., and Boone, R. D. (1998). Soil water content and temperature as independent or confounded factors controlling soil respiration in a temperate mixed hardwood forest. Global Change Biol. 4, 217 –227. Deichmann, U. (2000). An unholy alliance. Nature 405, 739. Deileman, J. A., Mortensen, D. A., and Buhler, D. D. (1997). Multivariate approaches for linking field-scale variability of soil properties and weed populations. Weed Sci. Soc. Am. Abstr. 37, 46. Derbruck, J. (1981). Der Einfluß der Fruchtfolge auf die Bodenfruchtbarkeit. Die Bodenkultur. 32, 215–222. Ditzler, C. A., and Tugel, A. J. (2002). Soil quality field tools: experiences of USDA–NRCS Soil Quality Institute. Agron. J. 94, 33–38. Doran, J. W. and Jones, A. J. (Eds.), (1996). Methods for assessing soil quality. SSSA Special Publication 49. In “Methods for Assessing Soil Quality”, 410 pages. SSSA, Madison, WI. Doran, J. W., and Parkin, T. B. (1994). Defining and assessing soil quality. SSSA Special Publication 35. In “Defining Soil Quality for a Sustainable Environment”. (J. W. Doran, D. C. Coleman, D. F. Bedzicek and B. A. Stewart, Eds.), pp. 3–21. SSSA, Madison, WI.

QUALITY SOIL MANAGEMENT

57

Doran, J. W., and Weinhold, B. J. (2001). Regarding agricultural accomplishments and impending concerns (A letter to the editor). Soil Sci. 166, 295. Doran, J. W., Coleman, D. C., Bedzicek, D. F. and Stewart, B. A. (Eds.), (1994). Defining soil quality for a sustainable environment. SSSA Special Publication 35, 244 pages. SSSA/ASA, Madison, WI. Doran, J. W., Sarrantonio, M., and Liebig, M. A. (1996). Advances in Agronomy. (D. L. Sparks, Ed.), Vol. 56, pp. 1–54. Academic Press, San Diego. Doyle, T. J., Zheng, W., Cerhan, J. R., Hong, C. P., Sellers, T. A., Kushi, L. H., and Folsom, A. R. (1997). The association of drinking water source and chlorination byproducts with cancer incidence among post menopausal women in Iowa: a prospective cohort study. Am. J. Public Health 87, 1168–1176. Edwards, N. T. (1975). Effects of temperature and moisture on carbon dioxide evolution in a mixed deciduous forest floor. Soil Sci. Soc. Am. Proc. 39, 361 –365. Edwards, C. A., and Bohlen, P. J. (1996). In “Biology and Ecology of Earthworms” (3rd ed.). Chapman and Hall, London. Edwards, C. A., and Lofty, J. R. (1977). In “Biology of Earthworms” (2nd ed.), 333 pages. Chapman and Hall, London. Edwards, W. M., Shipitalo, M. J., Owens, L. B., and Norton, L. D. (1989). Water and nitrate movement in earthworm burrows within long-term no-till cornfields. J. Soil Water Conserv. 44, 240 –243. Edwards, W. M., Shipitalo, M. J., Traina, S. J., Edwards, C. A., and Owens, L. B. (1992). Role of Lumbricus terrestris (L.) burrows on the quality of infiltrating water. Soil Biol. Biochem. 24, 1555–1561. Edwards, W. M., Shipitalo, M. J., Owens, L. B., and Dick, W. A. (1993). Factors affecting preferential flow of water and atrazine through earthworm burrows under continuous no-till corn. J. Environ. Qual. 22, 453 –457. Ehlers, W. (1975). Observations on earthworm channels and infiltration on tilled and untilled loess soil. Soil Sci. 119, 242–249. Ellis, J. R., Roder, W., and Mason, S. C. (1992). Grain sorghum–soybean rotation and fertilization influence in vesicular–arbuscular mycorrhizal fungi. Soil Sci. Soc. Am. J. 56, 789–794. Entry, J. A., Sojka, R. E., and Shewmaker, G. E. (2002). Management of irrigated agriculture to increase carbon storage in soils. Soil Sci. Soc. Am. J. 66, 1957–1954. Ewel, K. C., Cropper, W. P., Jr., and Gholz, H. L. (1987). Soil CO2 evolution in Florida Slash pine plantations. I. Changes through time. Can. J. Forest Res. 17, 325–329. Fausey, N. R., Brown, L. C., Belcher, H. W., and Kanwar, R. S. (1995). Drainage and water quality in Great Lakes and Cornbelt states. J. Irrig. Drainage Engng. 121, 282 –288. Fenelon, J. M., and Moore, R. C. (1998). Transport of agrichemicals to ground and surface water in a small Central Indiana watershed. J. Environ. Qual. 27, 884 –894. Fitter, A. H., Graves, J. D., Self, G. K., Brown, T. K., Bogie, D. S., and Taylor, K. (1998). Root production, turnover, and respiration under two grassland types along an altitudinal gradient: influence of temperature and solar radiation. Oecologia 114, 20–30. Flury, M. (1996). Experimental evidence of transport of pesticides through field soils—a review. J. Environ. Qual. 25, 25– 45. Flury, M., Flu¨hler, H., Jury, W. A., and Leuenberger, J. (1994). Susceptibility of soil to preferential flow of water: a field study. Water Resour. Res. 30, 1945– 1954. Focht, U. (1999). The effect of smoke from charcoal kilns on soil respiration. Environ. Monitor. Assess. 59, 73–80. Freeman, C., Evans, C. D., Monteith, D. T., Reynolds, B., and Fenner, N. (2001). Export of organic carbon from peat soils. Nature 412(6849), 785. Friend, J. J., and Chan, K. Y. (1995). Infuence of cropping on the population of a native earthworm and consequent effects on hydraulic properties of vertisols. Aust. J. Soil Res. 33, 995 –1006.

58

R. E. SOJKA, D. R. UPCHURCH AND N. E. BORLAUG

Fujii, R., Renalli, A. J., Aiken, G. R., and Bergamaschi, B. A (1998). Dissolved organic carbon concentrations and compositions, and trihalomethane formation potentials in waters from agricultural peat soils, SacramentoSan Joaquin Delta, California: implications for drinking water quality. In “USGS Water Resources Investigations Report 984147.” California Department of Water Resources, Sacramento, CA. Gagliardi, J. V., and Karns, J. S. (2000). Leaching of Escherichia coli O157:H7 in diverse soils under various agricultural management practices. Appl. Environ. Microbiol. 66, 877 –883. Erratum noted. Garcia, F. O., and Rice, C. W. (1994). Microbial biomass dynamics in tallgrass prairie. Soil Sci. Soc. Am. J. 58, 816 –823. Gaston, L. A., Locke, M. A., Zablotowicz, R. M., and Reddy, K. N. (2001). Spacial variability of soil properties and weed populations in the Mississippi Delta. Soil Sci. Soc. Am. J. 65, 449– 459. Gaudet, C. L., Wong, M. P., Brady, A., and Kent, R. (1997). How are we managing? The transition from environmental quality to ecosystem health. Ecosyst. Health 3, 3–10. Ghodrati, M., and Jury, W. A. (1992). A field study of the effect of soil structure and irrigation method on preferential flow of pesticides in unsaturated soil. J. Contam. Hydrol. 11, 101–120. Gilot, C (1994). Effets de l’introduction du ver geophage tropical Millsonia anomala Omodeo en systemes cultives sur les caracteristiques des sols et la production vegetale en Moyenne Cote d’Ivoire. PhD Dissertation University of Paris VI/Institut National Agronomique de ParisGringon. Gomes, W. R. (2000). Population change to challenge Californians. California Agric. 54, 4–8. Goolsby, D. A., Battaglin, W. A., Aulenbach, B. T., and Hooper, R. P. (1999). Nitrogen fluxes and sources in the Mississippi River Basin. In “U.S. Geological Survey Water-Resources Investigations Report 99-418B”. (D. W. Morganwalp and H. T. Buxton, Eds.), pp. 203 –213. Gordon, A. M., Schlenter, R. E., and Van Cleve, K. (1987). Seasonal patterns of soil respiration and CO2 evolution following harvesting in the white spruce forests of interior Alaska. Can. J. Forest Res. 17, 304 –310. Graber, E. R., Gerstl, A., Fischer, E., and Mingelgrin, U. (1995). Enhanced transport of atrazine under irrigation with effluent. Soil Sci. Soc. Am. J. 59, 1513–1519. Grahammer, K., Jawson, M. D., and Skopp, J. (1991). Day and night soil respiration from a grassland. Soil Biol. Biochem. 23, 77–81. Granatstein, D., and Bezdicek, D. F. (1992). The need for a soil quality index: local and regional perspectives. Am. J. Altern. Agric. 7, 12 –16. Gregorich, E. G., Carter, M. R., Angers, D. A., Monreal, C. M., and Ellert, B. H. (1994). Towards a minimum data set to assess SOM quality in agricultural soils. Can. J. Soil Sci. 74, 367–385. Grolchulska, J., and Kladivko, E. J. (1994). A two-region model of preferential flow of chemicals using a transfer function approach. J. Environ. Qual. 23, 498–507. Haberern, J. (1992). A soil health index. J. Soil Water Conserv. 47, 6. Hagedorn, F., Mohn, J., Schleppi, P., and Flu¨hler, H. (1999). The role of rapid flow paths for nitrogen transformation in a forest soil: a field study with micro suction cups. Soil Sci. Soc. Am. J. 63, 1915–1923. Hall, J. K., Murray, M. R., and Hartwig, N. L. (1989). Herbicide leaching and distribution in tilled and untilled soil. J. Environ. Qual. 18, 439–445. Hall, J. K., Mumma, R. O., and Watts, D. W. (1991). Leaching and runoff losses of herbicides in a tilled and untilled field. Agric. Ecosyst. Environ. 37, 303– 314. Hallberg, G. R. (1987). Agricultural chemicals in ground water: extent and implications. Am. J. Altern. Agric. 2, 3–15. Halvorson, J. J., Smith, J. L., and Papendick, R. I. (1996). Integration of multiple soil parameters to evaluate soil quality: a field example. Biol. Fert. Soils 21, 207 –214. Hampson, M. C., and Coombes, J. W. (1989). Pathogenesis of Synchytrium endobioticum. VII. Earthworms as vectors of wart disease of potato. Plant Soil 116, 147–150.

QUALITY SOIL MANAGEMENT

59

Haraldsen, T. K., Larsen, M. A., and Sveistrup, T. E. (1994). Effects of cattle slurry and soil compaction on the earthworm population in a silty clay loam soil in central Norway. Norwegian J. Agric. Sci. 8, 231–241. Harinikumar, K. M., and Bagyaraj, D. J. (1989). Effect of cropping sequence, fertilizers and farmyard manure on vesicular–arbuscular mycorrhizal fungi in different crops over three consecutive seasons. Biol. Fert. Soils 7, 173–175. Harris, R. F., Karlen, D. L., and Mulla, D. J. (1996). A conceptual framework for assessment and management of soil quality and health. SSSA Special Publication 49. In “Methods for Assessing Soil Quality”. (J. W. Doran, D. C. Coleman, D. F. Bedzicek and B. A. Stewart, Eds.), pp. 61–82. SSSA, Madison, WI. Hassett, J. P., and Anderson, M. A. (1982). Effects of dissolved organic matter on adsorption of hydrophobic organic compounds by river- and sewage-borne particles. Water Res. 16, 681–686. Hayes, M. H. B. (1970). Adsorption of triazine herbicides on soil organic matter, including a short review on soil organic matter chemistry. Res. Rev. 32, 131 –174. Heckrath, G., Brookes, P. C., Poulton, P. R., and Goulding, K. W. T. (1995). Phosphorus leaching from soils containing different phosphorous concentrations in the Broadbalk experiment. J. Environ. Qual. 24, 904 –910. Heisel, T., Ersbøll, A. K., and Andreasen, C. (1999). Weed mapping with co-kriging using soil properties. Precision Agric. 1, 39–52. Helling, B., and Larink, O. (1998). Contribution of earthworms to nitrogen turnover in agricultural soils treated with different mineral N-fertilizers. Appl. Soil Ecol. 9, 319 –325. Herrick, J. E., Brown, J. R., Tugel, A. J., Shaver, P. L., and Havstad, K. M. (2002). Application of soil quality to monitoring and management: paradigms from rangeland ecology. Agron. J. 94, 3–11. Hildesheim, M. E., Cantor, K. P., Lynch, C. F., Dosemeci, M., Lubin, J., Alavanja, M., and Craun, G. F. (1998). Drinking water source and chlorination byproducts in Iowa. II. Risk of colon and rectal cancers. Epidemiology 9, 29–35. Hilgard, E. W. (1914). In “Soils, their formation, properties, composition, and relations to climate and plant growth in the humid and arid regions”, 593 pages. MacMillan, New York. Hoffman, J. A., and Purdy, L. H. (1964). Germination of dwarf bunt teliospores after ingestion by earthworms. Phytopathology 54, 878–879. Ho¨gberg, P., Nordgren, A., Buchmann, N., Taylor, A. F. S., Ekblad, A., Ho¨gberg, M. N., Nyberg, G., Ottosson-Lo¨fvenius, M., and Read, D. J. (2001). Large-scale forest girdling shows that current photosynthesis drives soil respiration. Nature 411, 789 –792. Hortensius, D., and Welling, R. (1996). International standardization of soil quality measurements. Commun. Soil Sci. Plant Anal. 27, 387 –402. Howard, P. J. A. (1993). Soil protection and soil quality assessment in the EC. Sci. Total Enviorn. 129, 219 –239. Howard, P. J. A., and Howard, D. M. (1979). Respiration of decomposing litter in relation to temperature and moisture. Oikos 33, 457–465. Howarth, R. W., Billen, G., Swaney, D., Townsend, A., Jaworski, N., Lajtha, K., Downing, J. A., Elmgren, R., Caraco, N., Jorden, T., Berendse, F., Freney, J. R., Kudeyarov, V., Murdoch, P. S., and Zhao-Liang, Z. (1996). Regional nitrogen budgets and riverine N and P fluxes for the drainage to the North Atlantic Ocean: natural and human influences. Biogeochemistry 35, 75 –139. Hutchison, S. A., and Kamel, M. (1956). The effect of earthworms on the dispersal of soil fungi. J. Soil Sci. 7, 213 –218. Isensee, A. R., Nash, R. G., and Helling, C. S. (1990). Effect of conventional and no-tillage on pesticide leaching to shallow groundwater. J. Environ. Qual. 19, 434 –440. Jamieson, D. (1995). Ecosystem health: some preventative medicine. Environ. Values 4, 333 –344. Joergensen, R. G. (2000). Ergosterol and microbial biomass in the rhizosphere of grassland soils. Soil Biol. Biochem. 31, 647 –652.

60

R. E. SOJKA, D. R. UPCHURCH AND N. E. BORLAUG

Juma, N. G. (1993). Interrelationship between soil structure/texture, soil biota/soil organic matter and crop production. Geoderma 57, 3–30. Kapusta, L. A., and Landis, W. G. (1998). Ecology: the science versus the myth. Human Ecol. Risk Assess. 4, 829 –838. Karlen, D. L., Mausbach, M. J., Doran, J. W., Cline, R. G., Harris, R. F., and Schuman, G. E. (1997). Soil quality: a concept, definition, and framework for evaluation. Soil Sci. Soc. Am. J. 61, 4 –10. Karlen, D. L., Andrews, S. S., and Doran, J. W. (2001). Soil quality: current concepts and applications. Advances in Agronomy. (D. L. Sparks, Ed.), Vol. 74, pp. 1–40. Academic Press, San Diego, CA. Keeney, R. L. (1992). In “Value-focused Thinking”. Harvard University Press, Cambridge, MA. Keeney, D. R., and De Luca, T. H. (1993). Des Moines River nitrate in relation to watershed agricultural practices: 1945 versus 1980s. J. Environ. Qual. 22, 267 –660. Keeney, R. L., and Raiffa, H. (1976). In “Decisions with multiple objectives: preferences and value tradeoffs”. Wiley, New York, NY. Kellman, L., and Hillaire-Marcel, C. (1998). Nitrate cycling in streams: using natural abundances of NO3-15N to measure in-situ denitrification. Biogeochemistry 43, 273 –292. Kemper, W. D., Trout, T. J., Segeren, A., and Bullock, M. (1987). Worms and water. J. Soil Water Conserv. 42(6), 401 –404. Kendall, H. W., and Pimentel, D. (1994). Constraints on the expansion of the global food supply. Ambio 23, 198–205. Kennedy, A. C., and Papendick, R. I. (1995). Microbial characteristics of soil quality. J. Soil Water Conserv. 50, 243–248. Kennedy, A. C., and Smith, K. L. (1995). Soil microbial diversity and the sustainability of agricultural soils. In “The Significance and Regulation of Soil Biodiversity”. (H. P. Collins, Ed.), pp. 75 –86. Kluwer Academic Publishers, Amsterdam. Khambata, S. R., and Bhat, J. V. (1957). A contribution to the study of the intestinal microflora of Indian earthworms. Arch. Mikrobiol. 28, 69 –80. Kirschbaum, M. U. F. (1995). The temperature dependence of soil organic matter decomposition, and the effect of global warming on soil organic C storage. Soil Biol. Biochem. 27, 753–760. Kladivko, E. J., Van Scoyoc, G. E., Monke, E. J., Oates, K. M., and Pask, W. (1991). Pesticide and nutrient movement into subsurface tile drains on a silt loam soil in Indiana. J. Environ. Qual. 20, 264–270. Kladivko, E. J., Grochulska, J., Turco, R. F., Van Scoyoc, G. E., and Eigel, J. D. (1999). Pesticide and nitrate transport into subsurface tile drains of different spacings. J. Environ. Qual. 28, 997–1004. Koepf, H. H (1991). Das Konzept der Bodenfruchtbarkeit im o¨kologischen Landbau. Berichte u¨ber Landwirtschaft. Neue Folge. 203. Sonderheft, 46–58. Lackey, R. T. (1998). Ecosystem management: paradigms and prattle, people and prizes. Renew. Resour. J. 16, 8–13. Lackey, R. T. (1998). A patch of common ground. J. Forestry 96(4), 32–33. Lackey, R. T. (2001). Values, policy, and ecosystem health. Bioscience 51, 437– 443. Laird, D. A., Barriuso, E., Dowdy, R. H., and Koskinen, W. C. (1992). Adsorption of atrazine on smectites. Soil Sci. Soc. Am. J. 56, 62–67. Laird, D. A., Yen, P. Y., Koskinen, W. C., Steinheimer, T. R., and Dowdy, R. H. (1994). Sorption of atrazine on soil clay components. Environ. Sci. Tech. 28, 1054–1061. Lal, R., and Pierce, F. J. (1991). The vanishing resource. In “Soil Management for Sustainability”. (R. Lal and F. J. Pierce, Eds.), pp. 1–5. Soil and Water Conservation Society, Ankeny, IA. Lambert, D. H., and Weidensaul, T. C. (1991). Elemental uptake by mycorrhizal soybean from sewage-sludge-treated soil. Soil Sci. Soc. Am. J. 55, 393–398. Lancaster, J. (2000). The ridiculous notion of assessing ecological health and identifying the useful concepts underneath. Hum. Ecol. Risk Assess. 6, 213–222.

QUALITY SOIL MANAGEMENT

61

Larson, W. E., and Pierce, F. J. (1991). Conservation and enhancement of soil quality. Proceedings of the International Workshop, Chiang Rai, Thailand, 15–21 Sept. 1991. In “Evaluation for Sustainable Land Management in the Developing World”. Vol. 2, pp. 175 –203. International Board for Soil Research and Management, Bangkok, Thailand. Larson, W. E., and Pierce, F. J. (1994). The dynamics of soil quality as a measure of sustainable management. SSSA Special Publication 35. In “Defining Soil Quality for a Sustainable Environment”. (J. W. Doran, D. C. Coleman, D. F. Bedzicek and B. A. Stewart, Eds.), pp. 37–51. SSSA/ASA, Madison, WI. Le Bayon, R. C., and Binet, F. (1999). Rainfall effects on erosion of earthworm casts and phosphorus transfers by water runoff. Biol. Fert. Soils 30, 7–13. Lele, S., and Nogaard, R. B. (1996). Sustainability and the scientist’s burden. Conserv. Biol. 10, 354 –365. Leopold, A. (1941). Wilderness as a land laboratory. Living Wilderness 6(July), 3. Letey, J. (1985). Relationship between soil physical properties and crop production. Adv. Soil Sci. 1, 278 –294. Letey, J., Sojka, R. E., Upchurch, D. R., Cassel, D. K., Olson, Ken, Payne, Bill, Petrie, Steve, Price, Graham, Reginato, R. J., Scott, H. D., Smerthurst, Philip, and Triplett, Glover. (2003). Deficiencies is the soil quality concept and its application. Journal of Soil and Water Conservation. (in press). von Liebig, J. (1862). In “Chemistry in its Application to Agriculture and Physiology” (7th ed.). Viewig, Brunswick. Liebig, M. A., Doran, J. W., and Gardiner, J. C. (1996). Evaluation of a field test kit for measuring selected soil quality indicators. Agron. J. 88, 683–686. Liebig, M. A., Varvel, G., and Doran, J. (2001). A simple performance-based index for assessing multiple agroecosystem functions. Agron. J. 93, 313 –318. Lindqvist, R., and Enfield, C. G. (1992). Biosorption of dichlorodephenyltrichloroethane and hexachlorobenzene in groundwater and its implication for facilitated transport. Appl. Environ. Microbiol. 58, 2211–2218. Linser, H. (1965). Fassung und Bedeutung des Begriffes Bodenfruchtbarkeit. Zeitschrift fu¨r Pflanzenernahrung, Du¨ngung Bodenkunde 108, 115–122. Lloyd, J., and Taylor, J. A. (1994). On the temperature dependence of soil respiration. Funct. Ecol. 8, 315 –323. Ludwig, B., Heil, B., Flessa, H., and Beese, F. (2000). Dissolved organic carbon in seepage water— production and transformation during soil passage. Acta Hydrochim. Hydrobiol. 28, 77 –82. Mackay, A. L. (1991). In “A Dictionary of Scientific Quotations”, 297 pages. Institute of Physics Publishing, Bristol, UK. Mackey, B. G. (1999). Environmental scientists, advocacy, and the future of earth. Environ. Conserv. 26, 245–249. Magesan, G. N., White, R. E., and Scotter, D. R. (1995). The influence of flow rate on the concentration of indigenous and applied solutes in mole-pipe drain effluent. J. Hydrol. 172, 23 –30. Mallawatantri, A. P., McConkey, B. G., and Mulla, D. J. (1996). Characterization of pesticide sorption and degradation in macropore linings and soil horizons of Thatuna silt loam. J. Environ. Qual. 25, 227 –235. Mann, C. C. (2000). Earthmovers of the Amazon. Science 287, 786 –789. Marialigeti, K. (1979). On the community-structure of the gut-microbiota of Eisenia lucens (Annelida, Oligochaeta). Pedobiologia 19, 213 –220. Mausbach, M., and Tugel, A. (1995). A decision document for establishing a Soil Quality Institute. White Paper. Natural Resources Conservation Service 12 pp. Mayhew, M., and Alessi, S. (1998). Unraveling the stakeholder dialogue of pest management. In “Bugs in the System: Redesigning the Pesticide Industry for Sustainable Agriculture”. (W. Vorley and D. Keeney, Eds.), pp. 136 –168. Earthscan Publishing, London.

62

R. E. SOJKA, D. R. UPCHURCH AND N. E. BORLAUG

McMahon, P. B., and Bo¨hlke, J. K. (1996). Denitrification and mixing in a stream-aquifer system: effects on nitrate loading to surface water. J. Hydrol. 186, 105–128. McMullin, E. (1987). Scientific controversy and its termination. In “Scientific Controversies”. (T. H. Keeney A. L. Engelhardt, Ed.), pp. 49– 91. Cambridge University Press, Cambridge, UK. Medlin, C. R., Shaw, D. R., Cox, M. S., Gerard, P. D., Abshire, M. J., and Wardlaw, M. C., III (2001). Using soil parameters to predict weed infestations in soybean. Weed Sci. 49, 367 –374. Meek, B. D., MacKenzie, A. J., Donovan, T. J., and Spencer, W. F. (1974). The effect of large application of manure on movement of nitrate and carbon in an irrigated desert soil. J. Environ. Qual. 3, 253–258. Meek, B. D., Graham, L. E., Donovan, T. J., and Mayberry, K. S. (1979). Phosphorus availability in a calcareous soil after high loading rates of animal manure. Soil Sci. Soc. Am. J. 43, 741–744. Meek, B., Graham, L., and Donovan, T. (1982). Long-term effects of manure on soil nitrogen, phosphorus, potassium, sodium, organic matter, and water infiltration. Soil Sci. Soc. Am. J. 46, 1014–1019. Mendelsohn, E. (1987). The political anatomy of controversy in the sciences. In “Scientific Controversies”. (T. H. Engelhardt, Jr., and Caplan, A. L. Eds.), pp. 93 –124. Cambridge University Press, Cambridge, UK. Miller, F. P. (1999). Soil quality: science or political correctness? Ohio Pedol. 7. Milnear, R. A., and Amy, G. L. (Eds.), (1996). ACS Symposium Series 649. In “Water Disinfection and Natural Organic Matter Characterization and Control”, 649 pages. American Chemical Society, Washington, DC. Mitchell, J. K., Walker, S. E., Hirshi, M. C., and McIsaac, G. F. (1997). Nitrate losses under various nitrogen management systems. In “Management of Landscapes Disturbed by Channel Incision”. (S. S. Y. Wang et al., Ed.), pp. 197–202. The Center for Computational Hydroscience and Engineering, The University of Mississippi, Oxford. Morris, R. D., Audet, A. M., Angelillo, I. F., Chalmers, T. C., and Musteller, F. (1992). Chlorination, chlorination by-products, and cancer: a meta-analysis. Am. J. Public Health 82, 955 –963. Mudhun, Y. A., Young, J. L., and Freed, V. H. (1986). Binding of herbicides by water-soluble organic materials from soil. J. Environ. Qual. 15, 64– 68. Mulla, D. J., Davis, D. M., and Randall, G. W. (1998). Hypoxia in the Gulf of Mexico: effects of tile drainage and nitrogen fertilizer management. Agron. Abstr. 185. Muszkat, L., Raucher, D., Margaritz, M., Ronen, D., and Amiel, A. J. (1993). Unsaturated zone and ground-water contamination by organic pollutants in a sewage-effluent-irrigated site. Groundwater 31, 556 –565. Narasimhan, M. G. (2001). Controversy in science. J. Biosci. 26, 299–304. National Research Council, (1993). In “Soil and Water Quality: An Agenda for Agriculture”. National Academy Press, Washington, DC. Nelson, D. W. (1994). Foreword. SSSA Special Publication 35. In “Defining Soil Quality for a Sustainable Environment”. (J. W. Doran, D. C. Coleman, D. F. Bedzicek and B. A. Stewart, Eds.), pp. 7. Nelson, S. D., Letey, J., Farmer, W. J., Williams, C. F., and Ben-Hur, M. (1998). Facilitated transport of napropamide by dissolved organic matter in sewage sludge amended soil. J. Environ. Qual. 27, 1194–1200. Nielsen, N. O. (1999). The meaning of health. Ecosyst. Health 5, 65 –66. Nielsen, T. H., and Revsbech, N. P. (1998). Nitrification, denitrification, and N-liberation associated with two types of organic hot-spots in soil. Soil Biol. Biochem. 30, 611 –619. van Noorwijk, M., De Ruiter, P. C., Swart, K. B., Bloem, J., Moore, J. C., van Faassen, H. G., and Burgers, S. L. G. E. (1993). Synlocation of biological activity, roots, cracks and recent organic inputs in a sugar beet field. Geoderma 56, 265–276.

QUALITY SOIL MANAGEMENT

63

Nortcliff, S. (1997). Standardization for soil quality attributes. Soil Quality for Crop Production and Ecosystem Health, 25. (E. G. Gregorich and M. R. Carter, Eds.), pp. 187–201. Elsevier, Amsterdam. Norcliff, S. (2002). Standardization of soil quality attributes. Agriculture, Ecosystems and Environment. 88, 161–168. Nowotny, H. (1975). Controversies in science: remarks on the different modes of production of knowledge and their use. Zeitschr. fu¨r Sociol. 4, 37. Oliver, M. A. (1997). Soil and human health: a review. Eur. J. Soil Sci. 48, 573–592. Paarlberg, R. L. (1994). The politics of agricultural resource abuse. Environment 36, 7 –41. Papendick, R. I., and Parr, J. F. (1992). Soil quality—the key to sustainable agriculture. Am. J. Altern. Agric. 7, 2 –3. Parkin, T. B. (1987). Soil microsites as a source of denitrification variability. Soil Sci. Soc. Am. J. 51, 1194–1199. Parkin, T. B. (1993). Spacial variability of microbial processes in soil. J. Environ. Qual. 22, 409–417. Parkin, T. B., and Berry, E. C. (1994). Nitrogen transformations associated with earthworm casts. Soil Biol. Biochem. 26, 1233–1238. Pashanasi, B., Lavelle, P., and Alegre, J. (1996). Effect of inoculation with the endogeic earthworm Pontoscolex corethrurus on soil chemical characteristics and plant growth in a low-input agricultural system of Peruvian Amazonia. Soil Biol. Biochem. 28(6), 801 –810. Patzel, N., Sticher, H., and Karlen, D. L. (2000). Soil fertility—phenomenon and concept. J. Plant Nutr. Soil Sci. 163, 129–142. Peterjohn, W. T., Melillo, J. M., and Bowles, S. T. (1993). Soil warming and trace gas fluxes: experimental design and preliminary flux results. Oecologia 93, 18– 24. ˚ ., and Olesen, T. (1996). O2 uptake, C metabolism and Petersen, S. O., Nielsen, T. H., Frostega˚rd, A denitrification associated with manure hot-spots. Soil Biol. Biochem. 28, 341 –349. Pierce, F. J., and Larson, W. E. (1993). Developing criteria to evaluate sustainable land management. In “Proceedings of the Eighth International Soil Management Workshop: Utilization of Soil Survey Information for Sustainable Land Use, May 1993”. (J. M. Kimble, Ed.), pp. 7–14. USDASCS, National Soil Survey, Lincoln, NE. Pimentel, D. (2000). Soil erosion and the threat to food security and the environment. Ecosyst. Health 6, 221–226. Pinstrup-Anderson, and Pandya-Lorch, R. (2000). In “Unfinished Agenda: Perspectives on Overcoming, Hunger, Poverty, and Environmental Degradation”. IFPRI, Washington, DC. Pivetz, B. E., and Steenhuis, T. S. (1995). Soil matrix and macropore biodegradation of 2,4-D. J. Environ. Qual. 24, 564 –570. Potter, C. (1998). In ““Against the Grain.” Ari-Environmental Reform in the United States and the European Union”, 208 pages. CAB International, Wallingford, UK. Pouyat, R. V. (1999). Science and environmental policy—making them compatible. BioScience 49, 281 –286. Price, G. (2000). “Soil quality”—a warning. Profile 122, 22. Quade, J. W. (1985). Perspectives on Nonpoint Source Pollution. Proceedings of the National Conference, 19–22 May 1985, Kansas, MO. In “Agricultural land improvement and water quality in South Central Minnesota”, pp. 508–514. USEPA, Washington, DC. Quandt, L., and Waltman, S. W. (1997). Dominant Soil Orders for the U.S. . USDA-NRCS, National Soil Survey Center, Lincoln, NE. Rabalais, N. N., Wiseman, W. J., Jr., and Turner, R. E. (1994). Comparison of continuous records of near-bottom dissolved oxygen from the hypoxia zone along the Louisiana coast. Estuaries 14, 850 –861. Raich, J. W., and Potter, C. S. (1995). Global patterns of carbon dioxide emissions from soils. Global Biogeochem. Cycles 9, 23 –36. Raich, J. W., and Schlesinger, W. H. (1992). The global carbon dioxide flux in soil respiration and its relationship to vegetation and climate. Tellus 44, 81–89.

64

R. E. SOJKA, D. R. UPCHURCH AND N. E. BORLAUG

Randall, G. W., Huggins, D. R., Russelle, M. P., Fuchs, D. J., Nelson, W. W., and Anderson, J. L. (1997). Nitrate losses through subsurface tile drainage in Conservation Reserve Program, alfalfa, and row crop systems. J. Environ. Qual. 26, 1240–1247. Rapport, D. J. (2000). Extravagant claims: the polarization of environmental issues is counterproductive. Ecosyst. Health 6, 219 –220. Rapport, D. J., Bo¨hm, G., Buckingham, D., Cairns, J., Jr., Costanza, R., Karr, J. R., deKruijf, H. A. M., Levins, R., McMichael, A. J., Nielsen, N. O., and Whitford, W. G. (2000). Reply to Calow: critics of ecosystem health misrepresented. Ecosyst. Health 6, 5 –6. Reicosky, D. C., and Lindstrom, M. J. (1993). Fall tillage method: effect on short-term carbon dioxide flux from soil. Agron. J. 85, 1237–1243. Rew, L. J., and Cousens, R. D. (2001). Spatial distribution of weeds in arable crops: are current sampling and analytical methods appropriate? Weed Res. 41, 1– 18. Rhoades, J. D. (1972). Quality of water for irrigation. Soil Sci. 113, 277 –284. Rhoades, J. D. (1998). Use of saline and brackish waters for irrigation: implications and role in increasing food production, conserving water, sustaining irrigation and controlling soil and water degradation. Proceedings of the International Workshop on “The Use of Saline and Brackish Water for Irrigation”, July 23 –24, 1998, Bali, Indonesia. Ragab and Pearce, G. Eds.), pp. 261– 304. International Commission on Irrigation and Drainage, Salt Lake City, UT. Robinson, J. S., and Sharpley, A. N. (1995). Release of nitrogen and phosphorus from poultry litter. J. Environ. Qual. 24, 62–67. Rochette, P., Des Jardins, R. L., and Patty, E. (1991). Spatial and temporal variability of soil respiration in agricultural fields. Can. J. Soil Sci. 10, 189 –196. Rodale Institute, (1991). In “Conference Report and Abstracts. International Conference on the Assessment and Monitoring of Soil Quality”. Rodale Press, Emmaus, PA. Romig, D. F., Garlynd, J., Harris, R. F., and McSweeney, K. (1995). How farmers assess soil health and quality. J. Soil Water Conserv. 50, 229–236. Rose, C. J., and Wood, A. W. (1980). Some environmental factors affecting earthworm population and sweet potato production in the Tari Basin, Papua New Guinea highlands. Papp. New Guin. Agric. J. 31, 1–13. Ross, M. A., and Lembi, C. A. (1985), Applied Weed Science. Macmillan, New York. Russow, L. (1995). Ecosystem health: an objective evaluation? Environ. Values 4, 363–369. Rustad, P., and Fernandez, I. J. (1998). Experimental soil warming effects on CO2 and CH4 flux from a low elevation spruce–fir forest soil in Maine, U.S.A. Global Change Biol. 4, 597 –605. Rustad, L. E., Huntington, T. G., and Boone, R. D. (2000). Controls on soil respiration: implications for climate change. Biogeochemistry 48, 1–6. Ryder, R. A. (1990). Ecosystem health, a human perception: definition, detection, and the dichotomous key. J. Great Lakes Res. 16, 619– 624. Rykiel, E. J., Jr. (2001). Scientific objectivity, value systems, and policymaking. BioScience 51, 433–436. Rykiel, E. J., Jr. (2001). What is, what might be, and what ought to be. BioScience 51, 423. Sanchez, P. A., Palm, C. A., and Buol, S. W. (2003). Fertility capability soil classification–Version 4: A tool to help assess soil quality in the tropics. Geoderma. (in press). Sagoff, M. (1995). The value of integrity. In “Perspectives on Ecological Integrity”. (L. Westra and J. Lemons, Eds.), pp. 162–176. Kluwer Academic Publishers, Dordrecht, The Netherlands. Schaeffer, D. J., Herricks, E. E., and Kerster, H. W. (1988). Ecosystem health. I. Measuring ecosystem health. Environ. Manag. 12, 445 –455. Schlenter, R. E., and Van Cleve, K. (1985). Relationship between CO2 evolution from soil, substrate temperature, and substrate moisture in four mature forest types in interior Alaska. Can. J. Forest Res. 15, 97 –106. Schleser, G. H. (1982). The response of CO2 evolution from soils to global temperature changes. Z. Naturforsch. 37, 287–291.

QUALITY SOIL MANAGEMENT

65

Scho¨nberger, H., and Wiese, J (1991). Forschungsbedarf im Zusammenhang mit den Zielvorstellungen, der Meß und Voraussagbarkeit von Elementen und Prozessen der Bodenfruchtbarkeit. Berichte u¨ber Landwirtschaft. Neue Folge. 203, Sonderheft, ppp. 144 –157. Schwab, G. O., Fausey, N. R., Desmond, E. D., and Holman, J. R. (1985). Research Bulletin 1166. In “Tile and Surface Drainage of Clay Soils”. Ohio Agricultural Research and Development Centre, Wooster, OH. Senesi, N. (1992). Binding mechanisms of pesticides to soil humic substances. Sci. Total Environ. 123/124, 63–76. Sextone, A. J., Revsbech, N. P., Parkin, T. B., and Tiedje, J. M. (1985). Direct measurement of oxygen profiles and denitrification rates in soil aggregates. Soil Sci. Soc. Am. J. 49, 645 –651. Sharpley, A. N., and Smith, S. J. (1995). Nitrogen and phosphorus forms in soils receiving manure. Soil Sci. 159, 253–258. Sharpley, A. N., and Syers, J. K. (1976). Potential role of earthworm casts for the phosphorus enrichment of run-off waters. Soil Biol. Biochem. 8, 341 –346. Sharpley, A. N., and Syers, J. K. (1977). Seasonal variation in casting activity and in the amounts and release to solution of phosphorus forms in earthworms casts. Soil Biol. Biochem. 9, 227 –231. Shipitalo, M. J., and Butt, K. R. (1999). Occupancy and geometrical properties of Lumbricus terrestris L. burrows affecting infiltration. Pedobiologia 43, 782– 794. Shipitalo, M. J., and Gibbs, F. (2000). Potential of earthworm burrows to transmit injected animal wastes to tile drains. Soil Sci. Soc. Am. J. 64, 2103–2109. Shipitalo, M. J., Edwards, W. M., and Redmond, C. E. (1994). Comparison of water movement and quality in earthworm burrows and pan lysimeters. J. Environ. Qual. 23, 1345– 1351. Shipitalo, M. J., Dick, W. A., and Edwards, W. M. (2000). Conservation tillage and macropore factors that affect water movement and the fate of chemicals. Soil Tillage Res. 53, 167–183. Shrader, S., and Zhang, H. (1997). Earthworm casting: stabilization or destabilization of soil structure? Soil Biol. Biochem. 29, 469–475. Shuford, J. W., Fritton, D. D., and Baker, D. E. (1977). Nitrate-nitrogen and chloride movement through undisturbed field soil. J. Environ. Qual. 6, 255– 259. Silva, R. G., Cameron, K. C., Di, H. J., Smith, N. P., and Buchan, G. D. (2000). Effect of macropore flow on the transport of surface-applied cow urine through a soil profile. Aust. J. Soil Res. 38, 13 –23. Simpson, T. W., and Cunningham, R. L. (1982). The occurrence of flow channels in soils. J. Environ. Qual. 11, 29 –30. Simpson, K. L., and Hayes, K. P. (1998). Drinking water disinfection by-products: an Australian perspective. Water Res. 32, 1522–1528. Sims, J. T., Cunningham, S. D., and Sumner, M. E. (1997). Assessing soil quality for environmental purposes: roles and challenges for soil scientists. J. Environ. Qual. 26, 20– 25. Sinclair, H. R., Waltman, W. J., Waltman, S. W., Terpstra, H. P., and Reed-Margetan, D. (1996). Soil Ratings for Plant Growth (SRPG). USDA-NRCS, National Soil Survey Center, Lincoln, NE. Singer, M. J., and Ewing, S. A. (2000). Soil quality. In “Handbook of Soil Science”. (M. Sumner, Ed.), pp. G271–G298. CRC Press, Boca Raton, FL. Singer, M. J., and Sojka, R. E. (2001). Yearbook of Science 2002. In “Soil quality”, pp. 312–314. McGrawHill, New York, NY. Singh, J. S., and Gupta, S. R. (1977). Plant decomposition and soil respiration in terrestrial ecosystems. Bot. Rev. 43, 449 –528. Smil, V. (1999). Detonator of the population explosion. Nature 400, 415. Smil, V. (1999b). Long-range perspectives on inorganic fertilizers in global agriculture. Travis P. Hignett Memorial Lecture, IFDC, Muscle Shoals, AL. Smith, R. A., Schwartz, G. E., and Alexander, R. B. (1997). Regional interpretation of water-quality monitoring data. Water Resour. Res. 33, 2781–2798. Soil Science Society of America, (1998). Internet Glossary of Soil Science Terms. http://www.soils. org/sssagloss/.

66

R. E. SOJKA, D. R. UPCHURCH AND N. E. BORLAUG

Soil Survey Staff, (1975). Agricultural Handbook 436. In “Soil Taxonomy”. U.S. Government Printing Office, Washington, DC. Soil Survey Staff, (1997). In “The Soil Quality Concept”, http://www.statlab.iastate.edu/survey/SQI/. Soil Quality Institute, Natural Resources Conservation Service, Ames, IA. Soil Survey Staff, (1998). In “Soil Quality Card Design Guide”, http://www.statlab.iastate.edu/survey/ SQI/. Soil Quality Institute, Natural Resources Conservation Service, Ames, IA. Sojka, R. E. (1996). Irrigation. In “1997 Yearbook of Science and Technology”. (S. Parker, Ed.), pp. 258 –260. McGraw Hill, New York. Sojka, R. E. (1998). Understanding and managing irrigation-induced erosion. In “Advances in Soil and Water Conservation”. (F. J. Pierce and W. W. Frye, Eds.), pp. 21–37. Sleeping Bear Press/ Ann Arbor Press, Chelsea, MI. Sojka, R. E., and Upchurch, D. R. (1999). Reservations regarding the soil quality concept. Soil Sci. Soc. Am. J. 63, 1039–1054. Sommer, T. J. (2001). Suppression of scientific research: Bahramdipity and Nulltiple scientific discoveries. Sci. Engng Ethics 7, 77–104. Sparrow, L. A., Sharpley, A. N., and Reuter, D. J. (2000). Safeguarding soil and water quality. Commun. Soil Sci. Plant Anal. 31, 1717–1742. Sposito, G., Martin-Neto, L., and Yang, A. (1996). Atrazine complexation by soil humic acids. J. Environ. Qual. 25, 1203–1209. Stagnitti, F., Parlange, J. Y., Steenhuis, T. S., Boll, J., Pivetz, B., and Barry, D. A. (1995). Transport of moisture and solutes in the unsaturated zone by preferential flow. In “Environmental Hydrology”. (V. P. Singh, Ed.), pp. 193– 224. Kluwer Academic Publishers, Dordrecht. Stamm, C., Flu¨hler, H., Ga¨chter, R., Leuenberger, J., and Wunderli, H. (1998). Preferential transport of phosphorus in drained grassland soils. J. Environ. Qual. 27, 515–522. Starett, S. K., Christians, N. E., and Austin, T. A. (1996). Comparing dispersivities and soil chloride concentrations of turfgrass-covered undisturbed and disturbed soil columns. J. Hydrol. 180, 21029. Steenhuis, T. S., Staubitz, W., Andreini, M. S., Surface, J., Richard, T. L., Paulsen, R., Pickering, N. B., Hagerman, J. R., and Geohring, L. D. (1990). Preferential movement of pesticides and tracers in agricultural soils. J. Irrig. Drain. Engng 116, 50–66. Stenberg, B. (1998). Soil attributes as predictors of crop production under standardized conditions. Biol. Fert. Soils 27, 104 –112. Stenberg, B. (1999). Monitoring soil quality of arable land: microbiological indicators. Acta Agric. Scand. Sect. B. Soil Plant Sci. 49, 1 –24. Stevenson, F. J. (1972). Organic matter reactions involving herbicides in soils. J. Environ. Qual. 1, 333–343. Temminghoff, E. J. M., Van Der Zee, S. E. A. T. M., and De Haan, F. A. M. (1998). Effects of dissolved organic matter on the mobility of copper in a contaminated sandy soil. Eur. J. Soil Sci. 49, 617 –628. Tewary, C. K., Pandey, U., and Singh, J. S. (1982). Soil and litter respiration rates in different microhabitats of a mixed oak–conifer forest and their control by edaphic conditions and substrate quality. Plant Soil 65, 233 –238. Thompson, P. (1999). The ethics of truth telling and the problem of risk. Sci. Engng Ethics 5, 489–510. Thornton, M. L. (1970). Transport of soil-dwelling aquatic phycomycetes by earthworms. Trans. Br. Mycol. Soc. 55, 391–397. Tiedje, J. M., Sexstone, A. J., Parkin, T. B., and Revsbech, N. P. (1984). Anaerobic processes in soil. Plant Soil 76, 197–212. Toyota, K., and Kimura, M. (1994). Earthworms disseminate a soil-borne plant pathogen, Fusarium oxysporum f. spp. Raphani. Biol. Fert. Soils 18, 32–36. Trojan, M. D., and Linden, D. R. (1992). Microrelief and rainfall effects on water and solute movement in earthworm burrows. Soil Sci. Soc. Am. J. 56, 727 –733.

QUALITY SOIL MANAGEMENT

67

Trout, T. J., and Johnson, G. S. (1989). Earthworms and furrow irrigation infiltration. Trans. ASAE 32(5), 1594–1598. Trout, T. J., Kemper, W. D., and Johnson, G. S. (1987). Earthworms cause furrow infiltration increase. In “Proceedings of the International Conference on Infiltration Development and Application, January 6–9, Honolulu, Hawaii”. (Y.-S. Fok, Ed.), pp. 398 –406. Water Resources Research Centre, University of Hawaii, Honolulu. Turco, R. F., Kennedy, A. C., and Jawson, M. D. (1994). Microbial indicators of soil quality. SSSA Special Publication 35. In “Defining soil Quality for a Sustainable Environment”. (J. W. Doran, D. C. Coleman, D. F. Bedzicek and B. A. Stewart, Eds.), pp. 73 –90. SSSA, Madison, WI. Tyler, D. D., and Thomas, G. W. (1977). Lysimeter measurements of nitrate and chloride losses from soil under conventional and no-tillage corn. J. Environ. Qual. 6, 63– 66. Urba´nek, J., and Dolez˙al, F. (1992). Review of some case studies on the abundance and on the hydraulic efficiency of earthworm channels in Czechoslovak soils, with reference to the subsurface pipe drainage. Soil Biol. Biochem. 24, 1563–1571. USDA, (1980). In “Report and Recommendations on Organic Farming”. U.S. Government Printing Office, Washington, DC. USDA, (1987). Miscellaneous Publication No. 1455. In “Farm drainage in the United States: history, status, and prospects”. U.S. Government Printing Office, Washington, DC. USEPA (1993). US Environmental Protection Agency, Office of Groundwater and Drinking Water, USEPA Draft D/DPB Rule Language, September 2, 1993, Washington, DC. Vaudaux, B. (1998). Environmental health hazards: how about microbiological pollution? Environ. Int. 24, 249 –250. Vervoort, R. W., Radcliffe, D. E., and Wester, L. T. (1999). Soil structure development and preferential solute flow. Water Resour. Res. 35, 913– 928. Vinten, A. J.A., Yaron, B., and Nye, P. H. (1983). Vertical transport of pesticides into soil when adsorbed on suspended particles. J. Agric. Food Chem. 31, 662–664. Vinther, F. P., Eiland, F., Laind, A. M., and Elsgaard, L. (1999). Microbial biomass and numbers of denitrifiers related to macropore channels in agricultural and forest soils. Soil Biol. Biochem. 31, 603 –611. Visser, S., and Parkinson, D. (1992). Soil biological criteria as indicators of soil quality: soil microorganisms. Am. J. Altern. Agric. 7, 33– 37. Waggoner, P. E. (1994). Council on Agricultural Science and Technology Report 121. In “How much land can ten billion people spare for nature?”, 64 pp. Waldon, H., Gliessman, S., and Buchanan, M. (1998). Agroecosystem response to organic and conventional management practices. Agricu. Syst. 57(1), 65– 75. Wallace, H. A. (1938). Soils and Men, 1938 Yearbook of Agriculture. pp.v. House Document 398. In “Forward”. U.S. Government Printing Office, Washington, DC. Waller, K., Swan, S. H., DeLorenze, G., and Hopkins, B. (1998). Trihalomethanes in drinking water and spontaneous abortion. Epidemiology 9, 134–140. Wander, M. M., Walter, G. L., Nissen, T. M., Bollero, G. A., Andrews, S. S., and Cavanaugh-Grant, D. A. (2002). Soil quality: science and process. Agron. J. 94, 23–24. Wang, D., Norman, J. M., Lowery, B., and McSweeney, K. (1994). Nondestructive determination of hydrogeometrical characteristics of soil macropores. Soil Sci. Soc. Am. J. 58, 294– 303. Warkentin, B. P. (1995). The changing concept of soil quality. J. Soil Water Conserv. 50, 226 –228. Warkentin, B. P., and Fletcher, H. F. (1977). Proceedings of the International Seminar on Soil Environment and Fertility Management in Intensive Agriculture. In “Soil quality for intensive agriculture”. Society of Science of Soil and Manure, National Institute of Agriculture Science, Tokyo. Warman, P. R., and Havard, K. A. (1996). Yield, vitamin and mineral contents of organically and conventionally grown carrots and cabbage. Agric. Ecosyst. Environ. 61, 155 –162.

68

R. E. SOJKA, D. R. UPCHURCH AND N. E. BORLAUG

Watts, D. G., Fekersillassie, D., Schepers, J. S., Eisenhauer, D. E., Spalding, R. F., and Klocke, N. L. (2000). Integrated management of irrigation and nitrogen for groundwater quality improvement. In “National Irrigation National Irrigation Symposium. Proceedings of the Fourth Decennial Symposium, November 14– 16, 2000, Phoenix, AZ”. (R. G. Evans, B. L. Benham and T. P. Tooien, Eds.), pp. 144–152. American Society of Agricultural Engineers (ASAE), St Joseph, MI. Weber, M. G. (1990). Forest soil respiration after cutting and burning in immature aspen ecosystems. Forest Ecol. Manage. 31, 1–14. Weyer, P. J., Cerhan, J. R., Kross, B. C., Hallberg, G. R., Kantamneni, J., Breuer, G., Jones, M. P., Zheng, W., and Lynch, C. F. (2001). Municipal drinking water nitrate level and cancer risk in older women: the Iowa women’s health study. Epidemiology 11, 327–338. Wharburton, C. W., Manifold, C. B., Kellog, C. E., and Barnes, C. P. (1938). Soils and Men, 1938 Yearbook of Agriculture, House Document 398. In “The remedies: education and research”, pp. 198 –222. U.S. Government Printing Office, Washington, DC. Widman, N. L (1998). Field research and technology development needs request: manure leaching to subsurface drainage tile [online]. Accessible through USDA-NRCS/ARS Partnership Management Team website at http://www.nps.ars.usda.gov/pmt/, FY99 Needs Initiative, Record #291 (verified 18 July, 2000). Willems, J. J. G. M., Marinissen, J. C. Y., and Blair, J. (1996). Effects of earthworms on nitrogen mineralization. Biol. Fert. Soils 23, 57 –63. Winkler, J. P., Cherry, R. S., and Schlesinger, W. H. (1996). The Q10 relationship of microbial respiration in a temperate forest soil. Soil Biochem. 28, 1067–1072. Witkamp, M. (1969). Cycles of temperature and carbon dioxide evolution from litter and soil. Ecology 50, 922 –924. Yakovchenko, V., Sikora, L. J., and Kaufman, D. D. (1996). A biologically based indicator of soil quality. Biol. Fert. Soils 21, 245–251. Zachmann, J. E., and Linden, D. R. (1989). Earthworm effects on corn residue breakdown and infiltration. Soil Sci. Soc. Am. J. 53, 1846–1849. Zachmann, J. E., Linden, D. R., and Clapp, C. E. (1987). Macroporous infiltration and redistribution as affected by earthworms, tillage, and residue. Soil Sci. Soc. Am. J. 51, 1580–1586. Zaller, J. G., and Arnone, J. A. (1999). Earthworm and soil moisture effects on the productivity and structure of grassland communities. Soil Biol. Biochem. 31, 517–523. Zeide, B. (1998). Another look at Leopold’s land ethic. J. Forestry 96(1), 13– 19. Zeide, B. (1998). Forestry and ethics. J. Forestry 96(4), 25– 26. Zucker, L. A., and Brown, L. C. (1998). Agricultural drainage. Ohio State Univ. Ext. Bull. 871.

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

N. E. Borlaug (1), Texas A&M University, College Station, Texas, USA K. S. Dhillon (119), Department of Soils, Punjab Agricultural University, Ludhiana 141 004, India S. K. Dhillon (119), Department of Soils, Punjab Agricultural University, Ludhiana 141 004, India I. R. P. Fillery (69), Commonwealth Scientific and Industrial Research Organisation, P.O. Box 5, Wembley, Western Australia 6913, Australia K. W. T. Goulding (69), Agriculture and Environment Division, IACR, Rothamsted, Harpenden, Hertfordshire AL5 2JQ, UK L. A. Harrier (185), SAC, West Mains Road, Edinburgh EH9 3JG, UK D. J. Hatch (69), Institute of Grassland and Environmental Research, North Wyke, Okehampton, Devon EX20 2SB, UK L. S. Jensen (69), Plant Nutrition and Soil Fertility, The Royal Veterinary and Agricultural University, Thorvaldsensvej 40, DK-1871 Frederiksberg, Copenhagen, Denmark Lars Stoumann Jensen (227), Plant Nutrition and Soil Fertility Laboratory, Department of Agricultural Science, Royal Veterinary and Agricultural University, Thorvaldsensvej 40, DK-1871 Frederiksberg, Denmark Jacob Magid (227), Plant Nutrition and Soil Fertility Laboratory, Department of Agricultural Science, Royal Veterinary and Agricultural University, Thorvaldsensvej 40, DK-1871 Frederiksberg, Denmark D. V. Murphy (69), Centre for Land Rehabilitation, The University of Western Australia, Nedlands, Western Australia 6009, Australia P. M. Priyadarshan (351), Rubber Research Institute of India, Regional Station, Agartala 799 006, India S. Recous (69), Unite´ d’Agronomie, INRA, rue Fernand Christ, 02007 Laon Cedex, France Michael L. Smetham (301), Department of Plant Science, Lincoln University, P.O. Box 94, New Zealand R. E. Sojka (1), Northwest Irrigation and Soils Research Laboratory, USDA, Agricultural Research Service, Kimberly, Idaho, USA E. A. Stockdale (69), Agriculture and Environment Division, IACR, Rothamsted, Harpenden, Hertfordshire AL5 2JQ, UK Kristian Thorup-Kristensen (227), Department of Horticulture, Danish Institute of Agricultural Science, P.O. Box 102, DK-5792 Aarslev, Denmark D. R. Upchurch (1), Cropping Systems Research Laboratory, Lubbock, Texas, USA C. A. Watson (185), SAC, Craibstone Estate, Aberdeen AB21 9YA, UK

xi

Preface Volume 79 contains seven cutting-edge reviews on important topics in agronomy. Chapter 1 is a thought provoking and comprehensive review on quality soils management, authored by leading experts in the field including Nobel Laureate N. E. Borlaug. Chapter 2 discusses aspects of theory, measurement and application of 15N pool dilution techniques in assessing gross nitrogen fluxes in soil. Chapter 3 is a thorough treatment of distribution and management of seleniferous soils including their geochemistry and management, and effects of selenium on plants and ecosystems. Chapter 4 discusses the role of arbuscular mycorrhizal fungi (AM) in sustainable cropping systems. This review explains the link between AM fungi and agricultural management practices. Chapter 5 is a comprehensive review on the use of catch crops and green manures as biological tools in nitrogen management in temperate zones. Chapter 6 deals with subterranean clover (Trifoluim subterraneum L.) with emphasis on its ecology and use as a pasture legume in Australia. Included are discussions on characteristics of the species, physiological responses to the environment, management and productivity, and pests and diseases. Chapter 7 is a definitive review on breeding of rubber (Hevea brasiliensis) under environmental constraints including geoclimatic and biotic stresses. State-of-the-art breeding programs are also discussed. I appreciate the fine contributions of the authors. DONALD L. SPARKS

xiii

GROSS NITROGEN FLUXES IN SOIL : THEORY, MEASUREMENT AND APPLICATION OF 15N POOL DILUTION TECHNIQUES D. V. Murphy,1 S. Recous,2 E. A. Stockdale,3 I. R. P. Fillery,4 L. S. Jensen,5 D. J. Hatch6 and K. W. T. Goulding3 1 Centre for Land Rehabilitation, Faculty of Natural and Agricultural Sciences, The University of Western Australia, Nedlands, Western Australia 6009, Australia 2 Unite´ d’Agronomie, INRA, rue Fernand Christ, 02007 Laon Cedex, France 3 Agriculture and Environment Division, Rothamsted Research, Harpenden, Hertfordshire AL5 2JQ, UK 4 Commonwealth Scientific and Industrial Research Organisation, P.O. Box 5, Wembley, Western Australia 6913, Australia 5 Plant Nutrition and Soil Fertility, The Royal Veterinary and Agricultural University, Thorvaldsensvej 40, DK-1871 Frederiksberg, Copenhagen, Denmark 6 Institute of Grassland and Environmental Research, North Wyke, Okehampton, Devon EX20 2SB, UK

I. Introduction II. Microbial N Pathways III. Principles of 15N Pool Dilution A. Theory B. Assumptions of 15N Pool Dilution IV. 15N Pool Dilution Techniques A. Experimental Design B. Methods of Applying 15N to Soil C. Methodological Considerations V. Calculation of Gross N Fluxes A. Analytical Solutions B. Numerical Solutions C. Calculating Immobilisation VI. Applications of 15N Pool Dilution A. Factors That Regulate Gross N Fluxes B. Quantifying the Actual Magnitude of N Cycling C. C and N Mineralisation D. Competition for Substrate Availability E. Soil Disturbance F. Activity of Ammonia-oxidisers in Soil G. N Saturation Index

69 Advances in Agronomy, Volume 79 Copyright q 2003 by Academic Press. All rights of reproduction in any form reserved 0065-2113/02$35.00

70

D. V. MURPHY ET AL. VII. Future Research Acknowledgments References Isotopic pool dilution using 15N is proving to be a valuable tool for increasing our understanding of gross N cycling processes and our ability to both model these processes and link them to microbial function. However, not all applications are appropriate. Many of the questions asked by agronomists and soil scientists can often be addressed by simpler experiments in which measurements of the main parameters of inorganic and total N content of soil and plant components would suffice. In addition, the theory, assumptions and techniques associated with the calculation of gross N fluxes can lead to large errors if not applied correctly. Some preliminary assessment of the principle N transformation processes to be studied, followed by an optimisation of the experimental conditions are needed for the effective application of 15N pool dilution. When applied correctly under carefully controlled laboratory incubations, the technique has been used successfully to quantify gross N fluxes and to understand the fundamental processes that regulate individual microbial N pathways. This has improved our understanding of how C and N cycles are linked, and thus has led us to question the most appropriate structure of C and N cycling models. Field based 15N pool dilution studies have been used successfully to study the climatic influence on the soil N cycle and also to quantify the impact of external inputs. Further field-based studies are required to aid model development and evaluation. Linking soil microbial/molecular ecology with process-based studies of microbial nutrient cycling presents a new and exciting field of research that will benefit from the further application of isotopic pool dilution techniques for N and other nutrients. q 2003 Academic Press.

I. INTRODUCTION The importance of soil organic matter (SOM) as a source of inorganic N in natural and agricultural ecosystems is widely appreciated. Recent reviews have summarised knowledge on the chemistry of SOM (Schnitzer, 2000) and on the factors that affect the release of N from SOM (Jarvis et al., 1996; Silgram and Shepherd, 1999; Martens, 2000) and crop residues (Kumar and Goh, 2000). Nevertheless, prediction of the availability of ammonium (NHþ 4 ) and nitrate (NO2 ) in soil under specified environmental conditions remains inaccurate and 3 imprecise, because the net availability of inorganic N in soil results from the competition between several opposing soil processes, particularly mineralisation –immobilisation turnover (MIT) and nitrification, that together determine the net release.

GROSS NITROGEN FLUXES IN SOIL

71

Until recently, most estimates of N mineralisation and nitrification were net measurements, determined from the change in the size of all, or a part of the inorganic N pool over prolonged periods (e.g., Raison et al., 1987; Robertson, 1982; Binkley et al., 1992; Mary and Recous, 1994). In comparison, the total flux (i.e., gross rates) of N mineralisation and nitrification can be measured without the confounding influence of consumptive processes (e.g., immobilisation) using the15N pool dilution technique (Kirkham and Bartholomew, 1954). Although the principles of 15N pool dilution were elucidated early in the application of isotopes in soil research (Kirkham and Bartholomew, 1954), widespread use of this technique to study gross N fluxes has largely occurred in the last decade. Therefore, it is timely to evaluate the strengths, weaknesses and opportunities for applying this methodology to the studies of N turnover in soil (see also Nishio, 1991; Di et al., 2000). As is the case with many 15N-based methodologies, the application of 15N pool dilution is not straightforward, and incorrect use can result in serious errors in calculating gross N fluxes. The novelty of the 15N pool dilution technique has also led to applications that are not ideal or that do not contribute to new knowledge. For example, the determination of net mineralisation by difference from measurements of gross mineralisation and immobilisation seems to be an expensive and complicated way of obtaining data that can be measured simply and cheaply. On the other hand, the technique is ideal for examining the rapid changes in microbial activity commensurate with changes in availability of substrate and the environment, and for unravelling the complexities of microbial function. This chapter will critically evaluate and discuss various ways of using 15 N pool dilution techniques in the study of gross mineralisation, immobilisation and nitrification. Our aims are to review the theory behind 15N pool dilution, to summarise its strengths and weaknesses in studies of gross N fluxes, and to describe appropriate use of the methodology. Examples of novel applications of measurements of gross N fluxes including modelling are given to highlight the potential of the methodology.

II. MICROBIAL N PATHWAYS It is critical that before the technique of 15N pool dilution is applied, the user has a good understanding of the N cycling processes occurring at the microbial ˚ gren, 1995). These are described simply scale within soils (see Bosatta and A below to aid the reader of this paper. Soil microorganisms, as all living entities, derive the energy necessary for cellular function through oxidation reactions. The soil microbial biomass is primarily driven by the energy contained within photosynthetically fixed carbon (C), i.e., derived from plant materials. However, the continuous cycling

72

D. V. MURPHY ET AL.

and recycling of C results in a continuum of SOM stabilised against oxidation to varying degrees (Schnitzer, 2000). The enzymatic oxidation of SOM by a wide range of non-specific aerobic and anaerobic heterotrophic microorganisms (mineralisation) yields not just energy but also the bulk of the microorganism’s requirements for C and N, which are amongst the macronutrients required as the building blocks of microbial cells. As a result, the C and N cycles in soil are closely linked. Ammonification denotes the enzymatic processes by which soil nitrogenous compounds are transformed to yield NHþ 4 as a reaction product. Organic N substrates include amino acids, amino sugars, purine and pyrimidines derived from the enzymatic degradation of proteins, amino-polysaccharides and nucleic acids. In addition, soils may receive organic-N as urea; either from urine or from applied fertiliser. The enzymes involved in ammonification, such as hydrolases, oxidases, deaminases and lyases, originate from plant, animal and microbial sources. The enzymes may function endocellularly, in dead autolysing cells, free in the soil solution, and whilst adsorbed to soil colloids. The NHþ 4 formed is either absorbed by soil microorganisms to sustain their N requirements, or is used by nitrifying bacteria or may just accumulate in the soil. Ammonium can also be fixed in clay lattice and it can be bound to organic matter, processes that deplete þ exchangeable NHþ 4 . Under alkaline conditions, NH4 after conversion to gaseous NH3 can be lost from the soil. The process by which N is incorporated into a microbial cell depends on assimilative processes. Ammonia is absorbed passively by cells and assimilated either by the GDH or the GS-GOGAT pathways to form glutamate (Brown, 1980). Nitrate is also absorbed by cells and first reduced to NHþ 4 internally by assimilative nitrate and nitrite reductases and then assimilated to form glutamate (Dalton, 1979; Betlach et al., 1981). This additional step makes the energy cost higher for assimilating N from NO2 3 compared to NH3, and microorganisms assimilate N preferentially as NH3 which is in equilibrium with NHþ 4 in soil þ solution (Recous et al., 1990). Microorganisms assimilate NO2 3 when NH4 is limited (Azam et al., 1986; Recous et al., 1990). The release of NHþ 4 into soil solution during the decomposition of SOM, is commonly known as N mineralisation (strictly gross N mineralisation). Microbial 2 assimilation of NHþ 4 and NO3 from soil solution is commonly known as immobilisation (strictly gross immobilisation). As mineralisation and immobilisation of N occur simultaneously, N is continually transferred within the soil from inorganic to organic forms and back again (Fig. 1). The extent of N assimilation is closely linked to the concomitant availability of organic-C to sustain growth and energy requirement, i.e., to the amount of readily available carbonaceous materials. Where organic-N is converted to NHþ 4 prior to microbial assimilation, this pathway is known as MIT and dominantly regulates the availability of soilderived N in agricultural and natural ecosystems (Jansson and Persson, 1982; Drury et al., 1991; Hadas et al., 1992). Alternatively, microorganisms can utilise

GROSS NITROGEN FLUXES IN SOIL

Figure 1 dilution.

Schematic diagram of the N cycling processes that can be quantified using

73

15

N pool

a direct route where low molecular weight organic compounds (e.g., sugars, peptides) are taken up directly into the cell to meet microbial C requirements. Endogenous enzymes break down these compounds; only the excess N will be released (Jung and Luttge, 1980; Harper, 1984; Hadas et al., 1992; Barraclough, 1997). Ammonium can also be metabolised in other enzymatic reactions; NHþ 4 is 2 and NO as the result of the activity of both autotrophic oxidised to soil NO2 2 3 (inorganic N route for nitrification only) and heterotrophic (inorganic and organic-N routes for nitrification) microorganisms. Depending on the respective rates of mineralisation and nitrification in soil, either mineralisation or nitrification processes can control the accumulation of NO2 3 in soil. This will affect the potential for losses of N from soil –plant systems since NO2 3 is readily moved to drainage waters and may also be lost from soil in gaseous forms (N2, NOx) as a result of denitrification processes. The net change in the inorganic N pool in the soil (known widely as net N mineralisation/immobilisation) is therefore the outcome of the interacting gross 2 mineralisation, nitrification and NHþ 4 and NO3 immobilisation processes (Stark and Schimel, 2001). Whether net mineralisation (gross mineralisation . gross immobilisation) or net immobilisation (gross mineralisation , gross immobilisation) occurs is closely linked with C availability and has been related to the C/N ratio of the SOM (Paul and Juma, 1981; van Veen et al., 1984; Chaussod et al., 1988).

III. PRINCIPLES OF

15

N POOL DILUTION

A. THEORY Kirkham and Bartholomew (1954, 1955) first proposed equations based on the use of tracer data (using heavy or radioactive atoms) to measure the nutrient

74

D. V. MURPHY ET AL.

transformations in soil. They provided, as a working example, the dilution in abundance (at.%) of heavy N (i.e., 15N) to measure the rate of NHþ 4 production from the mineralisation of SOM, most of which will be as 14N. The use of what came to be called 15N pool dilution to measure gross N fluxes is now commonplace. However, as indicated by Kirkham and Bartholomew, these equations may also be applied to other elements; Di et al. (2000) report on the use of isotopic pool dilution to study phosphorus and sulphur transformations in soil. Gross mineralisation (m ) is measured by first enriching the soil NHþ 4 pool, which will contain 15N at natural abundance levels, with 15N to increase the 15Nenrichment above natural abundance. The dilution of the 15N-enrichment in the þ NHþ 4 pool, and change in the size of the NH4 pool is then traced through time as þ SOM is mineralised, releasing NH4 at natural abundance (Fig. 2). Equation (1) is used to calculate gross mineralisation (Kirkham and Bartholomew, 1954). Gross nitrification (n ) can also be calculated using Eq. (1) except that it is the NO2 3 pool that is enriched with 15N and the nitrification of soil NHþ 4 at natural abundance leads to a dilution in the 15N-enrichment of the NO2 3 pool. The “immobilisation”

15 2 Figure 2 Soil NHþ N isotopic excess (bottom) 4 -N and NO3 -N concentrations (top) and their over time for a sandy loam soil incubated at 3 and 158C. The soil NHþ 4 pool was labelled five times during 37 d (10 mg 15NH4-N kg21 at 4.72 at.% excess), and sampled initially (2 h), intermediately and after 3–5 d, points connected by a line belong to the same time interval. Bars show standard errors (n ¼ 3). Data modified from Andersen and Jensen (2001).

GROSS NITROGEN FLUXES IN SOIL

75

(i ) calculation (Eq. (2)) reported by Kirkham and Bartholomew (1954) should 2 more correctly be termed NHþ 4 or NO3 consumption since the calculation includes all processes that remove N from the 15N-labelled pool. Kirkham and Bartholomew (1955) proposed a subsequent equation, which allowed remineralisation of the applied labelled 15N. However, this was intended for use in a simple two-pool model and cannot account for the removal of N from the 15N2 labelled pool (NHþ 4 or NO3 ), except via immobilisation. This equation has not been widely adopted. As pointed out by Schimel (1996), the 15N pool dilution technique does not measure mineralisation and immobilisation per se; rather it 2 measures the production and consumption of the N pool (either NHþ 4 or NO3 ) 15 that has received the N tracer without regard for the specific processes. It is necessary therefore to have a sound understanding of the dominant microbial N processes occurring within the soil, so that correct partitioning of input and output fluxes is achieved. Equations (1) and (2) of Kirkham and Bartholomew (1954) are used when i – m and for conditions when i . m so that all components of the equation yield positive quantities. Kirkham and Bartholomew (1954) present alternative equations when m . i and i ¼ m. m¼ i¼

M0 2 M log H0 £ M=H £ M0 £ t log M0 =M M0 2 M log H0 =H £ t log M0 =M

when i – m

when i – m

ð1Þ ð2Þ

where m is the mineralisation rate per unit mass of soil per unit time (mg N kg21 d21); i, proposed as immobilisation but actually total N consumed from the N pool receiving 15N tracer per unit mass of soil per unit time þ (mg N kg21 d21); M, NHþ 4 total—mass of tracing plus non-tracing NH4 -N per 21 þ þ unit mass of soil (mg N kg ); H, NH4 tracer—mass of tracing NH4 -N per unit mass of soil (mg N kg21); t, time, refers to the unit of time (days) between the first (M0, H0) and subsequent (M, H ) soil analysis. These formulae are also equivalent to Eqs. (6) and (7) given by Wessel and Tietema (1992). A similar formulation can be derived for the nitrification rate (n ) in experiments where the NO2 3 pool has been labelled (Davidson et al., 1990, 1991; Barraclough, 1991) with the nomenclature of n, gross nitrification per unit mass of soil per unit time (mg N kg21 d21); M, NO2 3 total—mass of tracing plus 21 non-tracing NO2 ); H, NO2 3 -N per unit mass of soil (mg N kg 3 tracer—mass of 21 2 tracing NO3 -N per unit mass of soil (mg N kg ).

B. ASSUMPTIONS OF

15

N POOL DILUTION

The application of 15N pool dilution is based on a number of assumptions. Many of the problems associated with the application of this technique arise

76

D. V. MURPHY ET AL.

when one or more of the assumptions are not met. This section discusses assumptions that must be met to minimise potential sources of error in studies using 15N pool dilution. Assumptions pertinent only to analytical calculations of gross N fluxes, that are not relevant when using mass balance models to calculate gross N fluxes, are discussed elsewhere (Section “Analytical solutions”).

1. No Isotopic Discrimination The principle of 15N pool dilution assumes that microbial processes, which 2 14 consume N from the soil NHþ 4 or NO3 pool, do not discriminate between N and 15 N isotopes. Isotopic discrimination of N molecules has been shown to occur in microbially mediated processes at natural abundance levels (Cheng et al., 1964; Heaton, 1986). However, errors associated with isotopic discrimination are usually considered to be negligible at 15N-enrichment above natural abundance if incubation intervals are short (Davidson et al., 1991). Wessel and Tietema (1992) recommend that the 15N abundance of the enriched pool should remain several times that of natural abundance. Problems arising from isotopic discrimination can also be reduced if the dilution in 15N-enrichment over the incubation period is large.

2.

Uniform Distribution of

15

N

Uniform distribution of the applied 15N-label throughout the soil, to achieve uniform 15N-enrichment of the inorganic N pool in question (i.e., whether NHþ 4 15 or NO2 N pool dilution. However, a completely 3 ), is a prerequisite of homogeneous mixture of the isotope with the soil inorganic N pool is virtually impossible. Davidson et al. (1991) calculated that errors of approximately 10% are caused when less than 70% of mineralisation micro-sites received 15N. Larger errors can occur if a bias in distribution of 15N corresponds with either a gradient 2 of soil NHþ 4 or NO3 concentration and/or microbial N transformation rates (Davidson et al., 1991). Similarly, Monaghan (1995) showed that, with nonuniform 15N-labelling, the errors increased considerably when the N consumption rates equalled or were greater than N production rates (i.e., mineralisation or nitrification).

3. 15

Equilibrium between N Pools

N pool dilution implies that equilibrium has been achieved between applied and indigenous N pools. Equilibrium in this case refers to the concept that once

GROSS NITROGEN FLUXES IN SOIL

77

the 15N is applied to soil it is in the same chemical state and location within the soil as the indigenous N, and that any soil N transformation process that occur during the subsequent incubation period (e.g., diffusion, fixation, gaseous loss pathways, immobilisation, nitrification) would equally affect the applied and indigenous N pools. Given the heterogeneity of the soil matrix and of microorganisms within that matrix, it is highly unlikely that the applied N will form an immediate equilibrium with the indigenous soil N. Until equilibrium is 2 reached 15N-enriched NHþ 4 or NO3 will not be consumed at the same 14 proportional rate as indigenous N. Avoiding the preferential use of either the added label or the indigenous soil N has been described as the most intractable problem in isotopic dilution (Barraclough, 1995).

IV.

15

N POOL DILUTION TECHNIQUES

Different approaches can be used to measure gross N fluxes. The most appropriate method will depend on the N transformation of interest and whether in situ measurements are intended or whether comparative rates are acceptable. There are also methodological considerations concerning, for example, the time of initial soil sampling after 15N application and the length of the incubation period. Therefore, it is important that the user first establishes the primary objective of the experiment before selecting the most appropriate experimental design.

A. EXPERIMENTAL DESIGN 1.

Gross N Fluxes from SOM

15 The application of 15NHþ 4 [e.g., ( NH4)2SO4] only enables the measurement of gross mineralisation (input flow) and the sum of NHþ 4 consumption (output þ flow) which can include NHþ immobilisation, NH oxidisation, gaseous loss and 4 4 plant uptake. In this case, immobilisation of added 15NHþ can be as NHþ 4 4 or as 2 2 15 15 NO3 after nitrification. Application of NO3 (e.g., K NO3) enables the measurement of gross nitrification (input flow) as the sum of the autotrophic plus heterotrophic nitrification, and NO2 3 consumption (output flow) that can include NO2 immobilisation, gaseous loss and plant uptake. In non-planted systems 3 where denitrification is considered to be negligible, NO2 3 consumption is often reported directly as NO2 immobilisation. 3 2 15 15 Different NHþ 4 and NO3 salts [e.g., ( NH4)2SO4 versus K NO3] are often þ used as “paired treatments” to calculate NH4 immobilisation from the measured þ NHþ 4 consumption (NH4 immobilisation þ gross nitrification) minus gross nitrification. This approach assumes that N cycling by microorganisms is

78

D. V. MURPHY ET AL.

unaffected by the form of N added. However, the addition of different N forms 2 can cause changes in the partitioning between immobilisation of NHþ 4 and NO3 (Mary et al., 1998) and in other fluxes. “True” paired treatments involve the application of both 15NH4NO3 and NH15 4 NO3 to separate soil subsamples. This 2 approach enables all gross N fluxes affecting the NHþ 4 and NO3 pools to be calculated simultaneously, negating N cycling issues where different forms of N are applied. A major advantage of applying K15NO3 instead of NH15 4 NO3 in the studies of gross nitrification is that NHþ 4 is not provided as an additional substrate for the nitrifier populations under study (Davidson et al., 1991; Willison et al., 1998b). When NH15 4 NO3 is applied, the nitrification rate then becomes “potential” instead of “actual”. Whilst this may lead to an overestimation of gross nitrification, there is an interest in assessing the rate of potential nitrification in soils where it is crucial to identify whether mineralisation or nitrification is the limiting step of 15 N-labelling treatments is NO2 3 accumulation. The use of NH4NO3 in paired also required in studies that include measurements of plant uptake that may 2 deplete both the NHþ 4 and NO3 pools (Barraclough, 1991). 2. 15

Gross N Fluxes from Residues

N pool dilution can also be used to determine the rate of gross N fluxes attributed to the mineralisation of an added organic substrate. Several methods have been proposed to quantify gross N fluxes associated with the decomposition of freshly added organic matter. The difference method involves determining gross mineralisation in the presence or absence of residues. Organic residues are incorporated and, at intervals, 15NHþ 4 is injected both into the amended and nonamended soil (Watkins and Barraclough, 1996; Recous et al., 1999; Andersen and Jensen, 2001). If conditions are identical, differences in gross N fluxes between the two treatments will be due to the N release from the decomposition of the incorporated residues. This approach assumes that the basal N mineralisation rate is the same in the presence or absence of residues. One valid criticism (Stevenson et al., 1998; Schimel et al., 1992; Silgram and Shepherd, 1999) of the difference method is the potential for a decline in the decomposition rate of the indigenous SOM on addition of a more labile C source (i.e., the basal rates of gross N fluxes from the SOM in the amended soil are lower compared to non-amended soil). However, if soils are chosen where gross N fluxes are small prior to residue addition (as with most arable soils), and given that in most studies residue additions are relatively high (compared to field application rates), this potential problem can be minimised. It should also be recognised that 15N pool dilution does not actually distinguish between sources of the mineralised N. This is shown in Fig. 3 where the cumulative gross fluxes of N (using the difference method) were greater than

GROSS NITROGEN FLUXES IN SOIL 79

Figure 3 Cumulative gross N fluxes (mg N kg21 d21) attributable to the application of plant residue (403 mg N kg21; dashed line) to soil incubated at 2, 5, 10 and 158C: (a) gross mineralisation, (b) gross immobilisation (i.e., gross–net mineralisation), (c) gross nitrification, and (d) NO2 3 consumption. Data derived from the difference between reported rates (Cookson et al., 2002) for gross N fluxes in amended and non-amended soil.

80

D. V. MURPHY ET AL.

the amount of N contained within the residue; highlighting the fact that the microbial biomass is turning over and releasing the previously immobilised residue-N. Thus, it is difficult to quantify the actual release from the residue using the difference method since cycling of residue-derived N between soil N pools during incubation will also be quantified within these fluxes. It is therefore recommended to also quantify C mineralisation from the residue (e.g., Macdonald, 2000; Trinsoutrot et al., 2000) so that it is possible to determine when the residue-C has been depleted and the incubation becomes C-limited. Alternatively the direct method involves the reverse situation, i.e., adding homogeneous 15N-labelled organic substrate and adding at interval unlabelled NHþ 4 (Watkins and Barraclough, 1996; Hood et al., 1999). The most robust approach is thus to use paired treatments with unlabelled and 15N-labelled 15 residues and NHþ N-labelled or unlabelled, but otherwise identical soil 4 treatments. This is known as the mirror image approach (Monaghan and Barraclough, 1995, 1997; Hood and Wood, 1996; Watkins and Barraclough, 1996). The combination of equations allows the mineralisation from the SOM and the mineralisation derived from the residue to be calculated. This method was used to investigate the possibility of direct assimilation of organic-N by microorganisms versus the MIT route (Barraclough, 1997; Gibbs and Barraclough, 1998). Lastly, Hood et al. (2000) proposed an alternative method that requires prelabelling of the SOM by immobilising applied inorganic 15N in the presence of an available C substrate (e.g., cellulose). This pre-labelling must occur at a sufficient time period before adding unlabelled residues to ensure that both the soil inorganic N pool and newly produced inorganic N from basal mineralisation are of the same 15N-enrichment and thus not altered by N immobilisation due to residue addition. In theory, this will overcome the problems associated with pool substitution. Mineralisation of unlabelled residue N is then calculated from the extent of 15N pool dilution of the previously enriched mineral N pool. However, the applicability of this method needs further testing, particularly under field conditions (e.g., Hood, 2001).

B. METHODS OF APPLYING

15

N TO SOIL

Three approaches are available to apply 15N to soil: application of solutions 15 2 2 containing 15N-enriched NHþ N-enriched NHþ 4 or NO3 ions, adding 4 or NO3 15 salt with an inert solid carrier, and N-enrichment of chamber atmospheres or injection into soil of gaseous N compounds, most notably NH3 or NO. Because solutions are easier to apply quantitatively than gases and are better carriers than inert solids, they have been the preferred method. However, there are conditions where the application of solution to soil could modify rates of N transformations, necessitating the use of “dry” application methods.

GROSS NITROGEN FLUXES IN SOIL

1.

81

Solution Application

Typically gross N fluxes are measured in laboratory studies after mixing small volumes of 15N-labelled solution into soil samples or leaf litter held at near optimal soil water contents (e.g., Nishio et al., 1985; Myrold and Tiedje, 1986; Bjarnason, 1988; Crawford and Chalk, 1992; Barrett and Burke, 2000). Mixing ensures that the added 15N and microbial N transformation rates are distributed randomly. However, the physical disturbance of soil can potentially alter N fluxes under study (see Section “Soil disturbance”). Pre-incubating the soil prior to the application of 15N can minimise the effect of the initial flush of microbial activity after soil disturbance. This should only be done if it does not cause a C-limitation in the soil as a result of the increased total incubation period. To ensure relatively uniform distribution of 15N without further mixing, it is necessary to ensure that the soil depth is restricted to , 2 cm and that multiple drops of 15N solution are applied over the surface (Barraclough and Puri, 1995; Garcia-Montiel and Binkley, 1998; Scott et al., 1998; Willison et al., 1998b). Several studies (Nishio and Fujimoto, 1989; Davidson et al., 1991; Barraclough and Puri, 1995; Sparling et al., 1995; Murphy et al., 1997; Willison et al., 1998b) conclude that gross N fluxes may be overestimated when 15N is applied as solution. The flush of N mineralisation following the addition of water is most profound when applied to dry soil, the so-called Birch effect (Birch, 1958). Therefore, the use of solution to deliver 15N is unlikely to be an appropriate method in studies of gross N mineralisation in dry soil for any soil type (Davidson et al., 1991; Sparling et al., 1995; Stark and Firestone, 1995). Davidson et al. (1991) recommended that 15N pool dilution techniques using solution not be applied to soils with matric potentials below 2 1.5 MPa. Surprisingly, no actual data exists on critical threshold water contents for the application of 15N solutions. The use of the solution as a 15N carrier is, therefore, restricted to fine-textured soils that are not dry if the purpose is to obtain gross N fluxes that reflect actual in situ rates of N transformations. Solutions still have a wider application in laboratory studies where the actual magnitude of the gross N fluxes is of less interest and 15N pool dilution techniques are being employed to obtain a greater fundamental understanding of differences in N cycling between treatments (e.g., Gibbs and Barraclough, 1998; Tlustos et al., 1998; Mendum et al., 1999).

2.

Application of

15

N Gases

Stark and Firestone (1995) commented that exposure of soil to 15NH3 would provide a method for labelling the soil NHþ 4 pool without the application of water. However, they did not describe a practical method for applying low concentrations of 15NH3 uniformly to soil. Murphy et al. (1997) designed

82

D. V. MURPHY ET AL.

a multiple needle gas injector that can deliver small amounts of 15NH3 contained in air (Table I). This system was designed to measure field rates of gross mineralisation in coarse-textured soils where solution application was undesirable (Murphy et al., 1998a,b,c), but the applicator has since been used successfully with finer-textured agricultural soils (Murphy et al., 1999). These studies have demonstrated that injection of 15NH3 gas to soil cores can produce uniformly labelled NHþ 4 pools without major soil disturbance, and particularly without modification of soil water content. The alternative approach—addition of small amounts of 15N-labelled gas to the headspace above soil in sealed jars involves the use of either 15NO to label the 15 NH3 to label the NHþ soil NO2 3 pool (Stark and Firestone, 1995) or 4 pool (Murphy et al., 1999). Although this approach is simpler to apply than direct gas injection into soil cores, some doubt exists about the uniformity of distribution of 15 N where the infiltration of gas into soil is passive, and therefore the validity of the technique. For example, Murphy et al. (1999) found the headspace 15NH3 labelling technique to underestimate gross mineralisation when compared against 15 NH3 gas injection and 15N solution application.

3.

Dry Powder

15

N Delivery

Willison et al. (1998b) describe techniques where a silica flour – 15N labelled NO2 3 mixture was used as a dry method to measure gross nitrification in disturbed soil. They note that the method is especially applicable to N-limited soils where the addition of water was likely to cause an increase in microbial activity and/or a flush of mineralisation and thus an increase in NHþ 4 availability to ammonia oxidisers. For example, 15N solution application was found to increase gross nitrification rates by 13– 155% (average 84%) compared to systems amended with 15N-labelled silica flour. The need to incorporate silica flour – 15N labelled NO2 3 mixtures into soil restricts the application of the method to studies of gross nitrification rates in disturbed systems; injection of 15NO2 3 labelled solutions is the only current practical means of determining in situ gross nitrification in the field using 15N pool dilution techniques. The potential overestimation of gross nitrification noted by Willison et al. (1998b) does illustrate that reported absolute gross nitrification rates under field conditions may not be a true measure of nitrifying activity in some systems.

4.

Barometric Process Separation

A novel non-isotopic approach to estimate gross nitrification, based on barometric process separation (BaPS), has been proposed by Ingwersen et al. (1999).

Description of the Injection Systems Used to Deliver

Table I N into Undisturbed Soil Samples and Their Respective Needle Injection Density and Volume of Delivery

15

Single point injections made into 4.2 cm diameter cores—12 injections to a depth of 8 cm were made into 9 cm long soil cores using an 18-gauge spinal needle Single point injections made into 4.0 cm diameter cores—six injections made from the base of 9 cm long cores with a single 18-gauge sideport opening needle to within 1 cm of the soil surface Single point injections made into 10.0 cm diameter cores—10 injections made using 10.0 cm long spinal needles Single point injections made into 5.4 cm diameter cores—one injection made centrally at the top of the core and four 3 cm deep injections made into the sides at 5, 9, 13, and 17 cm from the surface Multiple point injector designed for 7.5 cm dia. cores—seven hypodermic needles 5.0 cm long with single sideport outlets at the tipp. All needles were connected to a single large detachable syringe Multiple point injector designed to cover one-fourth of a 10 cm diameter core—nine needles18-gauge and 9.0 cm injection depth with two sideport openings at the tip (i.e., 36 injection points per core). Each needle was attached to a separate syringe barrel (length 9 cm, volume 1 ml). This ensured solution was injected uniformly with depth and also uniformly between injection points Single point injections made into 4.8 cm diameter cores—eight injections into 15 cm long soil cores

Soil surface injection density

Soil texturea

Reference

1 per 1.2 cm2 soil

1 ml 15N solution per 20.8 cm3 soil

SiL

Schimel et al. (1989)

1 per 2.1 cm2 soil

1 ml 15N solution per 18.8 cm3 soil

SiL

Davidson et al. (1991)

1 per 7.9 cm2 soil

1 ml 15N solution per 78.6 cm3 soil 1 ml 15N solution per 91.6 cm3 soil

Riparian fen

Ambus et al. (1992) Stockdaleet al. (1994)

1 per 6.3 cm2 soil

1 ml 15N solution per 7.4–11.0 cm3 soil

C

Monaghan (1995)

1 per 2.2 cm2 soil

1 ml 15N solution per 19.6 cm3 soil

LS

Sparling et al. (1995)

1 per 2.3 cm2 soil

1 ml 15N solution per 17.0 cm3 soil

SL, L and SiL

NA

SL

83

Stark and Hart (1997) (continued on next page)

GROSS NITROGEN FLUXES IN SOIL

Injection system

Volume of 15N solution or NH3 gas delivered per unit of soil or forest material

84

Table I (continued)

Injection system

Single point injections made into 7.5 cm diameter soil sampling ring (cores were only inserted at time of soil collection)—six injections made using a 5.0 cm hypodermic needle with four sideport openings. 15N solution injected at three depths down to 5.0 cm Single point injections made with self-refilling syringe—33 injections made into 25 cm diameter scores at 2.5, 7.5, 12.5 and 17.5 cm from the surface. Total soil layer analysed 0–20 cm Multiple point injector designed for 5.0 cm dia. soil cores—12 needles 8.8 cm long with one sideport opening. Soil core was 10 cm long but injections were made between 1 and 8 cm

2

1 per 18.4 cm soil

1 per 3.4 cm2 soil

1 per 6.3 cm2 soil

15

1 ml N solution per 128.7 cm3 O forest material 1 ml 15NH3/air gas mixture per 7.3 cm3 soil

Soil texturea

Reference

Oe and Oa forest layer

Berntson and Bazzaz (1997) Murphy et al. (1997)

LS and SCL

1 ml 15N solution per 24.3 cm3 soil 1 ml 15N solution per 24.5 cm3 soil

SiCL

1 per 14.9 cm2

1 ml 15N solution per 62.0 cm3 soil

L

Recous et al. (1999)

1 per 1.6 cm2

1 ml 15N solution per 16.4 cm3 soil (assumes sampling depth 0–10 cm)

SL

Zaman et al. (1999)

1 per 7.4 cm2 soil

Calcareous

Ledgard et al. (1998) Jamieson et al. (1998)

D. V. MURPHY ET AL.

Multiple point injector consisting of 12 needles to be used 17 times within a microcosms (75 cm £ 50 cm)—hypodermic needles 22-gauge and 7.0 cm injection depth (i.e., 204 injections per microcosm) Multiple point gas injector system designed to deliver a mixture of 15NH3 and air into 5.5 cm dia. soil cores using seven 13 cm long 14-gauge stainless steel tubing containing two sideport openings at the end of each needle. The gas is injected over a 12 cm soil depth with the effective labelling and soil sampling depth being 0–10 cm. A 1:1 ratio between syringe and barrel lengths ensured uniform distribution of the label as the needles are withdrawn from soil Multiple point injector of similar design to Monaghan (1995)

Soil surface injection density

Volume of 15N solution or NH3 gas delivered per unit of soil or forest material

Table I (continued)

Injection system

Soil surface injection density

Volume of 15N solution or NH3 gas delivered per unit of soil or forest material

Soil texturea

1 per 3.3 cm2 soil

1 ml 15N solution per 49.1 cm3 soil

L

Multiple point injector designed for 10 cm diameter cores—13 needles 14-gauge with an injection depth of 15 cm with two sideport openings. All needles were connected to a single large detachable syringe. The mechanism ensured uniform delivery of the solution with depth. A template of 13 spikes was used to make holes in the soil prior to injection Multiple point injector designed for 6.0 cm dia. cores—seven hypodermic needles with sideport openings injected 3.0 cm into 3.5 cm long cores

1 per 6.7 cm2 soil

1 ml 15N solution per 29.5 cm3 soil

SiCL

1 per 4.0 cm2 soil

1 ml 15N solution per 31.4 cm3 soil

SL

a

L, loam; SL, sandy loam; LS, loamy sand; SCL, sandy clay loam; SiL, silt loam; SiCL, silty clay loam; C, clay.

Campbell and Gower (2000) Hatch et al. (2000b)

Andersen and Jensen (2001)

GROSS NITROGEN FLUXES IN SOIL

Single point injections made into 5.0 cm diameter cores—multiple (six assumed) injections made using a single 15.0 cm long sideport needle

Reference

85

86

D. V. MURPHY ET AL.

Changes in air pressure and O2 and CO2 concentrations above intact soil cores are attributed to the microbial activity of nitrification (pressure decrease), denitrification (pressure increase) and respiration (pressure neutral). Comparison of the BaPS system to 15N pool dilution yielded average gross nitrification values that were 40% higher when determined using BaPS (13.3 mg N kg21 d21) compared to 15N pool dilution (9.4 mg N kg21 d21; Ingwersen et al., 1999). This difference was not significant leading Ingwersen et al. (1999) to conclude that the BaPS system was suitable for the measurement of gross nitrification in wellaerated soils. However, the sensitivity of the BaPS technique at low rates of nitrification appeared to be poor in comparison to 15N pool dilution measurements given the 10-fold higher minimum rates obtained with the BaPS technique (Ingwersen et al., 1999). These authors do not address this discrepancy suggesting that the technique requires further critical evaluation against 15N pool dilution.

C. METHODOLOGICAL CONSIDERATIONS The apparent simplicity of 15N pool dilution has encouraged the use of this technique in many aspects of N cycling research, in some cases without thought to the correct methodological approach. This section addresses major methodological considerations and highlights difficulties with applying this technique to a soil.

1.

Amount and Enrichment of

15

N

Wessel and Tietema (1992) suggest that the 15N-labelling of the enriched pool should be of several times natural 15N abundance; since the precision of measurements is reduced as 15N-enrichments approach natural abundance values. New generation mass spectrometers make this less of an issue. Whilst higher 15N application rates can ensure sufficient enrichment in the labelled pool, it is generally recommended that as little total N as possible should be applied to avoid stimulating microbial processes that consume N (Davidson et al., 1991; Di et al., 2000). Defining a suitable 15N application rate for a soil is therefore a compromise between not increasing soil pool sizes unrealistically and attaining sufficient enrichment to follow the 15N pool dilution with precision. In soils with large indigenous NHþ 4 pool sizes (e.g., forest and some grassland systems) it is possible to obtain sufficient enrichment of the soil NHþ 4 pool with additions of highly enriched 15N equivalent to only 5 – 25% of the initial pool size (e.g., Wessel and Tietema, 1992; Berntson and Bazzaz, 1997). However, in many 21 ) and this pool has arable soils indigenous NHþ 4 levels are small (, 2 mg N kg

GROSS NITROGEN FLUXES IN SOIL

87

been shown to turnover rapidly in 15N pool dilution studies (often within 2 d; Murphy et al., 1998a, 1999). An N application rate of 1 –10 mg N kg21 (. 10 at.%) has been required to maintain sufficient enrichment during the incubation period in a number of agricultural soils (e.g., Rees et al., 1994; Hungate et al., 1997; Willison et al., 1998a; Murphy et al., 1999; Recous et al., 1999). Larger applications of highly enriched 15N have also been applied to temperate grassland soils (e.g., Jamieson et al., 1998; Hatch et al., 2000a) whilst smaller N applications (0.4 –0.8 mg N kg21; 12 at.%) were used by Murphy et al. (1997) in arable soils of low fertility.

2. Incubation Period It is necessary to conduct an initial soil extraction to ascertain the proportion of applied 15N that is actually involved in 15N pool dilution, i.e., to exclude NHþ 4 fixation and to ensure that equilibrium exists between indigenous and applied N pools. Calculations from the studies that have not conducted initial soil extractions can only be used to approximate the magnitude of true rates since the assumption that all applied 15N is microbially available but almost never holds (e.g., Davidson et al., 1991; Stockdale et al., 1994; Murphy et al., 1997, 1999). Davidson et al. (1991) highlighted the need to conduct a true initial soil extraction by showing that gross mineralisation rates were overestimated by 7% in a forest O2 layer and by 122% in a grassland mineral soil when an actual initial soil extraction was not used. Gross nitrification and N consumption rates were also overestimated. In general, the error associated with estimating initial 14N and 15 N pool sizes is larger in systems where the 15N at.% enrichment only declines slightly during the incubation period (Davidson et al., 1991). Since the time required for abiotic fixation and equilibrium to occur will vary between soils, the required incubation period before initial sampling is not constant. Murphy et al. (1997) found no difference in measured rates of gross mineralisation if initial soil measurements were taken at 1, 2 or 4-h after 15N application to coarse textured soils. In soils that are more aggregated with greater clay contents, a delay of up to 24-h is often used in an attempt to improve the uniformity of 15N distribution before initial soil extraction (e.g., Hatch et al., 2000a; Ledgard et al., 1998; Murphy et al., 1999). When applying silica flour – 15N labelled NO2 3 mixtures to measure gross nitrification, Willison et al. (1998b) found that 48-h was required to reach the equilibrium. Although very short periods may be valid for some soils, 24-h seems to be a sensible choice for the first extraction period given that the potential for remineralisation usually preclude longer initial time periods. We suggest that two subsequent soil extractions are made within the 2 –6 d after 15N application so as to ensure that 15N pool dilution can be quantified

88

D. V. MURPHY ET AL.

accurately. The length of incubation used should consider the anticipated rate of 15 N pool dilution so that the final enrichment does not decline to natural abundance.

3.

Rapid Consumption

A difficulty with applying 15N pool dilution in many soils is that ammonia oxidisers rapidly deplete (possibly preferentially) applications of 15NHþ 4 (Myrold and Tiedje, 1986; Mendum et al., 1999). Watson et al. (2000) observed that the 15 NHþ 4 applied to four grassland soils was rapidly nitrified, with 24– 55% of the 15 NHþ added label recovered as 15NO2 3 after 24-h. This rapid conversion of 4 to 15 2 NO3 occurred without a concurrent increase in the size of the unlabelled pool. This was interpreted to indicate non-uniform mixing of the 14N and 15N pools concomitant with a preferential nitrification of newly applied 15NHþ 4 , creating significant errors in estimated gross N mineralisation rates. However, Stark and Schimel (2001) have questioned this conclusion. They argued that NO2 3 consumption was occurring, causing 15N to accumulate in the NO2 3 pool at a disproportionately higher rate than 14N. In response, Watson et al. (2002) reinterpreted their data using a numerical model similar to the one described by Stark and Schimel (2001) that included NO2 3 consumption. These subsequent simulations supported their original hypothesis of preferential use of the applied N, but at lower rates than initially proposed. Use of nitrification inhibitors is recommended in soils where rapid nitrification could cause preferential consumption of applied NHþ 4 and/or rapid depletion of pool. Evidence from laboratory experiments suggests that nitrification the 15NHþ 4 inhibitors have little or no direct effect on gross mineralisation and immobilisation (Nishio et al., 1985; Myrold and Tiedje, 1986; Chalk et al., 1990; Crawford and Chalk, 1992; Barraclough and Puri, 1995). However, acetylene (C2H2) should be used with caution because it is a possible C-substrate for microbial activity and this inhibitor has been shown to increase NHþ 4 immobilisation (Hatch et al., 2000b). We recommend the effect of all nitrification inhibitors on gross N fluxes be tested before routine use in 15N pool dilution studies.

4.

Undisturbed Soil

Schimel et al. (1989) and Davidson et al. (1991) were the first to apply 15N to relatively undisturbed soil cores using a single needle-syringe assembly (Table I). Others have since developed injection equipment containing several needles to increase the speed and accuracy of 15N-labelling (Table I). Precise delivery of

GROSS NITROGEN FLUXES IN SOIL

89

solution can be achieved, as needles are withdrawn through soil by using a simple sliding mechanism (Sparling et al., 1995; Murphy et al., 1997) or through a geared mechanism (Hatch et al., 2000a). A more troublesome issue is the interaction of volume of 15N solution applied and needle injection density on the horizontal distribution of 15N in soil cores. Monaghan (1995) found a gradient of 15 NHþ 4 radiating out from a single injection point, with 32% of the recovered label in a sandy loam and 91% in a clay soil occurring within a horizontal radius of 1.5 cm (i.e., central 15% of the soil core). Such gradients in 15N concentration introduce large errors in estimates of gross N fluxes (Davidson et al., 1991). This unevenness in the distribution of 15N within soil cores is largely eliminated when using multiple injections, because of the overlap in delivery of 15N between injection points. However, we recommend that injection systems be assessed for both vertical and horizontal uniformity of 15N application of liquid before studies of gross N fluxes are attempted in a soil (e.g., Stockdale et al., 1994; Monaghan, 1995; Murphy et al., 1997; Andersen and Jensen, 2001). It is often assumed that a delay in initial soil extraction will aid the radial distribution of injected N. Nye and Tinker (1977) give solution diffusion rates for non-adsorbed ions such as chloride (Cl2) and NO2 3 in a sandy clay loam, containing 20% by volume of water as 1026 cm2 s21 and 40% by volume of water as 1025 cm2 s21. In this case, NO2 3 will diffuse , 1 cm from the injection site in 24-h. A strongly adsorbed ion such as NHþ 4 diffuses much more slowly, 1027 and 1028 cm2 s21 in the sandy loam soil under the same moisture conditions (i.e., , 0.1 mm in 24-h). Thus a better distribution of applied 15NO2 3 is attainable with delayed initial soil extraction, and reduced needle density may be possible. However, this is not the situation with 15NHþ 4 distribution and a high needle density should still be used. A few studies (Geens et al., 1991; Unkovich et al., 1998) have measured gross N fluxes without any soil disturbance by applying a 15N solution to soil surfaces. Subsequent irrigation is typically used to further distribute 15N into soil. However, we do not recommend use of this approach for two main reasons. First, it is difficult to determine the depth of infiltration of the 15N-label into soil, particularly if macropore flow occurs. Second, it is questionable whether this procedure provides satisfactory uniform vertical distribution of the 15N-label throughout the soil core, especially where 15NHþ 4 is applied to fine-textured soils. Nevertheless, in situations where considerable stone content or large tree roots preclude the use of injection systems, surface application of solutions may be the only practical approach available. Some of the limitations of the surface application of solutions can be minimised by aiding the downward distribution of the 15N-label by applying solution into holes augered at regular intervals into the soil volume under study. An example of such an application is described by Mikan et al. (2000) for studies of gross mineralisation in large root chambers. In this case, 40 –60 l of 15N-labelled solution was applied to four holes (1.2 cm diameter that were made to a depth of 1 m) in an area of 0.64 cm2.

90

D. V. MURPHY ET AL.

Some concentration of the 15N was likely to have occurred adjacent to these holes, but the ratio of 15N solution applied to the volume of soil used in this study (equivalent to 1 ml solution per 11 – 16 cm3) was much better than used in many injection systems described in Table I.

5.

Spatial Heterogeneity

Wessel and Tietema (1992) showed that large error terms are likely to be associated with the calculation of gross N fluxes, with the heterogeneity of soil parameters being the main source of error. This finding should not be unexpected. Calculations of gross N fluxes are based on the change over time of both N pool size and 15N-enrichment, using two distinct soil samples that were theoretically identical at the start of the incubation. In reality this will never be achieved. How significant is this problem? Stockdale et al. (1994) sampled 100 soil cores from an area of 100 m2 and showed that the coefficient of variation (CV) for NHþ 4 concentration was 60%. Bhogal et al. (1999) reported an even higher CV (600%) for net N mineralisation when 36 cores were taken from a 12 m2 area. Murphy et al. (1997) showed that gross mineralisation ranged from 0.89 to 2.50 mg N kg21 d21 when 5 cm diameter soil cores were collected from 1 m2. Expressing gross N mineralisation rates per unit of microbial biomass-N in each set of cores accounted for most of the variation in gross mineralisation (Murphy et al., 1997). Davidson et al. (1991) attempted to minimise problems of spatial variability by using a large cylinder placed around a smaller cylinder; the soil between the cylinders could then be removed and analysed as a representative sample of the soil core. Others (Murphy et al., 1998a,c) have attempted to minimise variability by inserting paired soil cores into the ground within centimetres of each other for initial and final destructive soil extraction. Given that “hot spots” of mineralisation, immobilisation, nitrification and denitrification are likely to frequently occur within undisturbed soils (Korsaeth et al., 2001), spatial variability of these processes within cores remains a major source of error. This can only be addressed through improved distribution of 15N label and increased replication of incubations.

V.

CALCULATION OF GROSS N FLUXES

The methods proposed to calculate gross N fluxes in soil differ in several aspects: the measured variables, the modelled system, and the N rates which are determined. Mary et al. (1998) describe the main features of the methods that

GROSS NITROGEN FLUXES IN SOIL

91

calculate N rates in aerobic soils. These calculations can either be analytical or numerical. Different methods of calculating N gross fluxes are discussed in this section.

A. ANALYTICAL SOLUTIONS The formulae of Kirkham and Bartholomew (1954) have been used in their original form (Davidson et al., 1991; Ambus et al., 1992; Smith et al., 1994; Sparling et al., 1995) but with an adjustment for 15N at natural abundance (Blackburn, 1979) or an equivalent expression (Barraclough et al., 1985; Nishio et al., 1985) to calculate gross N fluxes. Derivations used in the different analytical formulations were examined by Smith et al. (1994) and Wang et al. (2001), who concluded that equations presented by Blackburn (1979), Nishio et al. (1985) and Barraclough (1991) were similar and could be reduced to the same formula. Equations have also been described that enable “potential” gross nitrification to be calculated from the increase in 15N abundance in the NO2 3 pool where only 15NHþ 4 is applied (Nishio et al., 1985; Barraclough, 1991). Isotopic dilution calculations based on analytical solutions that follow zeroorder kinetics assume that the rate of processes remain constant during the incubation period. As shown by several authors (e.g., Bjarnason, 1988; Watson et al., 2000), this assumption only causes minor errors in the estimated gross mineralisation rates when short incubation periods (, 7 d) are used and the errors are probably insignificant relative to the experimental error. Alternative equations take into account the possibility that processes are non-linear with respect to time, particularly if the time interval is long, and consequently zeroorder equations are invalid. This is particularly true for nitrification, which is known to highly depend on substrate concentration. Consequently, first-order equations have been used to describe how nitrification depletes the NHþ 4 pool (Barraclough, 1991; Mary et al., 1998). Furthermore, by using a numerical solution (Myrold and Tiedje, 1986; Nason and Myrold, 1991; Wessel and Tietema, 1992; Mary et al., 1998) any kinetic order may be applied, and the assumption becomes irrelevant. The simple model of Kirkham and Bartholomew (1954) does not allow for remineralisation, leading to an underestimation of gross N fluxes where remineralisation is significant during incubation. Remineralisation can be assessed by determining the time for 15NHþ 4 to appear after the application of 15 NO2 or by calculating the incremental decline in gross mineralisation with 3 increasing incubation time (Davidson et al., 1991; Wessel and Tietema, 1992). 2 Given the preference of microbial immobilisation for NHþ 4 , compared to NO3 , þ the typically smaller NH4 pool in soils and the shorter residence time of NHþ 4 compared to NO2 3 , remineralisation is more likely to add errors to calculations of gross mineralisation than gross nitrification.

92

D. V. MURPHY ET AL.

Times when remineralisation could be expected are highlighted in the following examples. Stockdale et al. (1994) found remineralisation to occur within 7 d at 148C. Barraclough (1995) suggested that gross mineralisation could be measured up to 7 d in temperate soils (158C) and 3 –5 d in tropical soils (238C) before remineralisation caused errors. Results from studies in Mediterranean type climates support this view: significant amounts of remineralisation occurred only after 3 –7 d at 208C (D.V. Murphy, unpublished data) while at temperatures of 30 – 408C incubations of 1 –2 d were necessary to minimise the effect of remineralisation on computed gross mineralisation (Murphy et al., 1998a,c). Use of numerical solutions to calculate gross N fluxes, including remineralisation, removes remineralisation as a constraint on the length of incubation (Myrold and Tiedje, 1986; Mary et al., 1998).

B. NUMERICAL SOLUTIONS Numerical solutions have been proposed as an alternative to analytical solutions and enable the multiple N fluxes that simultaneously dilute or enrich the 15 N composition of a given pool, to be calculated (Myrold and Tiedje, 1986; Bjarnason, 1988; Nason and Myrold, 1991; Wessel and Tietema, 1992; Zagal et al., 1993; Smith et al., 1994; Mary et al., 1998). Numerical solutions are 2 particularly useful in soils where immobilisation of NHþ 4 and NO3 occur simultaneously or when remineralisation must be assessed. The various models are rather similar in their concept, as they combine a numerical integration of the partial differential mass balance equations describing the changes in N and 15N between pools over time, and a non-linear fitting program that searches for rate parameters giving the best approximation of modelled pools to measured values. For example the FLUAZ model developed 2 by Mary et al. (1998) consists of four pools (organic-N, NHþ 4 , NO3 and microbial 2 biomass N). FLUAZ enables mineralisation, immobilisation of NHþ 4 and NO3 , nitrification, humification, volatilisation, denitrification and remineralisation to be calculated. Rates of mineralisation and immobilisation are assumed constant (zero order) during each measurement interval. In contrast, nitrification, volatilisation and denitrification rates are allowed to follow first order kinetics 2 relative to either NHþ 4 concentration for the first two or to NO3 concentration for the last process. The solution is fitted to an experiment by a non-linear curve fitting computer program with the least sum of squares as criterion. The interest in using numerical procedures is that: (1) It does not make any approximation in the calculations. Indeed there are no 2 exact analytical solutions of the system if immobilisation of NHþ 4 and NO3 occur simultaneously, or if remineralisation occurs, or if reactions are not zero order. The numerical solution may account for any case without appreciable error as discussed by Nason and Myrold (1991).

GROSS NITROGEN FLUXES IN SOIL

93

(2) It enables testing the consistency of the data set relative to the modelled system since it combines mineral and organic-N and 15N measurements. This is particularly interesting in experiments with paired treatments that produce an over-determined system (e.g., eight measured variables compared to four or five calculated fluxes). (3) Numerical programmes calculate all rates simultaneously. Analytical methods that calculate rates successively, starting with gross mineralisation rate, may transmit errors and transfer all the variability onto the last fluxes calculated. 1.

Zero versus First-Order Rates

Several studies have discussed the impact of considering zero- or first-order rates for mineralisation and nitrification during each measurement interval. 15N pool dilution causes a non-linear decline in at.% 15N-enrichment. During shortterm incubation studies it can be shown that using a linear model does not cause any significant errors if mineralisation rates are low (Recous et al., 1995). However, when gross mineralisation rates increase or when the incubation period of the experiment is extended, the rate of 15N pool dilution becomes increasingly non-linear, and it is important to consider non-linear equations. Barraclough (1991) and Davidson et al. (1991) used a formula to calculate a mean 15N at.% excess enrichment of the NHþ 4 pool based on exponential decrease between the beginning and the end of the experiment. This mean excess of NHþ 4 is then used to calculate immobilisation and nitrification rates. Using the FLUAZ model, Recous et al. (1999) and Andersen and Jensen (2001) obtained a better prediction of the measured values of rates (lower mean weighted error, MWE) using first-order rates for nitrification. Recous et al. (1999) found that gross mineralisation rates were smaller when a first-order versus zero-order rate was assumed for nitrification, and the differences were marked particularly when NHþ 4 consumption was high. They showed that the calculation of gross mineralisation was more sensitive to the hypothesis on the kinetics of the nitrification reaction than to whether the analytical or the numerical procedures were used to calculate the gross N fluxes. When rates of NHþ 4 depletion were high it was important to assume a first-order rate for nitrification. Zero-order kinetics for nitrification resulted in a 65 –85% overestimate of gross mineralisation in such situations (Recous et al., 1999). Estimates of gross immobilisation were not affected by the assumptions concerning nitrification.

2.

Error Evaluation

A major source of error in calculating gross N fluxes can be attributed to the variability between the replicate soil samples, leading to large standard errors for

94

D. V. MURPHY ET AL.

the mean rates of gross N fluxes derived from erratic values for some N pools over time. Errors in calculating gross N fluxes can also be attributed to the changes in the rate of processes over time while they are assumed to be constant over the calculation period. Davidson et al. (1991) and Wessel and Tietema (1992) showed how a small change in a measured variable (e.g., initial or final value for NHþ 4 pool) might translate into a large change (as large as 10 times) in the calculated result, through multiplication factors. By examining the error multiplication factor for each variable, they showed that calculating the rates successively, starting with gross mineralisation, results in transmitting errors and transferring all the variability on the last fluxes calculated, i.e., consumption rates (mainly immobilisation). Davidson et al. (1991) analysed how error in initial pool sizes translate into error in mineralisation and consumption rates as a function of 15N pool decline. A 10% error in pool size caused a 5 –10% error in rate estimates when pool decline was high (75% decline) while the same 10% error caused a 100% error on rate estimates when the pool decline was low (25%). Multiplication factors depend strongly on the following variables: experiment time, initial and final size of N pool, 15N abundance of the N pools indicating that the setting of experimental design (duration of experiment, level of N addition, initial 15N abundance) is crucial in minimising the potential errors for calculation rates. Consequently, getting accurate values for size and enrichment of N pools is also essential before any calculation is run. The use of a numerical model solves some of the problems associated with errors, as the model calculates all rates simultaneously thus avoiding transfer of all of the variability onto the last flux calculated, as it is the case with analytical equations (Wessel and Tietema, 1992). FLUAZ (Mary et al., 1998) minimises the quadratic weighted error (QWE) instead of the usual sum of squares, which has two main advantages (Huet et al., 1992). First, it accounts for the variance of the measurements (those with the greatest variability have the lowest weight) and secondly it “normalises” the various variables, which can then be summed up. If different time intervals are available, one can calculate the global MWE, which takes the data variability into account. This avoids putting too much weight on variables having a high CV, such as the NHþ 4 pool size and favours the variables that are more precisely determined. Nason and Myrold (1991) emphasised the importance of weighting data and the differences between each observed and measured variable can be further tested using complementary statistical tests (e.g., Whitmore, 1991; Smith et al., 1997).

C. CALCULATING IMMOBILISATION Gross immobilisation can either be derived by difference between gross and net N mineralisation, by partitioning NHþ consumption or NO2 4 3 consumption rates into their component processes or from measurement of

GROSS NITROGEN FLUXES IN SOIL

95

the appearance of the added 15N in the total SOM or microbial biomass pools. Deriving NHþ 4 immobilisation by subtracting gross nitrification from NHþ 4 consumption (Davidson et al., 1991) necessitates conducting separate 15 NO2 soil incubations where 15NHþ 4 or 3 are applied to separate soil samples. It is also necessary to assume that immobilisation and nitrification are the only consumptive processes for NHþ 4 . A problem occurs with this approach if stimulates immobilisation (Barraclough and Smith, the application of 15NHþ 4 1987; Davidson et al., 1991; Norton and Firestone, 1996). For example, Andersen and Jensen (2001) compared net N mineralisation/immobilisation dynamics derived either from changes in mineral N or from the difference between gross N fluxes, and the results clearly indicate a larger apparent net N immobilisation when (15NH4)2SO4 labelling was used. Other examples of difficulties that can be encountered when using 15N pool dilution data to calculate gross immobilisation are shown in Fig. 4. Under continuous arable or grass-ley rotation NHþ 4 consumption was stimulated by the addition of substrate, leading to an overestimated gross immobilisation rate (Fig. 4). In contrast, gross immobilisation was underestimated when determined by the difference between NHþ 4 consumption and gross nitrification in reseeded grassland soils. Good agreement was only obtained where the difference between NHþ 4 consumption and gross nitrification approach was used to calculate gross immobilisation in grassland soils (Fig. 4). Overall these examples indicate that calculating immobilisation as the difference between

Figure 4 Comparison of two approaches used to estimate gross immobilisation. Description of soils used reported in Murphy et al. (1999). Nitrate consumption was zero in all soils (D.V. Murphy, unpublished data).

96

D. V. MURPHY ET AL.

NHþ 4 consumption and gross nitrification is likely to be a poor approximation of actual microbial NHþ 4 immobilisation. Measuring gross immobilisation by recovery of 15N in SOM may be done using either the microbial biomass (Ledgard et al., 1998; Hatch et al., 2000b) or the total soil organic-N pool (Mary et al., 1998; Recous et al., 1999; Andersen and Jensen, 2001). Hatch et al. (2000a) determined NHþ 4 immobilisation using both 15N tracing into the microbial biomass and by the difference in using NHþ 4 consumption minus gross nitrification; agreement between methods was close in four out of six treatments. It should be noted that one major advantage of measuring the 15N enrichment of the total soil organic-N pool is that this data is required to calculate gross N fluxes using numerical solutions such as FLUAZ (Mary et al., 1998; Recous et al., 1999). In this case, it is assumed that the increase in organic 15N-enrichment during the incubation period represents microbial 15N, which is probably an acceptable assumption for at least over short periods of time. One difficulty with measuring 15N assimilation directly into the microbial biomass is the problem of incomplete recovery of microbial-N by the chloroform fumigation– extraction method (Brookes et al., 1985). It is thus necessary to scale up microbial-N values by an extraction factor, kEN; otherwise gross N immobilisation rates will be underestimated (e.g., Ledgard et al., 1998). A kEN factor is typically in the order of 0.5 but is known to be somewhat variable (Joergensen and Mueller, 1996) making it difficult to determine accurate rates.

VI. APPLICATIONS OF 15N POOL DILUTION The intention of this section is not to review every application of gross N fluxes. Instead, examples have been chosen to illustrate how studies utilising 15N pool dilution have improved our fundamental understanding of the individual microbial pathways within the soil N cycle, at scales from soil micro-sites to global issues. Examples of how modelling of gross N fluxes has improved the understanding of the soil N cycle, including links between the C and N cycles, are also presented.

A. FACTORS THAT REGULATE GROSS N FLUXES The effect of season on gross N fluxes has been documented for a range of soils and production systems (Davidson et al., 1992; Coyne et al., 1998; Jamieson et al., 1998, 1999; Murphy et al., 1998a; Puri and Ashman, 1998; Recous et al., 1999; Stottlemeyer and Toczydlowski, 1999; Bonde et al., 2001). These studies confirm strong seasonality in gross N fluxes, but it is difficult to assign the effects to individual factors because of the concomitant changes in soil

GROSS NITROGEN FLUXES IN SOIL

97

temperature, soil water content and organic matter (residue) quality over time. Examples are provided where the response of gross N fluxes to individual factors has been studied.

1.

Temperature

Recous et al. (1999) used correction factors for daily changes in temperature and soil water content to express field measurements of gross N fluxes per “normalised” day. In doing so they were able to illustrate that in non-amended soil normalised gross N fluxes remained constant over the year, while in the straw-amended soil normalised gross N fluxes varied according to C dynamics in the residue, as would be expected. This finding illustrates that appropriate functions can correct for changes in temperature and/or soil-water content to reveal the true effect of other factors such as C decomposition on field-measured gross mineralisation. It is therefore surprising that only a few studies have quantified temperature relationships for gross N fluxes (e.g., Stark and Firestone, 1996; Stottlemeyer and Toczydlowski, 1999; Andersen and Jensen, 2001; Cookson et al., 2002). Stottlemeyer and Toczydlowski (1999) found gross mineralisation to respond more strongly to temperature over the range 5 –208C than NHþ 4 immobilisation in forest systems. Gross nitrification and NO2 3 consumption did not respond to temperature over the same temperature range. The effect of temperature on gross mineralisation, immobilisation and nitrification in four arable soils of contrasting soil texture also showed immobilisation to be less responsive to temperature than mineralisation (S. Recous and I. R. P. Fillery, unpublished data). In contrast, to the forest system, nitrification in the arable soils increased sharply between 5 and 308C, and this process exhibited a rapid decline above 308C. Similarly, Stark and Firestone (1996) found temperature optima between 32 and 368C for ammonia oxidiser populations under oak woodland and grassland. However, Binkley et al. (1994) found that temperature had a greater effect on gross immobilisation than gross mineralisation when arctic ecosystem soils were raised from 5 to 128C. A single function or algorithm in models obviously cannot describe such complexity in response of N processes to temperature change. The response of gross N fluxes to temperature can be complicated by C substrate availability. Jamieson et al. (1998, 1999) found that increasing soil temperatures by 38C (to simulate global warming) in natural grassland did not increase gross mineralisation during winter, but decreased gross mineralisation in spring. They hypothesised that the combined effect of changes in C and N inputs derived from plant residues, and the lower soil water contents as a result of heating, masked the effect of temperature. Temperature relationships for gross N fluxes, presented by Cookson et al. (2002), highlight the difficulty in interpreting

98

D. V. MURPHY ET AL.

temperature responses with declining substrate availability. Positive responses of gross N fluxes are apparent over the soil temperature range from 2 to 158C within 21 d of clover residue application when substrate availability was unlikely to be limiting (Fig. 3). In contrast, cumulative gross mineralisation and immobilisation was less at 10 and 158C compared to 2 and 58C after 56 d when substrate availability was limited (Fig. 3). Andersen and Jensen (2001) also studied temperature effects on gross N fluxes after incorporation of a ryegrass cover crop residue. It is clear from this work that Q10 values were not independent of residue addition or the time after residue application when measurements were made. Differences in Q10 were more marked in the 3 –98C range than in the 9– 158C range which implies that residue quality had a greater effect on temperature response at the lower compared to the higher temperature regime studied. Nicolardot et al. (1994) and MacDonald et al. (1995) likewise found C and N dynamics in soil to be affected by both substrate and substrate –temperature interactions.

2. Water Potential The sensitivity of the 15N pool dilution technique, and its short incubation period, makes this an ideal method to explore the effects of rapid wetting-drying cycles on the soil N cycle (e.g., Davidson et al., 1993; Murphy et al., 1998c; Bonde et al., 2001). Soil water was considered to be the main constraint in a study of seasonal trends on gross mineralisation in agricultural soil (Murphy et al., 1998a) and also in natural grasslands (Jamieson et al., 1999). The work of Pilbeam et al. (1993) and Pilbeam and Warren (1995) showed that the net size of the mineral N pool was unrelated to the soil matric potential. In contrast, effects on gross mineralisation of wetting dry soil were dependent upon the initial soil matric potential. This finding is in accordance with long-standing conclusions of Birch (1958) that the magnitude of the flush of mineralisation is dependent of the dryness of the soil.

3. Osmotic Potential The use of low quality irrigation water in agriculture can lead to high salt concentrations in surface soil. The effect of osmotic potential (Cs) on net N mineralisation and nitrification has been widely studied (see Low et al., 1997). Low et al. (1997) report that gross nitrification was dependent of Cs with þ added NHþ 4 but was independent of Cs without added NH4 , indicating that the þ effects of Cs on nitrification were secondary to NH4 concentration. Gross nitrification rates declined more when a single salt was used than when a mixture

GROSS NITROGEN FLUXES IN SOIL

99

of salts (that mimicked solute composition in soil solutions) was used. Another unexpected finding was that microbial assimilation of NO2 3 decreased more quickly than gross nitrification with declining Cs causing an increase in net nitrification with higher salt concentration. Gross mineralisation declined exponentially with Cs between 0 and 2 0.5 MPa but gross N mineralisation was independent of Cs between 2 0.5 and 2 1.75 MPa (Low et al., 1997).

B. QUANTIFYING THE ACTUAL MAGNITUDE OF N CYCLING 1.

MIT

Prior to the routine application of 15N pool dilution it was often thought that the difference between gross versus net N turnover was relatively small, especially in soils receiving little or no organic matter inputs. Use of 15N pool dilution has dispelled this view as shown in the following examples. Tlustos et al. (1998) compared gross and net N fluxes in soil taken from long-term arable and grassland systems (the Broadbalk and Park Grass experiments at the Rothamsted Experimental Station) that had only received atmospheric N deposition (15 kg N ha21year21; Goulding et al., 1998) during the last . 150 years. Although net N mineralisation rates were similar between systems, the permanent grassland soil exhibited MIT turnover rates 30 times higher than those in the arable soil (Fig. 5). In another study, Watkins and Barraclough (1996) found that incorporation of oilseed rape stems or wheat straw with a wide C/N ratio was immediately followed by substantial increases in gross mineralisation that were matched by increases in gross immobilisation which accompanied residue decomposition, resulting in minimal net mineralisation. This response, of large MIT fluxes with minimal net N mineralisation, was also found during the first 30 d after cattle slurry incorporation to the soil (Trehan, 2000). Incorporation of wheat straw into arable soil can cause even greater differences in MIT. Recous et al. (1999) showed rates of gross mineralisation to increase by 50% over the non-amended soil, whereas gross immobilisation increased more than 20 times in the first month of incorporation of wheat straw in an year long study. This field data has recently been used to validate a new one-dimensional mechanistic model (PASTIS; Garnier et al., 2001) that simulates C and N transformations, water and heat flow and solute transport using a daily time step (Garnier et al., 2003). The close agreement between simulated and observed N fluxes enabled the use of PASTIS to estimate the annual cumulative gross mineralisation and gross immobilisation for soil systems with and without straw application. Without straw addition, gross mineralisation amounted to 506 kg N ha21year21 with gross immobilisation at 318 kg N ha21year21, a net release of 188 kg N ha21 year21. In contrast, in the straw-amended soil gross mineralisation was 743 kg ha21 year21 and gross

100

D. V. MURPHY ET AL.

Figure 5 Comparison between gross N fluxes (mg N kg21 d21) in long-term (a) arable and (b) grassland soils. Figure derived from data reported in Tlustos et al. (1998).

immobilisation 580 kg ha 21 year21, a net release of 163 kg ha21 year21 (Garnier et al., 2003).

2.

Organic N Assimilation

The direct assimilation of low molecular weight organic-N substrates (e.g., amino acids) into microbial cells, the so-called direct route, has long been recognised as a N cycling pathway (Jansson, 1958). Barraclough (1997), using 15 N pool dilution, was the first to quantify the relative importance of MIT versus the direct route in the decomposition of low molecular weight amino acids. For example, only 44% of leucine-N and 82% of glycine-N was released as NHþ 4 (i.e., gross mineralisation) while 100% of both compounds was decomposed

GROSS NITROGEN FLUXES IN SOIL

101

(Barraclough, 1997). O’Dowd et al. (1999) found a similar partitioning between MIT and the direct route for leucine, but their data also indicated, for the first time, a difference in the C/N ratio of the microbial populations that decompose L - compared to D -enantiomers of leucine. Knowledge of the C/N ratio of organic compounds, the C/N ratio of the microbial biomass and the C assimilation efficiency of the microbial population involved in decomposition can theoretically be used to predict the percentage of N within an organic compound that will be released as NHþ 4 (Barraclough, 1997). 15 N pool studies have confirmed that this relationship holds for simple organic molecules (Barraclough, 1997; Gibbs and Barraclough, 1998; O’Dowd et al., 1999). A compilation of this data is shown in Fig. 6. The dominance of bacteria in the decomposer populations is clearly evident, along with an apparent difference in the relative importance of direct assimilation versus MIT for bacteria compared to fungi. 15 N pool dilution data has also been used along with mechanistic C and N models to confirm hypotheses concerning the fate and turnover of small organicN compounds in soil (Hadas et al., 1987; Barak et al., 1990; Molina et al., 1990). Comparisons of the fit between NCSOIL (Molina, 1996) and PASTIS (Garnier et al., 2001) with measured gross N fluxes in soil have shown that MIT

Figure 6 Relationship between C/N ratio of a compound and the proportion of N within the compound that was mineralised through to NHþ 4 (i.e., gross mineralised). The N that was not released as NHþ 4 is assumed to be directly assimilated. Lines are based on the equation of Barraclough (1997) and represent the proportion of N that would be decomposed through to NHþ 4 ions if the decomposer community was either bacterial (solid line; C/N ratio of bacteria ¼ 4 and microbial N efficiency ¼ 50%) or fungal (dashed line; C/N ratio of fungi ¼ 12 and microbial N efficiency ¼ 40%). Data compiled from 1Barraclough (1997), 2Gibbs and Barraclough (1998) and 3O’Dowd et al. (1999).

102

D. V. MURPHY ET AL.

and the direct route both occur concurrently in soil. The structure of PASTIS enables the microbial population to be split into an autochtonous microbial biomass that decomposes humified organic matter and a zymogeneous microbial biomass that decomposes fresh and soluble organic matter. This capacity, when used with 15N pool dilution data reported by Recous et al. (1999) for studies of wheat straw incorporation, indicates that N uptake from the residue by the substrate-specific population (zymogeneous microbial biomass) was by the direct route, whereas MIT was the dominant pathway for N uptake undertaken by the autochthonous soil population (Garnier et al., 2003). These examples of the coupling of 15N pool dilution data with carefully constructed and evaluated models highlight the capacity of the 15N pool dilution technique to critically test and refine hypotheses concerning pathways within the soil N cycle.

C. C

AND

N MINERALISATION

The close linkage of the C and N cycles in soil was highlighted in Section “Microbial N pathways”. Process-based models that link C and N mineralisation most commonly divide SOM into a number of pools and the process of gross N mineralisation is controlled by the decomposition rate of SOM (C mineralisation) and the C/N ratios of materials in the pool (Jenkinson, 1990). Gross immobilisation is determined by the microbial C efficiency (i.e., the proportion of C mineralisation retained by the biomass) and the microbial C/N ratio. This general logic underpins the modelling of MIT in most process-based models of C and N dynamics (McGechan and Wu, 2001). 15 N pool dilution studies have been invaluable for testing the hypotheses regarding the link between C and N cycling. For example, it is now possible to assess the relationship between total C mineralised, defined here as CO2 respired plus C assimilated, and the total N mineralised (i.e., gross mineralisation). Furthermore, if the C/N ratio of the decomposing material is known, it should be possible to confirm whether the amounts of total N mineralised can be determined by the total C mineralised. Table II summarises findings on CO2 respiration and gross N mineralisation for a range of organic amendments. In the majority of cases the predicted C/N ratio of the mineralised residue was similar to the measured C/N ratio indicating that the correct amount of N mineralisation is determined using 15N pool dilution. Given that the link between C and N mineralisation holds for gross fluxes, it should in theory be possible to predict gross N mineralisation from CO2 production with knowledge of the C/N ratio of the decomposing material and the microbial C efficiency. An example of how models that link C and N cycles can be used to predict gross mineralisation is illustrated here using the SUNDIAL model (Smith et al., 1997) to simulate values of soil CO2 production, the C/N ratio of the composite mineralising pool (derived from biomass, humus and crop debris within the model) and soil parameter values

Table II Data from Laboratory Experiments that have Incorporated Organic Amendments to Soil and Subsequently Measured Both CO2 Production and Gross N Mineralisation. Data Reported for the C and N Mineralisation from the Residue have been Adjusted for Rates from the Non-amended soil. The Ratio of Total Carbon Mineralisation (Derived from Measured CO2 Production and Assumed Microbial C Efficiency) to Gross N Mineralisation was Calculated for Comparison against the Actual C/N Ratio of the Residue Material

%C

C/N ratio of C/N ratio of Gross Nmin Total C mineralised mineralised organic CO2 respired material % N amendment (mg C kg21 d21) (mg C kg21 d21)a (mg N kg21 d21)

42.50 3.00 Sandy loamb Brassica napus Secale cereale 35.50 2.20 Sandy loamc Lolium multiflorum 41.60 1.60 Sandy loamd Rubisco

Rubisco þ sucrose Sandy loame Grape marc

–

–

– – 46.00 2.30

14:1 16:1 26:1

2.79 2.11 7.65

4.65 3.52 12.75

0.31 0.19 0.88

15:1 18:1 14:1

3:1

2.79

4.65

1.03

4.5:1

12:1 20:1

6.41 37.14

10.68 61.90

1.24 2.61

9:1 24:1

Source Macdonald (2000) Anderson and Jensen (2001) Gibbs and Barraclough (1998) B. Lalor (unpublished data)

a Total carbon mineralised ¼ CO2 respired/1 2 microbial C efficiency; where the microbial carbon assimilation efficiency was defined as 0.40 (i.e., 40% efficiency; van Veen et al., 1985). b Average C and N mineralisation rates calculated from a 52-d incubation at 108C. c Average C and N mineralisation rates calculated over a 37-d incubation at 158C. d Average C and N mineralisation rates calculated over a 29-d incubation at 108C. e Average C and N mineralisation rates calculated from a 70-d incubation at 158C.

GROSS NITROGEN FLUXES IN SOIL

Soil texture

Organic amendment

103

104

D. V. MURPHY ET AL.

Table III Comparison between Field Measurements and SUNDIAL Estimates (Eq. (3)) of Gross Mineralisation (kg N ha21 d21; 0 –23 cm) on 11 Occasions during the Growth of a Winter Wheat Crop from an Arable Rotation. (D.V. Murphy and E.A. Stockdale, Unpublished Data) Measured rangea

Sampling date 03/10/96 15/10/96 04/11/96 02/12/96 01/01/97 03/02/97 07/03/97 07/04/97 30/06/97 21/07/97 18/08/97

1.30–4.66 0.88–2.63 1.33–3.22 0.81–1.26 Soil frozen 0.95–2.07 0.77–1.51 1.16–1.75 2.38–3.43 1.89–2.45 0.88–2.98

Modelled rangeb 2.27–3.00 1.31–3.13 2.53–4.20 0.90–1.39 0.02–0.07 0.44–0.79 0.67–1.20 0.44–0.65 1.10–1.29 0.86–1.23 1.12–1.60

a

Mean minimum and maximum gross mineralisation rates measured from replicate arable field plots using a 2–3 d 15N pool dilution incubation period. b Modelled range of gross mineralisation rates over the same 2 –3 d incubation period used for 15N pool dilution.

controlling microbial efficiency (a and b ) (Eq. (3)). 

 Soil CO2 production 1 2 ða þ bÞ Gross mineralisation ¼ C=N ratio of mineralisation pool

ð3Þ

Use of this logic in a daily time-step version of SUNDIAL produced gross mineralisation rates that were similar to field measurements of gross mineralisation (Table III; D.V. Murphy and E.A. Stockdale, unpublished data). Therefore, in models similar to SUNDIAL that accurately simulate C mineralisation, it should be possible to predict gross N mineralisation from standard soil, crop and weather information typically used to initialise soil-crop models. This potentially overcomes the need for intensive measurements of gross N mineralisation under field conditions.

D. COMPETITION FOR SUBSTRATE AVAILABILITY It is generally assumed that microbial heterotrophs are more competitive for NHþ 4 than autotrophic nitrifiers (Vitousek et al., 1982; Tietema and Wessel, 1992). This assumption implies that sufficient C substrate exists to enable heterotrophic microbial growth (NHþ 4 immobilisation) to dominate over autotrophic growth (NHþ oxidation). The major limitation to autotrophic 4

GROSS NITROGEN FLUXES IN SOIL

105

nitrification is the supply of NHþ 4 (Alexander, 1965) and this process is therefore dependent on MIT in the absence of external N inputs such as fertiliser or atmospheric N deposition. 15N pool dilution measures the activity of specific microbial populations and therefore can be used to assess the competitiveness of each population for substrate. Hart et al. (1994) used the ratio of gross nitrification to gross mineralisation to assess the capacity of autotrophic nitrifiers to compete for NHþ 4 . This work showed that heterotrophs assimilate most of the available NHþ 4 during periods of microbial growth when demand for NHþ 4 was high and C was available. When C availability declined and the heterotrophic population became stable, autotrophic nitrifiers competed strongly for NHþ 4 . Thus C availability within specific soil microsites is likely to regulate the fate of NHþ 4 between immobilisation and nitrification. Another example of the use of gross N fluxes to distinguish activities of different microbial populations, in this case autotrophic from heterotrophic nitrification, is shown in the work of Hart et al. (1997) and Pedersen et al. (1999). Hart et al. (1997) found that 65– 72% of gross nitrification in conifer and coniferred alder stands was heterotrophic when determined in the presence of C2H2 which inhibits autotrophic nitrification. Similarly, Pedersen et al. (1999) used C2H2 to inhibit autotrophic nitrifiers in soil from a mixed conifer forest and an adjacent clear cut area, presumably with limited C supply. Autotrophic nitrification dominated NO2 3 production and accounted for one-third gross mineralisation in the forested sites with available C, while heterotrophic nitrification dominated NO2 3 production, but was equivalent to only 5% of gross mineralisation measured in the clear cut area (Pedersen et al., 1999). Such studies further illustrate the potential of 15N pool dilution measurements to examine the relative dominance of specific groups of microorganisms.

E. SOIL DISTURBANCE Soil disturbance is known to alter net N mineralisation/immobilisation (Silgram and Shepherd, 1999) by changing the balance of MIT. 15N pool dilution studies can be used to determine if this is due to a change in gross mineralisation, gross immobilisation or both. Concurrent measurements of MIT in a coarsetextured soil, following mixing, have shown that gross mineralisation was unaffected by soil disturbance and changes to net N mineralisation were a result of changing gross immobilisation rates (Fig. 7). This work suggests that gross mineralisation will not be affected in the short-term by soil disturbance, since microorganisms decomposing SOM and residue must already be located on or near the decomposing organic material. Therefore, mixing should not increase contact between the decomposer populations and substrate. Microbial immobilisation of N is affected by the spatial heterogeneity of NHþ 4, NO2 and C in soil (Schimel et al., 1989; Davidson et al., 1990; Drury et al., 1991; 3

106

D. V. MURPHY ET AL.

Figure 7 Nitrogen transformation rates (0–10 cm) measured in (a) undisturbed or (b) disturbed soil. Immobilisation was calculated as the difference between gross and net mineralisation. Mean rates are presented in bold as mg N kg21 d21. Values in parentheses are ^95% confidence intervals of the mean. The soil used was coarse textured (92% sand, 7% clay) under an unimproved, low production, annual grass/pasture with total C of 0.69%, total N of 0.05% and bulk density 1.1 g cm23 (D.V. Murphy, unpublished data).

Norton and Firestone, 1996). The availability of soil inorganic N is known to affect gross immobilisation rates, particularly in systems where organic materials of high C/N, coupled with low contents of soil inorganic N retard decomposition (Recous et al., 1995; Mary et al., 1996). Soil disturbance redistributes substrates (both C and N) and can improve physical and chemical conditions for their use by soil decomposers, increasing immobilisation as shown in Fig. 7. In a model simulation study, Korsaeth et al. (2001) showed how spatial segregation of hotspots with net immobilisation and net mineralisation is likely to be the rule rather than the exception in soils with large inputs of manures or

GROSS NITROGEN FLUXES IN SOIL

107

plant residues. This spatial segregation of N processes could have a profound impact on the N dynamics of the soil –plant systems, and should be taken into account in the interpretation of 15N pool dilution experiments, as well as in general models for the C and N dynamics of soil – plant systems, since the phenomenon favours plant root over microbes in the competition for inorganic N.

F. ACTIVITY OF AMMONIA-OXIDISERS IN SOIL One of the main areas of current interest in nutrient cycling is linking processes to differences in soil microbial ecology. Tlustos et al. (1998) examined methane (CH4) oxidation and gross nitrification rates under long-term (. 150 years) unfertilised arable and grassland soils. They interpreted the low gross nitrification rates (see Fig. 5) to indicate an N substrate limitation on the ammonia-oxidisers in the arable soil (where the gross mineralisation rate was slow) and a probable pH limitation on the ammonia-oxidisers in the grassland soil (where soil N supply was not limiting). Given the lack of gross nitrification in the grassland soil they were able to conclude that the role of ammonia-oxidisers in regulating CH4 oxidation at this site was negligible and must therefore be mediated by methanotrophs (Tlustos et al., 1998). By contrast, in the cultivated arable soil it was suggested that CH4 oxidation was due to ammonia-oxidisers. The greater CH4 oxidation rate, by order of magnitude, in the grassland compared to the arable soil was thus attributed to the fact that ammonia-oxidisers oxidise CH4 more slowly than methanotrophs (Jones and Morita, 1983). Mendum et al. (1999) subsequently measured gross nitrification and quantified ammona-oxidiser populations by competitive PCR assays based on amoA and ribosomal 16S genes from both unfertilised and fertilised plots (inorganic N only or farmyard manure plus inorganic N) of the same arable soil as used by Tlustos et al. (1998). Since the amoA gene has only been identified in microorganisms capable of oxidising NH3, and given that the ammonia monooxygenase enzyme is directly involved in NH3 oxidation, Mendum et al. (1999) were able to make a direct link between the population dynamics of the ammonia-oxidising bacteria and gross nitrification rates. They showed that the impact of such long-term inputs of inorganic and/or organic-N fertiliser on ammonia-oxidising populations and gross nitrification rates was surprisingly small compared to nil N plots. In contrast, a single application of NH4NO3 had a significant but short-lived effect on ammonia-oxidiser populations and gross nitrification rates. This suggested that the activity of the ammonia-oxidisers in all plots was N-limited during the majority of the year except immediately after recent N inputs. It was possible for Mendum et al. (1999) to use the short-term incubation period of 15N pool dilution to illustrate that gross nitrification increased rapidly (within 3 d) after application

108

D. V. MURPHY ET AL.

of NH4NO3. Mendum et al. (1999) were therefore able to conclude that this increase in activity was too rapid to be entirely accounted for by growth in the ammonia-oxidiser population, implying phenotypic changes in the ammoniaoxidising bacteria.

G. N SATURATION INDEX Aber (1992) and Tietema and Wessel (1992) suggested the use of the ratio of gross nitrification to gross NHþ 4 immobilisation (N/I) as a means of quantifying competition between heterotrophic microbes and nitrifiers for NHþ 4 and thus showed how much a soil is over-producing N relative to the ability of the microbial community to consume it (i.e., N saturation status). They suggested that when this ratio is . 1, NHþ 4 ions in soil are more likely to be transformed to the more leachable NO2 ion than when retained within MIT. 3 Tietema and Van Dam (1996) and Stark and Hart (1997) were the first to use 15 N pool dilution to test this theory. They found rapid gross nitrification rates in forest soils in North America and highlighted the risk of N saturation and loss from forests that receive anthropogenically enhanced amounts of N. Tietema (1998) found that microbial C and N cycling was characterised by lower gross NHþ 4 transformation rates, lower respiration, higher microbial C efficiencies and high microbial C/N ratios in two N-limited sites compared to three N-saturated sites. A comprehensive assessment of the problem in forested ecosystems across Europe and North America has since been made in the NITREX research programme (see papers by Tietema et al., 1998a,b; Gundersen et al., 1998). The concept of N saturation has recently been applied to agricultural soils (Goulding et al., 1998; Murphy et al., 2001; Stockdale et al., 2002) in an attempt to identify indicators of the potential for N loss, avoiding the need for lengthy and extensive measurements of leaching and gaseous losses. Under arable and grassland soils the N/I ratio determined on soil collected at the time of cereal crop harvest was significantly correlated with subsequent overwinter N losses (Fig. 8), and the ratio in spring with subsequent crop uptake in the absence of fertiliser (E.A. Stockdale, unpublished data). Fig. 8 also shows that the losses for comparable N/I indices in arable soils receiving manure were higher, and similar to those in grassland soils, than those in unmanured arable soils. This may reflect increased SOM contents and gross mineralisation rates. Whilst neither relationship was strong enough to use in the prediction of N losses or calculations of N fertiliser requirement, the data does illustrate the þ strong link between the relative dominance of NHþ 4 immobilisation and NH4 oxidation processes and N loss.

GROSS NITROGEN FLUXES IN SOIL

109

21 Figure 8 Relationship between the index of N saturation (N/I) and modelled NO2 ) 3 -N (kg N ha leaching for agricultural soils: (W) arable; (X) arable þ manure; (B) grassland. Data compiled from Stockdale et al. (2002).

VII. FUTURE RESEARCH 15

N pool dilution will continue to provide important data both for the generation of new hypotheses, which may subsequently be encapsulated within models, and also for the validation of the simulation of the soil internal N cycle within models of the soil – plant N cycle. Here, we highlight a few potential research directions that we see as providing further valuable contributions to our understanding and prediction of the soil N cycle. There have been attempts to increase the degree to which modelled pools relate to measurable fractions of SOM (Christensen, 1996; Sohi et al., 2001). This approach increasingly enables the internal cycling of the model to be validated against actual data. Such model structures also enable C and N fluxes to be modelled separately for each SOM fraction. However, there are few models that assume different temperature and moisture response functions between microbial processes (Ma and Shaffer, 2001; McGechan and Wu, 2001). Since gross N fluxes are not controlled in the same way by environmental factors, models that apply rate-adjusting parameters equally to both processes, cannot represent the complex dynamics actually occurring in the field. As our fundamental understanding of the controlling factors of the individual kinetics of these N transformations improves (e.g., Stark and Firestone, 1996), it is likely that modelling approaches will also develop and enable the hypothesised relationships to be evaluated robustly. This will assist the development of more mechanistically based models. There is also a need to evaluate more hypotheses and model structures which link gross N fluxes with bacterial kinetics (e.g., Smith et al., 1986) and also with

110

D. V. MURPHY ET AL.

the dynamics of bacterial predators (Osler, 2002). Molz et al. (1986) described processes occurring at a scale of 0.3 – 500 mm in saturated porous media and used the microcolony concept to describe the location and occurrence of bacteria in such media. This concept has not been applied to the modelling of gross N fluxes in soil. However, Drury et al. (1991), Norton and Firestone (1996) and Korsaeth et al. (2001) highlight the importance of the spatial-temporal distribution of microorganisms, plant roots and organic residues in soil in regulating the soil N cycle. Data from Mendum (2000) suggests that the spatial distribution of bacterial colonies is critical in regulating nitrification. The explicit inclusion of micro-sites in a model that seeks to make predictions at the soil or field scale would be difficult and probably unrealistic. However, more research is required at this scale to improve our fundamental understanding of the role of spatial variability in regulating the dominance of individual microbial N processes and to elucidate how such interactions influence processes operating at a larger scale. It is at the scale of field, farm and catchment that the application of our understanding of processes controlling the availability of N in soils is needed most critically.

ACKNOWLEDGMENTS The authors wish to thank Drs Andrew Macdonald (Rothamsted Research) and Richard Cookson (The University of Western Australia) for providing access to the specific data values presented within figures reporting their research findings. DVM completed this review whilst employed as an Australian Grains Research and Development Corporation Research Fellow. IACR and IGER receive grantaided support from the Biological and Biotechnology Scientific Research Council and funding from the UK Department for Environment, Food and Rural Affairs.

REFERENCES Aber, J. D. (1992). Nitrogen cycling and nitrogen saturation in temperate forest ecosystems. Tree 7, 220–224. Alexander, M. (1965). Nitrification. Agronomy. In “Soil Nitrogen”. (W. V. Bartholomew and F. E. Clark, Eds.), Vol. 10, pp. 307– 343. American Society of Agronomy, Madison, Wisconsin. Ambus, P., Mosier, A., and Christensen, S. (1992). Nitrogen turnover rates in a riparian fen determined by 15N dilution. Biol. Fert. Soils 14, 230– 236. Andersen, M. K., and Jensen, L. S. (2001). Low soil temperature effects on short-term gross N mineralisation– immobilisation turnover after incorporation for green manure. Soil Biol. Biochem. 33, 511–521. Azam, F., Malik, K. A., and Hussain, F. (1986). Microbial biomass and mineralisation – immobilisation of nitrogen in some agricultural soils. Biol. Fert. Soils 2, 157–163.

GROSS NITROGEN FLUXES IN SOIL

111

Barak, P., Molina, J. A. E., Haddas, A., and Clapp, C. E. (1990). Mineralisation of amino acids and evidence for direct assimilation of organic nitrogen. Soil Sci. Soc. Am. J. 54, 769–774. Barraclough, D. (1991). The use of mean pool abundances to interpret 15N tracer experiments. I. Theory. Plant Soil 131, 89–96. Barraclough, D. (1995). 15N isotope dilution techniques to study soil nitrogen transformations and plant uptake. Fert. Res. 42, 185– 192. Barraclough, D. (1997). The direct or MIT route for nitrogen immobilisation: a 15N mirror image study with leucine and glycine. Soil Biol. Biochem. 29, 101–108. Barraclough, D., and Puri, G. (1995). The use of 15N pool dilution and enrichment to separate the heterotrophic and autotrophic pathways of nitrification. Soil Biol. Biochem. 27, 17– 22. Barraclough, D., and Smith, M. J. (1987). The estimation of mineralisation, immobilisation and nitrification in nitrogen-15 field experiments using computer simulation. J. Soil. Sci. 38, 519 –530. Barraclough, D., Geens, E. L., Davies, G. P., and Maggs, J. M. (1985). Fate of fertiliser nitrogen. III. The use of single and double labelled 15N ammonium nitrate to study nitrogen uptake by ryegrass. J. Soil. Sci. 36, 593–603. Barrett, J. E., and Burke, I. C. (2000). Potential immobilisation in grassland soils across a soil organic matter gradient. Soil Biol. Biochem. 32, 1707–1716. Berntson, G. M., and Bazzaz, F. A. (1997). Nitrogen cycling in microcosms of yellow birch exposed to elevated CO2: simultaneous positive and negative below-ground feedbacks. Global Change Biol. 3, 247–258. Betlach, H. R., Tiedje, J. M., and Firestone, R. B. (1981). Assimilatory nitrate uptake of Pseudomonas fluorescens studied using Nitrogen 13. Arch. Microbiol. 139, 135–140. Bhogal, A., Hatch, D. J., Shepherd, M. A., and Jarvis, S. C. (1999). Comparison of methodologies for field measurement of net nitrogen mineralisation in arable soils. Plant Soil 207, 15 –28. Binkley, D., Bell, R., and Sollins, P. (1992). Comparison of methods for estimating soil nitrogen transformations in adjacent conifer and alder-conifer forests. Can. J. Res. 22, 858–863. Binkley, D., Stottlemyer, R., Saurez, F., and Cortina, J. (1994). Soil nitrogen availability in some arctic ecosystems in northwest Alaska: responses to temperature and moisture. Ecoscience 1, 64–70. Birch, H. F. (1958). The effect of soil drying on humus decomposition and nitrogen availability. Plant Soil 10, 9–31. Bjarnason, S. (1988). Calculation of gross nitrogen immobilisation and mineralisation in soil. J. Soil Sci. 39, 393 –406. Blackburn, T. H. (1979). Method for measuring rates of NHþ 4 turnover in anoxic marine sediments, using a 15N-NHþ 4 dilution technique. Appl. Environ. Microbiol. 37, 760 –765. Bonde, T. A., Nielen, T. H., Miller, M., and Sørensen, J. (2001). Arginine ammonification assay as a rapid index of gross N mineralisation in agricultural soils. Biol. Fert. Soils 34, 179– 184. Bosatta, E., and Agren, G. I. (1995). Theoretical analyses of interactions between inorganic nitrogen and soil organic matter. Eur. J. Soil Sci. 46, 109 –114. Brookes, P. C., Landman, A., Pruden, G., and Jenkinson, D. S. (1985). Chloroform fumigation and the release of soil nitrogen: a rapid direct extraction method to measure microbial biomass nitrogen in soil. Soil Biol. Biochem. 17, 837–842. Brown, C. M. (1980). Ammonia assimilation and utilization in bacteria and fungi. In “Microorganisms and Nitrogen Sources”. (J. W. Payne, Ed.), pp. 437– 511. Wiley, New York. Campbell, J. L., and Gower, S. T. (2000). Detritus production and soil N transformations in oldgrowth eastern hemlock and sugar maple stands. Ecosystems 3, 185 –192. Chalk, P. M., Victoria, R. L., Muraoka, T., and Piccolo, M. C. (1990). Effect of a nitrification inhibitor on immobilisation and mineralisation of soil and fertiliser nitrogen. Soil Biol. Biochem. 22, 533–538. Chaussod, R., Houot, S., Guiraud, G., and He´tier, J. M. (1988). Size and turnover of the microbial biomass in agricultural soils: laboratory and field measurements. In “Nitrogen in Agricultural Soils”. (D. S. Jenkison and K. A. Smith, Eds.), pp. 312–326. Elsevier, London.

112

D. V. MURPHY ET AL.

Cheng, H. H., Bremner, J. M., and Edwards, A. P. (1964). Variations on nitrogen-15 abundances in soil. Science 146, 1574–1575. Christensen, B. T. (1996). Matching measurable soil organic matter fractions with conceptual pools in simulation models of carbon turnover: revision of model structure. Proceedings of the NATO Advanced Research Workshop May 1995. NATO-ASI Series I Global Environmental Change. In “Evaluation of soil organic matter models. Using existing long-term datasets”. (D. S. Powlson, P. Smith and J. U. Smith, Eds.), Vol. 38, pp. 143–159. Springer, Berlin. Cookson, W. R., Cornforth, I. S., and Rowarth, J. S. (2002). Winter soil temperature (2–158C) effects on nitrogen transformations in clover green manure and unamended soils: a laboratory and field study. Soil Biol. Biochem. in press 34, 1401–1415. Coyne, M. S., Zhai, Q., Mackown, C. T., and Barnhisel, R. I. (1998). Gross nitrogen transformation rates in soil at a surface coal mine site reclaimed for prime farmland use. Soil Biol. Biochem. 30, 1099–1106. Crawford, D. M., and Chalk, P. M. (1992). Mineralisation and immobilisation of soil and fertiliser nitrogen with nitrification inhibitors and solvents. Soil Biol. Biochem. 24, 559 –568. Dalton, H. (1979). Utilisation of inorganic nitrogen by microbial cells. (J. R. Quayle, Ed.), Microbiol. Biochem. 21, 226–266. Davidson, E. A., Stark, J. M., and Firestone, M. K. (1990). Microbial production and consumption of nitrate in an annual grassland. Ecology 71, 1968–1975. Davidson, E. A., Hart, S. C., Shanks, C. A., and Firestone, M. K. (1991). Measuring gross nitrogen mineralisation, immobilisation, and nitrification by 15N isotopic pool dilution in intact soil cores. J. Soil Sci. 42, 335–349. Davidson, E. A., Hart, S. C., and Firestone, M. K. (1992). Internal cycling of nitrate in soils of a mature coniferous forest. Ecology 73, 1148–1156. Davidson, E. A., Matson, P. A., Vitousek, P. M., Riley, R., Dunkin, K., Garcia-Mendez, G., and Maass, J. M. (1993). Processes regulating soil emissions of NO and N2O in a seasonally dry tropical forest. Ecology 74, 130–139. Di, H. J., Cameron, K. C., and McLaren, R. G. (2000). Isotopic dilution methods to determine the gross transformation rates of nitrogen, phosphorus, and sulfur in soil: a review of theory, methodologies, and limitations. Aust. J. Soil Res. 38, 213– 230. Drury, C. F., Voroney, R. P., and Beauchamp, E. G. (1991). Availability of NHþ 4 -N to microorganisms and the soil internal N cycle. Soil Biol. Biochem. 23, 165–169. Garcia-Montiel, D. C., and Binkley, D. (1998). Effect of Eucalyptus saligna and Albizia falcataria on soil processes and nitrogen supply in Hawaii. Oecologia 113, 547–556. Garnier, P., Ne´el, C., Mary, B., and Lafolie, F. (2001). Evaluation of a nitrogen transport and transformation model in bare soil. Eur. J. Soil Sci. 52, 253– 268. Garnier, P., Ne´el, C., Aita, C., Recous, S., Lafolie, F., and Mary, B (2003). Modelling carbon and nitrogen dynamics in a bare soil with and without straw incorporation. Eur. J. Soil Sci. 54, 00 –00. Geens, E. L., Davies, G. P., Maggs, J. M., and Barraclough, D. (1991). The use of mean pool abundances to interpret 15N tracer experiments. II. Application. Plant Soil 131, 97–105. Gibbs, P., and Barraclough, D. (1998). Gross mineralisation of nitrogen during the decomposition of leaf protein I (ribulose 1,5-diphosphate carboxylase) in the presence or absence of sucrose. Soil Biol. Biochem. 30, 1821–1827. Goulding, K. W. T., Bailey, N. J., Bradbury, N. J., Hargreaves, P., Howe, M., Murphy, D. V., Poulton, P. R., and Willison, T. W. (1998). Nitrogen deposition and its contribution to nitrogen cycling and associated soil processes. New Phytol. 139, 49–58. Gundersen, P., Emmett, B. A., Kjonaas, O. J., Koopmans, C. J., and Tietema, A. (1998). Impact of nitrogen deposition on nitrogen cycling in forests: a synthesis of NIREX data. Forest Ecol. Manage. 101, 37–55.

GROSS NITROGEN FLUXES IN SOIL

113

Hadas, A., Molina, J. A. E., Feigenbaum, S., and Clapp, C. E. (1987). Simulation of 15N immobilisation by the model NCSOIL. Soil Sci. Soc. Am. J. 51, 102– 106. Hadas, A., Sofer, M., Molina, J. A. E., Barak, P., and Clapp, C. E. (1992). Assimilation of nitrogen by soil microbial population: NH4 versus organic-N. Soil Biol. Biochem. 24, 137 –143. Harper, J. E. (1984). Uptake of organic nitrogen forms by roots and leaves. In “Nitrogen in Crop Production”. (R. D. Hauck, Ed.), pp. 165 –170. ASA/CSSA/SSSA, Madison, WI. Hart, S. C., Nason, G. E., Myrold, D. D., and Perry, D. A. (1994). Dynamics of gross nitrogen transformations in an old-growth forest: the carbon connection. Ecology 75, 880 –891. Hart, S. C., Binkley, D., and Perry, D. A. (1997). Influence of red alder on soil nitrogen transformations in two conifer forests of contrasting productivity. Soil Biol. Biochem. 29, 1111–1123. Hatch, D. J., Jarvis, S. C., Parkinson, R. J., and Lovell, R. D. (2000). Combining field incubation with 15 N labelling to examine N transformations in low to high intensity grassland management systems. Biol. Fert. Soils 30, 492– 499. Hatch, D. J., Lovell, R. D., Antil, R. S., Jarvis, S. C., and Owen, P. M. (2000). Nitrogen mineralisation and microbial activity in permanent pastures amended with nitrogen fertiliser or dung. Biol. Fert. Soils 30, 288–293. Heaton, T. H. E. (1986). Isotopic studies of nitrogen pollution in the hydrosphere and the atmosphere: a review. Chem. Geol. 59, 87– 102. Hood, R. (2001). Evaluation of a new approach to the nitrogen-15 isotope dilution technique, to estimate crop N uptake from organic residues in the field. Biol. Fert. Soils 34, 1456–1461. Hood, R. C., and Wood, M. (1996). Estimating gross mineralisation of Alnus glutinosa residues, using 15 N mirror image experimentation. Proceedings of the 8th Nitrogen Workshop, University of Ghent, 5 –6 September, 1994. In “Progress in Nitrogen Cycling Studies”, pp. 53 –56. Kluwer Academic Publishers, Dordrecht, The Netherlands. Hood, R. C., Goran, K. N., Aigner, M., and Hardarson, G. (1999). A comparison of direct and indirect 15 N isotope techniques for estimating crop uptake from organic residues. Plant Soil 208, 259 –270. Hood, R. C., Merckx, R., Jensen, E. S., Powlson, D., Matijevic, M., and Hardarson, G. (2000). Estimating crop N uptake from organic residues using a new approach to the 15N isotope dilution technique. Plant Soil 223, 33– 46. Huet, S., Jolivet, E., and Messe´an, A. (1992). In “La re´gression Nonline´aire: Me´thods et Applications en Biologie”, pp. 256. INRA Editions, Paris. Hungate, B. A., Lund, C. P., Pearson, H. L., and Chapin, F. S., III (1997). Elevated CO2 and nutrient addition alter soil N cycling and N trace gas fluxes with early season wet-up in a California annual grassland. Biogeochemistry 37, 89–109. Ingwersen, J., Butterbachh-Bahl, K., Gasche, R., Richter, O., and Papen, H. (1999). Barometric process separation: new method for quantifying nitrification, denitrification, and nitrous oxide sources in soils. Soil Sci. Soc. Am. J 63, 117 –128. Jamieson, N., Barraclough, D., Unkovich, M., and Monaghan, R. (1998). Soil N dynamics in a natural calcareous grassland under a changing climate. Biol. Fert. Soils 27, 267–273. Jamieson, N., Monaghan, R., and Barraclough, D. (1999). Seasonal trends of gross N mineralisation in a natural calcareous grassland. Global Change Biol. 5, 423– 431. Jansson, S. L. (1958). Tracer studies on nitrogen transformations in soil with special attention to mineralisation– immobilisation relationships. Kungul. Lanbrukshogskolans Ann. 24, 101 –361. Jansson, S. L., and Persson, J. (1982). Mineralisation and immobilisation of soil nitrogen. Agronomy Monographs No. 22. In “Nitrogen in Agricultural Soils”. (F. J. Stevenson, J. M. Bremner, R. D. Hauck and D. R. Keeney, Eds.), pp. 229– 252. ASA/CSSA/SSSA, Madison, WI. Jarvis, S. C., Stockdale, E. A., Shepherd, M. A., and Powlson, D. S. (1996). Nitrogen mineralisation in temperate agricultural soils: processes and measurement. Adv. Agron. 57, 187 –235. Jenkinson, D. S. (1990). The turnover of organic carbon and nitrogen in soil. Philos. Trans. R. Soc. London B 329, 361 –368.

114

D. V. MURPHY ET AL.

Joergensen, R. G., and Mueller, T. (1996). The fumigation extraction method to estimate soil microbial biomass: calibration of the kEN value. Soil Biol. Biochem. 28, 33–37. Jones, R. D., and Morita, R. Y. (1983). Methane oxidation by Nitrosococcus oceanus and Nitrosomonas europaea. Appl. Environ. Microbiol. 45, 401–410. Jung, K. D., and Luttge, U. (1980). Effects of [the fungal toxin] fusicoccin and abscisic acid on sugar and ion transport from plant glands. Ann. Bot. 45, 339– 349. Kirkham, D., and Bartholomew, W. V. (1954). Equations for following nutrient transformations in soil, utilizing tracer data. Soil Sci. Soc. Am. Proc. 18, 33–34. Kirkham, D., and Bartholomew, W. V. (1955). Equations for following nutrient transformations in soil, utilizing tracer data: II. Soil Sci. Soc. Am. Proc. 19, 189–192. Korsaeth, A., Molstad, L., and Bakken, L. R. (2001). Modelling the competition for nitrogen between plants and microflora as a function of soil heterogeneity. Soil Biol. Biochem. 33, 215–226. Kumar, K., and Goh, K. M. (2000). Crop residues and management practices: effects on soil quality, soil nitrogen dynamics, crop yield, and nitrogen recovery. Adv. Agron. 68, 197–319. Ledgard, S. F., Jarvis, S. C., and Hatch, D. J. (1998). Short-term nitrogen fluxes in grassland soils under different long-term nitrogen management regimes. Soil Biol. Biochem. 30, 1233–1241. Low, A. P., Stark, J. M., and Dudley, L. M. (1997). Effects of soil osmotic potential on nitrification, ammonification, N-assimilation, and nitrous oxide production. Soil Sci. 162, 16–27. Ma, L., and Shaffer, M. J. (2001). A review of carbon and nitrogen processes in nine US soil nitrogen dynamics models. In “Modelling Carbon and Nitrogen Dynamics for Soil Management”. (M. J. Ma, L. Ma and S. Hansen, Eds.), pp. 55–102. Lewis Publishers, Boca Raton. Macdonald, A. J. (2000). The effects of cover crop on soil N transformations and losses from arable land. PhD Thesis, University of Reading, UK. MacDonald, N. W., Donald, R. Z., and Kurt, S. P. (1995). Temperature effects on kinetics of microbial respiration and net nitrogen and sulfur mineralisation. Soil Sci. Soc. Am. J. 59, 233 –240. Martens, D. A. (2000). Nitrogen cycling under different soil management systems. Adv. Agron. 70, 143–192. Mary, B., and Recous, S. (1994). Measurement of nitrogen mineralisation and immobilisation fluxes in soil as a means of predicting net mineralisation. Eur. J. Agron. 3, 291 –300. Mary, B., Recous, S., Darwis, D., and Robin, D. (1996). Interactions between decomposition of plant residues and nitrogen cycling in soil. Plant Soil 181, 71–82. Mary, B., Recous, S., and Robin, D. (1998). A model for calculating nitrogen fluxes in soil using 15N tracing. Soil Biol. Biochem. 30, 1963–1979. McGechan, M. B., and Wu, L. (2001). A review of carbon and nitrogen processes in European soil nitrogen dynamics models. In “Modelling Carbon and Nitrogen Dynamics for Soil Management”. (M. J. Shaffer, L. Ma and S. Hansen, Eds.), pp. 103 –171. Lewis Publishers, Boca Raton. Mendum, T. A. (2000). The molecular ecology of autotrophic ammonia oxidizing bacteria in agricultural soils. PhD Thesis. University of Nottingham, UK. Mendum, T. A., Sockett, R. E., and Hirsch, P. R. (1999). Use of molecular and isotope techniques to monitor the response of autotrophic ammonia-oxidizing populations of the beta subdivision of the class Proteobacteria in arable soils to nitrogen fertiliser. Appl. Environ. Microbiol. 65, 4155–4162. Mikan, C. J., Zak, D. R., Kubiske, M. E., and Pregitzer, K. S. (2000). Combined effects of atmospheric CO2 and N availability on the below ground carbon and nitrogen dynamics of aspen mesocosms. Oecologia 124, 432– 445. Molina, J. A. E. (1996). Description of the model NCSOIL. Proceedings of the NATO Advanced Research Workshop, May 1995. NATO-ASI Series I Global Environmental Change. In “Evaluation of soil organic matter models: using existing long-term datasets”. (D. S. Powlson, P. Smith and J. U. Smith, Eds.), Vol. 38, pp. 269 –274. Springer, Berlin. Molina, J. A. E., Hadas, A., and Clapp, C. E. (1990). Computer simulation of nitrogen turnover in soil and priming effect. Soil Biol. Biochem. 22, 349–353.

GROSS NITROGEN FLUXES IN SOIL

115

Molz, F. J., Widdowson, M. A., and Benefield, L. D. (1986). Simulation of microbial growth dynamics coupled to nutrient and oxygen transport in porous media. Water Resour. Res. 22, 1207–1216. Monaghan, R. (1995). Errors in estimates of gross rates of nitrogen mineralisation due to non-uniform distributions of 15N label. Soil Biol. Biochem. 27, 855– 859. Monaghan, R., and Barraclough, D. (1995). Contributions to gross N mineralisation from 15N-labelled soil macroorganic matter fractions during laboratory incubation. Soil Biol. Biochem. 27, 1623–1628. Monaghan, R., and Barraclough, D. (1997). Contributions to N mineralisation from soil macroorganic matter fractions incorporated into two field soils. Soil Biol. Biochem. 29, 1215–1223. Murphy, D. V., Fillery, I. R. P., and Sparling, G. P. (1997). Method to label soil cores with 15NH3 gas as a prerequisite for 15N isotopic dilution and measurement of gross N mineralisation. Soil Biol. Biochem. 29, 1731–1741. Murphy, D. V., Fillery, I. R. P., and Sparling, G. P. (1998). Seasonal fluctuations in gross N mineralisation, ammonium consumption, and microbial biomass in a Western Australian soil under different land uses. Aust. J. Agric. Res. 49, 523–535. Murphy, D. V., Sparling, G. P., and Fillery, I. R. P. (1998). Stratification of microbial biomass-C and N and gross N mineralisation with soil depth in two contrasting Western Australian agricultural soils. Aust. J. Soil Res. 36, 45 –55. Murphy, D. V., Sparling, G. P., Fillery, I. R. P., McNeill, A. M., and Braunberger, P. (1998). Mineralisation of soil organic nitrogen and microbial respiration after simulated summer rainfall events in an agricultural soil. Aust. J. Soil Res. 36, 231 –246. Murphy, D. V., Bhogal, A., Shepherd, M. A., Goulding, K. W. T., Jarvis, S. C., Barraclough, D., and Gaunt, J. L. (1999). Comparison of 15N labelling methods to measure gross nitrogen mineralisation. Soil Biol. Biochem. 31, 2015–2024. Murphy, D. V., Dise, N., Goulding, K. W. T., MacDonald, J., Peake, C., Redfern, P., and Stockdale, E. A. (2001). Can concepts of N saturation developed for forest systems be applied in arable soils? In “Sustainable Management of Soil Organic Matter”. (R. M. Rees, B. C. Ball, C. D. Campbell and C. A. Watson, Eds.), pp. 233 –238. CAB International, Wallingford, UK. Myrold, D. D., and Tiedje, J. M. (1986). Simultaneous estimation of several nitrogen cycle rates using 15 N: theory and application. Soil Biol. Biochem. 18, 559 –568. Nason, G. E., and Myrold, D. D. (1991). 15N in soil research: appropriate application of rate estimation procedures. Agric. Ecosyst. Environ. 34, 427 –441. Nicolardot, B., Fauvet, G., and Cheneby, D. (1994). Carbon and nitrogen cycling through soil microbial biomass at various temperatures. Soil Biol. Biochem. 26, 253 –261. Nishio, T. (1991). 15N-NHþ 4 isotope dilution method for analysing nitrogen transformations in upland soils. Jpn Agric. Res. Q. 25, 88–92. Nishio, T., and Fujimoto, T. (1989). Mineralisation of soil organic nitrogen in upland fields as determined by a 15NHþ 4 dilution technique, and absorption of nitrogen by maize. Soil Biol. Biochem. 21, 661– 665. Nishio, T., Kanamori, T., and Fujimoto, T. (1985). Nitrogen transformations in an aerobic soil as determined by a 15NHþ 4 dilution technique. Soil Biol. Biochem. 17, 149– 154. Norton, J. M., and Firestone, M. K. (1996). N dynamics in the rhizosphere of Pinus ponderosa seedlings. Soil Biol. Biochem. 28, 351–362. Nye, P. H., and Tinker, P. B. H. (1977). Studies in Ecology. In “Solute movement in the soil–root system”. Vol. 4, pp. 342. Blackwell, Oxford. O’Dowd, R. W., Barraclough, D., and Hopkins, D. W. (1999). Nitrogen and carbon mineralisation in soil amended with D - and L -leucine. Soil Biol. Biochem. 31, 1573–1578. Osler, G. H. R. (2003). Impact of fauna on chemical transformations in soil. In “Soil Biological Fertility: A Key to Sustainable Land Use in Agriculture”. (L. K. Abbott and D. V. Murphy, Eds.), in press. Kluwer Academic Publishers, Dordrecht, in press.

116

D. V. MURPHY ET AL.

Paul, E. A., and Juma, N. G. (1981). Mineralisation and immobilisation of soil nitrogen by microorganisms. Ecological Bulletin 33. In “Terrestrial Nitrogen Cycles. Processes, Ecosystems, Strategies and Management Impacts”. (F. E. Clark and T. Rosswall, Eds.), pp. 179–204. Swedish Natural Science Research Council, Stockholm. Pedersen, H., Dunkin, K. A., and Firestone, M. K. (1999). The relative importance of autotrophic and heterotrophic nitrification in a conifer forest soil as measured by N-15 tracer and pool dilution techniques. Biogeochemistry 44, 135–150. Pilbeam, C. J., and Warren, G. P. (1995). Use of 15N for fertiliser N recovery and N mineralisation studies in semi-arid Kenya. Fert. Res. 42, 123 –128. Pilbeam, C. J., Mahapatra, B. S., and Wood, M. (1993). Soil matric potential effects on gross nitrogen mineralisation in an orthic ferrasol from Kenya. Soil Biol. Biochem. 25, 1409–1413. Puri, G., and Ashman, M. R. (1998). Relationship between soil microbial biomass and gross N mineralisation. Soil Biol. Biochem. 30, 251 –256. Raison, R. J., Connell, M. J., and Khanna, P. K. (1987). Methodology for studying fluxes of soil mineral-N in situ. Soil Biol. Biochem. 19, 521 –530. Recous, S., Mary, B., and Faurie, G. (1990). Microbial assimilation of ammonium and nitrate in soil. Soil Biol. Biochem. 22, 597– 602. Recous, S., Robin, D., Darwis, D., and Mary, B. (1995). Soil inorganic N availability: effect on maize residue decomposition. Soil Biol. Biochem. 27, 1529–1538. Recous, S., Aita, C., and Mary, B. (1999). In situ changes in gross N transformations in bare soil after addition of straw. Soil Biol. Biochem. 31, 119–133. Rees, R. M., McTaggart, I. P., Smith, K. A., and Stockdale, E. A. (1994). Methodology for the study of N mineralisation in the field. Proceedings of a Symposium held at the Institute for Soil Fertility Research, 19–20 April 1993. In “Nitrogen mineralisation in agricultural soils”. (J. J. Neeteson and J. Hassink, Eds.), pp. 81 –94. Robertson, G. P. (1982). Nitrification in forested ecosystems. Philos. Trans. R. Soc. London 296, 445–457. Schimel, J. (1996). Assumptions and errors in the 15NHþ 4 pool dilution technique for measuring mineralisation and immobilisation. Soil Biol. Biochem. 28, 827 –828. Schimel, J. P., Jackson, L. E., and Firestone, M. K. (1989). Spatial and temporal effects on plantmicrobial competition for inorganic nitrogen in a California annual grassland. Soil Biol. Biochem. 21, 1059–1066. Schimel, J. P., Helfer, S., and Alexander, I. J. (1992). Effects of starch additions on N turnover in Sitka spruce forest floor. Plant Soil 139, 139– 143. Schnitzer, M. (2000). A lifetime perspective on the chemistry of soil organic matter. Adv. Agron. 68, 1–58. Scott, N. A., Parfitt, R. L., Ross, D. J., and Salt, G. J. (1998). Carbon and nitrogen transformations in New Zealand plantation forest soils from sites with different N status. Can. J. Forest Res. 28, 967–976. Silgram, M., and Shepherd, M. A. (1999). The effects of cultivation on soil nitrogen mineralisation. Adv. Agron. 65, 267– 311. Smith, J. L., McNeal, B. L., Cheng, H. H., and Campbell, G. S. (1986). Calculation of microbial maintenance rates and net nitrogen mineralisation in soil at steady state. Soil Sci. Soc. Am. J. 50, 332 –338. Smith, C. J., Chalk, P. M., Crawford, D. M., and Wood, J. T. (1994). Estimating gross nitrogen mineralisation and immobilisation rates in anaerobic and aerobic soil suspensions. Soil Sci. Soc. Am. J. 58, 1652–1660. Smith, P., Smith, J. U., Powlson, D. S., McGill, W. B., Arah, J. R. M., Chertov, O. G., Coleman, K., Franko, U., Frolking, S., Jenkinson, D. S., Jensen, L. S., Kelly, R. H., Klein-Gunnewiek, H., Komarov, A. S., Li, C., Molina, J. A. E., Mueller, T., Parton, W. J., Thornley, J. H. M., and Whitmore, A. P. (1997). A comparison of the performance of nine soil organic matter models using datasets from seven long-term experiments. Geoderma 81, 153 –225.

GROSS NITROGEN FLUXES IN SOIL

117

Sohi, S. P., Mahieu, N., Arah, J. R. M., Powlson, D. S., Madari, B., and Gaunt, J. (2001). A procedure for isolating soil organic matter fractions suitable for modelling. Soil Sci. Soc. Am. J. 65, 1121–1128. Sparling, G. P., Murphy, D. V., Thompson, R. B., and Fillery, I. R. P. (1995). Short-term net N mineralisation from plant residues and gross and net N mineralisation from soil organic matter after rewetting of a seasonally dry soil. Aust. J. Soil Res. 33, 961 –973. Stark, J. M., and Firestone, M. K. (1995). Isotopic labelling of soil nitrate pools using nitrogen-15nitric oxide. Soil Sci. Soc. Am. J. 59, 844–847. Stark, J. M., and Firestone, M. K. (1996). Kinetic characteristics of ammonium-oxidiser communities in a California oak woodland–annual grassland. Soil Biol. Biochem. 28, 1307–1317. Stark, J. M., and Hart, S. C. (1997). High rates of nitrification and nitrate turnover in undisturbed coniferous forests. Nature 385, 61–64. Stark, J., and Schimel, J. (2001). Errors in “overestimation of gross N transformation rates in grassland soils…”. Soil Biol. Biochem. 33, 1433–1435. Stevenson, F. C., Walley, F. L., and van Kessel, C. (1998). Direct vs. indirect nitrogen-15 approaches to estimate nitrogen contributions from crop residues. Soil Sci. Soc. Am. J. 62, 1327– 1334. Stockdale, E. A., Davies, M. G., Koch, M., Rees, R. M., and Vinten, A. J. A. (1994). Assessment of the problems associated with measuring gross mineralisation rates in soil cores. Proceedings of a Symposium Held at the Institute for Soil Fertility Research, Haren, Netherlands, 19–20 April 1993. In “Nitrogen Mineralisation in Agricultural Soils”. (J. J. Neeteson and J. Hassink, Eds.), pp. 95–110. Stockdale, E. A., Hatch, D. J., Murphy, D. V., Ledgard, S. F., and Watson, C. J. (2003). Verifying the nitrification to immobilisation ratio (N/I) as a key determinant of potential nitrate loss in grassland and arable and grassland soils. Agronomie´ 22, in press. Stottlemeyer, R., and Toczydlowski, D. (1999). Nitrogen mineralisation in a mature boreal forest, Isle Royale, Michigan. J. Environ. Qual. 28, 709– 720. Tietema, A. (1998). Microbial carbon and nitrogen dynamics in coniferous forest floor material collected along a European nitrogen deposition gradient. Forest Ecol. Manage. 101, 29–36. Tietema, A., and Van Dam, D. (1996). Calculating microbial carbon and nitrogen transformations in acid forest litter with 15N enrichment and dynamic simulation modelling. Soil Biol. Biochem. 28, 953 –965. Tietema, A., and Wessel, W. W. (1992). Gross nitrogen transformations in the organic layer of acid forest ecosystems subjected to increased atmospheric nitrogen input. Soil Biol. Biochem. 24, 943 –950. Tietema, A., Boxman, A. W., Bredemeier, M., Emmett, B. A., Moldan, F., Gundersen, P., Schleppi, P., and Wright, R. F. (1998). Nitrogen saturation experiments (NITREX) in coniferous forest ecosystems in Europe: a summary of results. Environ. Pollut. 102, 433 –437. Tietema, A., Emmett, B. A., Gundersen, P., Kjonaas, O. J., and Koopmans, C. J. (1998). The fate of N15 labelled nitrogen deposition in coniferous ecosystems. Forest Ecol. Manage. 101, 1–3. Tlustos, P., Willison, T. W., Baker, J. C., Murphy, D. V., Pavlikova, D., Goulding, K. W. T., and Powlson, D. S. (1998). Short-term effects of nitrogen on methane oxidation in soils. Biol. Fert. Soils 28, 64–70. Trehan, S. P. (2000). Mineralisation of nitrogen from cattle slurry decomposing in soil. J. Indian Soc. Soil Sci. 48, 463–468. Trinsoutrot, I., Recous, S., Mary, B., and Nicolardot, B. (2000). C and N fluxes of decomposing 13C and 15N Brassica napus L.: effects of residue composition and N content. Soil Biol. Biochem. 32, 1717–1730. Unkovich, M., Jamieson, N., Monaghan, R., and Barraclough, D. (1998). Nitrogen mineralisation and plant nitrogen acquisition in a nitrogen-limited calcareous grassland. Environ. Expp. Bot. 40, 209 –219. van Veen, J. A., Ladd, J. N., and Frissel, M. J. (1984). Modelling C and N turnover through the microbial biomass in soil. Plant Soil 76, 257–274.

118

D. V. MURPHY ET AL.

van Veen, J. A., Ladd, J. N., and Amato, M. (1985). Turnover of carbon and nitrogen through the microbial biomass in a sandy loam and a clay soil incubated with [14C(U)]glucose and [15N](NH4)2SO4 under different moisture regimes. Soil Biol. Biochem. 17, 747 –756. Vitousek, P. M., Gosz, J. R., Grier, C. C., Melillo, J. M., and Reiners, W. A. (1982). A comparative analysis of potential nitrification and nitrate mobility in forest ecosystems. Ecol. Monogr. 52, 155–177. Wang, W. J., Chalk, P. M., Chen, D., and Smith, C. J. (2001). Nitrogen mineralisation, immobilisation and loss, and their role in determining differences in net nitrogen production during waterlogged and aerobic incubation of soils. Soil Biol. Biochem. 33, 1305–1315. Watkins, N., and Barraclough, D. (1996). Gross rates of N mineralisation associated with the decomposition of plant residues. Soil Biol. Biochem. 28, 169 –175. Watson, C. J., Travers, G., Kilpatrick, D. J., Laidlaw, A. S., and O’Riordan, E. (2000). Overestimation of gross N transformation rates in grassland soils due to non-uniform exploitation of applied and native pools. Soil Biol. Biochem. 32, 2019–2030. Watson, C. J., Travers, G., Kilpatrick, D. J., Laidlaw, A. S., and O’Riordan, E. (2002). Response to Stark and Schimel’s letter to the editor published in Soil Biology and Biochemistry 33, 1433–1435, 2001. Soil Biol. Biochem. in press 34, 553 –555. Wessel, W. W., and Tietema, A. (1992). Calculating gross N transformation rates of 15N pool dilution experiments with acid forest litter: analytical and numerical approaches. Soil Biol. Biochem. 24, 931–942. Whitmore, A. P. (1991). A method for assessing the goodness of computer simulation of soil processes. J. Soil Sci. 42, 289–299. Willison, T. W., Baker, J. C., and Murphy, D. V. (1998). Methane fluxes and nitrogen dynamics from a drained fenland peat. Biol. Fert. Soils 27, 279– 283. Willison, T. W., Baker, J. C., Murphy, D. V., and Goulding, K. W. T. (1998). Comparison of a wet and dry 15N isotopic dilution technique as a short-term nitrification assay. Soil Biol. Biochem. 30, 661–663. Zagal, E., Bjarnason, S., and Olsson, U. (1993). Carbon and nitrogen in the root-zone of barley (Hordeum vulgare L.) supplied with nitrogen fertiliser at two rates. Plant Soil 157, 51 –63. Zaman, M., Di, H. J., Cameron, K. C., and Frampton, C. M. (1999). Gross nitrogen mineralisation and nitrification rates and their relationships to enzyme activities and the soil microbial biomass in soils treated with dairy shed effluent and ammonium fertiliser at different water potentials. Biol. Fert. Soils 29, 178– 186.

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS K. S. Dhillon and S. K. Dhillon Department of Soils, Punjab Agricultural University, Ludhiana 141 004, India

I. II. III. IV. V. VI. VII. VIII.

IX.

X.

XI. XII.

Introduction Historical Aspects Properties and Uses of Se Rocks and Minerals Containing Se Formation of Seleniferous Soils Se Additions to the Environment A. Natural Sources B. Anthropogenic Activities Geographical Distribution of Se Toxic Soils Se in the Seleniferous Environments A. Seleniferous Soils B. Se in Water C. Se in Plants D. Se in the Atmosphere Effects of Se Toxicity on the Components of the Ecosystem A. Plants B. Animals C. Selenium Toxicity in Humans Management of Seleniferous Soils A. Soil Mixing/Covering B. Soil Washing C. Thermal Treatment D. Immobilization of Se in the Toxic Environment E. Presence of Competitive Ions in Soil Solution F. Selecting Plants with Low Se Absorption Capacity G. Phytoremediation H. Bioremediation Conclusions Future Research Needs References

Selenium is one of the few elements absorbed by plants in sufficient amounts that can be toxic to livestock. Volcanic activity, weathering, precipitation of minerals and burning of fossil fuel control the distribution of Se in the environment. In addition, anthropogenic activities such as disposal of fly ash, utilization of underground water for raising of crops and mining operations have contributed substantially to the redistribution and cycling 119 Advances in Agronomy, Volume 79 Copyright q 2003 by Academic Press. All rights of reproduction in any form reserved 0065-2113/02$35.00

120

K. S. DHILLON AND S. K. DHILLON of Se. Changes in topographical features and leaching/erosion processes have played an important role in the development of seleniferous soils in the different parts of the world. Areas of toxic, adequate and deficient Se levels exist side by side. Soils with elevated levels of Se are found in many countries including Australia, China, India, Ireland and USA. Soils containing . 0.5 mg Se kg21 are considered as seleniferous as the forages produced on such soils absorb Se more than the maximum permissible level for animal consumption. Selenium binding onto soils and sediments depends upon the pH, Eh, Se species, competing anions, hydrous oxides of iron and type of clay minerals. Selenium in contaminated soil and water exists mainly as highly mobile toxic 22 6þ 4þ inorganic species such as selenate (SeO22 4 , Se ) and selenite (SeO3 , Se ). During the last decade, a number of technologies have been proposed to get rid of excessive Se from the contaminated environments. These are based on either immobilization of Se to biologically unavailable forms or complete removal of Se through phytoextraction or biomethylation. It is the concentration of mobile Se, usually selenate, that determines the need for adopting a specific technology and the extent of change in the mobile concentration is considered as the measure of success. Potential of different remedial technologies leading to the permanent removal of Se oxyanions from the seleniferous soils has also been discussed. q 2003 Academic Press.

I. INTRODUCTION In many parts of the world, there exist soils with elevated levels of Se where accumulation of Se in toxic levels in the food chain is resulting in serious health problems in plants, migratory birds, animals and human beings (Rosenfeld and Beath, 1964; Yang et al., 1983; Ohlendorf, 1989; Dhillon and Dhillon, 1991a, 1997a). The word “seleniferous” denotes high in selenium, which is sometimes misleading. The soils can be classified as seleniferous or non-seleniferous depending upon the Se level of non-accumulators (cultivated agricultural crops) grown on that soil. The soils containing as low as 0.1 – 0.5 mg Se kg21 (Ravikovitch and Margolin, 1957; Dhillon et al., 1992a) are considered as seleniferous because the forages grown on such soils contain . 4 mg Se kg21, i.e., the maximum permissible level for animal consumption. Consumption of feed containing , 0.05– 0.1 mg Se kg21 may result in severe deficiency diseases in animals and humans. Toxic effects of Se occur with exposure to levels of 2 – 5 mg Se kg21 depending upon the chemical form of Se (Gissel-Nielson et al., 1984). The extremely narrow gap between the essential and toxic levels has prompted Oldfield (1986) to refer Se as an essential poison. In spite of well-known toxic effects of Se, it was not acknowledged as a pollutant for a long time. Inclusion of Se in the list of inorganic carcinogenic agents

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 121

(Shubik et al., 1970) has prompted the researchers all over the world to examine the Se levels in every component of the environment. In 1985, the United States Environmental Protection Agency (USEPA) postulated that Se should receive closer scrutiny as a potential contaminant of the food chain. Parent material has been considered as an important factor controlling the level of Se in geoecosystems (Anderson et al., 1961; Rosenfeld and Beath, 1964; Brown and Shrift, 1982). Human activities contribute substantially to the redistribution and cycling of Se on a global scale. Anthropogenic activities which include disposal of coal generated fly ash, mine tailings and agricultural drainage water; use of fertilizers and underground water for crop production and domestic household sources such as dandruff shampoo have been linked to Se toxicity problems (Thomson and Heggen, 1982; Nriagu and Pacyna, 1988; Jacobs, 1989; Dhillon and Dhillon, 1990; Frankenberger and Benson, 1994). While assessing the long-term changes in Se deposition in southeast England, Haygarth et al. (1993) reported that the earlier half of the 20th century saw an increase in the Se content of herbage, that was probably related to increased atmospheric deposition following the increased use of fossil fuel; the decline in Se content of herbage during the later half probably reflects a change in the use of fuel usage from coal to oil or gas. Total worldwide input of Se into soils from anthropogenic activities has been estimated to be 6000 –76,000 t year21 (Nriagu and Pacyna, 1988). Concern over the alarming ecotoxicological impacts of Se has diverted the attention of researchers towards finding a just solution for the management of contaminated soils. The most effective remediation strategies should protect all components of the biosphere, i.e., land, air, surface water, ground water as well as health of the general public (McNeal and Waring, 1992). During the last decade or so, a number of useful biological technologies have been investigated for the removal of Se and regulating its movement into the food chain from contaminated soil and water. The major emphasis has been placed on permanent removal of Se by manipulating the biological transformations of toxic Se oxyanions into less toxic or biologically unavailable forms that are mediated through microorganisms as well as higher plants. This chapter reviews the distribution of Se, toxicological impacts and the management of seleniferous environments.

II.

HISTORICAL ASPECTS

J. J. Berzelius and J. G. Gahan discovered Se in 1817 while working with sediments of a sulfuric acid plant at Gripsholm, Sweden. It was isolated from the red deposits on the walls of lead chambers and given the name of selenium after the moon Goddess Selene. The toxic nature of Se became known even much before it was discovered. There are reports that chemists were called in after

122

K. S. DHILLON AND S. K. DHILLON

workers at the sulfuric acid factory had become sick probably due to inhalation of vapors. Thus, for the next 140 years Se was mainly considered as environmental toxicant till it was discovered as an essential nutrient for animal and human health (Schwarz and Foltz, 1957). Perhaps, Marco Polo was the first to come in contact with probable Se poisoning while traveling in the remote parts of Western China and eastern Turkestan in 13th century. Then in the 16th century, Father Pedro Simon described a disease occurring in Colombia with similar symptoms in animals and human beings as described by Marco Polo. However, the first documented report occurred in the USA during the 1850s in South Dakota, USA. A near prefect description of Se poisoning in animals (horses), as known to today’s scientists, was penned in 1856 by an army surgeon Dr Madison, who was stationed at Fort Randall located at the second terrace above the Missouri river in what is now known as Gregory County. He pointed out a strong relationship between the consumption of locally available forages and disease symptoms in horses. Later the disease was named as Alkali disease. Although the name resulted from a false assumption that saline water and salt crusts were the cause of trouble, it continues to be in use even today. Fream reported loss of hooves and hair from horses grazing on the “poisoned lands of Meath” in Ireland as early as 1890. Boon (1989) reported that the settlements in Boyd County, Nebraska, experienced apparent Se toxicities in 1891. In 1893, Se toxicity in horses (Equus caballus ) was identified as a result of ingestion of irrigated hay in the Shirley Basin, Carbon County, Wyoming. In 1907 and 1908, more than 15,000 sheep (Ovis aries ) mortalities occurred in the Shirley Basin area. With the establishment of State Agricultural Experimental Stations in South Dakota and Wyoming, research scientists continued to explore systematically the relationship of the disease occurring in animals with environmental factors. By 1922, they came to know that there is a strong relationship between the disease and certain soil types and it is further linked to soils derived from Pierre Shale. This was followed by the identification of selenium indicator plants, parent materials of toxic soils, cataloguing the symptoms of alkali disease and chronic selenosis in animals and probable cures. Toxic effects of Se in plants and animals continued to be well documented, although the analysis of samples revealed the presence of Se in toxic plant tissues for the first time in 1933 by W. O. Robinson. Armed with a Se analytical method, Se toxic areas were identified in Columbia (1940), Ireland (1951), Israel (1957), Australia (1958) and India (1991). There are no documented instances of naturally occurring Se causing damage to agricultural plants in the field (NAS – NRC, 1976). Selenium toxicity symptoms have been studied under laboratory and greenhouse conditions. Hurd-Karrer (1934) was the first to record snow-white chlorosis with pink coloration on the leaves of wheat plants. For the first time, Se toxicity symptoms of snow-white chlorosis with pink coloration on the lower side of leaves and sheath of wheat

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 123

were recorded under field conditions in the seleniferous areas of Punjab, India (Dhillon and Dhillon, 1991a). In a summary of work published in 1957, the research staff working at the South Dakota Experimental Station could not fully understand the pattern of Se distribution, but observed that Se concentrations in plants were not strictly related to the total Se concentration in soils on which they grow; Se bearing soils could occur at distances away from their parent materials, redeposited by wind and water erosion; seleniferous soils tended to be heavy textured or clayey. Once considered a poison, Se has become the most intensively studied element since its discovery in 1957 as an essential nutrient in animal and human diet. Presently, benefits some scientists claim seem just short of miraculous—a protectant against some of the most dreaded diseases such as cancer, AIDS, cardiovascular ailments and Alzheimer’s disease. In the course of developing useful Se applications it has generated an impressive research effort. Since its discovery in 1817, the list of scientific publications has passed 100,000 (Reilly, 1996) and it continues to grow still higher.

III. PROPERTIES AND USES OF Se Chemically, selenium is similar to S and both the elements along with oxygen, lithium and polonium are placed in group VI A of the periodic table. Selenium is a metalloid having chemical and physical properties that are intermediate between metals and non-metals. Important properties of selenium are reported in Table I. Selenium exists in four oxidation states, i.e., þ 6, þ 4, 0 and 2 2 and has a six-electron system of valence orbitals. It has six naturally occurring stable isotopes with varying degrees of abundance: 74Se (0.9%), 76Se (9.0%), 77Se (7.6%), 78Se (23.5), 80Se (49.7%) and 82Se (9.2%). A number of radioactive isotopes of Se with half-lives ranging from a few seconds to 104 years have been characterized including 77Se (17.7 s), 75Se (120 days), and 79Se (6.5 £ 104 years). Due to a sufficiently long half-life (120 days), 75Se is being used in radiology and tracer applications in soil –plant – animal systems. Annual world production of Se is between 1800 and 2000 t and is closely related to copper production. Over 80% of the Se used commercially is obtained as a by-product from electrolytic copper refinery slimes, which vary in their Se content from 2 to 55%. Selenium has a wide variety of industrial applications and is utilized commercially principally in elemental form. The percentage break down of its application is given in Fig. 1. Commercial grade Se is being utilized for improving machinability and porosity control in steel, decolorization and pigmentation in glass, in architectural glass and as a catalyst. Other applications include blasting caps, rubber vulcanizers and accelerators, chemical reagents, lubricants and medicines. Due to its unique semiconducting properties, high purity Se finds

124

K. S. DHILLON AND S. K. DHILLON Table I Important Physical and Chemical Properties of Selenium

Periodic position Atomic number Atomic weight Atomic radius (uncharged) Stable isotopes Radioactive isotopes Hardness (relative units) Allotropic forms Melting point (amorphous) Boiling point Oxidation slates Electronic configuration Electronegativity (Pauling scale) Entropy Enthalpies Fusion Vapourization Bond energy (Se–Se) Self-diffusion coefficients (Trigonal) Standard reduction potentials

Group VIA, Period 4 34 78.96 215 pm 6 14 2 Trigonal (grey), a monoclinic (red), b monoclinic (red), red amorphous, black amorphous, vitreous (black) 323 K 958 K 22, 0, þ 4, þ 6 2-8-18-6 2.54 10.15 cal g atom21 5.10 kJ mol21 26.32 kJ mol21 44 kcal mol21 3.8 £ 10212 cm2 s21 at 308 K

Se þ 2e2 ¼ Se22

0.78 V

Se þ 2Hþ þ 2e2 ¼ H2 Se ðaqÞ

0.36 V

Source: Zingaro and Cooper (1974).

Figure 1 Commercial utilization of Se [adapted from Oldfield (1990)].

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 125

application in xerography, rectifiers, photoelectric cells and solar cells. The antioxidant property of Se makes it suitable for inclusion in inks, mineral and vegetable oils. In the pharmaceutical industry, selenium sulfides are used in the treatment of dandruff and dermatitis. Trace quantities of sodium selenite or selenate are added to the feed of chicken, turkeys, swine, beef and dairy cattle and sheep to prevent its deficiency. Recently, in deficient areas, nutritional supplementation has also been recommended by direct topsoil dressing of grazing pastures and through addition to chemical fertilizers (Korkman, 1985). The estimated biological uses of Se in different parts of world are reported in Table II.

IV. ROCKS AND MINERALS CONTAINING Se Selenium is one of the most dispersed elements in the earth’s crust, ranking 69th in order of abundance and ranging from 0.03 to 4.08 mg kg21. It occurs in minute amounts in all materials of the crust, but is rarely concentrated in any of the materials in amounts above 100 mg kg21.The selenium concentration of rocks varies in different geological formations in different beds of the same formation and even in different parts of the same bed. Estimates made by geochemists of the average Se content for most of the common rock types are given in Table III. The concentration of Se in igneous rocks is much less than that of sedimentary rocks. Among sedimentary rocks, shales commonly contain more Table II Biological Uses of Selenium Estimated annual use of Se ( £ 103 kg) Continent North America South America Europe Australia Asia Finland New Zealand Others Total Grand total

Feed and veterinary

Fertilizer in agriculture

Human applicationa

49.5 1.5 12.9 1.4 42.0b – 0.6 2.0 109.9 164.0

– – – – – 15.0 15.0 – 30.0

10.0 4.0 7.0 1.0 2.0 – – 0.1 24.1

Source: Oldfield (1990). a Mainly as anti-dandruff shampoo. b Mainly in China only.

126

K. S. DHILLON AND S. K. DHILLON Table III Selenium Content of Rocks and Other Materials

Rocks/materials Igneous rocks (from different countries)a Sedimentary rocks Shales European Paleozoic shalesb Japanese Paleozoic shalesb Japanese Mesozoic shalesb Pyritic shales of early Carboniferous age Copper rich black shales of Permian age Shales of Selurian age from Baltic region Vanadiferous shale zone of Phosphoria formation Mudstone from Phosphoria formation Gray shales from seleniferous area in Columbia Sandstones Sandstones of Cretaceous to Tertiary age Sandstones adjacent to the shales Sandstones 10 in. away from shale–sandstone contact Sandstones of Tertiary age Phosphate rocks Phosphate rock Phosphate rock from Phosphoria formation Pyritiferous Phosphate rock Phosphoria formation Primary phosphate deposits Secondary phosphate deposits Limestones Devonian limestone (Germany) Fort Hays limestone (Kansas) Fort Hays limestone (S. Dakota) Calcareous marl of Niobara formation Limestone of Mississipian and Ordovician age Limestone of Phosphoria formation Limestone of Frontier formation Limestone of Thaynes formation Limestone from Ireland

Selenium content (mg kg21) 0.09–0.46

1.20 0.24 0.38 28.5 15 0.3–9.0 680 .1500 1–14 Nil 112.8 2.86 112 0.1 –55 300 0.9 –30 ,10.9 ,2 ,0.1 0.3–6 3 20 0.25–0.8 14.3 6.68 1.54 2

Source: Anderson et al. (1961). a Rock samples from USSR, USA and Germany. b Composite sample.

Se than sandstone, limestone and phosphate rocks. According to Krauskopf (1955), the processes responsible for enrichment of Se in sedimentary rocks and other geological materials are: mechanical enrichment, precipitation, adsorption, substitution and presence of organic material in deposits. Sandstones are usually more permeable than limestones and shales and their gross composition is more variable. Local enrichment of sandstone may occur because of precipitation of Se

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 127

from ground waters moving through beds long after their deposition. Of the total Se in sandstones, . 80% is water soluble, whereas in pyritic phosphoric rocks only 0.6% is water-soluble. Selenium easily forms compounds with metals and occurs in about 50 minerals, which occur in very low concentrations in soils. The most common are: clockmannite—CuSe; ferroselite—FeSe; clausthalite—PbSe; naumannite— Ag2Se and tiemannite—HgSe.

V. FORMATION OF SELENIFEROUS SOILS The soils that tend to be seleniferous have developed like the normal soils as a result of so called active and passive factors of soil formation. Toxic Se levels are a direct consequence of the geological origin of soils. The selenium content of soils is determined as a result of slight modifications in: (i) formations or rock outcrops, (ii) formations lying beneath the soil mantle, (iii) decomposition of parent rocks by wind and water and subsequent transport by ground or surface water, (iv) indicator plants, and (v) man made enrichment of the soil with Se in mining and other operations (Rosenfeld and Beath, 1964). Changes in topographical features and leaching/erosion processes have played an important role in the development of seleniferous soils in different parts of the world. Areas of toxic, adequate and deficient Se levels exist side by side. The water-soluble form of Se constitutes the major source of variations in the Se content of soils. Areas from where leaching has taken place, have developed into deficient regions and where the leachate has been deposited, has lead to the development of a Se enriched region. The sources of Se in soils and the distribution of seleniferous rocks in the United States were studied intensively during the 1930s and 1940s after Se in pastures was shown to cause fatal diseases in cattle and horses in South Dakota (Anderson et al., 1961). Most of the seleniferous soils in arid and semi-arid areas of the western states have developed in situ from weathering of underlying rocks. Investigations revealed that the soils that contain harmful quantities of Se are derived from materials of Cretaceous age. Extensive volcanic eruptions during Cretaceous time are thought to be the primary source of Se through deposition in Cretaceous seas that had invaded a considerable portion of the Western states of the USA (Davidson and Powers, 1959). Selenium was either eroded from igneous rocks or became incorporated in rainfall from volcanic gases and dust in the atmosphere. In either case Se was incorporated in sediments that have been uplifted over time and exposed to weathering and erosion. In the 17 western states, toxic soils are derived from Cretaceous sedimentary deposits of the Niobrara and portions of Pierre shale which outcrop or underlie . 700,000 km2. The chalky and calcareous marls and shales of the Niobrara formation are

128

K. S. DHILLON AND S. K. DHILLON

the most persistent seleniferous beds of the Great Plains region. The occurrence of elevated Se concentrations in other sedimentary rocks in the Western USA is localized. During the 1930s, irrigation was considered a remedial measure for seleniferous soils because the drain water removed soluble Se from soils. Little did people know at that time that consistent use of drainage water for irrigation purposes may lead to accumulation of Se in levels toxic to fish and wildlife as happened in the San Joaquin Valley in California, USA (Ohlendorf, 1989). Reconnaissance of the rock outcropping in the drainage basins to the contaminated area from the surrounding coastal ranges showed elevated levels of Se in the extensive surfacial exposures of the marine Upper Cretaceous– Paleocene Moreno and Eocene – Oligocene Kreyenhagen shales (Presser, 1994). The alternative source materials investigated in Californian coastal ranges included Cretaceous and Tertiary sandstones, and Pliocene – Pleistocene continental rocks and these were comparatively devoid of Se. Reduced Se present in elevated concentrations in the marine pyritic shales is weathered with S, concentrated by evaporation in soluble sulfate salts on farmland soils and mobilized as selenate by irrigation into sub-surface agricultural drains. Tidball et al. (1989) studied the elemental associations in soils of the San Joaquin Valley in California. Using a factor analysis technique, they observed that Se is being added to the valley sediments in moderate amounts all along the west side of the valley probably as detritus of shales. In the fan areas, Se may be diluted by barren sediments as well as be more readily dissolved and removed in a well-watered system. In the interfan areas, the solution of Se would be diminished by the lack of solvent, ineffective removal and possible excess evaporation. Seleniferous soils of Canada have developed from Cretaceous shales and spread over large areas in Saskatchewan, Alberta and a small area in Manitoba (Rosenfeld and Beath, 1964). In Alberta, seleniferous soils have also originated from glacial lacustrine parent material (Thorvaldson and Johnson, 1940). The development of seleniferous soils in Ireland form an interesting contrast to those in the Great Plains areas of the USA. The topography of the area has played a dominant role. According to Fleming (1962), the toxic soils have developed under quite humid conditions in depressions once occupied by old lakes. Consequently, the soils are poorly drained and rich in organic matter and their development may be compared to the initial stages of basin peat formation. Seleniferous soils occurring near Garristown (Dublin) have developed from Namurian black shales and those located in Limerick County have developed from Clare shales. While studying the formation of seleniferous soils in parts of England and Wales, Webb et al. (1966) observed that the principle bed rock sources of Se were certain marine shale facies of the Lower Carboniferous age in both the Staffordshire and Devon areas and of the Ordovician age in North Wales. In Columbia, the seleniferous soils have also developed in humid areas where

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 129

rainfall is high and are derived from the black slate of the Villeta formation (Rosenfeld and Beath, 1964). In northwestern Queensland in Australia, soils with high level of Se are associated with limestone shale. Vegetation grown on soils receiving the run-off from these limestones produced acute toxicities in livestock (Knott et al., 1958). Toxic soils in China have developed from carbon shale having a high Se content of 61.31 mg g21. Use of coal ash containing high Se levels as a fertilizer further increased Se concentrations in the soils. Topography and leaching processes seem to have played an important role in the development of seleniferous soils in China as well. On the basis of spatial distribution patterns and structures of Se in the geoecosystem in China, Tan et al. (1994) inferred that there is a low Se belt running from northwest to southwest. Flanked on both sides to the southeast and northwest are two relatively high Se belts. They concluded that (i) the low Se geoecosystem occurs mainly in and near the temperate forest and forest steppe landscape, (ii) relatively high Se concentrations in the geoecosystem usually appear in typical humid tropical and subtropical landscapes, (iii) in some mountain districts or elevated areas, e.g., in western Sichuan and Yunan provinces and eastern or south-eastern Tibet, low Se geoecosystems generally occur at altitudes above approximately 1300– 2000 m above sea level and is associated with vertical mountain forest, forest steppe and meadow steppe landscapes, and (iv) relatively high Se concentrations were found in the geoecosystem in some large accumulation plains such as Songlio, Weihe and Hua Bei plains, compared with washing areas within the same type of geographical zone. In contrast to the natural seleniferous soils, toxic soils have developed near the town of Irapuato (Mexico) in the Valley of Guanajuato River as a result of inadvertent induction of Se through ore processing in the mines in a neighboring area (Rosenfeld and Beath, 1964). Many of the areas in Wyoming classified as “Super toxic” are known today for their uranium deposits. The development of large-scale mining operations have raised additional concerns regarding potential Se contamination of soils, surface waters, and in ground waters and adjacent areas to these mining operations (Boon, 1989). The exact nature of the parent material of seleniferous soils in Punjab, India is not known, but convincing evidence is available that Se has been transported by rain water through seasonal rivulets from nearby hills of the Siwalik range and deposited in low lying areas (Dhillon and Dhillon, 1991a). The toxic sites are located at the dead ends of seasonal rivulets. Obviously the transported material was rich in Se. The parent material of the soils in the seleniferous region is derived from Upper Siwalik rocks that are mainly composed of polymictic conglomerates of variable composition containing many unstable materials (granite, basalt, limestone, etc.) and derived from the metamorphic terrain of the Himalayas (Karunakaran and Rao, 1979).

130

K. S. DHILLON AND S. K. DHILLON

VI. Se ADDITIONS TO THE ENVIRONMENT In addition to the role of parent material in the development of seleniferous soils, Se is being continuously added to the environment by processes such as volcanic activity, sea salt spray, forest wildfires, combustion of fossil fuels, incineration of municipal waste, weathering of rocks and soils, dust particles, soil leaching, metal smelting, fertilizers, groundwater transport, plant and animal uptake and release (Nriagu and Pacyna, 1988; Mayland et al., 1989; Nriagu, 1989; Dhillon and Dhillon, 2001). Estimated Se fluxes indicate that the natural sources of Se emission are as important as anthropogenic emissions. The possible processes leading to Se additions into the soil are presented in Fig. 2.

Figure 2 Schematic diagram of selenium inputs/outputs in the soil and possible impact on the environment.

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 131

A. NATURAL SOURCES While assessing the contribution of Se to the environment through natural processes, Nriagu (1989) reported that the total annual Se flux is (6 – 13) £ 103 t year21 (Table IV) and it is in close agreement with that reported by Mosher and Duce (1989). Each of the processes such as sea salt spray, volcanic emanation and suspended dust particles seem to account for , 10% of total Se emitted to the atmosphere from natural sources. The major Se form emitted as a result of volcanic activity and fossil fuel burning, as determined from thermodynamic calculations and chemical characteristics, is elemental Se (Andren et al., 1975; Suzuki, 1965). Marine biogenic sources can account for 60– 80% of the total Se released annually to the atmosphere from natural sources (Table IV). Biological release of dimethylselenide (DMSe) into the atmosphere could be considered as an important factor in the distribution of Se (Doran and Alexander, 1977). Supporting this view, La˚g and Steinnes (1978) observed a distinct decrease in Se concentration with increasing distance from the ocean, thereby indicating that Se is supplied to Norwegian forest soils only through precipitation. In southern districts of eastern Norway, relatively high concentrations of Se in soils may reflect a contribution from air pollution. In addition to this, Se has been detected in remote regions of the world such as Antarctica (Zoller et al., 1974) and the ice sheets of Greenland (Weiss et al., 1971). Biologically mediated volatilization processes can account for 30 – 50% of total Se emitted annually. While examining the exchange of Se in a grassland ecosystem in UK, Haygarth et al. (1994) found that wet deposition far exceeds dry deposition of Se, probably representing an input in excess of 200 mg m22 year21. Contributions through dry deposition only amounts to 2 mg m22 year21 whereas volatilization fluxes were estimated to be 100– 200 mg m22 year21.

Table IV Worldwide Emissions of Se ( £ 106 kg year21) from Natural Sources Source Wind borne soil particles Sea salt spray Volcanoes Wild forest fires Biogenic Continental particulates Continental volatiles Marine volatiles Total Source: Nriagu (1989).

Range

Median value

0.01–0.35 0–1 0.1–1.8 0–0.52

0.18 10.55 0.95 0.26

0–0.25 0.15–5.0 0.4–90 0.66–18.0

0.12 2.60 4.70 9.30

132

K. S. DHILLON AND S. K. DHILLON

B. ANTHROPOGENIC ACTIVITIES The natural fluxes of Se are small compared with emissions from industrial activities, implying that mankind has become the key agent in the global atmospheric distribution of Se. Total emission of Se into the atmosphere ranged from 2.5 to 24,000 t year21 which included 42% from anthropogenic sources (Nriagu, 1989). Inadvertent use of fly ash and sewage sludge as a cheap substitute for fertilizer and use of Se contaminated underground/drainage water for irrigation purposes has led to significant increases in the level of Se in the soil – plant system (Dhillon and Dhillon, 2001). The importance of anthropogenic activities in the distribution of Se can be further gauged from the following facts: (i)

The quantity of Se discharged to the atmosphere from the burning of coal represents 6 – 11% of the Se mobilized through weathering processes and river flow (Bertine and Goldberg, 1971; Andren et al., 1975). (ii) It has been estimated that 1.5– 2.3 times as much Se is mobilized through disposal of fly ash and slag wastes than by natural weathering and erosion of crystal materials. (iii) The addition of fly ash to soil or aquatic habitats can result in increased Se concentrations in plants, animals and other organisms of the ecosystem. Furr et al. (1978) reported that sweet clover plants growing directly on flyash disposal sites contained 205 mg Se kg21.

The loading rates of Se with terrestrial and aquatic ecosystems as calculated by Nriagu and Pacyna (1988) emphatically demonstrate that human activities now have major impacts on global and regional cycles of Se (Table V). Combustion of hard coal, lignites and brown coal in electric power plants and in industrial, commercial and residential business is the major source of air-borne Se. The major sources of Se contamination of the aquatic ecosystem including the ocean are: coal burning power plants (45%), non-ferrous metal smelters (28%), domestic waste water effluents (9%), and dumping of sewage sludge (4%) (Table V). Assuming that only 25% of the industrial effluents are discharged into lakes and rivers, the average concentration of Se in the water would be increased by about 800 ng l21. Most of the effluent discharges occur in Europe, North America and some Asian countries, implying that the contamination of fresh water resources in these regions may be more severe than is generally realized (Nriagu and Pacyna, 1988). The data reported in Table VI suggest that soils are receiving large quantities of Se from a wide variety of industrial wastes. The two principle anthropogenic sources of Se in soils are the disposal of ash residues from coal combustion (77%) and dumping of agricultural and municipal wastes on land (20%). In addition to direct effects, dumping of organic wastes on land favors the formation of volatile Se compounds resulting in losses of Se in gaseous forms and ultimately increasing

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 133 Table V Anthropogenic Inputs of Se ( £ 106 kg year21) into the Aquatic Ecosystem Source Domestic waste water Steam electric Metal mining and dressing Smelting and refining of non-ferrous metals Manufacturing processes of Metals Chemicals Pulp and paper Petroleum products Atmospheric fall-out Dumping sewage sludge Total input

Range

Median value

0– 7.5 6.0–30 0.25–1.0 3.0 –20

3.75 18.0 0.62 11.5

0– 5.0 0.02–2.5 0.01–0.9 0–0.09 0.54–1.1 0.26–3.8 10–72

2.50 1.26 0.46 0.045 0.82 1.67 41

Source: Nriagu and Pacyna (1988).

the atmospheric Se load (Kabata-Pendias and Pendias, 1984). There is no conclusive evidence that use of fertilizers has produced toxic crops; the only exception is in Japan (Suzuki et al., 1959). A typical situation has developed in California (USA) where normal soil and water management practices have led to contamination of the ecosystem with Se. In the San Joaquin valley, approximately 160,000 ha of farmland is affected by elevated levels of Se and salinity. Highly saline shallow ground water was collected by the tile drainage system from the farmland and transported through the Luis drain to the Kesterson Reservoir for storage and re-use for irrigation Table VI Anthropogenic Inputs of Se ( £ 106 kg year21) into Soils Source Food and agricultural wastes Wood wastes Municipal refuge and sewage sludge Solid waste—metal manufacturing Coal ash Fertilizer Peat (agriculture and fuel uses) Wastage of commercial products Atmospheric fallout Mine tailings Smelter slag wastes Total discharge on land Source: Nriagu and Pacyna (1988).

Range

Median

0.4–8.94 0–3.3 0.05– 4.14 0–0.19 4.1–60 0.02– 0.10 0–0.41 0.1– 0.2 1.3– 2.6 0.28– 0.41 0.1– 0.2 6.4–77

65 1.65 2.10 0.10 32.0 0.06 0.20 0.15 2.00 0.35 0.15 42

134

K. S. DHILLON AND S. K. DHILLON

purposes. The drainage water contained Se in the range of 250– 350 mg l21 (Presser, 1994). Following the discovery of Se accumulation in the Kesterson Reservoir, discharge of agricultural drainage water to the reservoir was halted. Within a short span of 6 years, from 1981 to 1986, about 9000 kg of Se and 3 £ 108 kg of salt were deposited into the evaporation ponds covering an area of 520 ha (Tokunaga et al., 1994). Another typical situation has been described by Dhillon and Dhillon (1990). In the seleniferous regions in northwestern India, underground water constitutes the only source for irrigating the farmland and drinking water, particularly for farm workers. The selenium content of underground water ranged from 2.5 to 69.5 mg l21. At certain locations in the seleniferous region, Se toxicity symptoms of snow-white chlorosis appeared in wheat grown in rotation with rice for 8– 10 years. This area was traditionally following a maize –wheat cropping sequence. But with the increased availability of water through installation of tube wells, the farmers shifted to a rice –wheat system. The addition of Se through irrigation water ranged from 49 to 1310 g ha21 year21 under a rice – wheat and from 18 to 520 g ha21 year21 under a maize – wheat system. The man-induced mobilization of Se into the biosphere has resulted in its increased circulation through soil, water and air and its inevitable transfer to the animal/human food chain has become an important environmental issue, which entails some unknown health risks for future generations.

VII. GEOGRAPHICAL DISTRIBUTION OF Se TOXIC SOILS The nature of Se distribution in soils is such that the preparation of maps showing clearly demarcated Se toxic areas is not an easy task. All over the world, Se toxic areas are interspersed with normal or even deficient soils. In spite of the large volume of research work carried out in the USA, the available maps still depict the distribution of seleniferous vegetation by dots to indicate seleniferous soils, which is misleading. Selenium accumulator plants may accumulate Se in toxic levels from a soil where cultivated agricultural plants may show deficiency. Very recently an excellent publication “World Selenium Atlas” has been published by a distinguished Se researcher, Oldfield (1999), that contains all the relevant information on Se levels in soils, plants, animals and humans along with maps. As early as 1936, W.O. Robinson confirmed the presence of Se in toxic levels in wheat grains from different countries of the world including North and South America, Australia, New Zealand, South Africa, Algeria, Morocco, Spain Bulgaria, France and Germany. Going through the literature, one finds that while research on Se toxicity has become a continuous process in the USA and Ireland

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 135

after the first reports appeared; in other countries, no follow up research work is traceable. Depending upon the relationship between the Se content of rocks and soils and the appearance of chronic and acute selenosis in animals and human beings, specific areas have been identified in a number of countries and are described here (Fig. 3). Australia. Se toxicity problems exist only in Queensland (Knott and McCray, 1959). Selenium accumulating plants, namely Morinda reticulata and Neptunia amplexicaulis, have been identified in Central Queensland which contained up to 123 mg Se per seed. “Change hoof disease” has been attributed to horses grazing on M. reticulata. The incidence of Se toxicity occurs in an area termed as the “Poison strip” that has been plotted on a map of Queensland. Ireland. Some of the most highly seleniferous soils in the world are located in Limerick, Tipperary and Meath Counties in Ireland (Fleming, 1962; Fleming and Walsh, 1957). The level of total Se in Irish soils (up to 1250 mg Se kg21) is about 10 times the level in the most highly seleniferous soils of the Western United States. As usual the location of Se toxic soils is sporadic. The location of most affected farms in different counties (Rogers et al., 1990) is summarized as below. County Carlow (near Castletown); Dublin (near Garristown); Kerry (near Ardfert); Kilkenny (near Piltown); Limerick (between Foynes and Ardgah); Meath (in Dunsany-Warrenstown area and near Athboy, Navan, Slane, Draogheda, Dunshauglin, Grauge and Trim) and Tipperary (near Boulick, Clonmel, Clogheen, Rathronan). In addition to these, small areas extend eastward from South County Limerick, through Clonmel (South County Tipperary) to Piltown (county Kilkenny). North America. Seleniferous soils and vegetation are widespread in the Rocky Mountains and the Great Plains of the Western United States extending from North Dakota to Texas and west to the Pacific region. They extend north into the Canadian provinces of Alberta, Saskatchewan and Manitoba and south into Mexico. The toxic soils have been mapped using Beath’s Se indicators and are concentrated especially in the states of Wyoming, South Dakota, North Dakota, Montana, Nebraska, Kansas, Colorado, New Mexico, Utah and Arizona and the San Joaquin valley in California (Rosenfeld and Beath, 1964). China. High selenium areas in the geoecosystem of China occur in some large accumulating plains such as Songlio, Weihe and Hua Bei provinces. However, endemic selenosis occurs mainly in Enshi county, in Hubei province and in Ziang county in Shanxi province (Ribang et al., 1992). Israel. Sporadic distribution of seleniferous soil has also been observed in the Huleh Valley of Israel where soils have originated from Limestone of the Cenomanian and Turonian formations of Cretaceous age (Ravikovitch and Margolin, 1959). India. The problem with Se toxicity has been identified in different states of India—in Haryana (Arora et al., 1975); Punjab (Dhillon and Dhillon, 1991a); West

136

K. S. DHILLON AND S. K. DHILLON

Figure 3 (A) The Tambo formation (hatched), a high Se area in Queensland in Australia [adapted from Oldfield (1990)]; (B) Selenium toxic area sporadically distributed in different counties in Ireland [(1) Kerry, (2) Limerick, (3) Tipperary, (4) Kilkenny, (5) Carlow, (6) Dublin, (7) Meath] (prepared on the basis of information by Rogers et al., 1990); (C) Geographical location of Se rich soils in Western United States of America and Canada (mapped on the basis of seleniferous vegetation) [Adapted from Rosenfeld and Beath (1964)]; (D) Location of Se toxic areas in Punjab, Haryana, West Bengal, Assam and Meghalaya states of India (Arora et al., 1975; Dhillon and Dhillon, 1991a,b; Ghosh et al., 1993; Dey et al., 1999); (E) Areas marked with cross lines indicate high Se ($0.4 mg kg21) belts in China where areas with toxic levels of Se ($3 mg kg21) are sporadically distributed [adapted from Tan et al., 1994].

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 137

Bengal (Ghosh et al., 1993) and Assam and Meghalaya (Dey et al., 1999). The largest area (. 1000 ha) affected by Se toxicity has been demarcated in the state of Punjab. The specific location of Se toxic soils is shown in Fig. 3. The toxic areas lie just on the boundary line between Hoshiarpur and Nawanshehar districts and are located in the villages of Panam, Nazarpur, Simbly, Barwa, Jainpur, Menhdpur, Rakkara Dhahan and Bhan Majara. In Haryana state, a few acres of land producing toxic fodders were identified in the village Chamarkhera near Karnal. In the subHimalayan West Bengal, soils growing Se toxic pasturesare located in the Jalpaiguri district. The affected areas remain flooded for 3 –4 months during the year. In northeast India investigations carried out by Dey et al. (1999) revealed that wild animal species from various locations of Assam and Meghalaya, viz. the Reserve Forest near Umkiang, Jaintia hills, Meghalaya; the Lailad Reserve forest, Ribhoi, Meghalaya and Rani Reserve Forest, Assam are suffering from known toxic effects of Se. The content of Se in the bones and hair exceeded (1.5 – 2.6 times) the toxic limits in all the animal species. At certain locations in this home range of animals; toxic levels of Se in soils (up to 17 mg g21) and water (up to 1.4 mg g21) were also recorded. Workers in different parts of the world have employed widely different procedures for estimating Se in soil and plant samples. Thus, many times, it becomes difficult to compare results from different regions. Keeping this in mind and recognizing the fragmented and limited nature of the information available, Sillanpaa and Jansson (1992) collected samples of soils and plants from some countries in Asia and the Pacific to establish the status of Se. Wheat and maize were used as indicator crops. The number of samples collected from each country varied from 25 (Malta) to 282 (India) for both plants as well as soils. The total sample population, including both soils and plants consisted of over 7600. All the samples were analyzed at one laboratory following the same methods of sample preparation and estimation. The relationships between mean Se content of soils (extractable with AAAc– EDTA) and plants are shown in Fig. 4. The highest percentage of low Se values was recorded from Finland (88%) and New Zealand (42%) where animal health is being seriously affected due to deficiency of Se. Other countries of low Se status are Peru (41%), Zambia (33%), Egypt (23%) and Malawi (17%). The highest Se contents approaching toxic levels were recorded from Pakistan (52%), Iraq (37%) and India (36%).

VIII.

Se IN THE SELENIFEROUS ENVIRONMENTS

Selenium is increasingly becoming an environmental threat and often has been described as the element with two faces of toxicity and deficiency existing side by side. Excellent reviews have been contributed by Rosenfeld and Beath (1964), Berrow and Ure (1989), Mayland et al. (1989) and Tan et al. (1994). Selenium

138

K. S. DHILLON AND S. K. DHILLON

Figure 4 Relationship between plant and soil Se medians of different countries [Ar, Argentina; Be, Belgium; Br, Brazil; Eg, Egypt; Et, Ethiopia; Fi, Finland; Gh, Ghana; Hu, Hungary; In, India; Ir, Iraq; It, Italy; Ko, Korea; Le, Lebanon; Mt, Malta; Mw, Malawi; Me, Mexico; Ne, Nepal; NZ, New Zealand; Ni, Nigeria; Pa, Pakistan; Pe, Peru; Ph, Philippines; Si, Sri Lanka; Sy, Syria; Ta, Tanzania; Th, Thailand; Tu, Turkey; Za, Zambia; adapted from Sillanpaa and Jansson (1992)].

occurs naturally in soils varying from 0.005 mg kg21 in a deficient area in Finland to 8000 mg kg21 in the Tuva area of Russia. This section will discuss the level of Se in soil, water, plants and atmosphere in the situations where a strong relationship existed between elevated levels of Se and health hazards.

A. SELENIFEROUS SOILS Distribution of Se in surface and subsurface soils is not uniform. Anderson et al. (1961) compiled early research on the Se content of seleniferous soils in the Great Plains of the USA. In highly seleniferous areas, the Se content of surface soils ranged from 1.5 to 20 mg kg21 and that of subsurface soils varied from 0.7 to 16 mg kg21. A maximum of 98 mg Se kg21 has been recorded in the toxic region in the Western United States (Rosenfeld and Beath, 1964). In the San Joaquin valley of California, average Se contents of soils yielding Se-laden drainage water ranged from 0.28 to 2.32 mg kg21 (Severson and Gaugh, 1992). At the Kesterson Reservoir and Lahontan valley, the Se content ranged from 4 to

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 139

25 and 0.7 – 1.5 mg kg21, respectively, in the upper 20 cm of soil (Tokunaga et al., 1994). In large accumulation plains of China such as Songliao, Weihe and Hua Bei plains, soils containing total Se $ 3.0 mg kg21 and water-soluble Se $ 0.02 mg kg21 are associated with Se poisoning. In typical seleniferous soils of China, the water-soluble Se concentration was as high as 42.9 mg kg21 (Tan et al., 1994). Total and water soluble Se in soils from the toxic region in Punjab, India ranged from 0.23 to 4.55 and 0.02– 0.16 mg kg21 (Dhillon et al., 1992a,b). In the state of Haryana, soils with as high as 10 mg Se kg21 have been reported (Singh and Kumar, 1976), but no cases of Se poisoning in animals and human beings have been reported so far. In the sub-Himalayan regions of West Bengal, the Se content of soils from the toxic pastures ranged from 1.45 to 2.25 mg kg21. Acute poisoning and chronic selenosis has been reported from the regions where total Se content in surface soils ranged from 0.3 to 0.7 mg kg21 in Canada, 0.3– 20 mg kg21 in Mexico, 1 –14 mg kg21 in Columbia, 1.2 –324.0 mg kg21 in Ireland and up to 6.0 mg kg21 in Israel (Rosenfeld and Beath, 1964).

1.

Forms of Se

Identification of the chemical forms of Se in soils has been very difficult because of the presence of Se in small amounts and the complex matrix of soils. Recent innovations in analytical chemistry including radio-tracer techniques have allowed scientists to determine different forms of Se existing even in minute quantities. Selenium exists in four oxidation states in the natural system: þ 6 [selenate, Se (VI)], þ 4 [selenite, Se (IV)], 0 [elemental Se] and 2 2 [selenide]. The chemical form of Se present in soils depends upon pH, oxidation – reduction potential, complexing ability of soluble and solid ligands and biological interactions (van Dorst and Peterson, 1984; McNeal and Balisterieri, 1989). The redox potential – pH relationship shown in Fig. 5 is based upon calculations of oxidation potential at varying pH for Se ions at a concentration of 1027 mol l21. Under most pH and redox conditions prevalent in normally productive cultivated soils of the world (area enclosed by solid lines), Se exists predominantly as the oxyanions, selenate and selenite. In poorly aerated acidic soils, inorganic Se predominates as the relatively insoluble selenide and elemental forms. Each line on the diagram represents equilibrium between the oxidized form written above the line and the reduced form written below it. The space between two lines is the stability field of the ion or molecule shown on the upper side of the lower line and the lower side of the upper line.

140

K. S. DHILLON AND S. K. DHILLON

Figure 5 Oxidation states of Se as a function of Eh–pH relationship for Se–H2O system at 1027 mol Se l21. Area enclosed by solid lines represents the normal range of characteristics of agricultural soils [adapted from Anderson et al. (1961)].

Speciation of saturation extracts from selected seleniferous San Joaquin Valley soils revealed that 98% of soluble Se was present as selenate (4 – 640 mg l21) followed by selenite (1 – 4 mg l21) and the hydrophobic organic Se (less than 1– 4 mg l21) associated with the humic acid part of the dissolved organic carbon (Fio and Fujji, 1990). Organic forms of Se such as seleno-amino acids represent an important source of plant available Se and selenomethionine is more bioavailable than selenocystine. In some Californian soils, nearly 50% of the Se may even be in the organic forms similar to analogues of S-amino acids (Abrams et al., 1990). Since the 1950s, the production of methylated derivatives of Se such as dimethyl selenide or dissolved organic selenide compounds through microbial processes has been noticed (Ganje and Whitehead, 1958). In the area where natural Se toxicosis has occurred, a significant portion of Se is present as selenate. In America, water soluble (selenate) varied from , 0.1 to 38 mg kg21 in seleniferous soils. Fractionation of native Se in seleniferous soils of Punjab, India, from where chronic selenosis has been reported, revealed that KCl extractable Se constitutes 6 –14% of the total Se, KH2PO4 extractable 11– 19%, 4 M HCl extractable Se 2– 7% and residual Se ranged from 67 to 76% (Dhillon and Dhillon, 1998).

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 141

2.

Geochemical Associations of Se and Mobility

The portion of total Se that is moved by water under toxic soil conditions and is available for plant uptake may be considered as mobile Se. Selenate-Se is soluble and, therefore mobile in soil at most pH values because it is weakly adsorbed by soil particles (Ahlrichs and Hossner, 1987) and thus is liable to leach out easily from alkaline soils. The mobility of organic Se compounds, such as seleno-amino acids and proteins has not been studied in detail. Only recently Gustafsson and Johnsson (1992), while studying the speciation of organic Se in relation to the degree of soil humification, reported that in Swedish forest soils most Se was associated with hydrophobic fulvates, which obviously leached down to lower layers. Polysulfides and thiols in mildly alkaline sulfidic pore water can increase Se mobility (Weres et al., 1989). The movement of Se to a depth of 0.66 m in the sediment at the Kesterson reservoir indicated that Se could be mobile even under strongly reducing conditions, possibly as a result of increased Se solubility in the presence of hydrogen sulfide. Clay minerals, organic matter and hydrous oxides of Fe, Al and Mn play an important role in chemical processes controlling the mobility of Se in soils. Minerals like vermiculite, gibbsite, geoethite are considered as main sinks for selenite-Se (Hamdy and Gissel-Nielsen, 1977). Christensen et al. (1989) observed that after 1 h of addition, maximum selenite-Se was fixed to clay particles (64 – 65%) followed by silt (45 – 61%) and sand (, 5%) particles. Selenium adsorption on clay minerals has been suggested to occur at the edges of clay particles (Bar-Yosef and Meek, 1987). Thus among the commonly found clay minerals in soils, kaolinite would be expected to adsorb more Se than other clay minerals as it tends to have a much greater fraction of its total surface area in the form of broken edges. While comparing the adsorption of selenite by different clay minerals, Hamdy and Gissel-Nielsen (1977) observed that kaolinite adsorbed more selenite than montmorillonite and vermiculite. While assessing solid phase speciation and geochemical transformations of soil Se, Sharmasarkar and Vance (1995) observed a positive relationship between total Se and clay content (r ¼ 0.81) and a negative relationship with sand content (r ¼ 2 0.69), thereby indicating an association of Se with clay sized particles in some range and mine soils from Wyoming. Among the oxides of Fe, Al and Mn, iron oxides, namely goethite, hematite, and gibbsite are the most abundant metal oxides in the soil. Goethite has been the most extensively studied metal oxide in relation to Se. Initially it was thought that Se in association with iron is present as basic ferric selenite, which is insoluble (Anderson et al., 1961). But, later on, Elrashidi et al. (1987) analyzed theoretically the equilibrium reactions and the formation constants for a wide range of Se minerals and found that metal selenite/selenate minerals, except MnSeO3, were too soluble and unstable to exist in most soils. In spite of this, ferric oxide – selenite complexes may play an important role in controlling Se

142

K. S. DHILLON AND S. K. DHILLON

solubility in alkaline soils where iron oxides adsorb selenate and selenite by forming inner- and outer-sphere complexes, respectively (Neal, 1995). Innersphere complexes are formed when an aqueous ligand exchanges for a surface hydroxyl group. This is described as specific adsorption (Eq. (1)). In the first step, the surface is protonated by a proton from the diprotic acid (also the source of ligand itself) and then the water molecule is exchanged for the ligand, e.g., selenite. Bonding in an inner-sphere complex is either ionic or covalent, and sometime can be a combination of both. It is this bonding that makes the complex stable. XOH ðSurface hydroxyl groupÞ

þ þ SeO22 ! 3 þ H Ligand

Proton

XSeO22 3 ðInner-sphere complexÞ

þ H2 O

ð1Þ

Water

Outer-sphere complexes are formed when a water molecule is retained between the surface site and the adsorbed ligand (Eq. (2)). This is also known as ion-pair formation and is described as non-specific adsorption. The water molecule in this case is incorporated into the complex itself. Bonding in an outer-sphere complex is usually of an electrostatic nature and is far weaker than the ionic or covalent bonding seen with inner-sphere complexes, thus resulting in a less stable complex. XOH ðSurface hydroxyl groupÞ

þ 22 þ SeO22 ! XOHþ 4 þ H 2 – SeO4 Ligand

Proton

ð2Þ

ðOuter-sphere complexÞ

While studying the adsorption of selenite on allophane and hydrous alumina, Rajan (1979) observed that increased adsorption of selenite results in a subsequent increase in the amount of hydroxyl ions released from the surface sites. In contrast to pure systems, the ligands in a soil are not restricted to OH2 and water molecules. Other anions may also play an important role. During selenite adsorption on alkali soils, there was a small increase in the pH of the equilibrium solution; but in acidic soils a conspicuous increase in pH of the equilibrium solution was observed (Fig. 6; Dhillon and Dhillon, 1999). The presence of anions and cations in the soil solution also affected the nature of Se adsorption by competing for the available binding sites. Benjamin (1983) observed an increase in selenate adsorption on amorphous iron oxide with the addition of cations, such as Cd, Cu, Co and Zn. Balisterieri and Chao (1987) investigated the effects of various anions on selenite adsorption by goethite and observed the following selectivity sequence: Phosphate . Silicate . Citrate . Molybdate . Bicarbonate/Carbonate . Oxalate . Fluoride . Sulfate. Hingston et al. (1971) concluded that both goethite and gibbsite had two types of binding sites: (i) the sites to which both selenite and phosphate competitively adsorbed, and (ii) the specific sites to which only either selenite or phosphate could adsorb. Studies on selenite adsorption in seleniferous soils of northwest India revealed that adsorption of Se decreased in the presence of different anions;

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 143

Figure 6 Percent Se sorbed as a function of equilibrium pH in some acidic soils of India [adapted from Dhillon and Dhillon (1999)].

the per cent decrease ranged from 3 to 21 at 10 mg SO4-S ml21, 8– 40% at 60 mg NO3-N ml21 and 32 –56% at 15 mg H2PO4-P ml21 (Dhillon and Dhillon, 2000b). The fact that selenite can be desorbed by phosphate over a large pH range suggests that application of phosphate fertilizer may seriously affect the Se status of soils. Desorbed Se could be easily oxidized to selenate that in turn could easily leach out due to the higher solubility of selenate. Probably this phenomenon will be weaker in low pH soils as phosphate competes more effectively at pH above 7.

B. S e

IN

WATER

Selenium naturally enters the food chain through water. The upper limit of Se set by the USEPA for drinking water is 10 mg Se l21 and that for water used for irrigation purposes is 20 mg l21. The current ambient water quality criteria for wildlife are 5 mg Se l21. However, keeping in mind the significance of bioaccumulation risks to wildlife, Peterson and Nebekar (1992) has proposed 1 mg Se l21 water-borne Se as the toxicity threshold for sensitive birds and 22 mammals. As with soils, under most pH and redox conditions, SeO22 3 and SeO4

144

K. S. DHILLON AND S. K. DHILLON

are the dominant forms of Se in water (Cutter and Bruland, 1984). This is also true for saline, alkaline evaporation pond waters in the San Joaquin Valley of California (Presser, 1994). However, stability of different states of Se is related to the electrochemical potential of the water. From the calculations based solely on pure Se mineral phases, White and Dubrovsky (1994) reported that solid-state elemental Se predominates at potentials less than þ 0.270 mV and that aqueous selenate and selenite would be stable at potentials greater than þ 270 mV. Marine studies have shown that both selenate and selenite are present in seawater, with higher concentrations in deep waters than at the surface. The National Academy of Sciences in the USA (1983) reported that Se is present in drinking water at concentrations ranging from , 0.1 to 100 mg l21. However, water derived from Cretaceous geological zones may contain as much as 1000 mg Se l21 (Mayland et al., 1989). In different geographic regions, the Se content in rainwater varied from , 0.001 to 2.5 mg l21 (Robberecht et al., 1983). Drainage water, essentially a soil leachate, formed an important source of Se in the San Joaquine Valley of California, USA, when it was being collected for re-use for irrigation purposes (Presser, 1994). Deveral et al. (1994) investigated the processes affecting the distribution and mobility of Se in groundwater at a regional scale and corroborated the results by studying the individual farm fields in the Diablo Range alluvium in the San Joaquin Valley. Regional scale studies demonstrated that Se is present in highly mobile selenate in the oxidized Diablo Range alluvium and concentrations of Se (up to 4200 mg l21) in the groundwater are the result of evapoconcentration of shallow ground water. In contrast, the oxidation– reduction reactions in Sierra Nevada sediments reduce Se mobility by transforming selenate to the reduced immobile form. Further studies on individual farm fields in Diablo Range alluvium revealed that when drainage systems were installed to stop the increase in soil salinity, evapoconcentrated groundwater with high Se concentration was displaced downward by good quality water that infiltrated after the installation of drainage systems.

C. S e

IN

PLANTS

Selenium is not an essential element for plant growth, but its concentration in fodder and grain crops is important to animal and human health. Discovery of Se as an essential element for animals has stimulated surveys on the Se concentration in field crops in several countries. For proper nutrition of both human and domestic animals, it is desirable to keep the Se concentration in food and feed between 0.1 and 1.0 mg kg21 (Allaway, 1968). The minimum nutritional requirement of animals is about 0.05 –0.10 mg Se g21 in dry forage feed and intake below that might cause severe deficiency diseases (Muth, 1963). Higher levels of 0.1– 1.0 mg Se g21 feed seem to offer protection against some diseases like cancer and cardiovascular diseases (Combs and Combs, 1984) and

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 145

exposure to still higher levels, 2 –5 mg Se g21 feed result in toxic effects (Gissel-Nielson et al., 1984). Accumulation of Se by different organisms in their tissues to concentrations that are substantially higher than those in the environment is termed bioaccumulation (Tinsley, 1979). In aquatic systems, bioaccumulation may occur by direct absorption or mainly by ingestion of contaminated foods. Accumulation of Se depends upon plant species, environmental conditions, age and phase of plant growth and the nature of Se compounds (Rosenfeld and Beath, 1964; Girling, 1984; Dhillon and Dhillon, 1997b). More recently, general biological interest in Se has focused on its metabolic functions in animals and humans. Comprehensive reviews on the biochemical role of Se have been published by Stadtman (1974), Brown and Shrift (1982) and Anderson and Scarf (1983).

1.

Plant Species

On the basis of their ability to accumulate Se, plants have been divided into three groups (Rosenfeld and Beath, 1964): (i)

Primary accumulators. Also known as Beath’s Se indicators or converter plants and may accumulate large amounts of Se ranging from 1000 to 10,000 mg kg21 (dry weight). These include species of Astragalus, Machaeranthera, Haplapoppus and stanleya genera. (ii) Secondary accumulators. Plants which rarely contain more than a few hundred milligrams of Se per kg (dry weight). Species of Astor, Atriplex, Castillega, Mentzelia and some species of Astragalus and Machaeranthra genera come under this category. (iii) Non-accumulators. Plants that may accumulate moderate amounts of Se, a maximum of 30 mg kg21. Most of the cultivated crop plants, fodders, grains crops and native grasses belong to this group. Typical agricultural crops have much lower tolerance to Se. Until recently there was no documented evidence showing damage to agricultural plants in the field, when grown on naturally occurring seleniferous soils. Recently, typical symptoms of Se toxicity—snow white chlorosis with pink coloration on the lower side of the leaf and sheath have been recorded on wheat (Triticum aestivum ) plants growing on naturally occurring seleniferous soils of Punjab, India (Dhillon and Dhillon, 1991a). Most plants contain , 1 mg Se kg21 when grown on nonseleniferous soils (Girling, 1984) but may accumulate several times more Se when grown on seleniferous soils. The available information generated through field surveys on relative accumulation of Se by agricultural crops growing on Se toxic soils is summarized in Table VII.

146

K. S. DHILLON AND S. K. DHILLON

Table VII Selenium Concentration (mg kg21, on dry weight basis) in Grains, Vegetables, Forages and Fruits Grown in Seleniferous Regions in Different Countries Crops

Irelanda

Grains Wheat Corn Rice Barley Oat Pearlmillet Cotton

Columbiab

155 40

Fruits Guava Galgal Lemon Nut (kernals) a

Israeld

USAe

Chinaf

Indiag

22

1–30 1–20

1.1h 11.6 1.2

13– 66 9– 35 5– 9 28– 38 4– 41 28– 33

2–17 2–15 25–137

Vegetables Radish Carrot Cabbage Peas/beans Potato Brinjal Bell pepper Red pepper Cucurbits Garlic Onion Forages Egyptian clover Mixed herbage Common vetch Sorghum Sugarcane leaves Corn Mustard (tops) Cereal strawsi

Mexicoc

0.1 –3.4 15

1.4 0.23–0.55 2.3 –4.5 0.2 –2.0 0.2 –0.9

15– 23 14–46 2.9h 0.4h

0.06–0.2 8– 48 8– 51 25– 36 0.4–18 3–30 1–450

11–44

1 –200 0 –84

21.9

0.6–30 40– 81

110–136 5– 12 8– 67 0.2– 9.0 80– 160 3– 58 1 6 11.9 0.01–0.2

Source: Walsh and Fleming (1952). Source: Ancizar-Sordo (1947). c Source: Williams et al. (1940). d Source: Ravikovitch and Margolin (1957). e Source: Rosenfeld and Beath (1964), Burau et al. (1988) and Carlson et al. (1988). f Source: Ribang et al. (1992) and Combs and Combs (1984). g Source: Dhillon and Dhillon (1991a,b, 1997a) and Ghosh et al. (1993). h Values on fresh weight basis. i Includes straw of wheat, oat, barley and rice at maturity. b

25– 28 0.9– 1.3 20– 51

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 147

2. Se Metabolism in Plants Selenium accumulating and non-accumulating plants seem to differ largely in the process of Se metabolism. Peterson and Butler (1962) observed very little incorporation of Se into amino acids in the proteins of Se accumulating plant species N. amplexicaulis; and extensive incorporation of selenoamino acids in ryegrass, wheat and clover (non-accumulating spp.). They suggested that Se accumulators have evolved a detoxification mechanism whereby Se may be excluded from protein incorporation. But non-accumulators do not have this mechanism. Thus incorporation of Se into proteins resulted in the alteration of protein structure, leading thereby to inactivation of the proteins and eventual poisoning of the plants. Apparently the toxic effects of Se to plants are caused by interference with S metabolism. The replacement of S containing cysteine and methionine with Se-amino acids can disrupt the normal biochemical reactions and enzyme functions within the cell. Due to sufficiently large differences between the chemical properties of selenocysteine, selenocystine and their S analogues, replacement of a Se analogue with a S analogue in a protein molecule must result in a marked alteration of the protein structure and function. Chow et al. (1971) studied the biosynthesis of S-methylcysteine and Se-methylselenocysteine by varying the concentration of Se and S in the growth medium. Increasing the concentration of selenate at a constant sulfate concentration resulted in increasing quantities of Semethylselenocysteine. The reverse was true when sulfate concentration was increased at a constant concentration of selenate. Keeping this in mind, they suggested that a common enzyme system is involved in the synthesis of both Smethyl cysteine and Se-methylselenocysteine. Anderson and Scarf (1983) suggested that the most sensitive components of the cell to Se toxicity are those that require S for some specific and essential functions, e.g., the –SH group of some enzymes loses its reactivity when S is replaced by Se. Selenate uptake has been recognized as an active transport process (Ulrich and Shrift, 1968). Selenite transportation is very complicated. Unlike selenate, a high proportion of the extractable Se accumulated by excised roots of Astragalus was no longer selenite (Shrift and Ulrich, 1969) and has been converted to other forms. It is assumed that metabolic steps that occur with selenate and sulfate assimilation are similar. However, the intermediate steps were not clearly understood. Only recently, Terry et al. (2000) have collected the relevant information available in the literature and constructed the possible pathways for Se assimilation and volatilization in Se accumulators and non-accumulators and also identified the rate limiting steps. Akagi and Campbell (1962) postulated that adenosine-3-phosphate (ATP) reacts with the selenate ion to produce adenosine50 -phosphoselenate (APSe). Studies conducted by Pilon-Smits et al. (1999a) at the molecular level have provided the first in vivo evidence that ATP sulfurylase is responsible for selenate reduction and this enzyme is rate limiting for both

148

K. S. DHILLON AND S. K. DHILLON

selenate reduction and Se accumulation. Selenate reduction may be described as below: ATP-sulfurylase

ATP þ SeO22 ¼¼) APSe þ PP 4 (¼¼¼¼¼

ð3Þ

By analogy to the S assimilation pathway (Eq. (4)), APSe could be reduced to selenite with GSH by APS reductase and then to selenide by the enzyme sulfite reductase using ferredoxin as a reductant. APS-kinase

APS þ ATP (¼¼¼¼)

PAPS #

þ ADP

ð4Þ

ð30 -phosphoadenosine-50 -phosphosulfateÞ

The selenium analogue PAPSe is expected to be synthesized in higher plants, but it has not been detected so far. Nissen and Bensen (1964) have suggested that selenate could be metabolized to amino acids through activation as adenosine phosphoselenate. On the basis of the information available in the literature, Terry and Zayed (1994) recently proposed a five step mechanism for the assimilation of selenate leading to the synthesis of selenoproteins and finally to volatile Se compounds by accumulating and non-accumulating plants. The steps are (i) reduction of selenate to selenite, (ii) reduction of selenite to selenide, (iii) conversion of selenide to selenocysteine, (iv) conversion of selenocysteine to selenomethionine, and (v) conversion of selenomethionine to the volatile Se compound DMSe. The simplified version of different steps in the form of a continuous process is given below: 22 NADPH 22 ATP GSH NADPH O-AS SeO22 4 ! APSe ! SeO3 ! GSSeSG ! GSSeH ! Se CH3

! Selenocysteine ! Selenocystathionine ! Selenohomocysteine

! Selenomethionine SAMð – CH!3 Þ MSeMS ! DMSe ðgasÞ 3.

Phytoavailability of Se in Seleniferous Soils

Bioavailability/phytoavailability of a chemical in the environment is defined as the fraction of the total contaminant in the interstitial water and soil particles that is available to the receptor organism with the extent of the bioavailable fraction varying with time, the nature of soil types, organisms and the environmental factors (Naidu et al., 2001). Studies conducted at CSIRO, Adelaide have amply demonstrated that there is a need for both biological and chemical assays for an accurate estimation of contaminant bioavailability (Megharaj et al., 2000). Various methods have been used to determine effective

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 149

indices of Se availability to plants by correlating total soil Se, Se extracted by particular extractants and plant uptake. Early investigations by research workers in South Dakota observed no significant relationship between total and watersoluble Se as well as plant uptake under field conditions; but working with homogenous soils in the greenhouse experiments, absorption of Se by plants was correlated with the amount of water-soluble Se in the soil (Rosenfeld and Beath, 1964). The greenhouse studies on Se availability to perennial rye grass (Lolium perenne L.) grown on some potentially toxic organic soils from Ireland revealed that EDTA extractable Se was highly correlated with plant concentration (Williams and Thornton, 1973). Soltanpour and Workman (1980) evaluated the use of ammonium bicarbonate-diethyleneamine pentaacetic acid (AB-DTPA) as an index of Se availability for alfalfa in North Dakota soils. Selenium concentration of alfalfa in the greenhouse showed a high degree of correlation with AB-DTPA extractable Se. Under field conditions, Soltanpour et al. (1982) reported a highly significant correlation coefficient (R 2 ¼ 0.82) between AB-DTPA extractable Se from the 0 to 90 cm depth and Se content of wheat grains. AB-DTPA may bring watersoluble Se into solution as a result of evolution of CO2 from the open flask and a rise in pH from an initial value of 7.6 – 8.5. Some organic constituents of low molecular weight may also get dissolved. In addition to this, the bicarbonate anion can also exchange with selenate or selenite. Jump and Sabey (1989) evaluated seleniferous soils for their Se supplying capacity using chemical extractants including AB-DTPA, DTPA, sodium carbonate, a saturation extract and hot water. Soil samples were collected from specific locations in Wyoming with documented histories of Se toxicity in livestock. Sodium carbonate extracted the largest fraction of total soil Se among the five soil tests. The selenium content of saturation extracts was found to be the most useful in predicting Se concentration in both four-wing saltbush (R 2 ¼ 0.66) and two-grooved milk vetch (R 2 ¼ 0.78). The results suggest that soils that yield . 0.1 mg Se l21 in a saturation extract may produce Se toxic saltbush plants, i.e., plant containing . 5 mg Se kg21. In soils collected from Se toxic sites located in the seleniferous region in Punjab (India), Dhillon et al. (2001) observed a highly significant relationship between hot water soluble Se and the Se content of raya (Brassica juncea var. compestris L.) (r ¼ 0.703pp ) and wheat (T. aestivum L.) (r ¼ 0.693pp ) in greenhouse experiments. These alkaline soils contain appreciable quantities of selenate-Se (Pareek et al., 1998); the selenate salts being highly soluble are easily available to plants. With increasing interest shown by consumers and producers towards the use of organic amendments as a substitute for conventional fertilizers, the use of compost in agriculture is increasing every year. Absorption and accumulation of compost borne Se by plants depends mainly on their concentration and speciation in the compost, media pH, and plant species. It is, however, not clear

150

K. S. DHILLON AND S. K. DHILLON

from the literature whether the application of compost to agricultural land would reduce the bioavailability of Se. Guidelines for the maximum concentration of Se in various composts have been established in a number of countries. Currently, the Canadian guidelines allow a maximum concentration of 2 mg Se kg21 in Type AA compost and 14 mg Se kg21 (air-dried) in type B compost. Warman et al. (1995) studied the effect of different levels of biosolid-amended compost (0, 25, 50, 75, 100%) on the bioavailability of Se to Swiss chard. The compost/biosolid mixture contained an average of 3.6 mg Se kg21. Selenium content in the plants grown on the control soil was undetectable. With an increasing application of biosolid –compost mixture, Se concentration in the plant tissue also increased, to an average of 0.29 mg kg21. The increase in tissue Se was, however, not proportional to the increase of Se concentration in the growth medium. Total and DTPA extractable Se increased with the increased application of compost –biosolid mixture to the medium and were highly correlated. Cappon (1987) investigated the uptake and speciation of Se in vegetable crops grown on garden plots exclusively treated with residential compost for 6 years and observed that Se is being readily absorbed by plants. Of the total Se, 20% was present in the hexavalent form and the rest was in divalent and tetravalent forms. Karam et al. (1998) observed a slight increase in the Se content of potato tubers when grown on municipal solid waste-amended substrates.

D. Se

IN THE

ATMOSPHERE

The global distribution of Se is relatively uniform with concentration ranging from 5 ng m23 in urban areas and 0.05 to 100.1 ng m23 in remote marine and continental areas (Mosher and Duce, 1989). Investigation into sources of atmospheric Se has indicated that much of this Se is produced by natural processes and anthropogenic activities such as fossil fuel burning or mining of sulfide ores or biological methylation (Mackenzie et al., 1979). The volatile forms of Se—the organoselenium species identified in the ambient atmosphere are mainly DMSe along with small quantities of dimethyle-di-selenide (DMDSe). DMSe and DMDSe have also been detected in the vicinity of moist seleniferous soils (Frankenberger and Karlson, 1989), seleniferous pond waters (Thompson-Eagle and Frankenberger, 1990) and various plant species (Terry et al., 1992). Demethylselenide is 500– 700 times less toxic than aqueous selenite and selenate ions. The inhaled DMSe vapor is non-toxic to rats at concentrations up to 34,000 mg m23 (Frankenberger and Karlson, 1994). As soon as the organoselenium species are emitted to the atmosphere, they get rapidly oxidized into fine particles for long-range transport. DMSe can be absorbed through the leaves of plants and transported to roots. It can also be sorbed on soil that may act

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 151

as an important natural sink for atmospheric DMSe (Zieve and Peterson, 1985). Recent analyses of archived soil samples in rural southeast England indicate that soil Se concentrations have increased by approximately 25% in the last 130 years solely due to atmospheric inputs (Haygarth et al., 1991)

IX. EFFECTS OF Se TOXICITY ON THE COMPONENTS OF THE ECOSYSTEM A. PLANTS Concern about selenosis in plants did not really develop until the identification of selenosis in livestock. General agricultural crops have a much lower tolerance of Se. Selenium toxicity symptoms have been studied mainly under laboratory and greenhouse conditions. Characteristic symptoms of selenate injury in soil, sand or solution culture in wheat plants are a snow-white chlorosis of the leaves (Rosenfeld and Beath, 1964). Till the 1970s, there were no documented instances of naturally occurring Se related damage to agricultural plants in the field (NAS– NRC, 1976). Recently, typical symptoms of Se toxicity—snow-white chlorosis with pink coloration on the lower sides of leaves and the sheath of 45– 60-day-old wheat plants growing under field conditions (Plate 1), have been

Plate 1 Selenium toxicity symptoms of snow-white chlorosis in wheat grown in a rice– wheat sequence under field conditions in seleniferous regions of northwestern India.

152

K. S. DHILLON AND S. K. DHILLON

reported by Dhillon and Dhillon (1991a). Wheat plants exhibiting varying degrees of toxicity symptoms contained Se ranging from 107.5 to 262.4 mg kg21 in shoots and from 29.2 to 66.5 mg kg21 in roots. It was also observed that leaves containing , 100 mg Se kg21 remained lush green in color. Toxic effects of Se to plants are caused largely by interference with S metabolism. Se readily replaces S in the amino acids thereby resulting in the formation of seleno-amino acids. The similarity of Se-amino acids to their S analogue of cysteine and methionine can disrupt the normal biological reactions and enzyme functions within the cell. Anderson and Scarf (1983) suggested that most sensitive components of the cell to Se toxicity are those requiring S for some specific and essential reaction, e.g., the – SH group of some enzymes loses its reactivity when Se replaces S.

B. ANIMALS Se intoxication to animals has been described by Rosenfeld and Beath (1964) as: (i)

Acute poisoning. Due to ingestion of excessive amounts of primary accumulators and indicator plants containing high amounts of Se leading ultimately to death within a short period. (ii) Chronic poisoning. When Se toxicity results from the consumption of forages containing excessive Se for relatively longer periods. It can be further divided into two categories: (a) Blind Staggers—results from consumption of moderate amounts of indicator plants over extended time periods. Initially the animal wanders in circles, becomes anorexic and vision gets impaired, followed by weakness of legs, paralysis of tongue, varying degree of blindness, abdominal pains, excessive salivation, emaciation and death. (b) Alkali disease—due to ingestion of plants such as grasses or small grains containing 5 –40 mg Se kg21 over a extended period and is characterized by dullness, lack of vitality, emaciation, rough coat, loss of hair, hoof changes, lameness and reduced reproductive performance. Chronic selenosis in domestic animals in Punjab, India has been recently reported by Dhillon and Dhillon (1991a). Loss of hair, horn and hoof abnormalities leading to cracks and detachment (Plate 2), tail necrosis, and a disturbed reproductive cycle were the main symptoms observed in domestic animals suffering from chronic selenosis when feeding on Se rich fodders containing 5– 160 mg Se kg21. Selenotic animals had 35% less hemoglobin and were suffering from macrocytic and hypchromic anemia (Dhillon et al., 1992b). Investigations on the pathophysiology of affected animals revealed that chronic

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 153

Plate 2 Selenium toxicity symptoms on horns and hoofs of animals as observed under field conditions in seleniferous regions of northwestern India.

selenosis results in chronic hepatitis and impairment of hepatic and renal functions (Randhawa et al., 1992). Post-mortem examination of various tissues of animals fed with seleno-urea (0.15 mg kg21 body weight) revealed deposition of Se in large amounts in the liver and kidney (Dhillon et al., 1996) and this must be responsible for malfunctioning of these vital organs.

154

K. S. DHILLON AND S. K. DHILLON

Recent studies on the bioaccumulation and effects of Se on wildlife at the Kesterson National Wildlife Refuge in Merced County, California (Ohlendorf, 1989) revealed a high incidence of embryotoxicity and mortality of adult birds. Ponded water in the reservoirs contained 50– 200 mg Se l21. The developmental abnormalities in embryos included missing or abnormal eyes, beaks, wings, legs and feet. The probability of embryo death or deformity increased significantly as Se concentration in the egg increased. Compared to the background level, the mean Se concentration was significantly higher in body tissues of several wildlife species including invertebrates, fish, frogs, snakes and mammals collected from the Kesterson reservoir. About 1000 migratory birds died at the Kesterson Reservoir during 1983 – 1985 as a probable result of feeding on plants, invertebrates and fish with a mean Se concentration from about 20– 120 times more than normal (20 – 170 mg kg21). Dey et al. (1999) studied the wild leopard cat (Felis bengalensis ), civet cat (Vivera zebitha ), flying squirrel (Petaurista magnificus ) and leopard (Panthera pardis ) from different locations in northeastern India. The content of Se was found to be high in the bones and in the hair it exceeded (1.5 – 2.6 times) toxic limits in all four species. As a result of increased Se content in the animal body, certain known toxic effects of Se such as brittleness of hair, appearance of blisters and eruptions on the skin, loss of long hair, loss of appetite and the tendency to wander in circles were also recorded in all four of the species of wild animals studied.

C. SELENIUM TOXICITY

IN

HUMANS

Human toxicosis resulting from dietary Se was unobserved up to 1980 (NRC, 1980). Klasing and Schenker (1988) described some cases of acute, sometimes fatal, Se toxicity as a consequence of self-medication or accidental consumption and also as a result of long-term industrial exposure. However, the best evidence of chronic Se toxicosis as a result of excessive dietary intake of Se was provided by Yang et al. (1983) from China. The authors have reported that a disease characterized by hair and nail loss in humans occurred in the 1960s in some villages of Enshi County of Hubei province. Corn was identified as the main toxic dietary constituent. The population suffering from chronic selenosis exhibited prominently the loss of hair and nails. In the affected individuals, hair was found to be brittle, easily broken and the new hair lacked pigment. Nails were also brittle with white spots and longitudinal streaks and eventually broke-off. Eruptive skin lesions occurred with reddish pigmentation that frequently remained after the lesions had healed. Tooth mottling was evident in approximately 1/3 of the affected population. Numerous neurological signs, frequently accompanied by gastrointestinal disturbances were observed in one village that had a particularly high prevalence of selenosis.

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 155

Hair, blood and urinary Se levels in the affected population were found to be 30– 100-fold higher in areas of chronic selenosis compared with Se adequate areas. Individual daily dietary intake of Se in this region ranged from 3.20 to 6.69 mg with an average value of 4.99 mg compared with 0.75 mg in high Se areas without selenosis and 0.116 mg in Se adequate areas. The factors responsible for this particular outbreak of selenosis as identified by the authors were: drought leading to increased consumption of high Se foods, increased use of plant ash as an amendment for increasing crop yields and decreased protein intake resulting in increased susceptibility to Se toxicity. In the seleniferous areas of northwestern India, out of 20 humans studied, 55% showed loss of hair from the body, particularly the head, malformation of fingers (Plate 3) as well as toe nails and progressive deterioration in general health (Dhillon and Dhillon, 1997a). Others complained of occasional severe headaches and nausea. In some of the affected humans, fingernails were completely damaged and blood was oozing out from the fingertips. Humans of all ages were affected by Se poisoning. The selenium content of hair and nails of affected persons was eight to nine and six to eight times higher than the healthy persons (Table VIII). Farmers in the seleniferous region did observe that even leaving the region temporarily for 3 – 4 months resulted in a remarkable recovery from Se poisoning. Similar observations have also been reported by Yang et al. (1983). Obviously, when the source of food was changed from seleniferous to nonseleniferous regions, the intake of Se was reduced. Recently, on the basis of the Se content of the daily diet being consumed by farm families in the seleniferous region of northwestern India, the average daily

Plate 3 Deformed finger nails of human beings as observed in seleniferous regions of northwestern India.

156

K. S. DHILLON AND S. K. DHILLON

Table VIII Se Content (mg kg21) of Hair and Nails of Humans Suffering from Chronic Selenosis in Northwestern India Hair

Nails

Humans

No. of samples

Se content

No. of samples

Se content

Selenotic Healthy

11 4

3.7–25.5 (13.1 ^ 9.7) 0.9–3.0 (1.9 ^ 1.1)

6 5

8.5 –25.7 (15.8 ^ 7.2) 1.6 –3.9 (2.3 ^ 0.8)

dietary intake by men and women was found to be 632 ^ 31 and 475 ^ 53 mg day21, respectively. The corresponding values from non-seleniferous areas were 65 ^ 2 and 52 ^ 1 mg day21, respectively. The US Food and Nutrition Board (1980) suggested 70 and 50 mg Se as the daily requirement of men and women.

X. MANAGEMENT OF SELENIFEROUS SOILS Until the 1960s, when soils with elevated levels of Se were confined to predominantly dry and non-agricultural regions, the management of toxic soils was limited to the mapping of seleniferous soils, withdrawing from cultivation all food plants and maintaining them as fenced farms, selection of safe routs for trailing of livestock, eradication of Se accumulators, etc. (Rosenfeld and Beath, 1964). Thereafter, the research emphasis shifted to the identification of sources of Se contamination, distribution in the environment and to understanding the mechanisms involved in regulating the transport and accumulation of Se in the soil – plant – animal/human system. More recently, the association of Se contamination with anthropogenic activities such as the use of fly ash (Adriano et al., 1980), metal refining (Nriagu and Wong, 1983), agricultural drainage water (Presser, 1994) and utilization of underground water (Dhillon and Dhillon, 1990), have caused research efforts of scientists to concentrate on finding out practical means of complete removal or immobilization of Se leading to its negligible movement out of the contaminated system. Seleniferous soils are otherwise highly productive except the fact that Se is being transported in toxic levels from soils through plants to animals and humans. Keeping this in mind, only those technologies will be suitable for the management of seleniferous soils which serve the twin objectives of gaining a complete understanding of chemical/biological processes in Se rich soils/waters for minimizing its movement into the food chain and maintaining/restoring the use of Se-laden soils for intensive agricultural production. Various remediation

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 157

technologies have been developed to detoxify the contaminated soils and waters and are discussed in the following sections. Technologies available for remediation of inorganic contaminated sites can be grouped into two categories: (i) offsite management through excavation of contaminated material and transport offsite, and (ii) onsite management in order to reduce possible exposure risks. As a rule, only remediation technologies that may prove effective on a long term basis, are those that are designed to protect all components of the biosphere—land, air, surface water and underground water. Onsite management of the contaminant provides an inexpensive and rapid solution (Jefferis, 1992), in contrast to the long-time problems of contaminant transport offsite. Although offsite burial of soil contaminated with heavy metals is being extensively practiced in Australia (Smith, 1993). But it does not seem to be feasible in the case of Se contaminated soils due to the simple reason that (a) contaminated sites (Fig. 3) are relatively large and (b) the contaminant (Se) is distributed throughout the soil profile and is highly mobile especially in the case of alkaline soils. Total soil Se concentration is not the issue of concern in adopting a particular remediation technology. It is the concentration of mobile Se, usually selenate, that determines the need for a specific treatment and the extent of change in the mobile concentration is the measure of success. Various technologies designed for onsite management of Se contaminated soils are discussed in the following sections.

A. SOIL MIXING/COVERING This is the simplest form of onsite containment of Se and has been designed to reduce the hazards associated with inorganic contaminants in the soils through dilution to levels below which exposure is not considered a risk. This technique has been employed for the reclamation of seleniferous soils at Kesterson Reservoir at California, USA. When the discharge of agricultural drainage water was terminated at Kesterson Reservoir, the evaporation ponds were covered with a clean soil layer of 15 cm thickness having a Se content between 15 and 20 mg kg21. Effectiveness of the clean cover on the contaminated soil was monitored for 2 years in comparison to the native soil (Wu, 1994). At the two monitoring points, the soil below the covering layer contained 273 and 233 mg Se kg21. For the first year, the Se content of the cover soil at one of the sites increased from 20 to 600 mg kg21 and thereafter, it decreased. At the other site, Se content of the cover soil remained almost unchanged. Because all the plant species established in the new environments had a deep rooting system, absorption of large amounts of Se from the underlying contaminated layer was observed. At another site within the Kesterson Reservoir covered with a 0.53 m deep imported non-seleniferous soil in 1988, a significant upward movement of

158

K. S. DHILLON AND S. K. DHILLON

Se6þ into non-seleniferous soil was observed (Tokunaga et al., 1994). Although no detrimental effect of Se on wildlife has been reported from the area covered with clean soil, it cannot be conceived as a long-term measure. Obviously, Se remains in place and is accessible to deep-rooted plants and is prone to upward and downward movement with time. Another dilution technique relies on deep ploughing. This process allows the vertical mixing of the surface contaminated soil with less contaminated sub-soil and thus certainly reduces the level of contamination of the surface soil (Thompson-Eagle and Frankenberger, 1992). Since the distribution of Se in a soil profile does not follow a definite pattern, this dilution technique may have a very limited application in the management of seleniferous soils. It may prove useful to some extent under specific situations, e.g., when the level of contamination remains low especially in the surface layer; or it is one time contamination and the site is not repeatedly accessible to contaminant sources.

B. SOIL WASHING The technology of soil washing is applied widely in Europe for the remediation of metal contaminated soils and can be carried out onsite using mobile equipment. Soil washing technology for removal of contaminants is practically an adaptation of mineral processing technologies (Williford and Bricka, 2001). It is rather a preconcentration stage prior to disposal or further treatment. Soil washing may prove useful for remediation of alkaline soils where an appreciable quantity of Se is present in water-soluble forms. Through repeated extractions with KCl, it is possible to remove even up to 90% of native Se present in seleniferous soils of Punjab. However, it may not be feasible at all to wash thousands of hectares of land containing excessive Se throughout the soil profile. Soil washing may generate a huge amount of water with an excessive level of Se and even its temporary storage before final processing may prove to be a Herculean task. Moreover, the dangerous consequences as a result of storage of drainage water containing elevated levels of Se (the scenario at the Kesterson Reservoir) is just a recent happening fresh in our minds.

C. THERMAL TREATMENT Heat treatment of soils on a relatively larger scale is a well-recognized technology for remediation of organic contaminated soils. It may be applied to seleniferous soils, as Se volatilization is a temperature dependent phenomenon. Temperature requirements for maximization of volatilization are not that large as is the case for heat treatment of organic contaminated soils. In field investigations, seasonal fluctuations have been recorded for volatilization of Se

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 159

with rates being greater in warmer months, i.e., spring and summer seasons, than in fall and winter seasons (Thompson-Eagle and Frankenberger, 1992). Biomethylation is strongly affected by temperature and increases 2.6-fold for every 108C rise in temperature. The rate of Se methylation by Alternaria alternata increased with an increase in incubation temperature from 5 to 308C (Thompson-Eagle et al., 1989). A temperature coefficient (Q10) of 2.60 has been reported for soil microbial volatilization from Kesterson sediment with a temperature optimum of 358C (Frankenberger and Karlson, 1989). Similarly, Se volatilization from evaporation pond water was observed to be 1.31-fold greater at 358C than at 258C and the Q10 values ranged from1.78 to 1.86 with an average of 1.82. In each of these investigations, maximum volatilization occurred at the maximum temperature tested. It is quite obvious that the optimum temperature for Se volatilization has not been reached yet. In Asia, especially in northwestern India, the soil temperature in the 0 –15 cm layer may reach up to 458C (Singh and Sandhu, 1979) and thus still higher Se volatilization rates are expected.

D. IMMOBILIZATION

OF

Se

IN THE

TOXIC ENVIRONMENT

The bioavailable fraction of an inorganic contaminant is mainly responsible for the toxicological or environmental risks. Bioavailability of a toxic element may be defined as the chemical form that will affect the life cycle of plants and other organisms. Mench et al. (1994) observed that the remediation technologies involving the immobilization of inorganic contaminants might become more attractive if bioavailability is considered as a key property for defining a particular soil as contaminated or not. Selenium exists in the soil in several forms that differ in their solubility and bioavailability to plants. Onsite immobilization of Se can be achieved by manipulating soil pH, composition of soil solution and the relationship between adsorption and desorption of contaminants. It has been discussed in section “Seleniferous soils”.

E. PRESENCE

OF

COMPETITIVE IONS

IN

SOIL SOLUTION

Selenium accumulation by plants is significantly influenced by the presence of sulfate ions in the growth medium. The antagonistic interaction between sulfate and selenate for plant uptake was observed as early as 1934 by Hurd-Karrer. In recent decades, this relationship has been confirmed in the greenhouse (Mikkelsen et al., 1988; Bawa et al., 1990) and in actual field situations in pastures (Pratley and McFarlane, 1974), sugarcane (Dhillon and Dhillon, 1991b) and a rice – wheat cropping sequence (Dhillon and Dhillon, 2000a). Reduction in Se absorption by 60 –70% in a number of crops has been achieved by application

160

K. S. DHILLON AND S. K. DHILLON

of S from gypsum in alkaline calcareous seleniferous soils of northwestern India. Farmers of the region have adopted this practice as a practical measure for reducing the transfer of Se from soil to food chain crops. In California, the seleniferous drainage waters generally contain high levels of sulfate salinity. Mikkelsen et al. (1988) observed that plant Se was reduced from 620 mg Se kg21 to less than 7 mg Se kg21 in the presence of sulfate salinity.

F. SELECTING PLANTS

WITH

LOW Se ABSORPTION CAPACITY

In order to reduce the movement of Se into the food chain, plants absorbing the least amount of Se may be recommended for cultivation in seleniferous regions (Bawa et al., 1992; Dhillon and Dhillon, 1997b). Investigations on the Se absorption capacity of fodder crops commonly grown in the seleniferous region revealed that the differences in the Se content of fodders was negligible up to a level of 0.25 mg Se kg21 soil. At higher Se levels, the differences in Se accumulation became apparent. Oat (Avena sativa ) and sorghum (Sorghum bicolor ) among cereals and senji (Melilotus parviflora ) among leguminous fodders absorbed the least amount of Se compared to other fodder crops. In the case of multi-cut fodders like berseem (Trifolium alexandrinum ) and lucerne (Medicago sativa ), the first one/two cuts contain two to three times more Se than the following cuts. The farmers may be advised to avoid feeding of the first cut of berseem and the first two cuts of lucerne to animals.

G. PHYTOREMEDIATION Although the concept of phytoremediation is not new, it has become the topic of extensive research only recently. Phytoremediation has been defined as the use of green plants to remove pollutants from the environment or to render them harmless (Raskin et al., 1997). Chaney (1983) introduced the idea of developing a “phytoremediation crop” to decontaminate polluted soils emphasizing that value of metals in the biomass might off-set a part or all of the cost of cleaning up the toxic site. Incidentally by that time, birth defects in waterfowl were linked to excessive Se build-up in the Kesterson reservoir in California (Ohlendorf, 1989), this incident provided a strong incentive to scientists to establish phytoremediation as a new technology for the clean up of Se polluted soil and water. Extensive research work has been done at the Phytoremediation Research Laboratory at the University of California, Berkley and the US Department of Agriculture at Fresno, California. Within a short period, to consolidate the developments, an international symposium was organized at the University of California, Berkley in 1997 and the proceedings were published in the form of a book entitled “Phytoremediation of Trace Elements in Soils and Waters”. Earlier, an excellent

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 161

review of various vegetation management strategies for remediation of Se contaminated soils was provided by Frankenberger and Benson (1994). The development of technology and the increased knowledge of phytoremediation processes have led to the identification of five important mechanisms by which plants are believed to remove, degrade or stabilize the environmental contaminants. The mechanisms are: Phytoextraction/phytoaccumulation—plant uptake and assimilation of contaminants. Phytofilteration/rhizofilteration—use of plant roots to remove contaminants from flowing water. Phytovolatilization—use of plants to make volatile chemical species of contaminants. Phytodegradation/phytovolatilization—use of plants to make volatile chemical species of contaminants. Rhizodegradation—biodegradation of environmental contaminants by plant exudates. Phytostabilization—use of plants to transform soil metals so as to reduce their bioavailability and prevent their entry into the food chain. Except for phytostabilization, which appears to have strong promise for toxic elements like Cr and Pb, all other forms of phytoremediation have been successfully employed for decontamination of Se polluted soils.

1.

Phytoaccumulation

Selenium hyperaccumulating plants are known to exist since the problem of Se toxicity was recognized (Rosenfeld and Beath, 1964). However, the practicality of including Se accumulators in remedial strategies is limited, because they are (i) genetically poor, (ii) susceptible to pests and diseases, (iii) not responsive to fertilizer application, and (iv) seed is not commercially available (Parker and Page, 1994). In fact the plants that are the best Se accumulators are very small plants and do not produce high biomass (Ban˜uelos et al., 1997). The ideal plant species for phytoremediation of Se must be able to accumulate and volatilize large amounts of Se; grow rapidly and accumulate large biomass on the contaminated soil; and tolerate salinity and other toxic conditions (Terry et al., 2000). Intensive screening of different cultivated agricultural plant species revealed that Brassica sp. have most of the desired attributes compared to others and notable among them were Indian mustard (B. juncea czern. L.) and canola (B. napus ). As soon as plant roots absorb Se, it is translocated to shoots and the harvested biomass can be removed away from the site; thus leading to reduction in the Se levels in the soil.

162

K. S. DHILLON AND S. K. DHILLON

A pilot experiment was conducted (Ban˜uelos et al., 1998) to demonstrate the phytoremediation capacity of canola for Se contaminated soil at the Kesterson reservoir. The plots were directly seeded with canola at a density of 25– 30 plants m22. Total Se concentration prior to seeding was 25.5, 6.8 and 1.8 mg kg21 soil, respectively, at 0– 30, 30– 60 and 60– 90 cm depths. The corresponding values for extractable Se were 0.45, 0.32 and 0.60 mg l21. It was reduced to 0.01 mg Se l21 at all three different depths when canola was harvested after 90 days of growth. In another study conducted for 3 years, Indian mustard, tall fescue (Festuca arundinacea L.), birdsfoot trefoil (Lotus corniculatus L.) and Kenaf (Hibiscus cannabinus L.) were compared for their Se absorption capacities (Ban˜uelos et al., 1995). Indian mustard accumulated an average of 1373 mg Se kg21 DM in its shoots and 805 mg Se kg21 DM in roots each season; it was two to six times more than other species. Obviously, Indian mustard accumulated the highest total quantity of Se and thereby lead to almost a 50% reduction in total soil Se concentration at a depth of 75 cm after 3 years of experimentation. In another study, Ban˜uelos et al. (2000) compared the efficacy of Brassica sp. and barley to extract Se from soils irrigated with Se-laden effluent. Results indicated that plant accumulation of Se accounted for at least 50% of the Se removed in soils planted to Brassica and up to 20% in soils planted to barley. Although cultivation of Brassica species led to a significant reduction in Se added to soil through irrigation with Se-laden effluent, additional plantings are necessary to further reduce Se content in the soil. Unlike the association of salinity problems with Se contamination in soils and drainage waters in the San Joaquin Valley of California (USA), seleniferous soils in Punjab (India) have all the characteristics of a normal and highly productive soil except that it contained elevated levels of Se. Likewise, the underground water used for irrigation purposes is also of good quality, but contained two to three times more Se than the maximum permissible level at toxic locations (Dhillon and Dhillon, 1990; Dhillon et al., 1992a). Among cereals, wheat accumulated more than 80 mg Se kg21 DM without showing any toxicity signs and raya (B. juncea ) accumulated up to 160 mg Se kg21 DM. Compared to legumes and cereals, Brassica sp. absorbed the highest amount of Se from a naturally occurring seleniferous soil containing 2.5 mg Se kg21 in the greenhouse and is being utilized in phytoremediation experiments under actual field conditions (Dhillon et al., 2001). Agro-forestry farming practices offer another novel phytoremediation technique that may provide a long-term solution to the Se problem in irrigated lands (Cervinka, 1994). By growing trees like eucalyptus and poplar, Se locked in woody stems can be easily carted away long distances without any immediate danger of Se entering the environment. Compared with tree species like eucalyptus and casuarina, selected populus clones may be more effective in capturing and extracting Se from Se-laden effluents at lower salinity levels (Ban˜uelos et al., 1999).

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 163

The plant material generated through Phytoremediation processes contains significant levels of Se and poses a serious disposal problem. The possibility of using Se-rich plant material as animal feed in deficient areas was recently explored by Ban˜uelos and Mayland (2000). Lambs were fed on freshly cut canola grown on soils irrigated with drainage waters containing 75 – 100 mg Se l21. After 64 days feeding, although Se content in animal tissues (including blood) increased by two to three times compared to the control, it remained within safe limits. Thus the amount of Se provided to lambs through canola was not harmful. The level of Se in canola fed to lambs (1.9 – 3.6 mg Se kg21 DM) cannot be considered as hazardous for animals as it was much below the widely accepted maximum permissible level of Se, i.e., 5 mg Se kg21. However, the results indicate that Se-rich materials can be utilized for meeting the dietary requirements of animals in deficient areas if the level of Se in animal feed is kept within safe limits through carefully blending/supplementing with non-Se accumulating forages. In fact, the plant material obtained through phytoremediation experiments may contain 10– 20 times more Se than the maximum permissible level of Se in animal feed.

2. Phytovolatilization In addition to phytoaccumulation, phytovolatilization of Se is an important component in the removal of Se. Volatilization is an attractive method for removal of Se from contaminated sites because toxic organic and inorganic forms of Se are converted into less harmful gaseous forms. In fact, this is the only method, which ensures that toxic forms of Se are permanently removed from the contaminated site and its associated food chain. Almost all the superior Se volatilizers belong to the Brassicaceae family: cabbage (B. oleracea var. capitata ), broccoli (B. oleracea var. botrytis cv. Green Valiant), cauliflower (B. oleracea var. cauliflora ), Indian mustard (B. juncea Czern. L.) and Chinese mustard (and B. campestris var. chinesis ). Other plants, such as rice (Oryza sativa L.; Zayed and Terry, 1994) and hybrid poplars (Populus tremula H alba L.; Pilon-Smits et al., 1998) have also been shown to have high rates of Se volatilization. Under optimum laboratory conditions, these plants can generally volatilize Se at rates of 1.5– 2.5 mg kg21 DM day21 when supplied nutrient solutions contained 1.6 mg Se l21 (Ban˜uelos et al., 1997). In growth chamber studies, the plant roots volatilized 7 – 20 times faster than the shoots. The rate of volatilization from the detopped roots in case of broccoli, rice, cabbage, cauliflower and mustard was determined to be 1.5– 5 times faster than that of the intact root and in the following 48 h it further increased up to 20– 30 times the intact root rate (Zayed and Terry 1994; Terry and Zayed, 1994). While studying factors controlling Se volatilization by Indian mustard, de Souza et al. (1998) observed that plants supplied with selenite showed a rate of Se

164

K. S. DHILLON AND S. K. DHILLON

volatilization at least twice as great than plants supplied with selenate. The rate of uptake, however, was significantly higher for selenate than for selenite. Selenate was rapidly translocated to the shoot, whereas only about 10% of the selenite was translocated. Analysis of X-ray absorption spectroscopy revealed that selenitesupplied plants accumulated organic Se, most likely in the form of selenomethionine, whereas selenate-supplied plants accumulated selenate. They concluded that the rate of selenate reduction limits the Se volatilization from selenate and that from selenite is limited by selenite uptake and the conversion of selenomethionine to dimethylselenide.

3.

Rhizofiltration

Few studies have been done on rhizofiltration of Se by wetland species. Investigations by Zayed et al. (1998) have shown that duckweed (Lemna minor L.), a floating aquatic plant, could accumulate Se in concentrations comparable to other known Se accumulating species. In a similar study, water hyacinth (Eichhornia crassipes ) was shown to be a moderate accumulator of Se. Both of these plants are commonly used in constructed wetlands for wastewater treatment (Zhu et al., 1999). The choice of a suitable species is important in the clean up of Secontaminated wastewater in the constructed wetlands because different plant species vary considerably in their ability to absorb, accumulate and volatilize Se. During screening of aquatic plants for accumulation and volatilization of Se from industrial or agricultural wastewater, Pilon-Smits et al. (1999b) observed that the rate of volatilization of selenite was twice that for selenate with more selenate being translocated into the plant tissues than selenite. Several plants including: parrot’s feather (Myriophyllum brasiliense Camb.), iris-leaved rush (Juncus xiphioides ), cattail (Typha latifolia L.), and saltmarsh bulrush (Scirpus robustus ) showed accumulation and volatilization rates comparable with B. juncea, the best known plant for phytoremediation of Se. Phytoremediation, a combination of phytoaccumulation and phytovolatilization, as a clean up technology for Se contaminated soils is relatively new. It appears to be highly cost effective, eco-friendly and easily adaptable/adjustable in the normal cultivation systems. The cost of growing a crop is minimal compared to that of soil removal and replacement, thus use of plants to remediate seleniferous soils holds a great promise (Chaney et al., 1997). It could generate a lot of enthusiasm among the environmentalists due to its success based on laboratory and greenhouse studies and a few field experiments. However, it needs to be tested on a larger scale by conducting multi-year and multi-location experiments under field conditions. Now, disposal of seleniferous plant material containing . 5 mg Se kg21 DM is becoming a major problem. There can be two options:

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 165

(i)

Although fodders containing Se less than the maximum permissible level, i.e., 5 mg Se kg21, DM have been considered safe for animal consumption. To be still in a safest position on the long-term basis, it must be diluted at least to 1 mg Se kg21 DM level before its consumption by animals. A few studies have been conducted to analyze the effect of feeding Se-rich material on animal health. However, in one of the studies conducted by Ban˜uelos and Mayland (2000), the forage used for feeding contained only 1.9 – 3.6 mg Se kg21 that is even otherwise considered safe for animal consumption. In another study, symptoms of Se poisoning started appearing in animals just within 2– 3 weeks of feeding wheat straw grown on seleniferous soils containing 8.5 mg Se kg21 DM (Singh et al., 2001). (ii) Use of phytoremediating materials as a raw material for extraction of Se metal.

At present, Se is being obtained mainly as a by-product from processing of copper ores. Depending upon the level of Se in soil, phytoremediation crops such as Indian mustard could absorb from 150 to 1500 mg Se kg21 DM under field as well as laboratory conditions (Dhillon and Dhillon, 1991a,b; Ban˜uelos et al., 1998) and may be processed for extraction of Se. Processing of voluminous plant material harvested from seleniferous fields with relatively low levels of Se compared to copper ores may, however, lead to an increase in its cost of production. Keeping in view numerous applications of Se in the industries that are highly profit oriented, it may be possible to absorb the extra burden due to increased cost of production. It has an extra advantage of recycling the already released Se due to weathering of parent materials, etc. and thereby saving the environment from further contamination. If the same amount of Se is to be recovered by mining operations, inadvertently a significant portion may escape to the atmosphere, lithosphere and groundwater, thus further aggravating the situation. The idea floated by Chaney (1983) for recycling the extracted metal by plants for money, needs to be seriously tested.

H. BIOREMEDIATION The utilization of the potential of microorganisms for the degradation of toxic pollutants is the main objective of bioremediation. Recent investigations have proved its effectiveness beyond doubt for the removal of inorganic contaminants such as As, Cr, Hg and Se from the contaminated soils and aquatic system. In a critical review of earlier investigations, Doran (1982) emphasized that the existing microbial population in the environment is capable of producing methylated Se compounds. Keeping in view these observations and the challenges posed by the Se contaminated Kesterson Reservoir, Frankenberger

166

K. S. DHILLON AND S. K. DHILLON

and co-workers enthusiastically and relentlessly pursued research for accelerating the natural microbial processes that lead to Se volatilization from soil sediments and water. Their efforts culminated into the development of a biotechnology prototype for microbial bioremediation of seleniferous environments (Frankenberger and Arshad, 2001). In situ bioremediation of seleniferous soils as proposed by the team led by Frankenberger seems to be the most suitable and practical alternative and that completely fits into the criteria listed by Alexander (1994) as given below: (a) Existence of microbial consortia possessing the requisite catabolic activity. (b) Ability to sustain adequate microbial transformation rates for decreasing contaminant concentrations to regulated levels within an acceptable period of time. (c) Absence of metabolites that are either toxic to microbial communities, or the absence of by-products that are themselves of environmental concern. (d) Targeted contaminants are bioavailable. (e) Presence of environmental conditions that are conducive to microbial activity, including concentrations of nutrients, electron acceptors, and energy sources. (f) Cost effective when compared to other competing technologies that may be used at the site. Among the microorganisms isolated from a silty clay loam soil, 11% fungi, 48% actinomycetes and 17% bacteria were capable of reducing selenate and 3% fungi, 71% actinomycetes and 43% bacteria could reduce selenite to elemental Se (Bautista and Alexander, 1972). The nature of Se reduction can be either dissimilatory (reduction of Se compounds as terminal electron acceptors in energy metabolism) or assimilatory (when Se compounds are reduced and used as a nutrient source). Studies have shown that microbial uptake of sulfate and selenate can occur via the same mechanism whereas selenite uptake may be separate from this system (Brown and Shrift, 1980; Bryant and Laishley, 1988). Due to excess Se in the system, indiscriminate substitution of Se for its analogue S results in the formation of selenoamino acids such as selenomethionine and selenocysteine. Because Se compounds are less stable and more reactive than S analogues, the organism begins to experience metabolic problems, which ultimately leads to death. The incorporation of selenoamino acids into proteins is perhaps the major mechanism of Se toxicity to the system. Thus, assimilatory reduction may be of very limited use in removing Se from the contaminated system. Although selenoamino acids are water-soluble and can be removed from the system, but selenoamino acids can be more toxic than Se oxyanions (Besser et al., 1989). Dissimilatory selenate and selenite reduction to insoluble elemental Se is an important microbial transformation having practical implications for the

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 167

remediation of seleniferous soils and water (Oremland, 1994). The reduced Se usually appears as red intracellular deposits. Selenate respiring bacteria are ubiquitous in nature, functioning even in highly saline soils and sediments and the reduction reactions are unaffected by sulfate concentration (Oremland et al., 1991). Thauera selenatis isolated from Se contaminated drainage water in California’s San Joaquin Valley has been the most intensively studied selenatereducing microorganism (Macy et al., 1993). It conserves energy to support growth by coupling the oxidation of acetate to the reduction of selenate to primarily selenite. In the presence of selenate and nitrate, the selenite produced from selenate reduction is further reduced to elemental Se (DeMoll-Decker and Macy, 1993). Another selenate reducing microorganism designated SES-3 grows in a specific medium with lactate as the electron donor and selenate as the electron acceptor (Oremland, 1994). Still another Se respiring organism, Enterobactor cloacae strain SLD1a-1 (Losi and Frankenberger, 1997) is a facultative anaerobe, respiring selenate when grown anaerobically and reducing selenate to elemental Se. It can be filtered out of the system, collected and purified to commercial grade Se. Dungan and Frankenberger (1999) have emphasized that assimilatory reduction, demethylation and re-oxidation of Se must also be considered while evaluating different bioremediating technologies; they could have a significant impact on remediation of seleniferous environments.

1.

Bioremediation of Seleniferous Water

Based on the results of investigations carried out during the last decade, several microbial treatment technologies for ex situ remediation of Se contaminated water have been suggested for practical utilization (Gerhardt et al., 1991; Macy et al., 1993; Oremland, 1994; Lortie et al., 1992; Owens, 1997). It is difficult to compare the effectiveness of different technologies because of the variable conditions used. However, a common feature is that contaminated water is treated before disposal and includes a pretreatment step to remove nitrogen oxyanions. Water is passed through a system containing selenatereducing microorganisms. After immobilization, the elemental Se is separated out. Using pilot studies, the technologies have been found to possess a potential to be economically feasible. A bench-scale plug-flow bioreactor inoculated with mixed Pseudomonas cultures has been designed and tested by Altringer et al. (1989). The reduction of selenate into elemental Se is a two step reaction in which selenate is reduced to selenite and then possibly to selenide and eventually to red amorphous granules of elemental Se. After over 1 year of operation, a steady state of rate of Se removal from simulated San Luis Drain water averaged up to 86%.

168

K. S. DHILLON AND S. K. DHILLON

Oremland et al. (1991) have patented a two stage reduction process in the first stage of which algae depletes nitrate concentration in the contaminated water to , 1 mM under aerobic conditions. Water is then fed to an anoxic bioreactor containing selenate respiring bacteria where selenate is reduced to insoluble elemental Se. On an overall basis, Se levels of more than 50 mg l21 as selenate were reduced to less than 0.2 mg Se l21 in 7 days of incubation. In spite of a drastic reduction in the Se level, the outflowing water still contained Se, which is 10 times more than the maximum permissible level for irrigation purposes. Lortie et al. (1992) characterized a Pseudomonas stutzeri isolate which is capable of rapidly reducing both selenite and selenate into elemental Se at initial concentrations of both oxyanions of Se up to 48.1 mM. Optimal Se reduction occurred under aerobic conditions. The “Owens Process” (Owens, 1997) proposed a technology based on the use of methanol as a carbon source for supporting denitrifying as well as Se-reducing microbes. Selenium reduction will not take place until nitrate is consumed. After the consumption of nitrate, Se reduction proceeds under anaerobic conditions from selenate to selenite to elemental Se. Adams et al. (1993) conducted a pilot study using a bioreactor system consisting of a rotating biological contactor (RBC). When effluent from a base metal smelter, containing 30 mg Se l21 was treated with Escherichia coli in the bioreactor, 97% of the Se could be removed within a 4 h contact period. In a bench scale RBC system, treatment of mining effluents with P. stutzeri resulted in up to 97% removal of Se in a 6 h retention period. The process of bioremediation as proposed by Macy (1994) involved T. selenatis gen. nov. sp. nov., which is able to reduce nitrate and selenate simultaneously using a different terminal reductase. The effectiveness of T. selenatis in detoxifying Se oxyanions in San Joaquin Valley drainage water was tested in a specifically designed biological reactor system. The experiment lasted for 507 days including 60 days for adaptation. Under optimum conditions in a bioreactor (e.g., correct pH and ammonia level) in 286 days, T. selenatis could reduce selenate and selenite in drainage waters from 350 to 450 to an average of 5 mg Se l21 and that of nitrate was reduced from 260 to 330 to , 1 mg N l21. The microbial count at the end of the experiment confirmed that T. selenatis reduced selenate to selenite and the selenite reduction to elemental Se was probably catalyzed by nitrate reductases of both T. slenatis and the denitrifying microbes. A three step biological treatment process called “Algal –Bacterial Selenium Removal System” (ABSRS) to remove Se and nitrate from drainage waters was proposed by Lundquist et al. (1994). The results obtained in the laboratory were tested under field conditions by constructing a pilot ABSRS and the system was patented as the “Oswald Process”. Aerobic algal growth removes nitrates to , 10 mg N l21. In an anoxic unit, denitrifying and selenate respiring bacteria

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 169

carry out the reduction of selenate to selenite in the biomass suspension before finally adding FeCl3 to precipitate out Se. Soluble Se levels in the drainage waters were reduced from 200– 400 to 7– 12 mg l21. Growth of algae in the drainage water served the twin objectives of decreasing nitrate concentration and providing an inexpensive carbon energy source for denitrifying and dissimilatory Se-reducing bacteria. A flow diagram of the proposed ABSRS is given below:

2.

Deselenification through Methylation

The process of microbial transformation of toxic Se species into less toxic methylated volatile Se compounds has been developed into an important mechanism responsible for reducing Se concentrations in the toxic environments. Bacteria and fungi are the two major groups of Se methylating organisms isolated from soils and sediments (Abu-Erreish et al., 1968; Doran, 1982); bacteria possibly play a more dominant role in water (Thompson-Eagle and Frankenberger, 1991). DMSe is found to be the predominant product of microbial methylation of Se, which is 500 – 700 times less toxic than selenite and selenate ions. Other volatile Se compounds produced in much smaller amounts are DMDSe, DMSeS, methaneselenone and methaneselenol. There is every possibility that aerobic or anaerobic demethylation of DMSe may take place and thereby retaining the Se as selenide or selenate in the system (Oremland, 1994). Thus, less toxic DMSe in the environment may again be changed to highly toxic and reactive forms of Se. The microorganisms responsible for methylation of Se into DMSe are naturally present in saline, alkaline drainage waters and soils (Table IX) and their activity can be dramatically accelerated by the addition of specific amendments. Reviewing various mechanisms of methylation through microbial metabolism, Doran (1982) proposed the following pathway for methylation of inorganic Se by a soil corynebacterium and also postulated that organic intermediates like selenoamino acids are not involved in this reaction: SeO22 3 ! selenite

Se0

elemental Se

! HSeX ! CH3 SeH ! ðCH3 Þ2 Se selenide

methane selenol

dimethyl selenide

170

K. S. DHILLON AND S. K. DHILLON Table IX Methylating Microorganisms Isolated from Seleniferous Environments

Microorganism Pencillium spp. C. humicola Cephalosporium spp., Fusarium spp., Scopulariopsis spp., Penicillium spp. Acremonium falciforme, Penicillium citrinum, Ulocladium tuberclatum A. alternata Aeromonas spp., Flavobacterium spp. Pseudomonas sp Corynebacterium sp P. fluorescens A. veroni Chlorella spp.

Source of isolation

Reference

Sewage Sewage Garden soil

Cox and Alexander (1974) Cox and Alexander (1974) Barkes and Fleming (1974)

Evaporation pond sediment

Karlson and Frankenberger (1988)

Evaporation pond water Lake sediment

Thompson-Eagle et al. (1989) Chau et al. (1976)

Seleniferous soil Evaporation pond sediment Seleniferous agricultural drainage water Saline evaporation pond

Doran and Alexander (1975) Chasteen et al. (1990) Rael and Frankenberger (1996) Fan et al. (1997)

Recently, Oremland (1994) has suggested that the process of methylation at first requires assimilatory reduction of extracellular selenate, selenite or elemental Se to the levels of an organoselenide compound as per the following reaction:

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 171

As soon as methylated Se species are evolved, it is released into the atmosphere, gets diluted and dispersed by air currents away from the contaminated source. The inhaled DMSe is found to be non-toxic to animals at concentrations up to 8034 mg kg21 or 34,000 mg m23. Field studies. Selenium biomethylation is of interest because it represents a potential mechanism for physical loss from Se contaminated environments. Although mechanisms are not clearly defined, microbial volatilization of Se has been successfully tested at the field scale as a bioremediation approach to remove toxic levels of Se in soils/sediments from the recently contaminated Kesterson Reservoir, California (Frankenberger and Karlson, 1995; Flurry et al., 1997). A field study was conducted for over 2 years at contaminated areas at the Kesterson Reservoir containing Se concentrations ranging from 10 to 209 mg kg21 (median 39 mg kg21). The experiment consisted of a number of treatments, i.e., application of water alone, water with cattail straw, cattle manure, citrus peel and protein sources such as casein and gluten. The decline in soil Se revealed that cattle manure was the least effective with 30% removal, while citrus peel þ Zn þ N and casein were the most effective treatments with 62 and 69% of the initial Se inventory reduction from the surface layer within a period of 2 years. Among different C sources tested, the Se methylation rate was found to be the highest with pectin resulting in Se removal up to 51.4% in 118 days (Frankenberger and Karlson, 1994). Surprisingly, in another field study conducted for 22 months at Summer Peck Ranche (near Fresno, California, USA) by Frankenberger and Karlson (1995), the most effective amendment was found to be the cattle manure as it could remove 59% of the Se inventory from sediment comprised mainly of clay. An average emission rate of 54 mg Se m22 h21 was recorded in cattle manure treated plots. Even the application of water plus tillage alone removed approximately 32% of the initial Se inventory in the top 15 cm layer of soil. In a long-term field study carried out by Flury et al. (1997), 68– 88% of the total Se (0 – 15 cm depth) was volatilized within a period of over 100 months. The total soil Se concentration varied from 40 to 60 mg kg21 in different plots. Casein amended soils resulted in the highest Se removal rates and the process of volatilization was more active in the warmer and drier months. Natural bioremediation by Se volatilization and precipitation processes in aquatic environments by a euryhaline green microalga has been reported by Fan et al. (1997). A species of Chlorella isolated from a saline evaporation pond was shown to transform Se aerobically into a variety of alkylselenides as well as elemental Se. In general, additions of organic amendments stimulate the evolution of DMSe as organic carbon serves as a source of energy for the processes of microbial methylation. In addition to this, Se volatilization also depends upon aeration, moisture and temperature. The physical, chemical and biological properties of

172

K. S. DHILLON AND S. K. DHILLON

soils also govern the potential of volatilization of Se. Thus it may not be possible to identify a single factor governing Se volatilization from a seleniferous environment.

XI. CONCLUSIONS Distribution of Se in soils greatly depends upon the composition of geological materials. Nevertheless, the load of Se in the environment is being further increased as a result of anthropogenic activities such as combustion of fossil fuel, metal production and certain agricultural practices. Seleniferous soils have developed like normal soils, but changes in topography and leaching/erosion processes have played a dominant role in the development of Se toxicity and deficiency in different parts of the world. The toxic effects of Se have been known for more than 180 years, but incidences like at the Kesterson Reservoir could make the general public aware of Se as a serious environmental contaminant. The narrow margin between the beneficial and harmful levels of Se has important implications for animal and human health. On the basis of appearance of chronic selenosis in animals and humans and the distribution of Se accumulating plants, seleniferous areas have been identified in Australia, Canada, China, Columbia, India, Ireland, Israel, Mexico and USA. All over the world, seleniferous soils lie interspersed with normal and even deficient areas and thus it is difficult to clearly demarcate toxic areas. In the seleniferous regions, the Se content of toxic soils ranged from 0.23 to 324 mg kg21 and is mainly present as selenate and selenite—the easily available forms for plant absorption. Clay minerals, organic matter and hydrous oxides of Fe, Al and Mn and the composition of the soil solution play an important role in controlling the solubility, mobility and phytoavailability of Se. The concentration of Se in the cultivated agricultural plant species in the toxic regions ranged from , 1 to 450 mg kg21. During the last two decades, a number of potentially useful technologies for the management of seleniferous soils and waters have been investigated. Among these, stabilization of Se in the soil matrix, use of soil cover, manipulating the composition of the soil solution or selecting low Se absorbing crops are only temporary remedial measures. Techniques such as phytoremediation and bioremediation are considered as more promising technologies because they allow for the permanent removal of Se from the soil matrix and have the potential for providing low cost and environmentally friendly treatment options. Laboratory and field studies have amply demonstrated the feasibility of these approaches for remediation of contaminated soils and waters. Utilization of cultivated agricultural plants and agroforestry practices hold a promise for remediation of seleniferous soils. Brassica spp. is an excellent phytoremediating crop. A number of promising microbial strains involved in the processes of Se reduction and volatilization have been isolated

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 173

from seleniferous environments and characterized. T. selenatis has been found to be one of the most efficient strains for reduction of Se oxyanions. The impact of dissimilatory reduction and volatilization processes has been successfully tested in a number of pilot studies for remediation of contaminated soil and water.

XII. FUTURE RESEARCH NEEDS Selenium is unique in that there exists a very narrow gap between deficiency and toxicity levels in animals and humans. As plants play a key role in regulating the delivery of Se into the food chain, there is a need to define more precisely the safe limits in plants consumed by animals, wildlife, and humans. Modern molecular biotechnological tools should be employed to establish Se requirement in the life cycle of higher plants. As the chemistry of Se in soils is easily altered by the nature of the soils and amendments, there is a need to refine tools that can precisely characterize different chemical forms of Se which contribute to its toxicity in a given soil – plant system. The role of native and applied organic matter in the soil in regulating the phytoavailability of Se in toxic levels needs to be studied more critically. Long-term effects of repeated organic matter application to agricultural soils on the fate of Se also need to be investigated. There should be uniformity in methodology for delineating and studying Se toxic regions, particularly in terms of defining the chemical behavior of Se in the soil –plant system. Creation of a network of scientists working on Se toxic soils is highly desirable in this context. Accumulating plants remove the bioavailable fraction of Se from soils. Obviously, there is a need to study the availability of Se from residual fractions of Se following phytoextraction and the kinetics of their reequilibration. Use of chelating agents may help in mobilizing residual Se toward plant roots. Brassica spp., when grown up to the full bloom stage, is probably the most suitable phytoremediation crop. However, when grown for grain production, it sheds almost all the leaves which remain in the field at the time of harvesting. The leaves may contain up to 50% of the total Se extracted from the soil and thus remarkably reduces the amount of Se actually extracted from the soil. Brassica, the most promising plant species for phytoremediation, is not able to support the economical recovery of Se with the present level of extraction and accumulation of Se. There is a need to augment Se absorption and the tolerance capacity of Brassica to a level where standard commercial metal smelting technologies can support a profitable Se extraction. Identification of genes responsible for volatilization and hyperaccmulation and then transferring them to

174

K. S. DHILLON AND S. K. DHILLON

high biomass accumulating plants might help in reducing the time period required for complete removal of Se from the soil. Bioremediation is a relatively new approach for clean up of Se from soil environments. In pilot studies, several technologies have been found to be of practical utility and economically feasible. To increase its social acceptability and confidence level among the probable users, bioremediation should be tested on a large scale by conducting multi-year and multi-location experiments under actual field conditions. Information regarding microbial transformations of Se is needed to accurately predict its behavior in the environment. Although a large number of microorganisms responsible for methylation of Se has been identified, the selenium volatilization efficiency of these organisms need to be further increased by evaluating the integrated effect of different factors such as carbon source, moisture, temperature and nutrient levels. The biomethylation of Se is carried out by a number of bacteria and fungi, but the exact mechanism by which this reaction occurs is still not clear. The biochemical pathway and enzymes used in the conversion of selenate and selenite to volatile organo-Se compounds is an area not appropriately explored. The disposal of Se rich vegetation generated through phytoremediation has become a major problem. One option can be the direct utilization of seleniferous plant materials to improve the Se status of animals in Se deficient areas. These materials can be used as a source of organic fertilizers applied to the soil for raising the Se level of forages in deficient areas on a long-term basis. Selenium is an essential nutrient for animals and humans, and organic sources of Se are more efficient than inorganic sources of Se for improving Se level of the affected population. Selenium is present in a variety of organic combinations in phytoremediating crops, which can be utilized to extract naturally produced organic Se compounds for use as supplements to the affected population.

REFERENCES Abrams, M. M., Burau, R. G., and Zasoski, R. J. (1990). Organic selenium distribution in selected Californian soils. Soil Sci. Soc. Am. J. 54, 979–982. Abu-Erreish, G. M., Whitehead, W. I., and Olson, O. E. (1968). Evolution of volatile selenium from soils. Soil Sci. 106, 415 –420. Adams, D. J., Altringer, P. B., and Fould, W. D. (1993). Bioremediation of selenate and selenite. In “Biohydrometallurgy”. (A. E. Torma, M. L. Apel and C. E. Brierley, Eds.), pp. 755– 771. Minerals, Metals and Materials Society, Warendale, PA. Adriano, D. C., Page, A. L., Elseewi, A. A., Chang, A. C., and Straughan, I. (1980). Utilization and disposal of fly ash and other coal residues in terrestrial ecosystems: a review. J. Environ. Qual. 9, 333–344. Ahlrichs, J. S., and Hossner, L. R. (1987). Selenate and Selenite mobility in overburden by saturated flow. J. Environ. Qual. 16, 95– 98.

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 175 Akagi, J. M., and Campbell, L. L. (1962). Studies on thermophillic sulphate reducing bacteria. II. Adenosine triphosphate-sulfurylase of Clostriduim nigrificans and Desulforibrio desulfuricans. J. Bacteriol. 84, 1194–1201. Alexander, M. (1994). In “Biodegradation and Bioremediation”. Academic Press, New York. Allaway, W. H. (1968). Controls on the environmental levels of selenium. Trace Subst. Environ. Health. 2, 181 –206. Altringer, P. B., Larsen, D. M., and Gardiner, K. R. (1989). Bench scale process development of selenium removal from wastewater using facultative bacteria. In “Biohydrometallurgy”. (J. McCready, R. J. L. McCready and W. H. Wichlacz, Eds.), pp. 643–657. CANMET, Ottawa. Ancizar-Sordo, J. (1947). Occurrence of selenium in soils and plants of Colombia, South America. Soil Sci. 63, 437–438. Anderson, J. W., and Scarf, A. R. (1983). Selenium and plant metabolism. In “Metals and Micronutrients Uptake and Utilization by Plants”. (D. A. Robb and W. S. Pierpoint, Eds.), pp. 241–275. Academic Press, New York. Anderson, M. S., Lakin, H. W., Beeson, K. C., Smith, F. F., and Thacker, E. (1961). USDA Handbook 200. In “Selenium in Agriculture”. U.S. Government Printimg Office, Washington, DC. Andren, A. W., Klein, D. H., and Talmi, Y. (1975). Selenium in coal-fired steam plant emissions. Environ. Sci. Technol. 9, 856–858. Arora, S. P., Kaur, P., Khirwar, S. S., Chopra, R. C., and Ludri, R. C. (1975). Selenium levels in fodders and its relationship with Degnala disease. Indian J. Dairy Sci. 28, 246–253. Balisterieri, L. S., and Chao, T. T. (1987). Selenium adsorption by goethite. Soil Sci. Soc. Am. J. 51, 1145–1151. Ban˜uelos, G. S., and Mayland, H. F. (2000). Absorption and distribution of selenium in animals consuming canola grown for selenium phytoremediation. Ecotoxic. Environ. Safety 46, 322 –328. Ban˜uelos, G. S., Terry, N., Zayed, A. M., and Wu, L. (1995). Managing high soil selenium with phytoremediation. In “Decades later: a time for reassessment. Proceedings of the 12th National Meeting of American Society of Surface Mining and Reclam,” pp. 394–405. Ban˜uelos, G. S., Ajwa, H. A., Terry, N., and Zayed, A. (1997). Phytoremediation of selenium-laden soils: a new technology. J. Soil Water Conserv. 52, 426 –430. Ban˜uelos, G. S., Ajwa, H. S., Wu, L., and Zambrzuski, S. (1998). Selenium accumulation by Brassica napus grown in Se-laden soil from different depths of Kesterson Reservoir. J. Contam. Soil 7, 481 –496. Ban˜uelos, G. S., Shannon, M. C., Ajwa, H., Drapex, J. H., Jordahl, J., and Licht, L. (1999). Phytoextraction by accumulation of boron and selenium by poplar (Populus ) hybrid clones. Int. J. Phytoremed. 1, 81–96. Ban˜uelos, G. S., Zambrzuski, S., and Mackey, B. (2000). Phytoextraction of selenium from soils irrigated with selenium-laden effluent. Plant Soil 224, 251 –258. Bar-Yosef, B., and Meek, M. (1987). Selenium sorption by kaolinite and montmorillonite. Soil Sci. 144, 11–19. Barkes, L., and Fleming, R. W. (1974). Production of dimethylselenide gas from inorganic selenium by eleven soil fungi. Bull. Environ. Contam. Toxicol. 12, 308–311. Bautista, E. M., and Alexander, M. (1972). Reduction of inorganic compounds by soil microorganisms. Soil Sci. Soc. Am. Proc. 36, 918–920. Bawa, S. S., Dhillon, S. K., and Dhillon, K. S. (1990). Effect of sulphur application on the absorption of selenium by different fodder crops. Indian J. Dairy Sci. 43, 564 –570. Bawa, S. S., Dhillon, K. S., and Dhillon, S. (1992). Screening of different fodders for selenium absorption capacity. Indian J. Dairy Sci. 45, 457–460. Benjamin, M. M. (1983). Adsorption and surface precipitation of metals on amorphous iron oxyhydroxide. Environ. Sci. Technol. 17, 686– 692.

176

K. S. DHILLON AND S. K. DHILLON

Berrow, M. L., and Ure, A. M. (1989). Geological materials and soils. In “Occurrence and Distribution of Selenium”. (M. Ihnat, Ed.), CRC Press, Boca Raton. Bertine, K. K., and Goldberg, E. D. (1971). Fossil fuel combustion and the major sedimentary cycle. Science 173, 233 –235. Besser, J. M., Huckins, J. N., Little, E. E., and La-Point, T. W. (1989). Distribution and bioaccumulation of selenium in aquatic microcosms. Environ. Pollut. 62, 1–12. Boon, D. Y. (1989). Potential selenium problems in Great Plains soils. SSSA Special Publication No. 23. In “Selenium in Agriculture and the Environment”. (L. W. Jacobs, Ed.), pp. 107– 121. SSSA, Madison, WI. Brown, T. A., and Shrift, A. (1980). Identification of selenocysteine in the proteins of selenate-grown Vigna radiata. Plant Physiol. 66, 758 –761. Brown, T. A., and Shrift, A. (1982). Selenium: toxicity and tolerance in higher plants. Biol. Rev. 57, 59–84. Bryant, R. D., and Laishley, E. J. (1988). Evidence for two transporters of sulphur and selenium oxyanions in Chlostridium pastuerianum. Can. J. Microbiol. 34, 700–703. Burau, R. G., McDonald, A., Jacobson, A., May, D., Grattan, S., Shennan, C., Swanton, B., Scherer, D., and Abrams, M. (1988). Selenium in tissues of crops sampled from west side of the San Joaquin Valley, California. In “Selenium Contents in Animal and Human Food Crops grown in California”. (K. K. Tanji, Ed.), pp. 61–68. University of California Publication No. 3330, Oakland, CA. Cappon, C. J. (1987). Uptake and speciation of mercury and Se in vegetable crops grown composttreated soil. Water Air Soil Pollut. 34, 353–361. Carlson, R. M., Uriu, K., and Kapustka, T. (1988). Selenium concentrations in deciduous nut tree tissue samples from western Fresno and Merced countries. In “Selenium Contents in Animal and Human Food Crops grown in California”. (K. K. Tanji, Ed.), pp. 57–60. University of California Publication No. 3330, Oakland, CA. Cervinka, V. (1994). Agroforestry farming system for the management of selenium and salt on irrigated farmland. In “Selenium in the Environment”. (W. T. Frankenberger, Jr. and S. Benson, Eds.), pp. 237–255. Marcel Dekker, New York. Chaney, R. L. (1983). Plant uptake of inorganic waste constituents. In “Land Treatment of Hazardous Wastes”. (J. F. Parr, P. B. Marsh and J. M. Kla, Eds.), pp. 50–76. Noyes Data, Park Ridge, NJ. Chaney, R. L., Malik, M., Li, Y. M., Brown, S. L., Brewer, E. P., Angle, J. S., and Baker, A. J. M. (1997). Phytoremediation of soil metals. Curr. Opin. Biotechnol. 8, 279–284. Chasteen, T. G., Silver, G. M., Birks, J. W., and Fall, R. (1990). Fluorine-induced chemoluminascence detection of biologically methylated tellurium, selenium and sulphur compounds. Chromatographia 30, 181 –185. Chau, Y. K., Wong, P. T. S., Silverberg, B. A., Luxon, P. L., and Bergert, G. A. (1976). Methylation of Se in the aquatic environment. Science 192, 1130–1131. Chow, C. M., Nigam, S. N., and McConnell, W. B. (1971). Biosynthesis of Se-methylselenocysteine and S-methylcysteine in Astrgalus bisulcatus: effect of selenium and sulphur concentrations in the growth medium. Phytochemistry 10, 2693–2698. Christensen, B. T., Bertelsen, F., and Gissel-Nielsen, G. (1989). Selenite fixation by soil particle-size separates. J. Soil Sci. 40, 641–647. Combs, G. F., Jr., and Combs, S. B. (1984). The nutritional biochemistry of selenium. Ann. Rev. Nutr. 4, 257– 280. Cox, D. P., and Alexander, M. (1974). Factors affecting trimethylarsine and dimethylselenide formation by Candida humicola. Microb. Ecol. 1, 136 –144. Cutter, G. A., and Bruland, K. W. (1984). The marine biogeochemistry of selenium: a re-evaluation. Limnol. Oceanogr. 29, 1179–1192. Davidson, D. F., and Powers, H. A (1959). The selenium content of some volcanic rocks from the western United States and the Hawaiian Islands. U.S. Geol. Survey Bull. 1084 C.

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 177 DeMoll-Decker, H., and Macy, J. M. (1993). The periplasmic nitrite reductase of Thauera selenatis may catalyze the reduction of selenite to elemental selenium. Arch. Microbiol. 160, 241 –247. De Souza, M. P., Pilon-Smits, E. A. H., Lytle, C. M., Hwang, S., Tai, J., Honma, T. S. U., Yeh, L., and Terry, N. (1998). Rate-limiting steps in selenium assimilation and volatilization by Indian mustard. Plant Physiol. 117, 1487–1494. Deveral, S. J., Fio, J. L., and Dubrovsky, N. M. (1994). Distribution and mobility of selenium in groundwater in the western San Joaquin Valley. In “Selenium in the Enviroment”. (W. T. Frankenberger, Jr. and S. Benson, Eds.), pp. 157–184. Marcel Dekker, New York. Dey, S., Stafford, R., Deb Roy, M. K., Bhattacharjee, C. R., Khathing, D. T., Bhattacharjee, P. C., and Dkhar, P. S. (1999). Metal toxicity and trace element deficiency in some wild animals species from north-east India, as revealed by cellular, bio-inorganic and behavioural studies. Curr. Sci. 77, 276–280. Dhillon, K. S., and Dhillon, S. K. (1990). Selenium toxicity in soil-plant-animal system—a case study. Trans. 14th Int. Cong. Soil Sci. 5, 300–305. Dhillon, K. S., and Dhillon, S. K. (1991a). Selenium toxicity in soils, plants and animals in some parts of Punjab, India. Int. J. Environ. Studies 37, 15–24. Dhillon, K. S., and Dhillon, S. K. (1991b). Accumulation of selenium in sugarcane (Sachharum officinarum Linn.) in seleniferous areas of Punjab, India. Environ. Geochem. Health 13, 165 –170. Dhillon, K. S., and Dhillon, S. K. (1997a). Distribution of seleniferous soils in north-west India and associated toxicity problems in the soil–plant–animal–human continuum. Land Contam. Reclam. 5, 313–322. Dhillon, K. S., and Dhillon, S. K. (1997b). Factors affecting level of selenium in plants grown in seleniferous soils of Punjab, India. Ecological Agriculture and Sustainable Development. (G. S. Dhaliwal, N. S. Randhawa, R. Arora and A. K. Dhawan, Eds.), 1, pp. 371–376. CRRID, Chandigarh. Dhillon, K. S., and Dhillon, S. K. (1998). Fractionation and adsorption–desorption reactions of Se in seleniferous soils of northwest India. Trans. 16th Int. Cong. Soil Sympp. 6. Dhillon, K. S., and Dhillon, S. K. (1999). Adsorption –desorption reactions of selenium in some soils of India. Geoderma 93, 19–31. Dhillon, K. S., and Dhillon, S. K. (2000a). Selenium accumulation by sequentially grown wheat and rice as influenced by gypsum application in a seleniferous soil. Plant Soil 227, 243– 248. Dhillon, S. K., and Dhillon, K. S. (2000b). Selenium adsorption in soils as influenced by different anions. J. Plant Nutr. Soil Sci. 163, 577–582. Dhillon, K. S., and Dhillon, S. K. (2001). Restoration of selenium-contaminated soils. In “Environmental Restoration of Metals-contaminates Soils”. (I. K. Iskandar, Ed.), pp. 199–227. CRC Press, Boca Raton. Dhillon, K. S., Bawa, S. S., and Dhillon, S. K. (1992a). Selenium toxicity in some plants and soils of Punjab. J. Indian Soc. Soil Sci. 40, 132 –136. Dhillon, K. S., Randhawa, S. S., Dhillon, S. K., Randhawa, C. S., and Nauriyal, D. C (1992b). Geomedical studies on selenium toxicity in bovines. In “Proceedings of the 17th World Biuatriics Congress and 25th American Association of Bovine Practitioners’ Conference,” vol. 3, pp. 351– 356. St Paul, Minnesota, USA. Dhillon, K. S., Dhillon, K. S., Dhillon, S. K., Singh, J., and Singh, T. J. (1996). Tissue selenium as an index of selenosis in buffaloes. Indian J. Anim. Sci. 66, 30–34. Dhillon, K. S., Dhillon, S. K., and Rani, N. (2001). Chemical fractions and bioavailability of selenium in seleniferous soils of northwest India. Paper presented at International Workshop on “Chemical Bioavailability in Terrestrial Environment” held at Adelaide, SA, 18–20 November. Doran, J. W. (1982). Microorganisms and the biological cycling of selenium. Adv. Microb. Ecol. 6, 1 –32.

178

K. S. DHILLON AND S. K. DHILLON

Doran, J. W., and Alexander, M. (1975). Microbial formation of dimethylselenide and biological cycling of selenium. In “Proceedings of the Annual Meeting of American Microbiologists,” pp. 188. Doran, J. W., and Alexander, M. (1977). Microbial formation of volatile Se compounds in soil. Soil Sci. Soc. Am. J. 40, 687– 690. van Dorst, S. H., and Peterson, P. J. (1984). Selenium speciation in the soil solution and its relevance to plant uptake. J. Sci. Food Agric. 35, 188 –192. Dungan, R. S., and Frankenberger, W. T., Jr. (1999). Microbial transformations of selenium and the bioremediation of seleniferous environments. Bioremed. J. 3, 171–188. Elrashidi, M. A., Adriano, D. C., Workman, S. M., and Lindsay, W. L. (1987). Chemical equilibria of selenium in soils: a theoretical development. Soil Sci. 144, 141–152. Fan, T. W. M., Lane, A. N., and Higashi, R. M. (1997). Selenium biotransformations by a euryhaline microalga isolated from a saline evaporation pond. Environ. Sci. Technol. 31, 569–576. Fio, J. L., and Fujii, R. (1990). Selenium speciation methods and application to soil saturation extracts from San Joaquin Valley, California. Soil Sci. Soc. Am. J. 54, 363 –369. Fleming, G. A. (1962). Selenium in Irish soils and plants. Soil Sci. 94, 28– 35. Fleming, G. A., and Walsh, T. (1957). Selenium occurrence in certain Irish soils and its toxic effects on animal. R. Irish Acad. Proc. 58, 151 –166. Flury, M., Frankenberger, W. T., Jr., and Jury, W. A. (1997). Long-term depletion of selenium from Kesterson dewatered sediments. Sci. Total Environ. 198, 259– 270. Frankenberger, W. T., Jr., and Arshad, M. (2001). Remediation of selenium-contaminated sediments and water. BioFactors 14, 241–254. Frankenberger, W. T., Jr., and Benson, S. (1994). In “Selenium in the Environment”. Marcel Dekker, New York. Frankenberger, W. T., Jr., and Karlson, U. (1989). Environmental factors affecting microbial production of dimethylselenide in a selenium contaminated sediment. Soil Sci. Soc. Am. J. 53, 1435–1442. Frankenberger, W. T., Jr., and Karlson, U. (1994). Microbial volatilization of selenium from soils and sediments. In “Selenium in the Environment”. (W. T. Frankenberger, Jr. and S. Benson, Eds.), pp. 369 –388. Marcel Dekker, New York. Frankenberger, W. T., Jr., and Karlson, U. (1995). Volatilization of selenium from a dewatered seleniferous sediment: a field study. J. Ind. Microbiol. 14, 226 –232. Furr, A. K., Parkinson, T. F., Heffron, C. L., Reid, J. T., Haschek, W. M., Gutenmann, W. H., Bache, C. A., St John, L. E., and Lisk, D. J. (1978). Elemental content of tissues and excreta of lambs, goats, and kids fed white sweet clover growing on fly ash. J. Agric.Food Chem. 26, 847–851. Ganje, T. J., and Whitehead, E. I. (1958). Evolution of volatile selenium from pierre shale supplied with 75Se as selenite or selenate. Proc. South Dakota Acad. Sci. 37, 81–84. Gerhardt, M. B., Green, F. B., Newman, R. D., Lundquist, T. J., Tresan, R. B., and Oswald, W. J. (1991). Removal of selenium using a novel algal bacterial process. Res. J. Water Pollut. Control Fed. 63, 799–805. Ghosh, A., Sarkar, S., Pramanik, A. K., Chowdhary, S. P., and Ghosh, S. (1993). Selenium toxicosis in grazing buffaloes and its relationship with soil and plant of West Bengal. Indian J. Anim. Sci. 63, 557–560. Girling, C. A. (1984). Selenium in agriculture and the environment. Agric. Ecosyst. Environ. 11, 37 –65. Gissel-Nielsen, G., Gupta, U. C., Lamand, M., and Westermarck, T. (1984). Selenium in soils and plants and its importance in livestock and human nutrition. Adv. Agron. 37, 397 –460. Gustafsson, J. P., and Johnsson, L. (1992). Selenium retention in the organic matter of Swedish forest soils. J. Soil Sci. 43, 461 –472. Hamdy, A. A., and Gissel-Nielsen, G. (1977). Fixation of selenium by clay minerals and iron oxides. Z. Pflanzener. Bodenkd. 140, 63–70.

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 179 Haygarth, P. M., Jones, K. C., and Harrison, A. F. (1991). Selenium cycling through agricultural grasslands in the U.K.: budgeting the role of the atmosphere. Sci. Total Environ. 103, 89 –111. Haygarth, P. M., Cooke, A. I., and Jones, K. C. (1993). Long-term change in the biogeochemical cycling of atmospheric selenium: depostion to plants and soils. J. Geophys. Res. 98(D9), 16, 769 –16,776. Haygarth, P. M., Fowler, D., Stu¨rup, S., Davison, B. M., and Jones, K. C. (1994). Determination of gaseous and particulate selenium over a rural grassland in the U.K. Atmos. Environ. 28, 3655–3663. Hingston, F. J., Posner, A. M., and Quirk, J. P. (1971). Competitive adsorption of negatively charged ligands on oxide surfaces. Disc. Faraday Soc. 52, 334–342. Hurd-Karrer, A. M. (1934). Selenium injury to wheat plants and its inhibition by sulfur. J. Agric. Res. 49, 343–357. Jacobs, L. W. (Ed.) (1989). SSSA Special Publication No. 23. In “Selenium in Agriculture and the Environment”. SSSA, Madison, WI. Jefferis, S. A. (1992). Remedial barriers and containment. In “Contaminated Land Treatment Technolgies”. (J. F. Rees, Ed.), pp. 58–82. Society of Chemical Industry/Elsevier, London. Jump, R. K., and Sabey, B. R. (1989). Soil test extractants for predicting selenium in plants. SSSA Special Publication No. 23. In “Selenium in Agriculture and the Environment”. (L. W. Jacobs, Ed.), pp. 95–105. SSSA, Madison, WI. Kabata-Pendias, A., and Pendias, H. (1984). In “Trace Elements in Soils and Plants”. CRC Press, Boca Raton,FL. Karam, N. S., Ereifej, K. I., Shibli, R. A., Abu-Kudais, H., Alkofahi, A., and Malkawi, Y. (1998). Metal concentration, growth and yield of potato produced from in situ plantlets or microtubers and grown in municipal solid waste-amended substrates. J. Plant Nutr. 21, 725–739. Karlson, U., and Frankenberger, W. T., Jr. (1988). Determination of gaseous 75selenium from soil. Soil Sci. Soc. Am. J. 52, 678–681. Karunakaran, C., and Rao, A. R. (1979). Geological Survey of India. Miscelaneous Publication No. 41, Part IV. In “Status for exploration for hydrocarbons in the Himalayan region—contribution to stratigraphy and structure”, pp. 1–66. Klasing, S. A., and Schenker, M. B. (1988). Public health implications of elevated dietary selenium. In “Selenium Contents in Animal and Human Food Crops grown in California”. (K. K. Tanji, Ed.), pp. 3–10. University of California Publication No. 3330, Oakland, CA. Knott, S. G., and McCray, C. W. R. (1959). Two naturally occurring outbreaks of selenosis in Queensland. Aust. Vet. J. 35, 161 –165. Knott, S. G., McCray, C. W. R., and Hall, W. T. K. (1958). Selenium poisoning in horses in North Queensland. Queensland Dept. Agric. Stock Div. Animal Ind. Bull. 41, 1 –16. Korkman, J. (1985). Bulletin 28. In “Selenium in Fertilizers”. Selenium–Telluruim Development Association, Darien, CT. Krauskopf, K. G. (1955). Sedementary deposits of rare metals. Econ. Geol. Anniv. 50, 411 –463. La˚g, J., and Steinnes, E. (1978). Regional distribution of selenium and arsenic in humus layers of Norwegian forest soils. Geoderma 20, 3– 14. Lortie, L., Gould, W. D., Rajan, S., McCready, R. G. L., and Chang, K. J. (1992). Reduction of selenate and selenite to elemental selenium by a Pseudomonas stutzeri isolate. Appl. Environ. Microbiol. 58, 4042–4044. Losi, M. E., and Frankenberger, W. T., Jr. (1997). Reduction of selenium oxyanious by Enterobactor cloacae strain SLD 1a-1. In “Environmental Chemistry of Selenium”. (W. T. Frankenberger, Jr. and R. A. Engberg, Eds.), pp. 515– 544. Marcel Dekker, New York. Lundquist, T. J., Green, F. B., Tresan, R. B., Newman, R. D., and Oswald, W. J. (1994). The algal– bacterial selenium removal system: mechanism and field study. In “Selenium in the Environment”. (W. T. Frankenberger, Jr. and S. Benson, Eds.), pp. 251–277. Marcel Dekker, New York.

180

K. S. DHILLON AND S. K. DHILLON

Mackenzie, F. T., Lantzy, R. J., and Peterson, V. (1979). Global trace metal cycles and predictions. J. Int. Assoc. Math. Geol. 11, 99–141. Macy, J. M. (1994). Biochemistry of selenium metabolism by Thauera selenatis gen. nov. spp. nov. and use of the organism for bioremediation of selenium oxyanions in San Joaquin drainage waters. In “Selenium in the Environment”. (W. T. Frankenberger, Jr. and S. Benson, Eds.), pp. 421 –444. Marcel Dekker, New York. Macy, J. M., Rach, S., Auling, G., Dorsch, M., Stackebtandt, E., and Sly, L. (1993). Thauera selenatis gen. nov. spp. nov., a member of the beta sub-class of proteobacteria with a noval type of respiration. Int. J. Syst. Bacteriol. 43, 135–142. Mayland, H. F., James, J. G., Panter, K. E., and Sonderegger, J. L. (1989). Selenium in seleniferous environments. SSSA Special Publication No. 23. In “Selenium in Agriculture and the Environment”. (L. W. Jacobs, Ed.), pp. 15– 50. SSSA, Madison, WI. McNeal, J. M., and Balisterieri, L. S. (1989). Geochemistry and occurrence of selenium: an overview. SSSA Special Publication No. 23. In “Selenium in Agriculture and the Environment”. (L. W. Jacobs, Ed.), pp. 1–13. SSSA, Madison, WI. McNeill, K. R., and Waring, S. (1992). Vitrification of contaminated soils. In “Contaminated Land Treatment Technologies”. (J. F. Rees, Ed.), pp. 143– 159. Society of Chemical Industry/Elsevier, London. Megharaj, M., Kantachota, D., Singleton, I., and Naidu, R. (2000). Effects of long-term contamination of DDT on soil microflora with special reference to soil algae algal metabolism of DDT. Environ. Pollut. 109, 35– 42. Mench, M. J., Didier, V. L., Loffer, M., Gomez, A., and Masson, P. (1994). A mimicked in situ remediation study of metal contaminated soils with emphasis on cadmium and lead. J. Environ. Qual. 23, 58–63. Mikkelsen, R. L., Page, A. L., and Haghnia, G. H. (1988). Effect of salinity and its composition on the accumulation of selenium by alfalfa. Pland Soil 107, 63–67. Mosher, B. W., and Duce, R. A. (1989). The atmosphere. In “Occurrence and Distribution of Selenium”. (M. Inhat, Ed.), pp. 295 –325. CRC Press, Boca Raton. Muth, B. H. (1963). White muscle disease, a selenium responsive myopahy. J. Am. Vet. Med. Assoc. 142, 272 –277. Naidu, R., Megharaj, M., Vig, K., and Kookana, R. S (2001). Bioavailability, definition and analytical techniques for assessment and remediation of contaminated (inorganic and organic) soils. Paper presented at the International Workshop on “Chemical Bioavailability in the Terrestrial Environment” held at Adelaide, SA, 18–20 November. National Academy of Science – National Research Council (NAS – NRC) (1976). Selenium. Committee on Medical and Biological Effects of Environmental Pollutants. NAS–NRC, Washington, DC. National Academy of Science-National Research Council (NAS– NRC) (1983). Selenium in Nutrition. NAS– NRC, Washington, DC. Neal, R. H. (1995). Selenium. In “Heavy Metals in Soils”. (B. J. Alloway, Ed.), . Blackwell, Glasgow. Nissen, P., and Bensen, A. (1964). Absence of selenate esters and selenolipid in plants. Biochem. Biophys. Acta 83, 400–402. Nriagu, J. O. (1989). A global assessment of natural sources of atmospheric trace metals. Nature 338, 47–49. Nriagu, J. O., and Pacyna, J. M. (1988). Quantitative assessment of worldwide contamination of air, water and soils by trace metals. Nature 333, 134–139. Nriagu, J. O., and Wong, H. K. (1983). Selenium pollution of lakes near the smelters of Sudbury, Ontario. Nature 301, 55–57. Ohlendorf, H. M. (1989). Bioaccumulation and effects of selenium in wildlife. SSSA Special Publication No. 23. In “Selenium in Agriculture and the Environment”. (L. W. Jacobs, Ed.), pp. 95– 105. SSSA, Madison, WI.

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 181 Oldfield, J. E. (1986). Selenium: an essential poison. Sci. Food Agric. 4, 22–26. Oldfield, J. M. (1990). In “Selenium: Its uses in Agriculture, Nutrition Health and the Environment”. Selenium–Tellurium Development Association, Grimbergen, Belgium. Oldfield, J. M. (1999). In “Selenium World Atlas”. Selenium–Tellurium Development Association, Grimbergen, Belgium. Oremland, R. S. (1994). Biogeochemical transformations of Se in anoxic environment. In “Selenium in the Environment”. (W. T. Frankenberger, Jr. and S. Benson, Eds.), pp. 369 –420. Marcel Dekker, New York. Oremland, R. S., Steinberg, N. A., Presser, T. S., and Miller, L. G. (1991). In situ bacterial selenate reduction in the agricultural drainage systems of western Nevada. Appl. Environ. Microbiol. 57, 615 –617. Owens, L. P. (1997). Bioreactors in removing selenium from agricultural drainage water. In “Environmental Chemistry of Selenium”. (W. T. Frankenberger, Jr. R. A. Engberg, Eds.), pp. 501–514. Marcel Dekker, New York. Pareek, N., Dhillon, K. S., and Dhillon, S. K. (1998). Adsorption/desorption behaviour of selenium in alluvial soils of Punjab. J. Indian Soc. Soil Sci. 46, 363 –367. Parker, D. R., and Page, A. L. (1994). Vegetation management strategies for remediation of selenium contaminated soils. In “Selenium in the Environment”. (W. T. Frankenberger, Jr. and S. Benson, Eds.), pp. 327– 342. Marcel Dekker, New York. Peterson, G., and Butler, G. (1962). The uptake and assimilation of selenite by higher plants. Aust. J. Biol. Sci. 15, 126–146. Peterson, J. A., and Nebeker, A. V. (1992). Estimation of waterborne Se concentation that are toxicity thresholds for wildlife. Arch. Environ. Contam. Toxicol. 23, 440 –452. Pilon-Smits, E. A. H., de Souza, M. P., Lytle, C. M., Shang, C., Lugo, T., and Terry, N. (1998). Selenium volatilization and assimilation by hybrid poplar (Populus tremula H. alba ). J. Expp. Bot. 49, 1889–1892. Pilon-Smits, E. A. H., Hwang, S., Lytle, C. M., Zhu, Y., and Tai, J. C. (1999). Overexpression of ATP sulfurylase in Indian mustard leads to increased selenate uptake, reduction and tolerance. Plant Physiol. 119, 123– 132. Pilon-Smits, E. A. H., de Souza, M. P., Hong, G., Amini, A., Bravo, R. C., Payabyab, S. T., and Terry, N. (1999). Selenium volatilization and accumulation by twenty aquatic plant species. J. Environ. Qual. 28, 1011–1018. Pratley, J. E., and McFarlane, J. D. (1974). The effect of sulphate on the selenium content of pasture plants. Aust. J. Expp. Agric. Anim. Husb. 14, 533–538. Presser, T. S. (1994). Geologic origion and pathways of selenium from the California coast ranges to the west-central San Joaquin Valley. In “Selenium in the Environment”. (W. T. Frankenberger, Jr. and S. Benson, Eds.), pp. 139–156. Marcel Dekker, New York. Rael, R. M., and Frankenberger, W. T., Jr (1996). Influence of pH. salinity and selenium on the growth of Aeromonas veronii in evaporation agriculture drainage water. Water Res. 30, 422–430. Rajan, S. S. S. (1979). Adsorption of selenite, phosphate and sulphate on hydrous alumina. J. Soil Sci. 30, 709–718. Randhawa, S. S., Randhawa, C. S., Brar, R. S., Dhillon, K. S., Dhillon, S. K., and Singh, B (1992). Biomedical profile of bovines fed on selenium rich fodders. In “Proceedings of the International Symposium on Nutrient Management for Sustained Productivity,” vol. 2, pp. 166–167. Ludhiana, India. Raskin, I., Smith, R. D., and Salt, D. E. (1997). Phytoremediation of metals using plants to remove pollutants from the environment. Curr. Opin. Biotechnol. 8, 221–226. Ravikovitch, S., and Margolin, M. (1957). Selenium in soils and plants. Agric. Res. Sta. Rehovot. 7, 41 –52. Ravikovitch, S., and Margolin, M. (1959). The effect of barium chloride and calcium sulphate in hindering selenium absorption by lucerne. Empp. J. Expp. Agric. 27, 235–240.

182

K. S. DHILLON AND S. K. DHILLON

Reilly, C. (1996). In “Selenium in Food and Health”. Chapman and Hall, London. Ribang, L., Jian’an, T., and Wuyi, W. (1992). A study on animal selenosis and its geographical environment in western Hubei province. In “Studies in Physical Geography and Environment”, pp. 220 –225. Zhongshan University Press, Guangdong. Robberecht, H., van Grieken, R., Sprundel, M. V., Berghe, D. V., and Deelstra, H. (1983). Selenium in environmental and drinking waters of Belgium. Sci. Total Environ. 26, 163–172. Rogers, P. A. M., Arora, S. P., Fleming, G. A., Crinion, R. A.P., and McLaughlin, J. G. (1990). Selenium toxicity in farm animals: treatment and prevention. Irish Vet. J. 43, 151 –153. Rosenfeld, I., and Beath, O. A. (1964). In “Selenium: Geobotany, Biochemistry, Toxicity and Nutrition”. Academic Press, New York. Schwarz, K., and Foltz, C. M. (1957). Selenium as an integral part of factor 3, against dietary liver necrosis degeneration. J. Am. Chem. Soc. 79, 3292–3293. Severson, R. C., and Gough, L. P. (1992). Selenium and sulphur relationship in alfalfa and soil under field conditions, San Joaqnin Valley, California. J. Environ. Qual. 21, 353 –358. Sharmasarkar, S., and Vance, G. F. (1995). Fractional partitioning for assessing solid phase speciation and geochemical transformations of soil selenium. Soil Sci. 160, 43–55. Shrift, A., and Ulrich, J. M. (1969). Transport of selenate and selenite into Astragalus roots. Plant Physiol. 44, 893 –896. Shubik, P., Clayson, D. B., and Terracini, B. (1970). “The Quantification of Environmental Carcinogens.” International Union Against Cancer. Technical Report Series 4. Sillanpaa, M., and Jansson, H. (1992). FAO Soils Bulletin 65. In “Status of cadmium, lead, cobalt and selenium in soils and plants of thirty countries”. United Nations Organization, Rome. Singh, M., and Kumar, P. (1976). Selenium distribution in soils of bio-climatic zones of Haryana. J. Indian Soc. Soil Sci. 24, 62–67. Singh, B., and Sandhu, B. S. (1979). Effect of irrigation, mulch and crop canopy on the soil temperature in forage maize. J. Indian Soc. Soil Sci. 27, 225–235. Singh, R., Randhawa, S. S., Dhillon, K. S., and Dhillon, S. K. (2001). Clinico-haematological studies on experimentally induced chronic selenosis in buffalo calves. Indian J. Anim. Sci. 71, 511 –514. Smith, B. (1993). Remediation update finding the remedy. Waste Manag. Environ. 4, 24–30. Soltanpour, P. N., and Workman, S. M. (1980). Use of NH4HCO3 –DTPA soil test to assess availability and toxicity of Se to alfalfa plants. Commun. Soil Sci. Plant. Anal. 11, 1147–1156. Soltanpour, P. N., Olsen, S. R., and Goods, R. J. (1982). Effect of nitrogen fertilization of dryland wheat on grain selenium concentration. Soil Sci. Soc. Am. J. 46, 430– 433. Stadtman, T. C. (1974). Selenium biochemistry. Science 183, 915–922. Suzuki, T. (1965). A geochemical study of selenium in volcanic exhalation and sulphur deposits. II. On the behaviour of selenium and sulphur in volcanic exhalation and sulphur deposits. Chem. Soc. Jpn Bull. 38, 1940–1946. Suzuki, Y., Nishiyama, K., Takano, Y., Tajeri, T., and Sakurayama, K. (1959). Studies on the selenium content of various foodstuffs, fertilizers and human hair. Tokushimi J. Expp. Med. 6, 243– 249. Tan, J. A., Wang, W. Y., Wang, D. C., and Hou, S. F. (1994). Adsorption, volatilization and speciation of selenium in different types of soils in China. In “Selenium in the Environment”. (W. T. Frankenberger, Jr. and S. Benson, Eds.), pp. 47–68. Marcel Dekker, New York. Terry, N., and Zayed, A. M. (1994). Selenium volatilization by plants. In “Selenium in the Environment”. (W. T. Frankenberger, Jr. and S. Benson, Eds.), pp. 343 –369. Marcel Dekker, New York. Terry, N., Carlson, C., Raab, T. K., and Zayed, A. M. (1992). Rates of selenium volatiliztion among crop species. J. Environ. Qual. 21, 341–344. Terry, N., Zayed, A. M., de Souza, M. P., and Tarun, A. S. (2000). Selenium in higher plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 51, 401–432.

DISTRIBUTION AND MANAGEMENT OF SELENIFEROUS SOILS 183 Thompson-Eagle, E. T., and Frankenberger, W. T., Jr. (1990). Volatilization of selenium from agricultural evaporation pond water. J. Environ. Qual. 19, 125– 131. Thompson-Eagle, E. T., and Frankenberger, W. T., Jr. (1991). Selenium biomethylation in an alkaline, saline environment. Water Res. 25, 231 –240. Thompson-Eagle, E. T., and Frankenberger, W. T., Jr. (1992). Bioremediation of soils contaminated selenium. Adv. Soil Sci. 17, 261– 310. Thompson-Eagle, E. T., Frankenberger, W. T., Jr., and Karlson, U. (1989). Volatilization of selenium by Alternaria alternata. Appl. Environ. Microbiol. 55, 1406–1413. Thomson, B. M., and Heggen, R. J. (1982). Water quality and hydrologic impacts of disposal of uranium mill tailings by back filling. In “Management of Wastes from Uranium Mining and Milling”, pp. 373–389. IAEA, Vienna. Thorvaldson, T., and Johnson, L. R. (1940). Selenium content of Saskatchewan wheat. Can. J. Res. B18, 138 –150. Tidball, R. R., Severson, R. C., Gent, C. A., and Riddle, G. O. (1989). Element association in soils of San Joaquin Valley of California. SSSA Special Publication No. 23. In “Selenium in Agriculture and Environment”. (L. W. Jacobs, Ed.), pp. 179–194. SSSA, Madison, WI. Tinsley, I. J. (1979). In “Chemical Concepts in Pollutant Behaviour”. Wiley, New York. Tokunaga, T. K., Zawislanski, P. T., Johannis, P. W., Lipton, D. S., and Benson, S. (1994). Field investigations of selenium speciation, transformation and transport in soils from Kesterson Reservoir and Lahontan Valley. In “Selenium in the Environment”. (W. T. Frankenberger, Jr. and S. Benson, Eds.), pp. 119 –138. Marcel Dekker, New York. Ulrich, J. M., and Shrift, A. (1968). Selenium absorption by excised Astragalus roots. Plant Physiol. 43, 14–20. Walsh, T., and Fleming, G. A. (1952). Selenium levels in rocks. Soils and herbage from a high selenium locality in Ireland. Trans. 5th Int. Cong. Soil Sci. 2(A), 178– 183. Warman, P. R., Muizelaar, T., and Termeer, W. C. (1995). Bioavailability of As, Cd, Co, Cr, Cu, Hg, Mo, Ni, Pb, Se and Zn from biosolids amended compost. Compost Sci. Utiliz. 3, 40 –50. Webb, J. S., Thornton, I., and Fletcher, K. (1966). Seleniferous soils in parts of England and Wales. Nature 211, 327. Weiss, H. V., Koide, M., and Goldberg, E. D. (1971). Selenium and S in a Greenland ice sheet: relation to fossil fuel combustion. Science 172, 261 –263. Weres, O., Jaouni, A. R., and Tsao, L. (1989). The distribution, speciation and geochemical cycling of selenium in a sedimentary environment, Kesterson Reservoir, California, USA. Appl. Geochem. 4, 543–563. White, A. F., and Dubrovsky, N. M. (1994). Chemical oxidation–reduction controls on selenium mobility in groundwater systems. In “Selenium in the Environment”. (W. T. Frankenberger, Jr. and S. Benson, Eds.), pp. 185–222. Marcel Dekker, New York. Williams, C., and Thornton, I. (1973). The use of soil extractants to estimate plant available molybdenum and selenium in potentially toxic soils. Plant Soil 39, 149–159. Williams, K. T., Lakin, H. W., and Byers, H. G. (1940). Selenium occurrence in certain soils in the United States with a discussion of related topics. Fourth Repp. US Dept. Agric. Tech. Bull. 702, 1 –59. Williford, C. W., Jr., and Bricka, R. M. (2001). Physical separation of metal contaminated soils. In “Environmental Restoration of Metals-contaminates Soils”. (I. K. Iskandar, Ed.), pp. 121–165. CRC Press, Boca Raton. Wu, L. (1994). Selenium accumulation and colonization of plants in soils with elevated selenium and salinity. In “Selenium in the Environment”. (W. T. Frankenberger, Jr. and S. Benson, Eds.), pp. 279–325. Marcel Dekker, New York. Yang, G., Wang, S., Zhou, R., and Sun, S. (1983). Endemic selenium intoxication of humans in China. Am. J. Clin. Nutr. 37, 872–881.

184

K. S. DHILLON AND S. K. DHILLON

Zayed, A. M., and Terry, N. (1994). Selenium volatilization in roots and shoots: effects of shoot removal and sulphate level. J. Plant Physiol. 143, 8– 14. Zayed, A. M., Gowthaman, S., and Terry, N. (1998). Phytoaccumulation of trace elements by wetland plants. I. Duckweed. J. Environ. Qual. 27, 715– 721. Zhu, Y. L., Zayed, A. M., Qian, J. H., de Souza, M. P., and Terry, N. (1999). Phytoaccumulation of trace elements by wetland plants. 1. Water hyacinth. J. Environ. Qual. 28, 339–344. Zieve, R., and Peterson, P. J. (1985). Sorption of dimethylselenide by soils. Soil Biol. Biochem. 17, 105–107. Zingaro, R. A., and Cooper, W. C. (1974). In “Selenium”. Van Nostrand Reinhold, New York. Zoller, W. H., Glandney, E. S., and Duce, R. A. (1974). Atmospheric concentration and source of trace metals at the North Pole. Science 183, 198 –200.

THE ROLE OF ARBUSCULAR MYCORRHIZAL FUNGI IN SUSTAINABLE CROPPING SYSTEMS L. A. Harrier1 and C. A. Watson2 1

SAC, West Mains Road, Edinburgh EH9 3JG, UK SAC, Craibstone Estate, Aberdeen AB21 9YA, UK

2

I. Introduction II. The Role of AM Fungi in Sustainable Farming Systems A. Crop Nutrition B. Crop Protection C. Crop Water Relations D. Plant Reproduction E. Root Architecture and Longevity F. Soil Structure G. Soil Microbial Populations H. Plant Populations III. Managing AM Fungi in Sustainable Agriculture IV. Direct Impact of Crop and Soil Management Practices on AM fungi A. Rotation Design B. Intercropping C. Crop and Varietal Selection D. Cultivation E. Nutrient Management F. Liming G. Crop Protection H. Grazing Livestock I. Inoculation V. Effects of Different Agricultural Systems on AM fungi A. Organic Versus Biodynamic Versus Conventional Farming B. Long-term Versus Short-term Effects VI. Managing AM fungi in Sustainable Agriculture—Prospects for the Future A. Economics and Product Quality B. Research and Development Needs Acknowledgments References Mycorrhizal associations vary widely in structure and function, but the most ubiquitous interaction is the arbuscular mycorrhizal (AM) symbiosis. This interaction forms between the roots of over 80% of all terrestrial plant 185 Advances in Agronomy, Volume xx Copyright q 2003 by Academic Press. All rights of reproduction in any form reserved 0065-2113/02$35.00

186

L. A. HARRIER AND C. A. WATSON species and Glomeromycete fungi. This ancient symbiosis confers benefits directly to the host plants growth and development through the acquisition of phosphate and other mineral nutrients from the soil by the fungus, while the fungus receives a carbon source from the host. The symbiosis may also enhance the plant’s resistance to biotic and abiotic stresses. Additionally, AM fungi develop an extensive external hyphal network, which makes a significant contribution to the improvement of soil structure. Therefore, these fungi constitute an integral and important component of agricultural systems. AM fungi are particularly important in organic farming systems and other sustainable systems which rely on biological processes, rather than agrochemicals, to supply nutrients and to control weeds, pests and diseases. The exploitation of AM fungi within sustainable cropping systems is dependent on a better understanding of the impact of agricultural management practices on AM fungal population dynamics and functioning. This review explores the links between AM fungi and agricultural management practices. q 2003 Academic Press.

I. INTRODUCTION Soil quality is an integral indicator of sustainable agricultural ecosystems (Herrick, 2000) and one of the key components of soil quality is the biota, in particular the microbial component. One of the most important groups of soil microorganisms is mycorrhizal fungi. Mycorrhizal associations vary widely in structure and function, but the most common interaction is the arbuscular mycorrhizal (AM) association. The AM symbiosis represents an ancient symbiosis (Pirozynski and Malloch, 1975; Pirozynski and Dalpe, 1989; Remy et al., 1994; Taylor et al., 1995; Redecker et al., 2000) with over 80% of all terrestrial flowering plants forming this type of relationship (Smith and Read, 1997). The symbiosis is formed when fungi belonging to the unique Phylum Glomeromycota (Schussler et al., 2001) grow within and around plant roots. These fungi are obligate symbionts, in that they need to obtain a source of carbon from the host plant to fulfil their lifecycle. The plant root system containing the fungi form a structure termed an arbuscular mycorrhiza and this is the normal state for the majority of plant root systems. However, the ability to form an AM symbiosis has been lost in about 10% of plants, and is completely absent in families, such as Chenopodiaceae and Brassicaceae (Tester et al., 1987). The development of an AM is a complex process and is characterised by distinct developmental stages in the growth of the AM fungus (Fig. 1): spore germination, hyphal differentiation, appressorium formation, root penetration, intercellular growth, intracellular arbuscule formation and nutrient exchange.

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 187

Figure 1 Schematic diagram of the formation of an arbuscular mycorrhiza. Developmental stages in the growth and development of an arbuscular mycorrhiza (Ap—appressoria; C—hyphal coils; Ar—arbusucle; BAS—branched absorbing structure; BASs—branched absorbing structure spore; S—spore; Rh—runner hyphae; V—vesicle).

188

L. A. HARRIER AND C. A. WATSON

For in-depth discussions on aspects of the development of the AM symbiosis, readers are referred to a number of comprehensive reviews (Smith and Gianinazzi-Pearson, 1988; Koide and Schreiner, 1992; Bonfante and Perotto, 1995; Harrison, 1997, 1998, 1999; Hirsch and Kalpulnik, 1998; Albrecht et al., 1999; Harrier, 2000). The AM fungus is a symbiotic organism, however, it also harbours structures called bacterium-like organisms (blos), which have been detected ultrastructurally in spores and germinating and/or symbiotic mycelium (Bianciotto et al., 1996). These bacteria have been shown to be group II Pseudomonads by DNA sequence analysis of small subunit rRNA genes. Moreover, these bacteria are known to contain nif genes. The nif genes are responsible for the production of nitrogenase, the key enzyme in nitrogen fixation (Minerdi et al., 2001). However, the significance of these bacteria, and the role these bacteria play in the beneficial interactions observed when AM fungi colonise plant root systems, remains unknown. However, this indicates that mycorrhizal systems can include plant, fungal and bacterial cells (Bianciotto et al., 1996). The AM symbiosis is generally beneficial to the growth and development of the host plant and is recognised as a key component of a healthy and productive agricultural ecosystem. Although AM fungi are recognised as beneficial symbionts, they are relatively unstudied, this is due to their obligate biotrophy, whereby completion of their lifecycle depends on their ability to colonise a host plant. Furthermore, fungal growth ceases after approximately 25 –30 days of culture in the absence of the host plant. AM fungi have not been cultured in the absence of the host plant and this has hampered their mass production and utilisation within crop systems (Jarstfer and Sylvia, 1992). However, it is possible to grow AM fungi in sterile culture with plant root explants (for example, Mosse and Hepper, 1975; Miller-Wideman and Watrud, 1984; Diop et al., 1994) and/or with hairy roots transformed with Agrobacterium rhizogenes (for example, Mugnier and Mosse, 1987; Becard and Fortin, 1988; Diop et al., 1992; Declerck et al., 1996, 1998; Pawlowska et al., 1999). AM fungi are adversely influenced by some of the practises of modern agriculture and the use of soluble fertilisers and pesticides. In the light of increasing cost of chemical inputs, recalcitrant pathogens resistant to chemical pesticides, etc., AM fungi may provide a more sustainable and environmentally acceptable alternative to these current practices. AM fungi will therefore have a key role to play in the health and productivity of lowinput agricultural systems. In this paper, we examine the role of the AM symbiosis within sustainable farming systems and we consider aspects of amendments and management practices that influence AM fungal dynamics.

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 189

II.

THE ROLE OF AM FUNGI IN SUSTAINABLE FARMING SYSTEMS A. CROP NUTRITION

Increases in plant growth following root colonisation by AM fungi are normally due to an improvement in the mineral nutrient status of plants. This is primarily achieved by the external hyphal (extra-radical hyphal) network (Fig. 2) in the soil matrix, which takes up the nutrients from outside the rhizosphere within the bulk soil. The nutrients are then transported via the hyphal network to the plant root whereby they are passed to the plant in exchange for carbon.

1. Phosphorus Phosphorous (P) is an important plant macronutrient, which constitutes 0.2% of a plant’s dry weight (Schactman et al., 1998). Low availability of P in bulk soil limits plant uptake. For this reason, AM fungi are key for P acquisition since fungal hyphae greatly increase the volume of bulk soil that the plant roots can explore. Quantitatively, P is the most important nutrient taken up by the extra-radical hyphae and influx of P in roots colonised by AM fungi can be three to five times higher than in an non-mycorrhizal root (Smith and Read, 1997). There is evidence that phosphatase activity is

Figure 2 Extra-radical hyphal network of an AM fungus around plant roots (BASs—branched absorbing structure spore; PR—plant root; Rh—runner hyphae).

190

L. A. HARRIER AND C. A. WATSON

increased in the rhizosphere soil around a mycorrhizal root (Dodd et al., 1987; Azco´n and Barea, 1997). However, there is no clear evidence that this is a fungal phenomenon. The phosphate within the soil is taken up via a phosphate transporter located in the extra-radical hyphae of the fungus (Harrison and van Buuren, 1995). The phosphate is then condensed into polyphosphate (poly-P) and translocated back towards the plant by cytoplasmic streaming into the intraradical hyphae (Cox et al., 1975; Cox and Tinker, 1976). Finally the poly-P may be hydrolysed and released as phosphate to the plant across the fungal membrane, probably at the arbuscule. Acquisition of P by mycorrhizal plants differs not only with plant species and cultivar, but also with the isolate of AM fungus (Clark and Zeto, 2000). With increasing soil P concentration, the growth enhancement effect of AM fungal colonisation decreases and may either be abolished or lead to growth depressions. Depression of AM fungal colonisation levels by high soil P concentrations are often observed (Abbott and Robson, 1984; Bolan, 1991; Marschner and Dell, 1994). In addition, AM colonisation provides little benefit to plants which have evolved other mechanisms for obtaining P from the soil, such as cluster roots, thin roots and long root hairs. Therefore, the benefits of AM fungal colonisation are often observed in soils depleted of nutrients and/or agricultural systems where the inputs are low.

2.

Nitrogen

Nitrogen (N) is the primary limiting nutrient in most terrestrial ecosystems and competition between plants and microbes for this nutrient is intense (Vitousek and Howarth, 1991; Hodge et al., 2000). Particular mycorrhizal associations, notably ectomycorrhizae and ericoid mycorrhizae improve N nutrition by accessing organic N that is inaccessible to roots alone (Chalot and Brun, 1998; Abuzinadah and Read, 1989; Ho¨gberg, 1990; Read, 1996; Lipson et al., 1999). Several studies have demonstrated the transport of inorganic N by AM fungi (Ames et al., 1983; Frey and Schu¨epp, 1993; George et al., 1992; Hawkins and George, 1999; Johansen et al., 1992, 1993, 1994; Hawkins et al., 2000) and there is indirect evidence for amino acid uptake by AM colonised plants in the field (Na¨sholm et al., 1998). Hyphae of AM fungi have been shown to take up amino acids and transport N to the plant roots in a form that is presently unknown. However, the uptake capacity for amino acids varied between AM fungal isolates, even of the same species (Hawkins et al., 2000). To date, little is known about the mechanism of N uptake by AM fungi and nothing is known about the N transfer from fungus to plant (Hawkins et al.,

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 191

2000). Although, Smith et al. (1994) suggested that N-rich amides probably asparagine and/or glycine, are possible candidates for the major form in which N is transported.

3.

Sulphur

By supplying radioisotopes such as 35SO4, it has been demonstrated that the extra-radical hyphae of AM fungi have the capacity to uptake and transport sulphur (Cooper and Tinker, 1978). However, the enhanced S nutrition of the mycorrhizal plant may not be solely attributed to the AM hyphal transport. Raju et al. (1990b) demonstrated that there were greater genotypic S uptake differences in non-mycorrhizal plants than mycorrhizal plants. Furthermore, population dynamics of sulphur cycling bacteria are known to after within the rhizosphere of Glomus mosseae and G. fasiculatum maize plants (Amora-Lazcano and Azco´n, 1997).

4.

Boron

Boron (B) is non-essential for the growth and development of fungi and variable effects on B acquisition in AM colonised plants have been observed (Marschner and Dell, 1994; Clark and Zeto, 2000). Observations have ranged from enhanced (Kothari et al., 1990) though not affected (Lu and Miller, 1989) to reduced acquisition (Clark et al., 1999b).

5.

Other Macronutrients

Reports of the acquisition of macronutrients such as calcium (Ca), magnesium (Mg), potassium (K) and sodium (Na) by mycorrhizal plants are varied, with increases, no effects and decreases having been observed (Lambert et al., 1979; Rhodes and Gerdemann, 1978; Azco´n and Barea, 1992; Clark and Zeto, 1996; Clark et al., 1999a). The enhancements of these macronutrients is dependent on soil pH, host plant and isolate of AM fungus (Saif, 1987; Sieverding and Toro, 1989; Lambais and Cardoso, 1993; Medeiros et al., 1994a,b,c; Clark and Zeto, 1996; He et al., 1997). Enhancements in the acquisition of K, Ca and Mg is often observed in mycorrhizal plants grown in acidic soils. This contrasts with mycorrhizal plants grown in neutral to alkaline soils where few enhancements in the acquisition of these elements are observed. Often the acquisition is lower in mycorrhizal as compared to non-mycorrhizal plants (Bethlenfalvay et al., 1989;

192

L. A. HARRIER AND C. A. WATSON

Hamel and Smith, 1991; Raju et al., 1990a,b; Espinoza-Victoria et al., 1993; Khalil et al., 1994; Tarafdar and Marschner, 1994; Bermudez and Azco´n, 1996).

6.

Micronutrients

Enhanced acquisition has been reported for mycorrhizal as compared to nonmycorrhizal roots of a number of micronutrients. These include zinc (Zn) and copper (Cu) (Marschner and Dell, 1993); chlorine (Cl) and bromine (Br) (Buwalda et al., 1983; Ellis et al., 1995); caesium (Cs) and cobalt (Co) (Rogers and Williams, 1986); molybdenum (Mo) (Raju et al., 1987); nickel (Ni) (Killham and Firestone, 1983; Rogers and Williams, 1986); cadmium (Cd) (Cooper and Tinker, 1978; Dehn and Schu¨epp, 1989; Guo et al., 1996) and lead (Pb) (Diaz et al., 1996). AM fungal colonised plants tend to have lower Mn acquisition compared to non-colonised plants (Arines et al., 1989; Kothari et al., 1990; Azaizeh et al., 1995). This mycorrhizal effect can contribute to higher Mn tolerance in plants, for example, in soybean grown in soils with high concentrations of Mn (Bethlenfalvay and Franson, 1989). However, tolerances to Mn are thought to be attributed to Mn oxidation and reduction in the rhizosphere and/or alteration in the dynamics of Mn-oxidising microbes within the rhizosphere (Kothari et al., 1991) rather than a direct effect.

7.

Interplant Transfer of Nutrients

Hyphal links between plants offer potential networks for the movement of soil-derived nutrients, and this transfer could play a key role in interplant and interspecies competition and re-distribution of nutrients within ecosystems. The development of a mycelial network is dependent on the genera of AM fungi as different genera have different patterns of development (Boddington and Dodd, 1998). Furthermore, different AM fungi have varying capacities for P transfer to the host. It has been suggested that AM fungal hyphae within senescent roots could lead to a rapid transfer and re-distribution of P released by autolysis, or by the activity of microbes to plants linked by mycelial networks. Chiariello et al. (1982) demonstrated that when 32P was applied to the leaves of a donor Plantago erecta plant grown in serpentine annual grassland, the labelled P could be detected in the shoots of neighbouring plants at a distance of approximately 45 mm after 6 – 7 days. In undisturbed ecosystems the AM fungal hyphal network forms a permanent external mycelial network and plants are linked by a common mycelial network (CMN; Hodge, 2000). The benefits to the plant of being connected to the CMN

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 193

are numerous and may help it to access and exploit nutrient rich zones in the soil. Furthermore, it may allow the fungus to access carbon sources from several host plants. The interplant nutrient transfer by the CMN may be sufficient to sustain growth of a receiver plant, for several weeks (Francis et al., 1986). Conversely, this route of nutrient transfer may be insignificant in some cases (Ikram et al., 1994).

B. CROP PROTECTION AM fungi and their associated interactions with plants can reduce damage caused by soil-borne pathogens (Schonbeck, 1979; Bartschi et al., 1981; Dehne, 1982; Bagyaraj, 1984; Rosendahl, 1985; Schenck, 1987; Smith, 1987; Caron, 1989; Jalali and Jalali, 1991; Paulitz and Linderman, 1991; Bethlenfalvay and Linderman, 1992; Sharma et al., 1992; Hooker et al., 1994; Linderman, 1994; Barea and Jefferies, 1995; Azco´n-Aguilar and Barea, 1996; Cordier et al., 1996, 1998; Mark and Cassells, 1996; Norman et al., 1996; Pozo et al., 1996, 1999; Trotta et al., 1996; Murphy et al., 2000). Biological control of pathogenic, species such as Phytophthora, Gauemannomyces, Fusarium, Chalara (Thielaviopsis ), Pythium, Rhizoctonia, Sclerotium, Verticillium and Aphanomyces (Rosendahl, 1985; Kjoller and Rosendahl, 1997; Slezack et al., 1999, 2000) by AM fungi have been consistently demonstrated in many cases. This protection is not, however, effective for all plant pathogenic fungi and the level of biological control conferred by the AM fungal colonisation is plant species and AM fungal isolate specific. The ability of the AM symbiosis to enhance resistance or tolerance within root systems is not equal and may range from disease reduction through no observable disease reduction (Ba¨a¨th and Hayman, 1983, 1984) to the occasional increase in disease severity (Davies et al., 1979). Moreover, enhancement of disease symptoms of soil borne and/or other diseases is often witnessed within the foliar tissues of mycorrhizal plants (Dehne, 1982; Shaul et al., 1999). Colonisation by AM fungi can alter the host plants susceptibility to insect herbivores (Gange and West, 1994; Gehrig and Whitham, 1994; Borowicz, 1997; Gange, 2001). There is increasing evidence that AM fungal colonisation of plant roots can affect the growth and reproduction of insect feeding on the host plant (Gange and Bower, 1997). Chewing insects have been found to be negatively affected by the presence of AM fungi, while sucking insects appear to be positively affected. Chewing larvae of the root feeding insect Otiorhynchus sulcatus were also negatively affected by colonisation of plant roots with AM fungi. The negative affects were quite dramatic with a 40% reduction in the survival and size of larvae being reported. Colonisation by AM fungi may, however, result in chemical changes within the roots that may be detrimental to the larvae (Gange, 2001). A range of chemicals have been isolated from

194

L. A. HARRIER AND C. A. WATSON Table I Mechanisms by which an AM Fungal Association may Control Plant Root Pathogens

Mechanism

References

Improvement of mineral nutrition

Davis and Menge (1980), Graham and Dehne (1982), Trotta et al. (1996) and Cordier et al. (1996)

Root damage compensation Competition for photosynthesis and/or colonisation and infection sites Production of anatomical and morphological changes in the root system Changes in the microbial composition of the mycorrhizosphere Activation of plant defence mechanisms

Caron et al. (1985) Atkinson et al. (1994)

Citernesi et al. (1986) Cordier et al. (1998) and Pozo et al. (1996, 1998, 1999, 2002)

mycorrhizal roots which may be active against insect herbivores, such as phenolics (Morandi, 1996), and terpenoids (Maier et al., 1995), etc. Disease reduction within host plants is the outcome of complex interactions between plant, pathogen and AM fungi. The mechanisms proposed to explain the bioprotection vary, and are detailed within Table I. Activation of plant defence responses by AM fungi have been shown to be transient and uncoordinated in comparison to those observed when pathogens attack plants. However, it has been suggested that the AM symbiosis could predispose the plant to respond more rapidly to pathogenic attacks (Dehne, 1982; Zambolim and Schenck, 1983; Rosendahl, 1985; Caron, 1989; Gianinazzi, 1991; Lindermann, 1999; Azco´n-Aguilar and Barea, 1996; Gianinazzi-Pearson, 1996).

C. CROP WATER RELATIONS AM fungi can affect the water balance of both amply watered and droughted host plants (reviewed by Reid, 1979; Fitter, 1985; Read and Boyd, 1986; Nelson, 1987; Gupta, 1991; Koide, 1993; Sa´nchez-Dı´az and Honrubia, 1994; Auge´, 2000, 2001). The different effects and potential mechanisms of AM fungal colonisation on the host plants water relations are presented in Table II. The drought responses of over 90 host plant species representing 69 fungal genera to AM fungal colonisation have been investigated (reviewed in Auge´, 2001) and the effects are not consistent. Reid (1979) noted that AM fungi appear to benefit droughted plants through both direct drought avoidance and tolerance.

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 195 Table II Host Plant Effects and Possible Mechanisms for Alteration in the Water Relations of AM Fungal Colonised plants Effects

Mechanisms

Stomatal conductance and transpiration Altered photosynthetic rates Altered leaf hydration status Modified root hydration status Hydraulic conductivity and hyphal water transport Soil drying rates and moisture relations Growth and nutrient uptake during drought Metabolic effects Morphological effects

Total biomass Plant size relationships Enhanced nutrition Rates of water absorption and soil drying Hydraulic conductances Soil water relations Soil –root potential gradients Plant water potential components Non-hydraulic root signals

In terms of sustainable agriculture, characterisation of drought hardiness should be measured in terms of growth, yield and survival (Auge´, 2001). AM fungal colonisation has also been shown to increase crop yields and survival rates in many studies in dry soil conditions. Moreover, colonisation by AM fungi has also been shown to enhance tolerance to other water-related environmental stresses, such as salinity, chilling and soil compaction (for example, Charest et al., 1993; Azco´n and El-Atrach, 1997; Yano et al., 1998; El-Tohamy et al., 1999).

D. PLANT REPRODUCTION AM fungi have been shown to alter several growth characteristics associated with plant reproduction through differentially affecting reproductive success of male and female functions (Pendleton, 2000). Flowering parameters of mycorrhizal plants have been investigated in several plant species (Schenck and Smith, 1982; Dodd et al., 1983; Koide et al., 1988; Bryla and Koide, 1990a; Koide and Lu, 1992; Ganade and Brown, 1997; Stephenson et al., 1998). Reproductive responses include earlier flowering, prolonged flowering period and an increase in the number of flower buds, inflorescences and fruits. Increased seed production of mycorrhizal plants has been demonstrated. This enhancement in seed production was brought about by changes in plant phenology, shoot architecture, inflorescence production, fruit set, number of fruits and number of seeds per fruit (Koide et al., 1988, 1994; Bryla and Koide, 1990a,b; Lewis and Koide, 1990; Koide and Lu, 1992; Stanley et al., 1993; Lu and Koide, 1994; Shumway and Koide, 1994a,b, 1995). Phosphorous content of the seed produced was significantly higher for mycorrhizal plants of all the species studied, resulting in enhanced growth, reproduction and competitive

196

L. A. HARRIER AND C. A. WATSON

fitness of subsequent generations (Lewis and Koide, 1990; Koide and Lu, 1992; Lu and Koide 1994; Shumway and Koide, 1994a,b; Heppel et al., 1998).

E. ROOT ARCHITECTURE AND LONGEVITY Colonisation of a plant root system by AM fungi can alter the morphology of the root system in a structural, spatial, quantitative and temporal manner (Atkinson, 1992). These responses are dependent on the species of host plant and isolate of colonising AM fungus (Schellenbaum et al., 1991; Tisserant et al., 1991; Berta et al., 1993; Hooker et al., 1995). AM fungi also alter host plant root system development, with colonised roots being more highly branched i.e., the root system contains shorter more branched adventitious roots of larger diameters and lower specific root lengths (Schellenbaum et al., 1991; Berta et al., 1993; Atkinson et al., 1994; Hooker et al., 1995). The mechanism by which AM fungi facilitates changes within root architecture and longevity are numerous (Atkinson et al., 1994). However, enhanced plant nutrition in particular, P nutrition was originally thought to have an effect (Drew and Sakar, 1978). AM fungi have been shown to effect lateral root primordial formation, root apices size, DNA synthesis and the length of the cell cycle (Berta et al., 1993; Fusconi et al., 1994; Guido et al., 2001; Tahiri-Alaoui et al., 2002). These effects will cause the production of a more highly branched root system, larger root apices, lower DNA synthesis and meristematic activity. Therefore, the effects of AM fungal colonisation on root architecture and longevity is more fundamental than growth promotion response mediated through enhanced mineral nutrition.

F. SOIL STRUCTURE AM fungi develop an extensive extra-radical hyphal network, that grows away from the root, through the rhizosphere and into the surrounding bulk soil matrix. This network makes a significant contribution to the improvement of soil texture and water relations, with particular benefits to soil aggregation (Tisdall and Oades, 1982; Miller and Jastrow, 1990; Tisdall, 1991). Fig. 2 shows a typical example of a hyphal network growing around the root of a strawberry plant. Aggregate stability is a necessary pre-requisite for a healthy managed agricultural ecosystem (Tisdall and Oades, 1982; Miller and Jastrow, 1992), particularly in soils that are prevalent to erosion. The extra-radical hyphae of AM fungi have been shown to be more important than root length and/or bacterial populations in stabilising soil aggregates (Schreiner et al., 1997). Recent studies indicate that AM fungi produce a glycoprotein, glomalin that acts as an insoluble glue to stabilise soil aggregates (Wright, 2000; Wright and Upadhyaya, 1996,

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 197

1998, 1999; Wright et al., 1996, 1998). Glomalin may play a role as an insoluble biofilm that is involved in the aggregation process. When glomalin is sloughed from hyphae, other organic matter and microbes then become attached to nearby soil particles by the protective glomalin coating. As a consequence, a localised environment is created for bacteria to degrade organic matter and to produce extracellular polysaccharides, which in turn further stabilises aggregates. Glomalin production varies between AM fungal species and in different soil types from 2.8 to 14.8 mg/g (Wright and Upadhyaya, 1998). Studies of aggregate stability and glomalin production in soils under different management practices and different crop rotations demonstrated that aggregate stability and glomalin were linearly correlated (Wright and Upadhyaya, 1998; Wright and Anderson, 2000). Moreover, aggregate stability was greater in grass soils than crop rotation and/or highly disturbed soils with comparable amounts of glomalin produced (Wright and Anderson, 2000). Therefore, management practices that disturb the AM fungal hyphal networks and glomalin production will cause a decrease in aggregate stability.

G. SOIL MICROBIAL POPULATIONS The diversity and functionality of microbes is a key component of sustainable systems. Microbial populations in soils actively develop around the roots within the rhizosphere and the mycorrhizosphere where they are stimulated by root exudates, plant residues and other organic substrates. AM fungi may influence soil microbial dynamics; they themselves may be influenced by soil microbial populations (Hodge, 2000). Colonisation of a plant root system causes qualitative and quantitative changes in the exudation patterns of the plant root system. Moreover, colonisation by AM fungi is known to alter the root architecture and longevity of root systems, thus modifying the amount of organic material within the soil system. Alteration in the root exudation patterns is known to effect the growth and development of particular root pathogenic fungi. For example, Phytopthora cinnamomi zoospore production was reduced in the presence of exudates from Zea mays colonised by G. fasiculatum. Furthermore, the amino acid content of root exudates of mycorrhizal tomato is known to be different from that of nonmycorrhizal root exudates. Soil microbes may influence AM fungal growth and development in a positive, negative or neutral manner. AM fungi may themselves directly influence soil microbial populations and AM fungi may be subject to grazing by other soil organisms, such as collembola, earthworms and mammals. Interactions between AM fungi and nitrogen fixing bacteria is beneficial for all symbionts and is critical for biological N input into the plant and soil system.

198

L. A. HARRIER AND C. A. WATSON

H. PLANT POPULATIONS Several studies show that AM fungi alter plant community structure by affecting the relative abundance of plant species and plant-species diversity (Grime et al., 1987; Gange et al., 1990; Helgason et al., 1998; van der Heijden et al., 1998). The original microcosm-based studies of Grime et al. (1987) demonstrated that the presence of AM fungi leads to an increase of plant species. However, the microcosm and field-based studies of van der Heijden et al. (1998) provided definitive proof that mycorrhizal diversity determines plant community structure, plant diversity and ecosystem variability. The studies also demonstrated that plant productivity and the diversity of plants are closely correlated with AM fungal diversity. Furthermore, different AM fungal species promoted the growth of different plant species.

III. MANAGING AM FUNGI IN SUSTAINABLE AGRICULTURE There are many definitions of sustainable agriculture in the literature (see Pretty (1995) for discussion of this topic). The common feature of most of these definitions is that sustainable systems work with biological processes to achieve acceptable levels of productivity and food quality with minimal adverse environmental impact. Therefore, they rely on crop rotation and other cultural practices for the conservation and supply of nutrients and the control of weeds, pests and diseases. The integration of livestock and crop production is often a key feature of sustainable systems. This differs from more intensive farming systems, which are often specialist crop or livestock systems, and rely mainly on the use of added manufactured products, in the form of readily soluble fertilisers and pesticides. Many different forms of sustainable agriculture are practised and referred to by different names including integrated crop management (ICM), integrated pest management (IPM), low external input sustainable agriculture (LEISA) and organic. Organic farming is the only sustainable farming system that is legally defined. Within the EU, crop and livestock products sold as organic must be certified as such under EC Regulation 2092/91 and 1804/99. This guarantees the production system by which food is produced rather than the quality of the products themselves. It is now widely recognised that the soil conditions prevalent in sustainable agriculture are likely to be more favourable to AM fungi than those under conventional agriculture (Bethlenfalvay and Schu¨epp, 1994; Smith and Read, 1997). Mosse (1986) stated that “It is as normal for the roots of plants to be mycorrhizal as it is for the leaves to photosynthesise. Any agricultural operation that disturbs the natural ecosystem will have repercussions on the mycorrhizal

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 199

Figure 3 Management factors that influence AM fungi within soils, and the major interactions between them.

system.” Within this review, we examine how the agricultural management practises that contribute to the development of sustainable forms of agriculture can foster the positive, and minimise the negative repercussions. Much reductionist research has addressed how individual management factors affect the form and function of AM fungi. However, sustainable systems rely on multiple practices to achieve their goals. Figure 3 highlights some of these management practices and their possible interactions in relation to AM fungi. Ma¨der et al. (2000) highlight the importance of the interactions between individual management practises and the need to understand the long-term impacts of agricultural systems. This review also investigates the published evidence on the effects of specific forms of sustainable agriculture (e.g., organic, biodynamic) on AM fungi. The complexity of agricultural systems in the field also means that it can be difficult to extrapolate from experiments carried out under controlled environmental conditions to field situations, thus wherever possible the review focuses on research carried out under field conditions.

IV. DIRECT IMPACT OF CROP AND SOIL MANAGEMENT PRACTICES ON AM FUNGI A. ROTATION DESIGN Crop rotation is a system where different plants are grown in a recurring, defined sequence. Crop rotation is a very powerful tool for managing nutrient supply,

200

L. A. HARRIER AND C. A. WATSON

through the alternation of legumes and nitrogen demanding crops, as well as managing weeds, pests and diseases. It is widely accepted that preceding crops affect the growth and yield of subsequent crops (Altieri, 1995; Karlen et al., 1994) and that this cannot be explained entirely by nutritional effects. This has been demonstrated in both small plot trials and in commercial scale agriculture (Bourgeois and Entz, 1996). Bagayoko et al. (2000) also suggested that AM fungi play a role in the so-called “rotation effect,” as they observed that the AM colonisation rate was higher in early season within cereals grown in rotation than in a monoculture. Vivekanden and Fixen (1991) found that AM colonisation was higher following soybean than following barley. Furthermore, Baltruschat and Dehne (1988) showed that the inoculum potential in winter wheat was significantly reduced in winter wheat grown in monoculture than as a component of a rotation. Although AM fungi are considered to have low host plant specificity with regard to crop species, several workers have now shown that crop species and rotations can influence AM fungal communities (e.g., Johnson et al., 1991; Hendrix et al., 1995; Schenck and Kinloch, 1980). Johnson et al. (1992) aiming to understand the importance of this observation, suggested that continuous cropping tended to select rapidly growing AM fungal species which formed inefficient symbioses with the host plant, leading eventually to a yield decline under monoculture. The inclusion of non-mycorrhizal crops within rotations has been shown to decrease the early growth, P uptake (Arihawa and Karasawa, 2000; Vivekanden and Fixen, 1991), AM fungal colonisation (Douds et al., 1997; Gavito and Miller, 1998a) and yield (Arihawa and Karasawa, 2000; Gavito and Miller, 1998b) of subsequent crops. In order to examine the role of soil characteristics on the effects of preceding crops, Karasawa et al. (2001) studied the shoot weight and P uptake of maize after either sunflower or mustard in 17 soils with different P status. In 14 of the 17 soils shoot weight and P uptake were higher following sunflower, and the growth differences could not be explained by soil P availability. The effects of the preceding crop were eliminated if the soils were sterilised; suggesting that AM fungi were playing a key role in the growth of the following crop. Non-mycorrhizal crops included within a monoculture have been shown to have a more-detrimental effect on the following crop than non-host crops within a crop rotation (Baltruschat and Dehne, 1988). The inclusion of fallow periods can have a similar effect to the inclusion of nonmycorrhizal crops on the growth and mycorrhizal infection of the subsequent crops (Black and Tinker, 1979; Ocampo et al., 1980). This reflects the decline in the viability of both total and metabolically active AM hyphae within the absence of host plants (Chilvers and Daft, 1982; Kabir et al., 1999) and the importance of extra-radical hyphae as a source of inoculum within the soil (Sylvia, 1992). Over winter mycorrhizal cover crops have been shown to increase the AM fungal colonisation of the following crop compared to fallow (Boswell et al., 1998; Kabir and Koide, 2000) due to the maintenance of fungal inoculum over the winter period (Dodd and Jefferies, 1986). Harinikumar and Bagyaraj (1988) found that fallow

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 201

reduced mycorrhizal propagules, by 40% compared with growing a mycorrhizal host plant. The length of the fallow period is also likely to influence the efficiency of the AM symbiosis in the following crop. In pot experiments Kabir et al. (1999) showed AM efficiency decreased with increased length of fallow. Non-crop species can also act as effective cover crops; Kabir and Koide (2000) showed that dandelion grown as a cover crop had a more beneficial effect than winter wheat on the growth of the following maize crop. Thus in rotations that include nonmycorrhizal crops, the degree of weed control may be a key factor in the effect of the non-host crop on the following crop. In organic systems, where the aim of weed control is to maintain the weed population at a level where it does not have a detrimental effect on yields (Stockdale et al., 2001), as opposed to eliminating weeds altogether, weeds potentially act as a mycorrhizal bridge between crops.

B. INTERCROPPING The growing of two or more crops together (intercropping) has the potential to improve resource use. This results from differences in competitive ability for resources between above and below ground crop components in space and time (Willey, 1979). Intercropping is commonly used in forage crops (e.g., grassclover leys) but is less common in arable crops in temperate zones. Several effective intercrop combinations of cereals and legumes have, however, been developed (Jensen, 1996). AM fungal mycorrhizal colonisation can affect the relative performance of different plant species, altering both the species diversity and growth of individuals (Smith and Read, 1997). Intercropping of vegetables and legumes is becoming a more common practice in agriculture (see, for example, Theunissen, 1997), and this raises the question as to whether colonisation of host plants is affected by the presence of non-host plants. In pot experiments, colonisation of the host plant has been found to be unaffected in some studies (Ocampo et al., 1980) and reduced in others (Hayman et al., 1975).

C. CROP AND VARIETAL SELECTION Variation in AM fungal development has been shown within a number of crop species including alfalfa (Douds et al., 1998), maize (Liu et al., 2000), pea (Estau´n et al., 1987), onions (Sharma and Adholeya, 2000) and wheat (Hetrick et al., 1996). Selection of varieties most suited to environmental and soil conditions is a key aim in sustainable agriculture. Breeding and selection of crop varieties has progressed for many years with conventional high inputs of fertilisers and herbicides. This progression has led to the development of modern varieties that are able to produce high yields from relatively small plants, in the absence of weeds. In other words, the varieties are adapted to efficient use of

202

L. A. HARRIER AND C. A. WATSON

soluble fertiliser and are not competitive against weeds. In lower input systems where available soil nutrients may be limited, below-ground characteristics are likely to be more important than in conventional systems. These include rooting depth, root architecture and root length (Atkinson et al., 1995). Hetrick et al. (1992) and Zhu et al. (2001) have demonstrated that modern cultivars are less responsive to AM fungal colonisation than older cultivars. Similarly Manske (1989) showed under low P conditions that AM inoculation increased spring wheat yields of land races much more than high-yielding varieties and under high P conditions AM fungal inoculation depressed the yield of high-yielding varieties significantly more than land races. Hetrick et al. (1993) demonstrated that cultivars released after 1950 have reduced dependence on AM fungal symbiosis. Disease resistance is also a feature of breeding programmes which may impact the mycorrhizal symbiosis; Toth et al. (1990) found that maize inbred lines with high overall resistance to a number of diseases also had low AM fungal colonisation.

D. CULTIVATION McGonigle and Miller (2000) have summarised the effects of tillage on AM fungi as reducing colonisation and P uptake of the following crop as a result of reducing the effectiveness of the mycorrhizal symbiosis. Plant P uptake in early season is generally greater in undisturbed than disturbed systems which may be linked to increased AM fungal colonisation within undisturbed systems. McGonigle and Miller (2000) found that the extent of the detrimental effect of tillage on AM fungi in the following crop depended on the inoculum density available—if the inoculum density is sufficiently high then tillage does not affect colonisation of the following crop. Kabir et al. (1997) found AM fungal hyphal density to be greater in no till than reduced till and least in a conventional tillage system. Reduced tillage has potential advantages in sustainable farming as it creates greater ground cover, less erosion, increased soil organic matter and better soil structure than conventional tillage in many soils (Brady, 1990). The downside of no-till for sustainable systems is that it may increase the need for biocides to control weeds, pests and diseases. The increased temperature in no-till systems may also favour the development of pathogenic fungi (e.g., Yarham, 1979) requiring the use of fungicides detrimental to AM fungi. The implication of interactions between rotation design and cultivation on AM fungi are clear, in that the combined use of excess tillage and non-mycorrhizal crops or fallow will be additive. Johnson and Pfleger (1992) hypothesise that tillage is likely to place a strong selection pressure on AM fungal communities; one of the very few studies that investigates this hypothesis is by Douds et al. (1995). The importance of AM fungi in soil structure formation was discussed earlier. Jastrow (1987) demonstrated that soil disturbance was likely to have a long-term

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 203

effect on aggregate stability. Wright and Upadhyaya (1998) found at 37 different sites that aggregate stability was correlated with components of the glycoprotein glomalin produced by AM fungi. Thus when hyphal networks are disrupted by tillage there is bound to be a negative effect on aggregate stability. Although the effects of tillage on AM fungi have been widely studied, the interactions between mycorrhizal fungi and soil compaction have received less attention. Soil compaction, whether caused by machinery or livestock, can result in poor root growth and reduced nutrient uptake (e.g., Shierlaw and Alston, 1984). Under some conditions reduced tillage can also result in soil compaction (Vyn and Raimbault, 1993). Pasture management has also recently been shown to influence soil microbial biomass, with lower stocking rates promoting higher biomass C (Bannerjee et al., 2000). Entry et al. (1996) and Mulligan et al. (1985) both showed a reduction in root growth and AM fungal colonisation with increased soil bulk density. Yano et al. (1998) demonstrated for the first time that AM formation promoted root elongation in compacted soils. Under experimental conditions Li et al. (1997) demonstrated that AM fungal hyphae were more efficient at obtaining P from compacted soils than either mycorrhizal or nonmycorrhizal roots of red clover.

E. NUTRIENT MANAGEMENT In agricultural systems, a number of practices are used for modifying soil nutrient status including the use of mineral fertilisers with a range of solubilities, and organic materials from a variety of on-farm and off-farm sources. This includes manures, crop residues and urban wastes. The use of these materials results both in changes in the available nutrient content of soils during the growing season and also annual changes on the balance between nutrient inputs and offtake on a field basis. The use of readily soluble fertilizers is unacceptable in organic and biodynamic farming (IFOAM, 2000), although products that release nutrients over a period of years rather than weeks, such as rock phosphate (Rajan et al., 1996), are acceptable. Furthermore, such products are not usually applied every year but only in certain phases of the rotation, thus their use is planned to build soil fertility in the longer-term. Mosse (1973) estimated that 75% of all P fertiliser applied to crops is not used in the first year and reverts to forms unavailable to plants. The evidence for the impact of soluble fertilisers on colonisation and function of AM fungi is contradictory. The application of soluble P has been shown to increase root colonisation by AM fungi (Gryndler et al., 1990) but more usually to decrease it (e.g., Abbott and Robson, 1984; Jensen and Jakobsen, 1980; Liu et al., 2000; Mosse, 1973). Effects of P fertilisation can persist for some time, 11 years after P fertilisation was stopped in a long-term field trials in the Netherlands, AM fungal colonisation was still lowest on plots which had received

204

L. A. HARRIER AND C. A. WATSON

the highest P fertilizer dressing (Dekkers and van der Werff, 2001). Similarly contradictory results have been reported with N fertilizer (e.g., Azco´n et al., 1982; Baltruschat and Dehne, 1988; Buwaldah and Goh, 1982; Liu et al., 2000). In designing sustainable systems, it is also important, to understand the effect of added nutrients on the function of the AM symbiosis. The AM fungal contribution to plant nutrient uptake has been shown to be reduced at high levels of readily available P (Thingstrup et al., 1998, 2000). When less soluble forms of P, such as rock phosphate are used, however, mycorrhizal plants may take up more P than non-mycorrhizal plants (Fabig et al., 1989). The variability in the reports on the effects of fertilizers on AM mycorrhizal associations is probably linked to the P status of both the plant and the soil conditions. Thomson et al. (1991) reported that the impact of P fertilizers on the AM symbiosis was mediated by the plant under low and medium soil P levels, but by the soil at high levels of P. This is further complicated by the fact that isolates of AMF differ in their sensitivity to soil and plant P levels and therefore fertilizer application may alter the activity of the symbiosis (Sylvia and Schenck, 1983). Due to the variable nature of organic manures and other organic materials added to soils, it is even more difficult to generalise about the response of AM fungi to these materials. Harinikumar and Bagyaraj (1989) and Tarkalson et al. (1998a) report positive influences of organic manure on AM fungi but Tarkalson et al. (1998b) and Joner (2000) report negative effects. The use of sewage sludge and composted urban wastes in agriculture is considered to be a way of closing the nutrient cycle between urban and rural communities. These materials may, however, contain elements such as P, N and heavy metals that can reduce mycorrhizal activity. There are few reports of the effects of these materials on mycorrhiza in the literature, although Lambert and Weidensaul (1991) confirmed that sewage sludge could be detrimental to mycorrhizally mediated nutrient uptake of soybean. Sainz et al. (1998) demonstrated negative effects of composted urban wastes on AM colonisation of red clover. Joner (2000) suggests that moderate quantities of farm yard manure have a less detrimental effect on AM fungi than equivalent amounts of nutrients as NPK fertilizer. Similar trends have been reported for farmyard manure by Harinikumar and Bagyaraj (1989) and for broiler litter and leaf compost (Douds et al., 1997). Christie and Kilpatrick (1992) reported contradictory results for slurry on grassland. There are also reports within the literature of mycorrhiza formation and spore numbers being unaffected by organic amendments (e.g., Hafner et al., 1993). The difference in effect between fertilizer and different organic materials may relate to the temporal difference in P availability from the different materials and the more gradual release of P from manures being more synchronised to plant demand. Soil texture can also be important. Kabir et al. (1997) found that source of nutrients did not affect mycorrhizal colonisation or AM fungal hyphal abundance in sandy loam but within clay soil, hyphae were more abundant with manure application than mineral fertilizer.

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 205

Soil analysis for advisory purposes relies on the interpretation of the chemical extraction of different nutrient pools from the soil to predict nutrient release to crops (Edwards et al., 1997). It is often difficult to reconcile soil analysis and plant nutrient uptake because of the dependence of uptake on biological processes, AM fungal activity undoubtedly plays a part in this. In more sustainable systems, which are less dependent on the import of readily soluble forms of nutrients, trends in soil analysis may be more important than snapshots.

F. LIMING The tolerance of different agricultural crops to soil pH is well documented (e.g., Brady, 1990) and lime is generally applied every few years to maintain suitable pH levels. Soil pH affects AM fungal colonisation (Mamo and Killham, 1987; Wang et al., 1985), species distribution (Porter et al., 1987) and the effectiveness of the mycorrhizal symbiosis (Hayman and Tavares, 1985).

G. CROP PROTECTION There are many reports on the effects of crop protection chemicals on sustainable agriculture which have been reviewed by Trappe et al. (1984), Johnson and Pfleger (1992), Pe´rez-Moreno and Ferrera-Cerrato (1994) and Thompson (1994). From these reports, it is difficult to generalise on the impact of pesticides on AM fungi and their symbiotic interactions since both beneficial and detrimental effects have been reported. As sustainable agriculture is driven by both environmental and human health concerns, it is likely that in the future cultural methods of weed, pest and disease control will be used to decrease the use of crop protection chemicals. The effects of crop rotation on AM fungi are discussed above, but choice of crops and rotation length will also influence the incidence and severity of soil-borne pathogens. For example, Clark et al. (1998) showed reduced disease levels associated with longer gaps between susceptible crops. The potential of AM fungi for disease control has been discussed earlier, Jordan et al. (2000) has recently reviewed their potential as weed control agents.

H. GRAZING LIVESTOCK The effect of grazing livestock on soil structure is discussed above but livestock may influence AM fungi through the removal and return of nutrients. Removal of photosynthetic tissue as a result of grazing, or cutting for silage,

206

L. A. HARRIER AND C. A. WATSON

may affect AM fungi as it involves the alteration of C allocated to the roots and between the symbionts (Wang et al., 1989). Eom et al. (2001) found that increased grazing intensity resulted in increased root colonisation, and also changes in AM fungal species composition in tallgrass prairie. Differences in AM fungal responses to defoliation were demonstrated in grass species by Allsop (1998). In grasses tolerant to grazing e.g., Lolium perenne, root loss following defoliation is in part compensated for by maintenance of the external hyphal network. More defoliation intolerant species such as Themeda trianda are unable to maintain the hyphal network following defoliation. This suggests that AM fungi have a role in changing grassland species composition under grazing.

I. INOCULATION Despite the influence of AM fungi on plant growth, there are still relatively few examples of inoculation being used on a commercial basis in farming systems in developed countries (Smith and Read, 1997). This reflects the difficulties associated with the expense and bulk of the inoculum. There is also a suggestion that there may be a negative response to inoculation with nonindigenous fungi compared with indigenous species (Dhillion, 1992). There are, however, certain situations where inoculation has been shown to yield positive benefits. These are mostly where plants are transplanted rather than grown from seed in the field (e.g., Cantrell and Linderman, 2001), or grown in controlled conditions in nursery beds or glasshouses (e.g., Duffy and Cassells, 2000; Sharma and Adholeya, 2000).

V. EFFECTS OF DIFFERENT AGRICULTURAL SYSTEMS ON AM FUNGI A. ORGANIC VERSUS BIODYNAMIC VERSUS CONVENTIONAL FARMING A number of observational studies have now been carried out on the differences between spore populations (e.g., Douds et al., 1993; Galvez et al., 2001; Kahiluoto and Vestberg, 1998; Kurle and Pfleger, 1994), colonisation levels (e.g., Ryan et al., 1994; Sattelmacher et al., 1991), and species richness and diversity (e.g., Douds et al., 1993; Franke-Snyder et al., 2001; Kurle and Pfleger, 1996) in agricultural systems managed in different ways. The differences in system management between studies means that these results can only be used to generalise on effects of farming system within studies rather than to make

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 207 Table III Spore Populations of AM Fungi in Alternative Agricultural Systems Presented as Ratio of Spores in Alternative Compared with Conventional Comparisons System

Crop

Sampling time

Spore populationsa,b

Reference

Low input Low input Low input Low input Organic

Rotation Rotation Corn Soybean Grassland

Spring Autumn AV AV AV

1.2:1–2.8:1 1.6:1–2.7:1 1.9:1 1.5:1 3:1

Galvez et al. (2001) Galvez et al. (2001) Kurle and Pfleger (1994) Kurle and Pfleger (1994) Eason et al. (1999)

a

Alternative:conventional. Recalculated from published literature. AV—average values. b

comparisons between studies. Generally, AM fungal spore populations are higher in lower input systems (Table III). Several studies show higher AM fungal colonisation rates in systems with lower inputs of fertilizers and pesticides (Table IV), with Douds et al. (1993) finding over seven times the infection rate in soybeans grown in low input green manure based systems compared with conventional. Using the Shannon – Wiener index of diversity, Kurle and Pfleger (1996) and Franke-Snyder et al. (2001) found similar fungal communities in

Table IV Levels of AM Fungal Colonisation of Field Crops in Alternative Systems presented as a Ratio Compared with Conventional System

Crop

Colonisation (%)a

Reference

Low input (green manure based) Low input (animal manure based) Minimum input Organic manure input Low input (green manure based) Low input (animal manure based) Minimum input Organic manure input Organic farming Organic farming Organic farming Biodynamic farming Organic farming Organic farming Biodynamic farming Organic conversion

Maize Maize Maize Maize Soybean Soybean Soybean Soybean Wheat Wheat Wheat Wheat Grass Grass Grass Strawberry

0.87:1 1.23:1 1.26:1 0.8:1 7.4:1 4.3:1 1.1:1 1.26:1 2.1:1– 3.1:1 1.3:1– 2.8:1 1.6:1 1.6:1 1.3:1 1.3:1 1.4:1 2.5:1– 3:1

Douds et al. (1993) Douds et al. (1993) Kurle and Pfleger (1994) Kurle and Pfleger (1994) Douds et al. (1993) Douds et al. (1993) Kurle and Pfleger (1994) Kurle and Pfleger (1994) Ryan et al. (1994) Dann et al. (1996) Ma¨der et al. (2000) Ma¨der et al. (2000) Ma¨der et al. (2000) Eason et al. (1999) Ma¨der et al. (2000) Werner et al. (1990)

a

Calculated from published literature.

208

L. A. HARRIER AND C. A. WATSON

systems receiving different inputs. The published studies of mycorrhizal communities have, however, so far relied on the use of spore morphology to characterise AM fungi which cannot give a complete picture of the systems (Douds and Millner, 1999; Kurle and Pfleger, 1996). In many of the published studies, it is difficult to interpret the cause of the observed differences as there are multiple factors involved. The Swiss DOC trial, however, compares conventional, organic and biodynamic management systems using identical crop rotations and tillage but different fertilisation and plant protection strategies (Ma¨der et al., 2000). Root colonisation was highest in the low input treatments, which was in part attributable to soil chemical properties in the different systems. It is important that future comparisons between farming systems take into account the efficacy of the mycorrhizal communities in enhancing plant growth (Franke-Snyder et al., 2001). One of the underlying principles of biodynamic agriculture is the use of preparations based on material of plant and animal origin to stimulate soil biological activity. Scientific studies of this form of agriculture are rare although both Penfold et al. (1995) and Carpenter-Boggs et al. (2000) found that microbial biomass was unaffected by biodynamic preparations. It is perhaps more likely that these preparations stimulate the biomass and/or activity of specific organisms; no published studies of their effect on AM fungi are known.

B. LONG-TERM VERSUS SHORT-TERM EFFECTS Changing from conventional to more sustainable agricultural practices often takes place over a number of years. In conversion to organic agriculture, it is common to find a temporary dip in yield following cessation of fertilizer and pesticides (Lampkin, 1990). Ma¨der et al. (2000) observed that during the second 7 years of a crop rotation experiment, yields of crops receiving 70% less available N and 50% less P and K input were reduced only by 19 –24%. They hypothesise that the AM fungi and their associated symbiotic interactions may symbiosis play an important role in these unexpectedly high yields. Conflicting information has been published on the temporal response of AM fungi to these changes. Dekkers and van der Werff (2001) found that the impact of fertilizer use on AM fungal colonisation persisted 11 years after cessation of fertilizer application. Limonard and Ruissen (1989), however, found a large increase in AM colonisation 4 years after conversion to a low input system. Douds et al. (1993) reported increased spore numbers in low input compared with conventional in a 10-year-old field trial. In the same trial after 15 years, Franke-Snyder et al. (2001) showed that there were few differences in the composition and structure of the AM fungal community between systems.

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 209

VI. MANAGING AM FUNGI IN SUSTAINABLE AGRICULTURE—PROSPECTS FOR THE FUTURE A. ECONOMICS AND PRODUCT QUALITY In a review of the results of field experiments on AM fungi, McGonigle (1988) concluded that there was insufficient evidence for a mutualistic function of the AM symbiosis. If farmers and agricultural advisors are to give serious consideration to the management of AM fungi, they must be convinced that there is an economic benefit. Economic analysis is rarely included in experimental protocols although Miller et al. (1994) suggest an economic analysis approach to assess the potential benefits of increased AM effectivity. They also suggest that environmental benefits could be factored into this analysis. Adding value to agricultural production is an important consideration within current agriculture. Premiums are currently paid for food produced organically or with other perceived quality benefits. The role of AM fungi in improving product quality has received little attention, except in terms of nutrient content, although Charron et al. (2001) showed that inoculation of onions with AM fungi improved bulb firmness.

B. RESEARCH AND DEVELOPMENT NEEDS The multifunctional role of AM fungi within agro-ecosystems is being increasingly recognised. In some cases, where a plant is colonised but no benefit is demonstrated the net cost of the symbiosis may exceed the net benefit to the plant (Johnson et al., 1997), but there may be a benefit to the agroecosystem, e.g., soil structure and microbial populations. Investigating the multifunctional aspects of the AM fungal symbiosis presents challenges in terms of experimental design, and may require measurement of parameters that have previously been ignored. For example, Mozafar et al. (2000) suggest that in field studies the presence of non-mycorrhizal root parasites, especially non-filamentous obligate fungi, should be taken into account. One of the difficulties in interpreting data from comparative studies of farming systems has been the reliance on the use of spore morphology to characterise AM fungi which cannot give a complete picture of the systems (Douds and Millner, 1999). The nature of the comparisons made can further increase the difficulty in interpreting results, for example, Kahiluoto and Vestberg (1998) compared conventionally managed continuous cereals with an organically managed ley/ arable rotation. Similarly many studies which compare the use of nutrients in different forms (e.g., organic manure and superphosphate) have failed to make comparisons on the basis of equivalent amounts of nutrients.

210

L. A. HARRIER AND C. A. WATSON

Translating the results of research into practical recommendations is difficult because of the interaction between factors involved in the plant mycorrhizal symbiosis, and the separation of cause and effect. For example, plants grown in fumigated soil show P deficiency in the early season as a result of reduced mycorrhizal infection (Jawson et al., 1993). It is also clear that measuring colonisation levels of crops by AM fungi in the field is no longer enough, it is important to understand function (Thingstrup et al., 1998). Basic information on seasonal variation in the development of AM fungi in different crop species is still lacking. Much of the available information on seasonal variation is based on spore counts that may or may not correlate with colonisation and function. Kling and Jakobsen (1998) noted the need for reliable tools for the assessment of mycorrhizal form and function. Daniell et al. (2001) have recently used DNA studies to confirm seasonal variation in colonisation of arable crops. Molecular techniques, used together with a knowledge of mycorrhizal ecology and crop physiology, hold the key to improved understanding of the functioning of AM fungi in sustainable crop production systems. The development of a diverse AM fungal population which can adapt to management and environmental changes is likely to be a key factor in improving the sustainability of low input and organic cropping systems. The future success of agriculture depends on improving our understanding of the dynamic relationships between agricultural practices and soil biology, and our ability to exploit these.

ACKNOWLEDGMENTS SAC receives financial support from the Scottish Executive Environment and Rural Affairs Department (SEERAD). The authors would like to thank Professor Martin Wolfe for helpful discussion on plant breeding programmes.

REFERENCES Abbott, L. K., and Robson, A. D. (1984). The effect of mycorrhizae on plant growth. In “VA Mycorrhiza”. (C. L. Powel and D. J. Bagyaraj, Eds.), pp. 113–130. CRC Press, Boca Raton, FL. Abuzinadah, R. A., and Read, D. J. (1989). The role of proteins in the nitrogen nutrition of ectomycorrhizal plants. IV. The utilisation of peptides by birch (Betula pendula L.) infected with different mycorrhizal fungi. New Phytol. 112, 55 –60. Albrecht, C., Geurts, R., and Bisseling, T. (1999). Legume nodulation and mycorrhizae formation two extremes in host specificity meet. EMBO J. 18, 281 –289. Allsop, N. (1998). Effects of defoliation on the arbuscular mycorrhizas of three perennial pasture and rangeland grasses. Plant Soil 202, 117–124.

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 211 Altieri, M. A. (1995). In “Agroecology: The Science of Sustainable Agriculture”. Intermediate Technology Publications, London. Ames, R. N., Reid, C. P. P., Porter, L. K., and Cambardella, C. (1983). Hyphal uptake and transport of nitrogen from two 15N-labelled sources by Glomus mosseae, a vesicular arbuscular mycorrhizal fungus. New Phytol. 95, 381 –396. Amora-Lazcano, E., and Azco´n, R. (1997). Response of sulphur cycling micro-organisms to arbuscular mycorrhizal fungi in the rhizosphere of maize. Appl. Soil Ecol. 6, 217–222. Arihawa, J., and Karasawa, T. (2000). Effect of previous crops on arbuscular mycorrhizal formation and growth of succeeding maize. Soil Sci. Plant Nutr. 46, 43– 51. Arines, J., Vilarino, A., and Sainz, M. (1989). Effect of different inocula of vesicular–arbuscular mycorrhizal fungi on manganese content and concentration in red clover (Trifolium pratense L.) plants. New Phytol. 112, 215–219. Atkinson, D. (1992). Tree root development: the role of models in understanding the consequences of arbuscular endomycorrhizal infection. Agronomie 12, 817–820. Atkinson, D., Berta, G., and Hooker, J. E. (1994). Impact of mycorrhizal colonisation on root architecture, root longevity and formation of growth regulators. In “Impact of Arbuscular Mycorrhizas on Sustainable Agriculture and Natural Ecosystems”. (S. Gianinazzi and H. Schu¨epp, Eds.), pp. 89–99. Birkhauser, Basel, Switzerland. Atkinson, D., Watson, C. A., Hooker, J. E., and Black, K. E. (1995). Nutrient conservation in organic crop production systems. In “Soil Management in Sustainable Agriculture”. (H. F. Cook and H. C. Lee, Eds.), pp. 54–60. Wye College Press, Ashford. Auge´, R. M. (2000). Stomatal behaviour of arbuscular mycorrhizal plants. In “Mycorrhizal Symbiosis: Molecular Biology and Physiology”. (Y. Kapulnik and D. Douds, Eds.), pp. 201–237. Kluwer Academic Publishers, Dordrecht. Auge´, R. M. (2001). Water relations, drought and vesicular–arbuscular mycorrhizal symbiosis. Mycorrhiza 11, 3–42. Azaizeh, H. A., Marschner, H., Ro¨mheld, V., and Wittenmayer, L. (1995). Effects of a vesicular– arbuscular mycorrhizal fungus and other soil microorganisms on growth, mineral nutrient acquisition and root exudation of soil grown plants. Mycorrhiza 5, 321–327. Azco´n, R., and Barea, J. M. (1992). The effect of vesicular arbuscular mycorrhizae in decreasing Ca acquisition by alfalfa plants in calcareous soils. Biol. Fert. Soils 13, 155 –159. Azco´n, R., and Barea, J. M. (1997). Mycorrhizal dependency of a representative plant species in mediterranean shrublands (Lavendula spica L.) as a key factor to its use for re-vegetation strategies in desertification threatened areas. Appl. Soil Ecol. 7, 83–92. Azco´n, R., and El-Atrach, F. (1997). Influence of arbuscular mycorrhizae and phosphorous fertilization on growth, nodulation and N2 (N-) in Medicago sativa at four salinity levels. Biol. Fert. Soils 24, 81–86. Azco´n-Aguilar, C., and Barea, J. M. (1996). Arbuscular mycorrhizas and biological control of soil-borne plant pathogens—an overview of the mechanisms involved. Mycorrhiza 6, 457 –464. Azco´n, R., Gomez-Ortega, M., and Barea, J. M. (1982). Comparative effects of foliar- or soil-applied nitrate on vesicular–arbuscular mycorrhizal infection in maize. New Phytol. 92, 553–559. Ba¨a¨th, E., and Hayman, D. S. (1983). Plant growth responses to vesicular arbuscular mycorrhiza. XIV. Interactions with Verticillium Wilt on tomato plants. New Phytol. 95, 419–426. Ba¨a¨th, E., and Hayman, D. S. (1984). No effect of VA mycorrhiza on red core disease of strawberry. Trans. Br. Mycol. Soc. 82, 534 –536. Bagayoko, M., Buerkert, A., Lung, G., Bationo, A., and Ro¨mheld, V. (2000). Cereal/legume rotation effects on cereal growth in Sudano-Sahelian West Africa: soil mineral nitrogen, mycorrhizae and nematodes. Plant Soil 218, 103 –116. Bagyaraj, D. J. (1984). Biological interactions with VA mycorrhizal fungi. In “VA Mycorrhiza”. (C. L. Powell and D. J. Bagyaraj, Eds.), pp. 163–198. CRC Press, Boca Raton, FL.

212

L. A. HARRIER AND C. A. WATSON

Baltruschat, H., and Dehne, H. W. (1988). The occurrence of vesicular–arbuscular mycorrhiza in agro-ecosystems. 1. Influence of nitrogen fertilizer and green manure in continuous monoculture and in crop rotation on the inoculum potential of winter wheat. Plant Soil 107, 279–284. Bannerjee, M. R., Burton, D. L., McCaughey, W. P., and Grant, C. A. (2000). Influence of pasture management on soil biological quality. J. Range Mgmt 53, 127–133. Barea, J. M., and Jefferies, P. (1995). Arbuscular mycorrhizas in sustainable plant–soil systems. In “Mycorrhizae: Function, Molecular Biology and Biotechnology”. (B. Hock and A. Varma, Eds.), pp. 521–560. Springer, Berlin. Bartschi, H., Gianinazzi-Pearson, V., and Vegh, I. (1981). Vesicular arbuscular mycorrhiza formation and root rot disease Phytophthora cinnamomi development in Chamaecyparis lawsonia. Phytopath. Z. 102, 213 –218. Becard, G., and Fortin, J. A. (1988). Early events of vesicular–arbuscular mycorrhiza formation on Ri T-DNA transformed roots. New Phytol. 108, 211–218. Bermudez, M., and Azco´n, R. (1996). Calcium uptake by alfalfa as modified by a mycorrhizal fungus and liming. Symbiosis 20, 175 –184. Berta, G., Fusconi, A., and Trotta, A. (1993). VA mycorrhizal infection and the morphology and function of root systems. Environ. Expp. Bot. 33, 159–173. Bethlenfalvay, G. J., and Franson, R. L. (1989). Manganese toxicity alleviated by mycorrhizae in soybean. J. Plant Nutr. 12, 953–970. Bethlenfalvay, G. J., and Linderman, R. G. (1992). In “Mycorrhizae in Sustainable Agriculture”. ASA Special Publication, Madison, WI. Bethlenfalvay, G. J., and Schu¨epp, H. (1994). Arbuscular mycorrhizas and agrosystem stability. In “Impact of Arbuscular Mycorrhizas on Sustainable Agriculture and Natural Ecosystems”. (S. Gianinazzi and H. Schu¨epp, Eds.), pp. 117– 131. Birkhauser, Basel. Bethlenfalvay, G. J., Franson, R. L., Brown, M. S., and Mihara, K. L. (1989). The Glycine– Glomus– Bradyrhizobium symbiosis. IX. Nutritional, morphological and physiological responses of nodulated soybean to geographic isolates of the mycorrhizal fungus Glomus mosseae. Physiol. Planta 76, 226–232. Bianciotto, V., Bandi, C., Minerdi, D., Sironi, M., Volker-Tichy, H., and Bonfante, P. (1996). An obligately endosymbiotic mycorrhizal fungus itself harbours obligately intracellular bacteria. Appl. Environ. Microbiol. 62, 3005–3010. Black, R., and Tinker, P. B. (1979). The development of endomycorrhizal root systems. II. Effect of agronomic factors and soil conditions on the development of vesicular–arbuscular mycorrhizal infection in barley and on the endophyte spore density. New Phytol. 83, 401–413. Boddington, C. L., and Dodd, J. C. (1998). A comparison of the development and metabolic activity of mycorrhizas formed by arbuscular mycorrhizal fungi from different genera on two tropical forage legumes. Mycorrhiza 8, 149 –157. Bolan, N. S. (1991). A critical review of the role of mycorrhizal fungi in the uptake of phosphorous by plants. Plant Soil 134, 189–207. Bonfante, P., and Perotto, S. (1995). Strategies of arbuscular mycorrhizal fungi when infecting host plants. New Phytol. 130, 3 –21. Borowicz, V. A. (1997). A fungal root symbiont modifies plant resistance to an insect herbivore. Oecologia 112, 534– 542. Boswell, E. P., Koide, R. T., Shumway, D. L., and Addy, H. D. (1998). Winter wheat cover cropping, VA mycorrhizal fungi and maize growth and yield. Agric. Ecosyst. Environ. 67, 55–65. Bourgeois, L., and Entz, M. H. (1996). Influence of previous crop type on yield of spring wheat: analysis of commercial field data. Can. J. Plant Sci. 76, 457– 459. Brady, N. C. (1990). In “The Nature and Properties of Soils” (10th ed.). Macmillan, New York. Bryla, D. R., and Koide, R. T. (1990). Regulation of reproduction in wild and cultivated Lycopersicon esculentum Mill. by vesicular arbuscular mycorrhizal infection. Oecologia 84, 74– 81.

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 213 Bryla, D. R., and Koide, R. T. (1990). Role of mycorrhizal infection in the growth and reproduction of wild vs. cultivated plants. II. Eight wild accessions and two cultivars of Lycopersicon esculentum Mill. Oecologia 84, 82–92. Buwalda, J. G., Stribley, D. P., and Tinker, P. B. (1983). Increased uptake of anions by plants with vesicular–arbuscular mycorrhiza. Plant Soil 71, 463–467. Buwaldah, J. G., and Goh, K. M. (1982). Host–fungus competition for carbon as a cause of growth depression in vesicular–arbuscular mycorrhizal ryegrass. Soil Biol. Biochem. 14, 103–106. Cantrell, I. C., and Linderman, R. G. (2001). Preinoculation of lettuce and onion with VA mycorrhizal fungi reduces deleterious effects of soil salinity. Plant Soil 233, 269–281. Caron, M. (1989). Potential use of mycorrhizae in control of soil borne diseases. Can. J. Plant Pathol. 11, 177–179. Caron, M., Fortin, J. A., and Richard, C. (1985). Influence of substrate on the interaction of Glomus intraradices and Fusarium oxysporum f. spp. radicis-lycopersici on tomatoes. Plant Soil 87, 233 –239. Carpenter-Boggs, L., Kennedy, A. C., and Reagnold, J. (2000). Organic and biodynamic management: effects on soil biology. Soil Sci. Soc. Am. J. 64, 1651– 1659. Chalot, M., and Brun, A. (1998). Physiology of organic nitrogen acquisition by ectomycorrhizal fungi and ectomycorrhizas. FEMS Microbiol. Rev. 22, 21–44. Charest, C., Dalpe, Y., and Brown, A. (1993). The effect of two arbuscular mycorrhizae and chilling on two hybrids of Zea mays L. Mycorrhiza 4, 89– 92. Charron, G., Furlan, V., Bernier-Cardou, M., and Doyon, G. (2001). Response of onion plants to arbuscular mycorrhizae. 2. Effects of nitrogen fertilization on biomass and bulb firmness. Mycorrhiza 11, 145– 150. Chiariello, N., Hickman, J. C., and Mooney, H. A. (1982). Endomycorrhizal role for interspecific transfer of phosphorous in a community of annual plants. Science 217, 941 –943. Chilvers, M. T., and Daft, M. J. F. (1982). Effects of low temperature on development of the vesicular–arbuscular mycorrhizal association between Glomus caledonium and Allium cepa. Trans. Br. Mycol. Soc. 79, 153 –157. Christie, P., and Kilpatrick, D. J. (1992). Vesicular–arbuscular mycorrhiza infection in cut grassland following long-term slurry application. Soil Biol. Biochem. 24, 325–330. Clark, R. B., and Zeto, S. K. (1996). Mineral acquisition by mycorrhizal maize grown on acid and alkaline soil. Soil Biol. Biochem. 28, 1495–1503. Clark, R. B., and Zeto, S. K. (2000). Mineral acquisition by arbuscular mycorrhizal plants. J. Plant Nutr. 23, 867–902. Clark, M. S., Ferris, H., Klobsky, K., Lanini, W. T., van Bruggen, A. H. C., and Zalom, F. G. (1998). Agronomic, economic and environmental comparison of pest management in conventional and alternative tomato and corn systems in northern California. Agric. Ecosyst. Environ. 68, 51– 71. Clark, R. B., Zobel, R. W., and Zeto, S. K. (1999). Arbuscular mycorrhizal fungal isolate effectiveness on growth and root colonisation of Panicum virgatum in acidic soil. Soil Biol. Biochem. 31, 1757–1763. Clark, R. B., Zobel, R. W., and Zeto, S. K. (1999). Effects of mycorrhizal fungus isolates on mineral acquisition by Panicum virgatum in acidic soil. Mycorrhiza 9, 167– 176. Cooper, K. M., and Tinker, P. B. (1978). Translocation and transfer of nutrients in vesicular arbuscular mycorrhizas. II. Uptake and translocation of phosphorous, zinc and sulphur. New Phytol. 81, 43 –52. Cordier, C., Gianinazzi, S., and Gianinazzi-Peaarson, V. (1996). Colonisation patterns of root tissues by Phytophthora nicotiania var parasitica related to reduced diseases in mycorrhizal tomato. Plant Soil 185, 223 –232. Cordier, C., Pozo, M. J., Gianinazzi, S., and Gianinazzi-Pearson, V. (1998). Cell defense responses associated with localized and systemic resistance to Phytophthora parasitica induced in tomato by an arbuscular mycorrhizal fungus. Mol. Plant—Microbe Interact. 11, 1017–1028.

214

L. A. HARRIER AND C. A. WATSON

Cox, G., and Tinker, P. B. (1976). Translocation and transfer of nutrients in vesicular–arbuscular mycorrhizas. I. The arbuscule and phosphorous transfer: a quantitative ultrastructural study. New Phytol. 77, 371. Cox, G., Sanders, F. E., Tinker, P. B., and Wild, J. A. (1975). Ultrastructural evidence relating to host– endophyte transfer in a vesicular–arbuscular mycorrhiza. In “Endomycorrhizas”. (F. E. Sanders, B. Mosse and P. B. Tinker, Eds.), pp. 297 –312. Academic Press, London. Daniell, T., Husband, R., Fitter, A. H., and Young, J. P. W. (2001). Molecular diversity of arbuscular mycorrhizal fungi colonising arable crops. FEMS Microbiol. Ecol. 36, 203 –209. Dann, P. R., Derrick, J. W., Dumaresq, D. C., and Ryan, M. H. (1996). The response of organic and conventionally grown wheat to superphosphate and reactive rock phosphate. Aust. J. Expp. Agric. 36, 71– 78. Davies, R. M., Menge, J. A., and Erwin, D. C. (1979). Influence of Glomus fasiculatum and soil phosphorous on Verticillium wilt of cotton. Phytopathology 69, 453–456. Declerck, S., Strullu, D. G., and Plenchette, C. (1996). In vitro mass-production of the arbuscular mycorrhizal fungus Glomus versiforme, associated with Ri T-DNA transformed carrot roots. Mycol. Res. 100, 1237–1242. Declerck, S., Strullu, D. G., and Plenchette, C. (1998). Monoxenic culture of the intraradical forms of Glomus sp isolated from a tropical ecosystem: a proposed methodology for germplasm collection. Mycologia 90, 579– 585. Dehn, B., and Schu¨epp, H. (1989). Influence of VA mycorrhizae on the uptake and distribution of heavy metals in plants. Agric. Ecosyst. Environ. 29, 79 –83. Dehne, H. W. (1982). Interaction between vesicular arbuscular mycorrhizal fungi and plant pathogens. Phytopathology 72, 1115–1119. Dekkers, T. B. M., and van der Werff, P. A. (2001). Mutualistic functioning of indigenous arbuscular mycorrhizae in spring barley and winter wheat after cessation of long-term phosphate fertilization. Mycorrhiza 10, 195 –201. Dhillion, S. S. (1992). Host–endophyte specificity of vesicular–arbuscular mycorrhizal colonization of Oryza sativa L. at the pre-transplant stage in low or high phosphorus soil. Soil Biol. Biochem. 24, 405 –411. Diaz, G., Azco´n-Aguilar, C., and Honrubia, M. (1996). Influence of arbuscular mycorrhiza on heavy metal (Zn and Pb) uptake and growth of Lygeum spartum and Anthyllis cytisoides. Plant Soil 180, 241–249. Diop, T. A., Becard, G., and Piche, Y. (1992). Long term in vitro culture of an endomycorrhizal fungus Gigaspora margarita on Ri T-DNA transformed roots of carrot. Symbiosis 12, 249– 259. Diop, T. A., Plenchette, C., and Strullu, D. G. (1994). Dual axenic culture of sheared root inocula of vesicular–arbuscular mycorrhizal fungi associated with tomato roots. Mycorrhiza 5, 17– 22. Dodd, J. C., and Jefferies, P. (1986). Early development of vesicular– arbuscular mycorrhizas in autumn-sown cereals. Soil Biol. Biochem. 18, 149–154. Dodd, J. C., Krikun, J., and Haas, J. (1983). Relative effectiveness of indigenous populations of vesicular arbuscular mycorrhizal fungi from four sites in the Negev. Isr. J. Bot. 32, 10– 21. Dodd, J. C., Burton, C. C., Burns, R. G., and Jefferies, P. (1987). Phosphatase activity associated with the roots and rhizosphere of plants infected with vesicular arbuscular mycorrhizal fungi. New Phytol. 107, 163–172. Douds, D. D., and Millner, P. (1999). Biodiversity of arbuscular mycorrhizal fungi in agroecosystems. Agric. Ecosyst. Environ. 74, 77 –93. Douds, D. D., Janke, R. R., and Peters, S. E. (1993). VAM fungus spore populations and colonization of roots of maize and soybean under conventional and low-input sustainable agriculture. Agric. Ecosyst. Environ. 43, 325–335. Douds, D. D., Galvez, L., Janke, R. R., and Wagoner, P. (1995). Effect of tillage and farming system upon populations and distribution of vesicular–arbuscular mycorrhizal fungi. Agric. Ecosyst. Environ. 52, 111– 118.

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 215 Douds, D. D., Galvez, L., Franke-Snyder, M., Reider, C., and Drinkwater, L. E. (1997). Effect of compost addition and crop rotation point upon VAM fungi. Agric. Ecosyst. Environ. 65, 257 –266. Douds, D. D., Galvez, L., Becard, G., and Kapulnik, Y. (1998). Regulation of arbuscular mycorrhizal development by plant host and fungus species in alfalfa. New Phytol. 138, 27–35. Drew, M. C., and Sakar, L. R. (1978). Nutrient supply and the growth of the seminal root system in barley III compensatory increases in growth of lateral root systems, and in rates of phosphate uptake in response to a localised supply of phosphate. J. Expp. Bot. 29, 435 –451. Duffy, E. M., and Cassells, A. C. (2000). The effect of inoculation of potato (Solanum tuberosum L.) microplants with arbuscular mycorrhizal fungi on tuber yield and tuber size distribution. Appl. Soil Ecol. 15, 137–144. Eason, W. R., Scullion, J., and Scott, E. P. (1999). Soil parameters and plant responses associated with arbuscular mycorrhizas from contrasting grassland management regimes. Agric. Ecosyst. Environ. 73, 245–255. Edwards, A. C., Withers, P. J. A., and Sims, T. J. (1997). Are current fertiliser recommendation systems for phosphorus adequate? Fert. Soc. Proc. 404. Ellis, J. R., Watson, D. M. H., Varvel, G. E., and Jawson, M. D. (1995). Methyl bromide soil fumigation alters plant element concentrations. Soil Sci. Soc. Am. J. 59, 848–852. El-Tohamy, W., Schnitzler, W. H., El-Behairy, U., and El-Betagy, M. S. (1999). Effect of VA mycorrhizas on improving drought and chilling tolerance of bean plants. J. Appl. Bot. 73, 178 –183. Entry, J. A., Reeves, D. W., Lee, W. J., Guertal, E., and Raper, R. L. (1996). Influence of compaction from wheel traffic and tillage on arbuscular mycorrhizae infection and nutrient uptake by Zea mays. Plant Soil 180, 139 –146. Eom, A. H., Wilson, G. W. T., and Hartnett, D. C. (2001). Effects of ungulate grazers on arbuscular mycorrhizal symbiosis and fungal community structure in tallgrass prairie. Mycologia 93, 233 –242. Espinoza-Victoria, D., Quintero-Ramos, M., Ferrera-Cerrato, R., and Bethlenfalvay, G. J. (1993). Fitting plants to soil through mycorrhizal fungi. Plant nutrition in host endophyte combinations evaluated by the diagnosis and recommendation integrated system. Biol. Fert. Soils 15, 96–102. Estau´n, V., Calvet, C., and Hayman, D. S. (1987). Influence of plant genotype on mycorrhizal infection: response of three pea cultivars. Plant Soil 103, 295–298. Fabig, B., Moawad, A. M., and Achtnich, W. (1989). Effect of VA mycorrhiza on dry weight and phosphorus content in shoots of cereals fertilized with rock phosphates at different soil pH and temperature levels. Z. Pflanzenernahr Bodenkd. 152, 255–259. Fitter, A. H. (1985). Functioning of vesicular–arbuscular mycorrhizas under field conditions. New Phytol. 99, 257–265. Francis, R., Finlay, R. D., and Dear, D. J. (1986). Vesicular–arbuscular mycorrhiza in natural vegetation systems. IV. Transfer of nutrients in inter- and intra-specific combinations of host plants. New Phytol. 102, 103–111. Franke-Snyder, M., Douds, D. D., Galvez, L., Phillips, J. G., Wagoner, P., Drinwater, L., and Morton, J. B. (2001). Diversity of communities of arbuscular mycorrhizal (AM) fungi in conventional versus low-input agricultural sites in eastern Pennsylvania, USA. Appl. Soil Ecol. 16, 35–48. Frey, B., and Schu¨epp, H. (1993). Acquisition of nitrogen by external hyphae of vesicular–arbuscular mycorrhizal fungus associated with Zea mays L. New Phytol. 124, 221–230. Fusconi, A., Berta, G., Tagliasacchi, A. M., Scannerini, S., Trotta, A., Gnavi, E., and De Padova, S. (1994). Root apical meristems of arbuscular mycorrhizae of Allium porrum L. Environ. Expp. Bot. 33, 159–173. Galvez, L., Douds, D. D., Drinkwater, L. E., and Wagoner, P. (2001). Effect of tillage and farming system upon VAM fungus populations and mycorrhizas and nutrient uptake of maize. Plant Soil 228, 299–308.

216

L. A. HARRIER AND C. A. WATSON

Ganade, G., and Brown, V. K. (1997). Effects of below-ground insects, mycorrhizal fungi and soil fertility on the establishment of Vicia in grassland commnities. Oecologia 109, 374–381. Gange, A. C. (2001). Species specific responses of a root and shoot-feeding insect to arbuscular mycorrhizal colonisation of its host plant. New Phytol. 150, 611–618. Gange, A. C., and Bower, E. (1997). Interactions between insects and mycorrhizal fungi. In “Multitrophic Interactions in Terrestrial Systems”. (A. C. Gange and V. K. Brown, Eds.), pp. 115 –131. Blackwell Science, Oxford. Gange, A. C., and West, H. M. (1994). Interactions between arbuscular mycorrhizal fungi and foliar feeding insects in Plantago lanceolata L. New Phytol. 128, 79–87. Gange, A. C., Brown, V. K., and Farmer, L. M. (1990). A test of mycorrhizal benefit in an early successional plant community. New Phytol. 115, 85– 91. Gavito, M. E., and Miller, M. H. (1998). Changes in mycorrhiza development in maize induce by crop management practices. Plant Soil 198, 185– 192. Gavito, M. E., and Miller, M. H. (1998). Early phosphorus nutrition, mycorrhizae development, dry matter partitioning and yield of maize. Plant Soil 199, 177–186. Gehrig, C. A., and Whitham, T. G. (1994). Interactions between above ground herbivores and the mycorrhizal mutualists of plants. Trends Ecol. Evol. 9, 251 –255. George, E., Haussler, K., Vetterlein, D., Gorgus, E., and Marschener, H. (1992). Water and nutrient translocation by hyphae of Glomus mosseae. Can. J. Bot. 70, 2130–2137. Gianinazzi, S. (1991). Vesicular arbuscular (endo-) mycorrhizas—cellular, biochemical and genetic aspects. Agric. Ecosyst. Environ. 35, 105–119. Gianinazzi-Pearson, V. (1996). Plant cell responses to arbuscular mycorrhizal fungi: getting to the roots of the symbiosis. Plant Cell 8, 1871–1883. Grime, J. P., Mckey, J. M. L., Hillier, S. H., and Read, D. J. (1987). Floristic diversity in a model system using experimental microcosms. Nature 328, 420–422. Gryndler, M., Lestina, J., Moravec, V., Prikyl, Z., and Lipavsky, J. (1990). Colonisation of maize roots by VAM under conditions of long-term fertilisation of varying intensities. Agric. Ecosyst. Environ. 29, 183– 186. Guo, Y., George, E., and Marchner, H. (1996). Contribution of an arbuscular mycorrhizal fungus to the uptake of cadmium and nickel in bean and maize plants. Plant Soil 184, 195 –205. Gupta, R. K. (1991). Drought response in fungi and mycorrhizal plants. Handb. Appl. Mycol. 1, 55–75. Hafner, H., George, E., Bationo, A., and Marschner, H. (1993). Effect of crop residues on root growth and phosphorus acquisition of pearl millet in an acid sandy soil in Niger. Plant Soil 150, 117–127. Hamel, C., and Smith, D. L. (1991). Interspecific N-transfer and plant development in a mycorrhizal field grown mixture. Soil Biol. Biochem. 23, 661– 665. Harinikumar, K. M., and Bagyaraj, D. J. (1988). Effect of crop rotation on native vesicular arbuscular mycorrhizal propagules in soil. Plant Soil 110, 77–80. Harinikumar, K. M., and Bagyaraj, D. J. (1989). Effect of cropping sequence, fertilizers and farmyard manure on vesicular–arbuscular mycorrhizal fungi in different crops over three consecutive seasons. Biol. Fert. Soils 7, 173 –175. Harrier, L. A. (2000). Arbuscular mycorrhizal (AM) symbiosis: a review of signalling and molecular aspects of root colonisation. Bot. J. Scotl. 52, 159–169. Harrison, M. J. (1997). The arbuscular mycorrhizal symbiosis: an underground association. Trends Plant Sci. 2, 54–60. Harrison, M. J. (1998). Development of the arbuscular mycorrhizal symbiosis. Curr. Opin. Plant Biol. 1, 360– 365. Harrison, M. J. (1999). Molecular and cellular aspects of the arbuscular mycorrhizal symbiosis. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50, 361–389.

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 217 Harrison, M. J., and van Buuren, M. L. (1995). A phosphate transporter from the mycorrhizal fungus Glomus versiforme. Nature 378, 626–629. Hawkins, H., and George, E. (1999). Effect of nitrogen status on the contribution of arbuscular mycorrhizal hyphae to plant nitrogen uptake. Physiol. Planta 105, 694 –700. Hawkins, H., Johansen, A., and George, E. (2000). Uptake and transport of organic and inorganic nitrogen by arbuscular mycorrhizal fungi. Plant Soil 226, 275–285. Hayman, D. S., and Tavares, M. (1985). Plant growth responses to vesicular–arbuscular mycorrhiza. XV. Influence of soil pH on the symbiotic efficiency of different endophytes. New Phytol. 100, 367 –377. Hayman, D. S., Johnson, A. M., and Ruddlesdin, I. (1975). The influence of phosphate and crop species on Endogone spores and vesicular–arbuscular mycorrhiza under field conditions. Plant Soil 43, 489–495. He, S., Fang, D., Wang, S., Wu, G., Wang, D., and Bei, Z. (1997). Effects of VA mycorrhizal fungus on phosphorous and potassium uptake in tea seedlings. Acta Agric. Nucl. Sin. 11, 45–48. van der Heijden, M. G. A., Boller, T., Wiemken, A., and Sanders, I. R. (1998). Different arbuscular mycorrhizal fungal species are potential determinants of plant community structure. Ecology 79, 2082–2091. Helgason, T., Daniell, T. J., Husband, R., Fitter, A. H., and Young, B. D. (1998). Ploughing up the wood-wide web. Nature 394, 431. Hendrix, J. W., Guo, B. Z., and An, Z.-Q. (1995). Divergence of mycorrhizal communities in crop production systems. Plant Soil 170, 131–140. Heppel, K. B., Shumway, D. L., and Koide, R. T. (1998). The effect of mycorrhizal infection of Abutilon theophrasti on competitiveness of offspring. Funct. Ecol. 12, 171– 175. Herrick, J. E. (2000). Soil quality an indicator of sustainable land management. Appl. Soil Ecol. 15, 75 –83. Hetrick, B. A. D., Wilson, G. W. T., and Cox, T. S. (1992). Mycorrhizal dependence of modern wheat cultivars, land races and ancestors. Can. J. Bot. 70, 2032–2040. Hetrick, B. A. D., Wilson, G. W. T., and Cox, T. S. (1993). Mycorrhizal dependence of modern wheat cultivars and ancestors: a synthesis. Can. J. Bot. 71, 512 –518. Hetrick, B. A. D., Wilson, G. W. T., and Todd, T. C. (1996). Mycorrhizal response in wheat cultivars: relationship to phosphorus. Can. J. Bot. 74, 19 –25. Hirsch, A. M., and Kalpulnik, Y. (1998). Signal transduction pathways in mycorrhizal associations: comparisons with the Rhizobium –legume symbiosis. Fungal Genet. Biol. 23, 205–212. Hodge, A. (2000). Microbial ecology of the arbuscular mycorrhiza. FEMS Microbiol. Ecol. 32, 91–96. Hodge, A., Robinson, D., and Fitter, A. (2000). Are microorganisms more effective than plants at competing for nitrogen. Trends Plant Sci. 5, 304 –308. Ho¨gberg, P. (1990). 15N natural abundance as a possible marker of the ectomycorrhizal habit of trees in mixed African woodlands. New Phytol. 115, 483–486. Hooker, J. E., Jaizme-Vega, M., and Atkinson, D. (1994). Biocontrol of plant pathogens using arbuscular mycorrhizal fungi. In “Impact of Arbuscular Mycorrhizas on Sustainable Agriculture and Natural Ecosystems”. (S. Gianinazzi and H. Schu¨epp, Eds.), pp. 191 –200. Birkhauser, Basel, Switzerland. Hooker, J. E., Black, K. E., Perry, R. L., and Atkinson, D. (1995). Arbuscular mycorrhizal induced alteration to root longevity of poplar. Plant Soil 172, 327–329. IFOAM, (2000). In “IFOAM Basic Standards for Organic Production and Processing”. IFOAM. Ikram, A., Jensen, E. S., and Jakobsen, I. (1994). No significant transfer of N and P from Pueraria phaseoloides to Hevea brasiliensis via hyphal links of arbuscular mycorrhiza. Soil Biol. Biochem. 26, 1541–1547. Jalali, B. L., and Jalali, I. (1991). Mycorrhiza in plant disease control. In “Handbook of Applied Mycology”. (K. Arora, B. Rai, K. G. Mukerjee and G. R. Knudsen, Eds.), pp. 131–154. Dekker, New York.

218

L. A. HARRIER AND C. A. WATSON

Jarstfer, A. G., and Sylvia, D. M. (1992). Inoculum production and inoculation technologies of vesicular arbuscular mycorrhizal fungi. In “Soil Technologies: Applications in Agriculture, Forestry and Environmental Management”. (B. Metting, Ed.), pp. 349– 377. Marcel Dekker, New York. Jastrow, J. D. (1987). Changes in soil aggregation associated with tallgrass restoration. Am. J. Bot. 74, 1656–1664. Jawson, M. D., Franzluebbers, A. J., Galusha, D. K., and Aike, R. M. (1993). Soil fumigation within monoculture and rotations: Response of corn and mycorrhizae. Agron. J. 85, 1174–1180. Jensen, E. S. (1996). Grain yield, symbiotic N2 fixation and interspecific competition for organic-N in a pea–barley intercrop. Plant Soil 182, 25– 38. Jensen, A., and Jakobsen, I. (1980). The occurrence of vesicular– arbuscular mycorrhiza in barley and wheat grown in some Danish soils with different fertiliser treatments. Plant Soil 55, 403–414. Johansen, A., Jakobsen, I., and Jensen, E. S. (1992). Hyphal transport of 15N-labelled nitrogen by a vesicular–arbusuclar mycorrhizal fungus and its effect on depletion of inorganic soil N. New Phytol. 122, 281 –288. Johansen, A., Jakobsen, I., and Jensen, E. S. (1993). External hyphae of vesicular–arbusuclar mycorrhizal fungus associated with Trifolium subterraneum L. III Hyphal transport of 32P and 15N. New Phytol. 124, 61–68. Johansen, A., Jakobsen, I., and Jensen, E. S. (1994). Hyphal N transport by a vesicular–arbusuclar mycorrhizal fungus associated with cucumber grown at three nitrogen levels. Plant Soil 160, 1 –9. Johnson, N. C., and Pfleger, F. L. (1992). Vesicular–arbsucular mycorrhizae and cultural stresses. ASA Special Publication No. 54. In “Mycorrhizae in Sustainable Agriculture”. (G. J. Bethlenfalvy and R. G. Lindermann, Eds.), pp. 71 –100. ASA, Madison, WI. Johnson, N. C., Pfleger, F. L., Crookston, R. K., Simmmons, S. R., and Copeland, P. J. (1991). Vesicular–arbuscular mycorrhizas respond to corn and soybean cropping history. New Phytol. 117, 657 –663. Johnson, N. C., Copeland, P. J., Crookston, R. K., and Pfleger, F. L. (1992). Mycorrhizae: possible explanation for yield decline with continuous corn and soybean. Agron. J. 84, 387–390. Johnson, N. C., Graham, J. H., and Smith, F. A. (1997). Functioning of mycorrhizal associations along the mutualism–parasitism continuum. New Phytol. 135, 575–586. Joner, E. J. (2000). The effect of long-term fertilization with organic or inorganic fertilizers on mycorrhiza mediated phosphorus uptake in subterranean clover. Biol. Fert. Soils 32, 435– 440. Jordan, N. R., Zhang, J., and Huerd, S. (2000). Arbuscular-mycorrhizal fungi: potential roles in weed management. Weed Res. 40, 397 –410. Kabir, Z., and Koide, R. T. (2000). The effect of dandelion or a cover crop on mycorrhiza inoculum potential, soil aggregation and yield of maize. Agric. Ecosyst. Environ. 78, 167–174. Kabir, Z., O’Halloran, I. P., Fyles, J. W., and Hamel, C. (1997). Seasonal changes of arbuscular mycorrhizal fungi as affected by tillage practices and fertilization: hyphal density and mycorrhizal root colonization. Plant Soil 192, 285–293. Kabir, Z., O’Halloran, I. P., and Hamel, C. (1999). Combined effects of soil disturbance and fallowing on plant and fungal components of mycorrhizal corn. Soil Biol. Biochem. 31, 307–314. Kahiluoto, H., and Vestberg, M. (1998). The effect of arbuscular mycorrhiza on biomass production and phosphorus uptake from sparingly soluble sources by leek (Allium porrum L.) in Finnish field soils. Biol. Agric. Hortic. 16, 65– 85. Karasawa, T., Kasahara, Y., and Takebe, M. (2001). Variable response of growth and arbuscular mycorrhizal colonization of maize plants to preceding crops in various types of soils. Biol. Fert. Soils 33, 286 –293. Karlen, D. L., Varvel, G. E., Bullock, D. G., and Cruse, R. M. (1994). Crop rotations for the 21st century. Adv. Agron. 53, 1 –45. Khalil, S., Loynachan, T. E., and Tabatabai, M. A. (1994). Mycorrhizal dependency and nutrient uptake by improved and unimproved corn and soybean cultivars. Agron. J. 86, 949 –958.

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 219 Killham, K., and Firestone, M. K. (1983). Vesicular–arbuscular mycorrhizal mediation of grass response to acid and heavy metal deposition. Plant Soil 72, 39– 48. Kjoller, R., and Rosendahl, S. (1997). The presence of the arbuscular mycorrhizal fungus Glomus intraradices influences enzymatic activities of the root pathogen Aphanomyces euteiches in pea roots. Mycorrhiza 6, 487–491. Kling, M., and Jakobsen, I. (1998). Arbuscular mycorrhiza in soil quality assessment. Ambio 27, 29 –34. Koide, R. T. (1993). Physiology of the mycorrhizal plant. Adv. Plant Pathol. 9, 33–54. Koide, R. T., and Lu, X. (1992). Mycorrhizal infection of wild oats: maternal effects on offspring growth and reproduction. Oecologia 90, 218–226. Koide, R. T., and Schreiner, R. P. (1992). Regulation of the vesicular–arbuscular mycorrhizal symbiosis. Annu. Rev. Plant Physiol. Plant Mol. Biol. 43, 557– 581. Koide, R. T., Li, M., Lewis, J., and Irby, C. (1988). Role of mycorrhizal infection in the growth and reproduction of wild vs. cultivated plants. I. Wild vs. cultivated oats. Oecologia 77, 537 –543. Kothari, S. K., Marschner, H., and Ro¨mheld, V. (1990). Direct and indirect effects of VA mycorrhizal fungi and rhizosphere microorganisms on acquisition of mineral nutrients by maize (Zea mays L.) in calcareous soil. New Phytol. 116, 637 –645. Kothari, S. K., Marschner, H., and Ro¨mheld, V. (1991). Effect of a vesicular arbuscular mycorrhizal fungus and rhizosphere microorganisms on manganese reduction in the rhizosphere and manganese concentrations in maize (Zea mays L.). New Phytol. 117, 649 –655. Kurle, J. E., and Pfleger, F. L. (1994). Arbuscular mycorrhizal fungus spore populations respond to conversions between low-input and conventional management practices in a corn–soybean rotation. Agron. J. 86, 467–475. Kurle, J. E., and Pfleger, F. L. (1996). Management influences on arbuscular mycorrhizal fungal species composition in a corn–soybean rotation. Agron. J. 88, 155–161. Lambais, M. R., and Cardoso, E. J. B. N. (1993). Response of Stylosanthes guianensis to endomycorrhizal fungi inoculation as affected by lime and phosphorous applications. II. Nutrient uptake. Plant Soil 150, 109–116. Lambert, D. H., and Weidensaul, T. C. (1991). Element uptake by mycorrhizal soybean from sewagesludge-treated soil. Soil Sci. Soc. Am. J. 55, 393 –398. Lambert, D. H., Baker, D. E., and Cole, H. (1979). The role of mycorrhizae in the interactions of phosphorous, zinc, copper and other elements. Soil Sci. Soc. Am. J. 43, 976– 980. Lampkin, N. (1990). In “Organic Farming”. Farming Press, Ipswich. Lewis, J. D., and Koide, R. T. (1990). Phosphorous supply, mycorrhizal infection and plant offspring vigour. Funct. Ecol. 4, 695–702. Li, X.-L., George, E., Marschner, H., and Zhang, J.-L. (1997). Phosphorus acquisition from compacted soils by hyphae of a mycorrhizal fungus associated with red clover (Trifolium pratense ). Can. J. Bot. 75, 723–729. Limonard, T., and Ruissen, M. A. (1989). The significance of VA-mycorrhiza to future arable farming in the Netherlands. Neth. J. Plant Pathol. 95(Suppl. 1), 129–135. Lindermann, R. G. (1994). Role of VAM in biocontrol. In “Mycorrhizae and Plant Health”. (F. L. Pfleger and R. G. Linderman, Eds.), pp. 1–26. APS, St Paul. Lingua, G., Fusconi, A., and Berta, G. (2001). The nucleus of differentiated root plant cells: modifications induced by arbuscular mycorrhizal fungi. Eur J Histochem 45, 9–20. Lipson, D. A., Schadt, C. W., Schmidt, S. K., and Monson, R. K. (1999). Ectomycorrhizal transfer of amino acid nitrogen to the alpine sedge Kobresia myosuroides. New Phytol. 142, 163–167. Liu, A., Hamel, C., Hamilton, R. I., and Smith, D. L. (2000). Mycorrhizae formation and nutrient uptake of new corn (Zea mays L.) hybrids with extreme canopy and leaf architecture as influenced by soil N and P levels. Plant Soil 221, 157 –166. Lu, X., and Koide, R. T. (1994). The effects of mycorrhizal infection on components of plant growth and reproduction. New Phytol. 128, 211–218.

220

L. A. HARRIER AND C. A. WATSON

Lu, S., and Miller, H. M. (1989). The role of VA mycorrhizae in the absorption of P and Zn by maize in field and growth chamber experiments. Can. J. Soil Sci. 69, 97–109. Ma¨der, P., Edenhofer, S., Boller, T., Wiemken, A., and Niggli, U. (2000). Arbuscular mycorrhizae in a long-term field trial comparing low-input (organic, biological) and high-input (conventional) farming systems in a crop rotation. Biol. Fert. Soils 31, 150–156. Maier, W., Peipp, H., Schmidt, J., Wray, V., and Strack, D. (1995). Levels of a terpenoid glycoside (bluemin) and cell wall bound phenolics in some cereal mycorrhizas. Plant Physiol. 109, 465–470. Mamo, T., and Killham, K. S. (1987). Effect of soil liming and vesicular–arbuscular-mycorrhizal inoculation on the growth and micronutrient content of the teff plant. Plant Soil 102, 257– 259. Manske, G. G. B. (1989). Genetical analysis of the efficiency of VA mycorrhiza with spring wheat. Agric. Ecosyst. Environ. 29, 273 –280. Mark, G. L., and Cassells, A. C. (1996). Genotype dependence in the interaction between Glomus fistulosum, Phytophthora fragariae and the wild strawberry (Fragaria vesca ). Plant Soil 185, 233–239. Marschner, H., and Dell, B. (1994). Nutrient uptake in mycorrhizal symbiosis. Plant Soil 159, 89–102. McGonigle, T. P. (1988). Numerical analysis of published field trials with vesicular–arbuscular mycorrhizal fungi. Funct. Ecol. 2, 473–478. McGonigle, T. P., and Miller, M. H. (2000). The inconsistent effect of soil disturbance on colonization of roots by arbuscular mycorrhizal fungi: a test of the inoculum density hypothesis. Appl. Soil Ecol. 14, 147–153. Medeiros, C. A. B., Clark, R. B., and Ellis, J. R. (1994). Effects of excess aluminium on mineral uptake in mycorrhizal sorghum. J. Plant Nutr. 17, 1399– 1416. Medeiros, C. A. B., Clark, R. B., and Ellis, J. R. (1994). Effects of excess manganese on mineral uptake in mycorrhizal sorghum. J. Plant Nutr. 17, 2203–2219. Medeiros, C. A. B., Clark, R. B., and Ellis, J. R. (1994). Growth and nutrient uptake of sorghum cultivated with vesicular–arbuscular mycorrhiza isolates at varying pH. Mycorrhiza 4, 185– 191. Miller, R. M., and Jastrow, J. D. (1990). Hierarchy of root and mycorrhizal fungal interactions with soil aggregation. Soil Biol. Biochem. 22, 579–584. Miller, R. M., and Jastrow, J. D. (1992). The role of mycorrhizal fungi in soil conservation. ASA Special Publication No. 54. In “Mycorrhizae in Sustainable Agriculture”. (G. J. Bethlenfalvy and R. G. Lindermann, Eds.), pp. 29– 44. ASA, Madison, WI. Miller, M., McGonigle, T., and Addy, H. (1994). An economic approach to evaluate the role of mycorrhizas in managed ecosystems. Plant Soil 159, 27–35. Miller-Wideman, M. A., and Watrud, L. S. (1984). Sporulation of Gigaspora margarita on root cultures of tomato. Can. J. Microbiol. 30, 642–646. Minerdi, D., Fani, R., Gallo, R., Boarino, A., and Bonfante, P. (2001). Nitrogen fixation genes in an endosymbiotic Burkholderia strain. Appl. Environ. Microbiol. 67, 725–732. Morandi, D. (1996). Occurrence of phytoalexins and phenolic compounds in endomycorrhizal interactions and their potential role in biological control. Plant Soil 185, 241 –245. Mosse, B. (1973). Advances in the study of vesicular– arbuscular mycorrhizae. Annu. Rev. Phytopathol. 11, 171 –196. Mosse, B. (1986). Mycorrhiza in a sustainable agriculture. Biol. Agric. Hortic. 3, 191–209. Mosse, B., and Hepper, C. (1975). Vesicular–arbuscular mycorrhizal infections in root organ cultures. Physiol. Plant Pathol. 5, 215 –223. Mozafar, A., Anken, T., Ruh, R., and Frossard, E. (2000). Tillage intensity, mycorrhizal and nonmycorrhizal fungi and nutrient concentrations in maize, wheat and canola. Agron. J. 92, 1117–1124. Mugnier, J., and Mosse, B. (1987). Vesicular–arbuscular mycorrhizal infections in transformed root inducing T-DNA roots grown axenically. Phytopathology 77, 1045–1050.

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 221 Mulligan, M. F., Smucker, A. J. M., and Safir, G. F. (1985). Tillage modifications of dry edible bean root colonization by VAM fungi. Agron. J. 77, 140 –144. Murphy, J. G., Rafferty, S. M., and Cassells, A. C. (2000). Stimulation of wild strawberry (Fragaria vesca ) arbuscular mycorrhizas by addition of shellfish waste to the growth substrate: interaction between mycorrhization, substrate amendment and susceptibility to red core (Phytophthora fragariae ). Appl. Soil Ecol. 15, 153–158. Na¨sholm, T., Ekblad, A., Nordin, A., Giesler, R., Hogberg, M., and Hogberg, P. (1998). Boreal forest plants take up organic nitrogen. Nature 392, 914–916. Nelson, C. E. (1987). The water relations of vesicular – arbuscular mycorrhizal systems. In “Ecophysiology of VA Mycorrhizal Plants”. (G. R. Safir, Ed.), pp. 71 –91. CRC Press, Boca Raton, FL. Norman, J. R., Atkinson, D., and Hooker, J. E. (1996). Arbuscular mycorrhizal fungal-induced alteration to root architecture in strawberry and induced resistance to the root pathogen Phytophthora fragariae. Plant Soil 185, 191 –198. Ocampo, J. A., Martin, J., and Hayman, D. S. (1980). Influence on plant interactions on vesicular– arbuscular mycorrhizal infections. I. Host and non-host plants grown together. New Phytol. 84, 27 –35. Paulitz, T. C., and Linderman, R. G. (1991). Lack of antagonism between the biocontrol agent Gliocladium virens and vesicular arbuscular mycorrhizal fungi. New Phytol. 117, 303 –308. Pawlowska, T. E., Douds, D. D., and Charvat, I. (1999). In vitro propagation and life cycle of the arbuscular mycorrhizal fungus Glomus etunicatum. Mycol. Res. 103, 1549–1556. Pendleton, R. L. (2000). Pre-inoculation by an arbuscular mycorrhizal fungus enhances male reproductive output of Cucurbita foetidissima. Int. J. Plant Sci. 161, 683–689. Penfold, C. M., Miyan, M. S., Reeves, T. G., and Grierson, I. T. (1995). Biological farming for sustainable agricultural production. Aust. J. Expp. Agric. 35, 849–856. Pe´rez-Moreno, J., and Ferrera-Cerrato, R. (1994). Mycorrhizal interactions with plants and soil organisms in sustainable agroecosystems. In “Soil Ecology in Sustainable Agricultural Systems”. (L. Brussard and R. Ferrera-Cerrato, Eds.), pp. 91–112. CRC Press, Boca Raton, FL. Pirozynski, K. A., and Dalpe, Y. (1989). Geological history of the Glomaceae with particular reference to mycorrhizal symbiosis. Symbiosis 7, 1– 36. Pirozynski, K. A., and Malloch, D. W. (1975). The origin of land plants: a matter of mycotrophism. Biosystems 6, 153 –164. Porter, W. M., Robson, A. D., and Abbott, L. K. (1987). Factors controlling the distribution of vesicular–arbuscular mycorrhizal fungi in relation to soil pH. J. Appl. Ecol. 24, 663–672. Pozo, M. J., Dumas-Gaudot, E., Slezack, S., Cordier, C., Asselin, A., Gianinazzi, S., GianinazziPearson, V., Azco´n-Aguilar, C., and Barea, J. M. (1996). Induction of new chitinase isoforms in tomato roots during interactions with Glomus mosseae and/or Phytophthora nicotianae var parasitica. Agronomie 16, 689– 697. Pozo, M. J., Azco´n-Aguilar, C., Dumas-Gaudot, E., and Barea, J.M. (1998). Chitosanase and chitinase activities in tomato roots during interactions with arbuscular mycorrhizal fungi or Phytophthora parasitica. J. Exp Bot. 49, 1729–1739. Pozo, M. J., Azco´n-Aguilar, C., Dumas-Gaudot, E., and Barea, J. M. (1999). Beta-1-3 glucanase activities in tomato roots inoculated with the arbuscular mycorrhizal fungi and/or Phytophthora parasitica and their possible involvement in bioprotection. Plant Sci. 141, 149–157. Pozo, M. J., Cordier, C., Dumas-Gaudot, E., Gianinazzi, S., Barea, J.M. and Azcon-Aguilar, C. (2002). Localized versus systemic effect or arbuscular mycorrhizal fungi on defence responses to Phytophthora infection in tomato plants J. Exp Bot. 53, 525–534. Pretty, J. (1995). In “Regenerating Agriculture—Policies and Practice for Sustainability and Selfreliance”. Earthscan Publications, London. Rajan, S. S. S., Watkinson, J. H., and Sinclair, A. G. (1996). Phosphate rocks for direct application to soils. Adv. Agron. 57, 77–159.

222

L. A. HARRIER AND C. A. WATSON

Raju, P. S., Clark, R. B., Ellis, J. R., and Maranville, J. W. (1987). Vesicular arbuscular mycorrhizal infection effects on sorghum growth, phosphorous efficiency and mineral nutrient uptake. J. Plant Nutr. 10, 1331–1339. Raju, P. S., Clark, R. B., Ellis, J. R., and Maranville, J. W. (1990). Effects of species of VAmycorrhizal fungi on growth and mineral uptake of sorghum at different temperatures. Plant Soil 121, 165 –170. Raju, P. S., Clark, R. B., Ellis, J. R., and Maranville, J. W. (1990). Mineral uptake and growth of sorghum colonised with VA mycorrhiza at varied soil phosphorous levels. J. Plant Nutr. 13, 843–859. Read, D. J. (1996). The structure and function of the ericoid mycorrhizal root. Ann. Bot. 77, 365– 374. Read, D. J., and Boyd, R. (1986). Water relations of mycorrhizal fungi and their host plants. In “Water, Fungi and Plants”. (P. Ayres and L. Boddy, Eds.), pp. 287 –303. Cambridge University Press, Cambridge, UK. Redecker, D., Kodner, R., and Graham, L. E. (2000). Glomalean fungi from the Ordovacian. Science 289, 1920–1921. Reid, C. P. P (1979). Mycorrhizae and waters stress. In “Root Physiology and Symbiosis” (A. Reidacher and G. Gagnaire-Michard, Eds.), pp. 392–408. IUFRO Proc, Nancy France. Remy, W., Taylor, T. N., Hass, H., and Kerp, H. (1994). Four hundred million year old vesicular arbuscular mycorrhizae. Proc. Natl Acad. Sci. USA 91, 11841– 11843. Rhodes, L. H., and Gerdemann, J. W. (1978). Translocation of calcium and phosphate by external hyphae of vesicular arbuscular mycorrhizae. Soil Sci. 126, 125 –126. Rogers, R. D., and Williams, S. E. (1986). Vesicular–arbuscular mycorrhizae: influence on plant uptake of cesium and cobalt. Soil Biol. Biochem. 18, 371 –376. Rosendahl, S. (1985). Interactions between the vesicular–arbuscular mycorrhizal fungus Glomus fasiculatum and Aphanomyces euteiches root rot of pea. Phytopathol. Z. 114, 31–40. Ryan, M. H., Chilvers, G. A., and Dumaresq, D. C. (1994). Colonisation of wheat by VA-mycorrhizal fungi was found to be higher on a farm managed in an organic manner than on a conventional neighbour. Plant Soil 160, 33–40. Saif, S. R. (1987). Growth responses of tropical forage plant species to vesicular arbuscular mycorrhizae. Plant Soil 97, 25–35. Sainz, M. J., Taboada-Castro, M. T., and Vilarino, A. (1998). Growth, mineral nutrition and mycorrhizal colonization of red clover and cucumber plants grown in a soil amended with composted urban wastes. Plant Soil 205, 85–92. Sa´nchez-Dı´az, M., and Honrubia, M. (1994). Water relations and alleviation of drought stress in mycorrhizal plants. In “Impact of Arbuscular Mycorrhizas on Sustainable Agriculture and Natural Ecosystems”. (S. Gianinazzi and H. Schu¨epp, Eds.), pp. 167 –178. Birkhauser, Basel. Sattelmacher, B., Reinhard, S., and Pomikalko, A. (1991). Differences in mycorrhizal colonization of rye (Secale cereale L.) grown in conventional or organic (biological-dynamic) farming systems. J. Agron. Crop Sci. 167, 350 –355. Schactman, D., Reid, R. J., and Ayling, S. M. (1998). Phosphorous uptake by plants: from soil to cell. Plant Physiol. 116, 447–453. Schellenbaum, L., Berta, G., Ravolanirina, F., Tisserant, B., Gianinazzi, S., and Fitter, A. H. (1991). Influence of endomycorrhizal infection on root morphology in a micropropagated woody plant species (Vitis vinifera L.). Ann. Bot. 68, 135–141. Schenck, N. C. (1987). Vesicular arbuscular mycorrhizal fungi and the control of fungal root disease. In “Innovative Approaches to Plant Disease Control”. (I. Chet, Ed.), pp. 271–280. Wiley, New York. Schenck, N. C., and Kinloch, R. A. (1980). Incidence of mycorrhizal fungi on six field crops in monoculture on a newly cleared woodland site. Mycologia 72, 445– 456. Schenck, N. C., and Smith, G. S. (1982). Responses of six species of vesicular arbuscular mycorrhizal fungi and their effects on soybean at four soil temperatures. New Phytol. 92, 193– 201.

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 223 Schonbeck, F. (1979). Endomycorrhiza in relation to plant diseases. In “Soil-borne Pathogens”. (B. Schippers and W. Gams, Eds.), pp. 271–280. Academic Press, New York. Schreiner, R. P., Mihara, K. L., McDaniel, H., and Bethlenfalvay, G. J. (1997). Mycorrhizal fungi influence plant and soil functions and interactions. Plant Soil 188, 199–209. Schussler, A., Schwarzott, D., and Walker, C. (2001). A new fungal phylum, the Glomeromycota: phylogeny and evolution. Mycol. Res. 105, 1413–1421. Sharma, M. P., and Adholeya, A. (2000). Enhanced growth and productivity following inoculation with indigenous AM fungi in four varieties of onion (Allium cepa L.) in an alfisol. Biol. Agric. Hortic. 18, 1–14. Sharma, A. K., Johri, B. N., and Gianinazzi, S. (1992). Vesicular–arbuscular mycorrhizae in relation to plant disease. World J. Microbiol. Biotechnol. 8, 550– 563. Shaul, O., Galili, S., Volpin, H., Ginzberg, H., Elad, Y., Chet, I., and Kalpulnik, Y. (1999). Mycorrhiza induced changes in disease severity and PR protein expression in tobacco leaves. Mol. Plant– Microbe Interact. 12, 1000–1007. Shierlaw, J., and Alston, A. M. (1984). Effect of soil compaction on root growth and uptake of phosphorus. Plant Soil 77, 15–28. Shumway, D. L., and Koide, R. T. (1994). Reproductive responses to mycorrhizal colonisation of Abutilon theophrasti Medic. Plants grown for two generations in the field. New Phytol. 128, 219 –224. Shumway, D. L., and Koide, R. T. (1994). Within season variability in mycorrhizal benefit to reproduction in Abutilon theophrasti Medic. Plant Cell Environ. 17, 821–827. Shumway, D. L., and Koide, R. T. (1995). Size and reproductive inequality in mycorrhizal and nonmycorrhizal populations of Abutilon theophrasti. J. Ecol. 83, 613–620. Sieverding, E., and Toro, S. (1989). Influence of soil water regimes on VA mycorrhiza. V. Performance of different VAM fungal species with cassava. J. Agron. Crop Sci. 161, 322–332. Slezack, S., Dumas-Gaudot, E., Rosendahl, S., Kjoller, R., Paynot, M., Negrel, J., and Gianinazzi, S. (1999). Endoproteolytic activities in pea roots inoculated with the arbuscular mycorrhizal fungus Glomus mosseae and/or Aphanomyces euteiches in relation to bioprotection. New Phytol. 142, 517 –529. Slezack, S., Dumas-Gaudot, E., Paynot, M., and Gianinazzi, S. (2000). Is a fully established arbusuclar mycorrhizal symbiosis required for bioprotection of Pisum sativum roots against Aphanomyces euteiches? Mol. Plant Microbe Interact. 13, 238–241. Smith, S. (1987). Interactions of nematodes with mycorrhizal fungi. In “Vistas on Nematology”. (J. A. Veetch and D. W. Dickon, Eds.), pp. 292–300. Society of Nematology, Hyattsville, MD. Smith, S. E., and Gianinazzi-Pearson, V. (1988). Physiological interactions between symbionts in vesicular arbuscular mycorrhizal plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 39, 221–244. Smith, S. E., and Read, D. J. (1997). In “Mycorrhizal Symbiosis” 2nd edition. Academic Press, California. Smith, S. E., Gianinazzi-Pearson, V., Koide, R., and Cairney, J. W. G. (1994). Nutrient transport in mycorrhizas: structure, physiology and consequences for efficiency of the symbiosis. Plant Soil 159, 103–113. Stanley, M. R., Koide, R. T., and Shumway, D. L. (1993). Mycorrhizal symbiosis increases growth, reproduction and recruitment of Abutilon theophrasti Medic. in the field. Oecologia 94, 30–35. Stephenson, A. G., Poulton, J. L., Lau, T. C., and Koide, R. T. (1998). Effects of soil phosphorus level and mycorrhizal infection on the male function of plants. In “Phosphorous in Plant Biology: Regulatory Roles in Molecular, Cellular, Organismic and Ecosystem Processes”. (J. P. Lynch and J. Deikman, Eds.), pp. 52–67. American Society of Plant Physiologists, Rockville, Madison. Stockdale, E. A., Lampkin, N. H., Hovi, M., Keatinge, R., Lennartsson, E. K. M., MacDonald, D. W., Padel, S., Tattersall, F. H., Wolfe, M. S., and Watson, C. A. (2001). Agronomic and environmental implications of organic farming systems. Adv. Agron. 70, 261 –327.

224

L. A. HARRIER AND C. A. WATSON

Sylvia, D. M. (1992). Quantification of external hyphae of vesicular–arbuscular mycorrhizal fungi. In “Methods in Microbiology 24: Techniques for the Study of Mycorrhiza”. (J. R. Norris, D. J. Read and A. K. Varma, Eds.), pp. 53–65. Academic Press, London. Sylvia, D. M., and Schenck, N. C. (1983). Application of superphosphate to mycorrhizal plants stimulates sporulation of phosphorus tolerant vesicular–arbuscular mycorrhizal fungi. New Phytol. 95, 655 –661. Tahiri-Alaoui, A., Lingua, G., Avrova, A., Sampo, S., Fusconi, A., Antoniw, J.F., and Berta, G. (2002). A cullin gene is induced in tomato roots forming arbuscular mycorrhizae. Can J. Bot. 80, 607–616. Tarafdar, J. C., and Marschner, H. (1994). Phosphatase activity in the rhizosphere and hyphosphere of VA-mycorrhizal wheat supplied with inorganic and organic phosphorous. Soil Biol. Biochem. 26, 387–395. Tarkalson, D. D., Jolley, V. D., Robbins, C. W., and Terry, R. E. (1998). Mycorrhizal colonization and nutrient uptake of dry bean in manure and composted manure treated subsoil and untreated topsoil and subsoil. J. Plant Nutr. 21, 1867–1878. Tarkalson, D. D., Jolley, V. D., Robbins, C. W., and Terry, R. E. (1998). Mycorrhizal colonization and nutrition of wheat and sweetcorn in manure treated topsoil and subsoil. J. Plant Nutr. 21, 1985–1999. Taylor, T. N., Remy, W., Hass, H., and Kerp, H. (1995). Fossil arbuscular mycorrhizae from the early Devonian. Mycologia 87, 560 –573. Tester, M., Smith, S. E., and Smith, F. A. (1987). The phenomenon of non-mycorrhizal plants. Can. J. Bot. 65, 419– 431. Theunissen, J. (1997). Application of intercropping in organic agriculture. Biol. Agric. Hortic. 15, 251–259. Thingstrup, I., Rubaek, G., Sibbesen, E., and Jakobsen, I. (1998). Flax (Linum usitatissimum L.) depends on arbuscular mycorrhizal fungi for plant growth and P uptake at intermediate but not high soil P levels in the field. Plant Soil 203, 37 –46. Thingstrup, I., Kahiluoto, H., and Jakobsen, I. (2000). Phosphate transport by hyphae of field communities of arbuscular mycorrhizal fungi at two levels of P fertilization. Plant Soil 221, 181–187. Thompson, J. P. (1994). What is the potential for management of mycorrhizas in agriculture? In “Management of Mycorrhizas in Agriculture, Horticulture and Forestry”. (A. D. Robson, L. K. Abbott and N. Malajczuk, Eds.), pp. 191–200. Kluwer Academic Publishers, Netherlands. Thomson, B. D., Robson, A. D., and Abbott, L. K. (1991). Soil mediated effects of phosphorus on the formation of mycorrhizas by Scutellospora calospora (Nicol & Gerd.) Walker & Sanders on subterranean clover. New Phytol. 118, 463– 469. Tisdall, J. M. (1991). Fungal hyphae and structural stability of soil. Aust. J. Soil Res. 29, 729–743. Tisdall, J. M., and Oades, J. M. (1982). Organic matter and water stable aggregates in soils. J. Soil Sci. 33, 141 –163. Tisserant, B., Schellenbaum, L., Gianinazzi-Pearson, V., Gianinazzi, S., and Berta, G. (1991). Influence of infection by an endo-mycorrhizal fungus on root development and architecture in Platanus acerifolia. Allonia 30, 171 –181. Toth, R., Toth, D., Starke, D., and Smith, D. R. (1990). Vesicular arbuscular mycorrhizal colonization in Zea mays affected by breeding for resistance to fungal pathogens. Can. J. Bot. 68, 1039– 1044. Trappe, J. M., Molina, R., and Castellano, M. (1984). Reactions of mycorrhizal fungi and mycorrhiza formation to pesticides. Annu. Rev. Phytopathol. 22, 331–359. Trotta, A., Varese, G. C., Gnavi, E., Fusconi, A., Sampo, S., and Berta, G. (1996). Interactions between the soilborne root pathogen Phytophtora nicotianae var. parasitica and the arbuscular mycorrhizal fungus Glomus mosseae in tomato plants. Plant Soil 185, 199– 209. Vitousek, P. M., and Howarth, R. W. (1991). Nitrogen limitation on land and in the sea: how can it occur? Biogeochemistry 13, 87–115.

THE ROLE OF FUNGI IN SUSTAINABLE CROPPING SYSTEMS 225 Vivekanden, M., and Fixen, P. E. (1991). Cropping system effects on mycorrhizal colonization, early growth, and phosphorus uptake of corn. Soil Sci. Soc. Am. J. 55, 136 –140. Vyn, T. J., and Raimbault, B. A. (1993). Long-term effects of five tillage systems on corn response and soil structure. Agron. J. 85, 1074–1079. Wang, G. M., Stribley, D. P., Tinker, P. B., and Walker, C. (1985). Soil pH and vesicular–arbuscular mycorrhizas. In “Ecological Interactions in Soil”. (A. H. Fitter, Ed.), pp. 219–224. Blackwell Scientific, Oxford. Wang, G. M., Coleman, D. C., Freckman, D. W., Dyer, M. I., McNaughton, S. J., Acra, M. A., and Goeschl, J. D. (1989). Carbon partitioning patterns of mycorrhizal versus non-mycorrhizal plants: real-time dynamic measurements using 11CO2. New Phytol. 112, 489–493. Werner, M. R., Kluson, R. A., and Gliesssman, S. R. (1990). Colonization of strawberry roots by VA mycorrhizal fungi in agroecosystems under conventional and transitional organic management. Biol. Agric. Hortic. 7, 139 –151. Willey, R. W. (1979). Intercropping—its importance and research needs. Part 1. Competition and yield advantages. Field Crop Abstr. 32, 1–10. Wright, S. F. (2000). A fluorescent antibody assay for hyphae and glomalin from arbuscular mycorrhizal fungi. Plant Soil 226, 171–177. Wright, S. F., and Anderson, R. L. (2000). Aggregate stability and glomalin in alternative crop rotations for the central West Plains. Biol. Fert. Soils 31, 249 –253. Wright, S. F., and Upadhyaya, A. (1996). Extraction of an abundant and unusual protein from soil and comparison with hyphal protein of arbuscular mycorrhizal fungi. Soil Sci. 161, 575–586. Wright, S. F., and Upadhyaya, A. (1998). A survey of soils for aggregate stability and glomalin, a glycoprotein produced by hyphae of arbuscular mycorrhizal fungi. Plant Soil 198, 97–107. Wright, S. F., and Upadhyaya, A. (1999). Quantification of arbuscular mycorrhizal fungi activity by the glomalin concentration on hyphal traps. Mycorrhiza 8, 283–285. Wright, S. F., Franke-Snyder, M., Morton, J. B., and Upadhyaya, A. (1996). Time course study and partial characterisation of a protein on the hyphae of arbuscular mycorrhizal fungi during active colonisation of plant roots. Plant Soil 181, 193–203. Wright, S. F., Upadhyaya, A., and Buyer, J. S. (1998). Comparison of n-linked oligosaccharides of glomalin from arbuscular mycorrhizal fungi and soils by capillary electrophoresis. Soil Biol. Biochem. 30, 1853–1857. Yano, K., Yamauchi, A., Iijima, M., and Kono, Y. (1998). Arbuscular –mycorrhizal formation in undisturbed soil counteracts compacted soil stress in pigeon pea. Appl. Soil Ecol. 10, 95 –102. Yarham, D. J. (1979). The effect on soil-borne diseases of changes in crop and soil management. In “Soil-borne Plant Pathogens”. (B. Schippers and W. Gams, Eds.), pp. 371 –383. Academic Press, London. Zambolim, L., and Schenck, N. C. (1983). Reduction of the effects of pathogenic root infecting fungi on soybean by the mycorrhizal fungus, Glomus mosseae. Phytopathology 73, 1402–1405. Zhu, Y. G., Smith, S. E., Barritt, A. R., and Smith, F. A. (2001). Phosphorus (P) efficiencies and mycorrhizal responsiveness of old and modern wheat cultivars. Plant Soil 237, 249–255.

CATCH CROPS AND GREEN MANURES AS BIOLOGICAL TOOLS IN NITROGEN MANAGEMENT IN TEMPERATE ZONES Kristian Thorup-Kristensen,1 Jacob Magid2 and Lars Stoumann Jensen2 1

Department of Horticulture, Danish Institute of Agricultural Science, P.O. Box 102, DK-5792 Aarslev, Denmark 2 Plant Nutrition and Soil Fertility Laboratory, Department of Agricultural Science, Royal Veterinary and Agricultural University, Thorvaldsensvej 40, DK-1871 Frederiksberg, Denmark

I. Why Use Catch Crops—How Do We Want Them to Affect the System? A. N Effects of Catch Crops, Solving Environmental Problems B. N Effects of Catch Crops in Agriculture II. Uptake of N and Soil Depletion A. Growth and Uptake Potential B. Root Growth C. Relationship Between Soil Depletion and Nitrate Leaching Loss D. Leguminous Green Manures III. Catch Crop Effect on N Supply for Subsequent Crops A. Calculating the N Effect, NEff B. What Determines Pre-emptive Competition C. What Determines N Mineralisation D. Depth Distribution of Inorganic N E. Gaseous Losses of N F. Field Effects IV. Other Effects of Catch Crops A. Effects on Other Nutrients Than N B. Effects on Soil Microbiological and Faunal Activity C. Effects on Soil Physical Properties D. Effects on Soil Water Content E. Effects on Pests, Pathogens and Weeds V. Making the Most of Catch Crops in Cropping Sequences and Whole Crop Rotations A. Placing Catch Crops in the Crop Rotation B. Establishing Catch Crops C. Incorporation Time D. Choosing Catch Crop Species 227 Advances in Agronomy, Volume 79 Copyright q 2003 by Academic Press. All rights of reproduction in any form reserved 0065-2113/02$35.00

228

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN E. Model Simulation of Catch Crops in the Crop Rotation F. Placing Catch Crops at the “policy Scale” VI. Perspectives References During the last decades a lot of research have been made on the use of cover crops. Cover crops are grown for many purposes, but most of the resent interest have focused on their effects on nitrogen. Studies have been made on catch crops grown to catch N from the soil and prevent leaching losses to the environment and on legume green manure crops grown to improve the N supply for succeeding crops. Many of the experiments have been agronomic studies, where choise of plant species or management strategies have been tested to identify the optimal way to grow cover crops in a specific situation. Other experiments have aimed at gaining more basic understanding of the effects of catch crops or green manure crops on N dynamics. These studies include subjects as catch crop growth, root growth, N uptake and soil depletion, kill-date, N mineralisation and pre-emptive competition, and how these factors interact with soil, climatic conditions, and the main crops in the cropping system, both in the short term and in the longer term. Together, the results from these studies have given a more comprehensive understanding of the mechanisms by which a catch crop or a green manure affect N leaching losses and N supply for succeeding crops. The principles governing the effect of catch crops on N supply for succeeding crops have been found to differ basically from the effects N effects of added organic matter. This is mainly due to the fact that a catch crop do not add N to the soil, the N which is incorporated with the catch crop has first been taken from the soil. In the review, we discuss this new knowledge of catch crops and green manures, and how it helps us to understand why the effects obtained by catch crops are so variable. We also discuss how it can be used to develop strategies which will improve the results we obtain from catch crops and green manures, and to make them more predictable. Many studies have been made on other effects of cover crops, on soil borne diseases, pests, weeds, soil structure, erosion, soil biology, and other nutrients than N. Though there are many studies, they are scattered over a large number of themes, and research in cover crop effects in most of these themes can be said to be at a very early stage. However, many very interesting effects have been observed, an there seems to be a significant potential for development of cover cropping also for other objectives than improved N husbandry. q 2003 Academic Press.

I. WHY USE CATCH CROPS—HOW DO WE WANT THEM TO AFFECT THE SYSTEM? This review deals with the use of catch crops in temperate climatic zones, i.e., the areas where the summer period is used for crop production, whereas

CATCH CROPS AND GREEN MANURES

229

the winters are too cold for most crops to grow. In these conditions, the soil is often left with no plant cover during the winter period. The winter period is normally also the part of the year where the highest percolation of water through the soil occurs, due to low evaporation and in many areas high autumn and winter precipitation. The plant nutrients available in the soil can therefore be leached downwards with percolating water and eventually be lost from the root zone. During the autumn period, after harvest of the main crop temperature and light conditions allow some plant growth, though not enough to produce commercial crops. Many attempts have been made to use this period to grow plants, which prevent nutrient leaching or improve the soil in various other ways. Such crops are often termed catch crops, cover crops or green manures. The term cover crop is used broadly to describe non-commercial crops grown for a number of reasons, as just mentioned. In this review we use the terms catch crops and green manure crops as more specific terms. We use the term catch crops when dealing with cover crops which are grown to catch available N in the soil and thereby prevent N leaching losses, and the term green manure when dealing with cover crops which are grown mainly to improve the nutrition of the subsequent main crops. The research interest in growing catch crops and green manures is old, but the use of such crops have decreased during the 20th century (see Renius and Entrup, 2002). The use and relevance of catch crops in farming systems developing from very extensive grassing systems or slash and burn systems to modern intensive farming has recently been analysed by Pound et al. (1999). They concluded that catch crops or green manures became highly important to maintain productivity as farming developed from the very extensive early forms, to systems, which were more intensive but still used little external input. During the last century, green manures and catch crops lost importance as the use of pesticides and chemical fertilisers became widespread, but are now regaining importance to cope with environmental constraints on agricultural production. Hendrix et al. (1992) concluded that one of the reasons for environmental problems in conventional agricultural systems is that they have higher throughflow, lower storage capacity and less recycling of nutrients. In an attempt to design more environmental-friendly cropping systems, catch crops may add some of the storage capacity and recycling needed without requiring dramatic changes of the cropping systems. An overview of desired and undesired effects of catch crops and green manures in high and low input agro-ecosystems is given in Table I. These will be discussed in more detail in later sections. In accordance with this analysis, the interest in catch crops is now increasing again, after the long period of decline. During the last decades much research has been made on the use of catch crops. The renewed interest is mainly due to environmental concerns such as problems with nitrate leaching losses and soil degradation due to erosion and loss of organic matter.

230

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Table I Catch Crops and Green Manures in High Input and Low Input Systems; Desired and Undesired Effects Desired effects

Undesired effects

Catch crops in high input systems

Reduced N-leaching amount at high leaching intensities Reduced NO3 concentrations at low leaching intensities Improved soil fertility Erosion control Improved tilth Pest control

Increased N fertiliser requirement due to: Pre-emptive competition by cover crop Immobilisation of N during cover crop decomposition

Green manures in low input systems

Increased stability of N-supply Consistent cash crop yields Soil organic matter building Increased base N mineralisation Improved soil fertility Erosion control Improved tilth Improved crop rooting depth Pest and weed control

Increased weed and pest pressure Increased N-loss due to poor synchrony Loss of cash crop when undersowing green manures

In this review we focus on the use of catch crops to prevent nitrogenleaching losses during the winter period, and on green manures including nitrogen fixing legume species grown during the winter period to add nitrogen to the soil. We will discuss the mechanisms by which the catch crops affect nitrogen losses and nitrogen supply for succeeding crops, and based on this we will discuss how they can be used in strategies to reach agronomic and environmental goals.

A. N EFFECTS OF CATCH CROPS, SOLVING ENVIRONMENTAL PROBLEMS To reduce nitrogen leaching losses to the environment is a very general description of the environmental effects wanted from catch crops. Nitrate lost from agriculture causes two sorts of environmental problems: (1) It increases the nitrate concentration in the deeper aquifers used for drinking water, and (2) it increases the nitrate concentration in surface waters such as streams, lakes and

CATCH CROPS AND GREEN MANURES

231

coastal waters. This can lead to eutrofication with various negative effects on the ecosystems. If the goal is to keep the nitrate concentration in a ground water reservoir below the limit of 50 mg nitrate kg21, a rather high nitrate loss can be tolerated, as long as it is diluted in enough water to keep the concentration below 50 mg kg21. On the other hand, this means that if the discharge of water is low, even quite low amounts of nitrate lost by leaching may be enough to bring the concentration in the leaching water above acceptable levels. This is the situation in many drier areas, where the nitrogen leaching loss is relatively low, but may still lead to unacceptably high nitrate concentration in ground water and streams. When protecting lakes or coastal waters, the total nitrate load rather than the nitrate concentration in the water coming from rural areas is the most important factor (e.g., Rask et al., 2002). Thus, when eutrophication of lakes or coastal waters is the main problem, the high amounts of nitrate lost from areas with high precipitation will be a greater problem than the high nitrate concentrations in the leaching water from dry areas. Therefore, the specific effect we want from a catch crop, or any other measures we apply to reduce nitrate-leaching losses depend on the local situation and problem.

B. N EFFECTS OF CATCH CROPS IN AGRICULTURE Catch crops are expected to reduce N leaching losses from the soil, but also to improve the N nutrition of subsequent crops, an effect, which can make them attractive to farmers. However, the price of fertiliser N is low relative to the cost of growing catch crops (e.g., Bollero and Bullock, 1994; Stute and Posner, 1995a), and the introduction of cheap fertilisers is probably a main reason why catch crops have been grown so little for many years. In the future, the ability of catch crops to supply N to main crops may again become attractive to farmers, due to an increase in organic farming, and increasing regulations also on N use also in conventional farming. In Denmark, regulations already limit the amount of N fertilisers farmers may use to 90% of the optimum supply, and strict regulations are imposed locally in water protection areas in many countries. In organic farming systems, where inorganic fertilisers are not used, the crops are often N limited. Therefore, the economic returns for improving N-supply through the use of catch crops and green manures make them attractive (Lu et al., 1999). Gaining 1 kg N extra in an organic cereal crop that will generally be Nlimited (Bulson et al., 1996), will typically increase grain yield with 50 kg ha21. With the prevailing premium prices on organic products, the “shadow price” of 1 kg N taken up would be in the range of US$10, based on the market value of organically grown grain. Taking the maximum efficiency of N transfer from catch

232

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

crops or green manures to the following crop into account, the value of 1 kg N (additional) retained in the system can be estimated at US$3 – 4. When we grow catch crops, we want them to remove nitrate from the soil water. Through this reduction of the nitrate concentration in the soil water, they will reduce the nitrate content in the water percolating from the soil. It is easy to show that catch crops do this, though the extent of the effect will depend on many factors. However, a catch crop will affect the soil also after its growth, and nitrogen mineralised from the catch crop material after its incorporation may add to subsequent leaching (Thomsen and Christensen, 1999; Wander et al., 1994). Though N is mineralised from the catch crop material after its incorporation, especially the first-year effect of the catch crops may often be a reduced N supply for the succeeding crop (Chapter 3). If this leads to increased N fertiliser inputs to the system, the catch crops have impaired the N balance of the field rather than improved it. Therefore, it is important to make sure that catch crops improve the N balance of the field, either by reduced fertiliser N input or by increased crop N utilisation. Only in this way we can be sure that the overall effect of a catch crop is a reduced N leaching loss. How to grow catch crops to obtain optimal environmental and agronomic benefits will depend on factors such as local climate, soil types, main crops and farming systems and the environmental problems encountered in the local area. A more detailed discussion of management options and strategies will be developed in sections below.

II.

UPTAKE OF N AND SOIL DEPLETION

In order to reach the goal of decreased environmental impact, the primary requisite for a catch crop is to take up nitrogen from the soil, and thereby reduce the nitrate content in the water percolating from the soil. Nitrogen uptake and soil depletion by catch crops vary strongly, and results show N uptake by non-legume catch crops to vary from only around 10 (Jensen, 1991; Richards et al., 1996; Ranells and Wagger, 1997c) to 200 kg N ha21 (Mu¨ller and Sundman, 1988; Sørensen and Thorup-Kristensen, 1993; Thorup-Kristensen, 1993b, 1994b, 2001; Jackson et al., 1993; Francis, 1995; Sørensen, 1992), and in extreme examples to 300 kg N ha21 (Francis et al., 1998). There are three main sources of this large variation: † Variable catch crop growth and N-uptake potential under the prevailing climatic conditions. † Variable catch crop root growth and contact to available soil nitrogen. † Variable amounts of available nitrogen in the soil.

CATCH CROPS AND GREEN MANURES

233

A. GROWTH AND N UPTAKE POTENTIAL Catch crops normally grow during periods of the year where conditions are not optimal for crop growth. Therefore, their growth is often limited, and this can limit their uptake capacity for nitrogen. In spite of this, it would seem that catch crop N uptake is normally limited by soil N availability rather than by N uptake capacity. For a catch crop well supplied with N more than 1 Mg dry matter ha21 can be produced in 2 weeks of active growth (Vos and Van der Putten, 1997) with 3– 4% N in the dry matter. Based on such results, a few weeks of active growth would be enough for most catch crops to take up the amount of N available to them. Vos and Van der Putten (1997) found that catch crops could take up 3 – 4 kg N ha21 day21. In most experiments only the aboveground plant material is sampled, but the roots may constitute up to approximately 50% of the total biomass production though this seems to vary strongly with plant species (Breland, 1996b). Though the N concentration in root matter is normally lower than in aboveground plant material, it may still add significantly to the total N uptake capacity of a catch crop. In most published results with catch crops, especially when grown after cereals, the catch crop biomass production and N uptake is much lower than this. At the same time the tissue N concentration is also relatively low [1.0 – 2.5% of dry matter (Martinez and Guiraud, 1990; Andersen and Olsen, 1993)]. That catch crops often have higher N uptake capacity than necessary is also indicated in results showing that when fertiliser N is added to catch crops they are able to produce more biomass and take up more N than catch crops grown at lower N supply (Fig. 1 and Andersen and Olsen, 1993; Schro¨der et al., 1997). Breland (1996a) added 40 kg N ha21 to a ryegrass catch crop which had been undersown in barley, and found that the catch crop removed almost all of this from the soil within only 1 week. Further, in many studies (e.g., Ranells and Wagger, 1997c; Vyn et al., 2000; Mueller and Thorup-Kristensen, 2001) legume catch crops have been found to produce more biomass and take up much more N than non-legumes; again indicating that the N uptake by the non-legumes is limited by soil N supply. Vyn et al. (2000) found that while non-legume catch crops grown after wheat took up 12 –31 kg N ha21, red clover produced more biomass and took up between 44 and 98 kg N ha21. In experiments made at high N supply nonlegumes are often found to take up as much N as legumes, and to have high N concentrations in their biomass (Francis, 1995; Thorup-Kristensen, 1994b, 2001), indicating that in these special situations the uptake capacity may be a limiting factor. The conclusion must be that provided a catch crop is not sown too late, the risk that it does not grow to acquire sufficient N uptake capacity is generally low.

234

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Figure 1 Catch crop performance at high or low N availability. Data are from two separate experiments (Thorup-Kristensen, 2001; Mueller and Thorup-Kristensen, 2001), but both experiments were made in the same 2 years (1996 and 1997) at adjacent sites, and the catch crops in the two experiments were sown at the same time. The main difference between the two experiments were the pre-crop (green pea or barley) and thereby soil N availability for the catch crops.

When it occurs, it will normally relate to adverse conditions such as dry soil at the time of establishment or unusually high N supply from the soil. When catch crops due to sowing time or germination conditions get a too short growing season, their ability to deplete the soil inorganic N content is reduced. The results of Elers and Hartmann (1987), who established catch crops at five different dates during the autumn, showed that their N uptake was reduced with approximately 1 kg N ha21 day21 for monocot catch crops and 2 kg N ha21 day21 for crucifer catch crops. Also Vos and Van der Putten (1997) found the N uptake by crucifers to be much more sensitive to sowing date than N uptake by monocots. When sown early a crucifer catch crop seems to be able to deplete the soil faster than grasses or cereals (e.g., Aufhammer et al., 1992). It is not possible from these results to distinguish whether the reduced N uptake is mainly due to reduced uptake capacity or to reduced rooting depth at late sowing (Section “Root growth”). In situations where much N is available for the catch crop or alternatively for leaching, e.g., after many horticultural crops, insufficient nitrogen uptake capacity is more likely to become limiting. If 200 kg N ha21 or more is available for leaching, it may take a biomass production of more than 5 Mg ha21 to make the crop able to assimilate the available N.

CATCH CROPS AND GREEN MANURES

235

Under such high-N conditions, a catch crop will normally take up much N, and thereby have a strong effect against N leaching losses (Schro¨der et al., 1996) even though it will not deplete the soil completely.

B. ROOT GROWTH To be able to deplete the soil, a catch crop must develop its root system to bring it into contact with the available N in the soil. Most of the soil inorganic N is normally present in the form of nitrate, and since nitrate is one of the most mobile plant nutrients in the soil, a high root length density is not necessary for plants to deplete the soil effectively (e.g., Robinson et al., 1994; Robinson et al., 1991). Accordingly, attempts to correlate soil nitrate depletion by catch crops to their root length density have not been successful (Sainju et al., 1998; Van Dam and Leffelaar, 1998; Vos et al., 1998). On the other hand, some nitrate will be available also from deep soil layers, sometimes in quite large quantities and nitrate will often be moving downwards during the growing period of the catch crops. Therefore, developing a deep root system will bring the catch crop into contact with more soil nitrate and thus allow them to increase their N uptake. Accordingly, Thorup-Kristensen (2001) found that soil N depletion by catch crop species was highly correlated to their rooting depth, whereas it showed little correlation to root intensity. Differences in root growth have been shown to affect the amount of nitrate left in the soil, particularly in deep soil layers (Thorup-Kristensen, 1993a, 2001; Fig. 2). This ability to deplete especially the deep soil layers better, makes deeprooted catch crops especially valuable (Thorup-Kristensen and Nielsen, 1998), as the N taken up from deep soil layers is otherwise at larger risk of being lost by leaching. Rooting depth is determined by a number of factors. Plant species and the duration of growth are two of the most important factors determining catch crop rooting depth (Thorup-Kristensen, 2001), and at the same time these are factors which the farmer can control. Large differences have been observed among crop species (Fig. 3). Thorup-Kristensen, 2001 estimated that 1000 day degrees after sowing crucifer catch crops would have a rooting depth of 1.5 m, winter rye and oats 0.9 – 1.0 m and ryegrass only 0.6 m. Also Grindlay (1995) found crucifer catch crops to have deeper rooting than monocots. The deep rooting of crucifer catch crops is observed even though they are also found to allocate a much smaller fraction of their biomass to the root system than grasses and rye (Laine´ et al., 1993). At the same time Laine´ et al. (1993) found that crucifer catch crops had much higher maximum nitrate uptake rates than monocots.

236 K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Figure 2 Effects of differences in rooting depth on the ability of catch crops to deplete the soil. (a) shows the close relationship found between temperature sum used by catch crop species to reach a rooting depth of 1.0 m and the amount of nitrate-N left in the 0.5–1.0 m soil layer. (b) shows the nitrate distribution in the soil profile and the different ability of ryegrass and fodder radish to deplete the deeper soil layers, in accordance with the observed differences in root growth [Reproduced from Plant and Soil (Thorup-Kristensen, 2001, figure 4d) with kind permission from Kluwer Academic Publishers].

CATCH CROPS AND GREEN MANURES

237

Figure 3 Root development of catch crops, showing very different rates of rooting depth development. The depth development rates varied from ca. 1 mm day21 per 8C for rye, ryegrass and oats to more than 2 mm day21 per 8C for the two crucifer crops [Reproduced from Plant and Soil (Thorup-Kristensen, 2001, figure 3) with kind permission from Kluwer Academic Publishers].

The duration of the growing season is as important as the choice of catch crop species. Postponing the sowing of a catch crop will reduce its rooting depth and its N uptake especially from deeper soil layers. The significance of nitrate depletion from deep soil layers is unfortunately often overlooked in research results. Very often the soil sampling is not deep enough to reveal the differences which may be there (e.g., Sainju et al., 1998; Vos et al., 1998). Even in experiments where soil has been sampled to 1.0 m or more, the total amount of inorganic N left in the soil is often shown, without showing the effect on specific soil layers thereby disguising possible important differences in deep soil layers.

C. RELATIONSHIP BETWEEN SOIL DEPLETION AND NITRATE LEACHING LOSS To understand the effects of catch crops on N leaching loss, it is important to keep in mind that N leaching and N leaching loss is not the same. N leaching is the process of N compounds (mostly nitrate) moving down the soil profile with percolating water. N leaching loss is the special case of N leaching where N is carried across the bottom of the rooting zone or into drains, and is thereby lost from the cropping system. One important problem about this is, that the bottom of

238

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

the rooting zone is not well defined; to what depth in the soil must N be leached, before it is actually lost? In mechanistic terms, a catch crop reduces N leaching loss through three mechanisms. (1) As catch crops take up N they reduce the nitrate concentration in the soil water. Thus, water leaching through the soil will carry less N with it. (2) Nitrogen is actively transported upwards in the catch crop roots, and in this way some of the downwards leaching which had already occurred is undone. (3) By its water use, a catch crop will reduce the amount of water percolating from the soil, and thereby the amount of N leached downwards. The two first effects are normally the main effects, but they do not automatically reduce the N leaching loss. Basically, it can be said, that the effect of catch crops is to reduce the potential for N leaching loss rather than to reduce the leaching loss directly. The extent to which the potential for reduced N loss is turned into actual reductions in leaching loss depend on a number of factors. Reductions in the nitrate content in the soil water in the root zone will only affect leaching loss if the surplus precipitation is high enough to leach the nitrate depleted soil water across the bottom of the root zone. Therefore, under drier conditions, a catch crop may reduce the nitrate content in the soil water strongly with little effect on actual leaching loss. One example can be seen in the results of Willumsen and Thorup-Kristensen (2001) where a catch crop experiment was performed in 2 years with very different winter precipitation. In the wet year the difference in soil nitrate content observed between catch cropped plots and uncovered plots in the autumn almost disappeared before spring since all nitrate was lost from the uncovered (control) plots by leaching during winter. In the dry year little nitrate was lost during winter, and the differences found in the late autumn were upheld until the spring. Such differences in climatic factors will greatly affect N availablility in the uncovered plots (Fig. 4). Another problem occurs especially under wet conditions, where the leaching may start in the early autumn. While the catch crop is growing and taking up nitrate from upper soil layers, surplus precipitation may create downward water movement, which causes leaching losses at the bottom of the rooting zone. Fig. 5 show an example of this effect. The catch crops were undersown in the wheat and pea/barley crops, and were well established already at the time of main crop harvest in August. In spite of this, they had little effect on nitrate-N concentration at 1.0 m depth and leaching across 1.0 m depth until January approximately 5 months later. In this example, considerable leaching was observed before they started to reduce the nitrate content in the deeper soil layers. Similarly, Aufhammer et al. (1992) found that catch crops undersown in field beans had little effect on subsoil 0.6 –0.9 m) until early January next year. An effect of this can be seen in the results of Shepherd (1999), who found that in years where drainage started early in the autumn, catch crops were much

CATCH CROPS AND GREEN MANURES

239

Figure 4 Spring soil nitrate-N content without catch crops as dependent on winter season precipitation. Only after one very dry season was any nitrate retention indicated within the upper 0.5 m of the soil whereas significant nitrate retention was indicated in several of the drier winters when the full 1.0 m soil layer was considered (data from various experiments made at the Danish Institute of Agricultural Science, Department of Horticulture during the period 1985– 2001).

less effective in preventing leaching loss than in years where drainage started later. This effect is one reason why deep-rooted catch crops can be expected to be especially effective; they have a better chance to reduce the nitrate concentration at the bottom of the rooting zone directly, and to do such earlier during the leaching season. This can also be seen in the results of Aufhammer et al. (1992) where winter rape reduced subsoil nitrate-N content earlier and stronger than winter barley or ryegrass. To have maximum effect the catch crop should reduce the N concentration at the bottom of the rooting zone already before the main leaching period starts, and should keep the concentration low until the end of the leaching period. The last demand may conflict with the demand for catch crops to release N again early enough to make it available for the succeeding main crop (see Section “incorporation time”).

240

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Figure 5 Effect of catch crops on temporal pattern of N leaching. It is clearly seen that the reduction of leaching is postponed relative to the active growth period of the catch crop in the autumn. In both years, the effect of a catch crop grown in the autumn were not observed until around February 1, but then continued until early summer more than 2 months after the catch crop had been killed and the next crop were established [data from the experiment of Olesen et al. (2000)].

D. LEGUMINOUS GREEN MANURES In many experiments where legumes are compared to non-legume catch crops, the legumes show higher N uptake, sometimes much higher (Ranells and Wagger, 1996; Torbert et al., 1996). The amount of N added to the system by biological N fixation when legume catch crops are grown varies strongly, even within experiments. Mueller and Thorup-Kristensen (2001), Ranells and Wagger (1996) and Torbert et al. (1996) estimated N fixation from around 30 kg N ha21 to almost 150 kg N ha21 depending on year and legume species. As the N uptake by legumes is not limited by N availability, the significance of a longer growth season may be bigger, as the legume can continue to take up N also after the soil has been depleted. In experiments with high N availability, the difference in N uptake between legumes and non-legumes is generally small or absent (Thorup-Kristensen, 2001), and Torbert et al. (1996) showed that the advantage of legumes in N uptake disappeared as the N supply for the catch crops were increased. Legume catch crops acquire their N both through biological N fixation and by taking inorganic N from the soil, and they reduce soil inorganic N content during the autumn as non-legumes (Mueller and Thorup-Kristensen, 2001). Breland

CATCH CROPS AND GREEN MANURES

241

(1996a) added 40 kg N ha21 to already established catch crops, and found that two clover species were unable to remove this from the soil, whereas ryegrass removed almost all of it from the soil within a week. This result is in accordance with Laine´ et al. (1993), who found that unlike grasses, the nitrate uptake of two legume species was 100% dependent on the inducible uptake system. Though most experiments show less or even no difference between legumes and nonlegumes in soil N depletion (e.g., Vyn et al., 2000), the general conclusion is that legumes are not as effective as non-legumes. One way to utilise the N-fixing capacity of legumes without the risk of inefficient soil N depletion is to grow them in mixtures with non-legumes (Ranells and Wagger, 1997a; Janzen and Schaalje, 1992). Such mixtures often deplete the soil inorganic N pool as effectively as pure non-legume crops (Thorup-Kristensen, 2001; Willumsen and Thorup-Kristensen, 2001), but they can still add substantial amounts of N to the system by N fixation (Ranells and Wagger, 1996).

III. CATCH CROP EFFECT ON N SUPPLY FOR SUBSEQUENT CROPS Whereas it is well documented that catch crops can considerably reduce nitrate leaching loss, their effect on N supply for succeeding crops is less clear. Legume catch crops normally increase the N supply for a succeeding crop, but looking only at non-legume catch crops, the results are less clear. In some experiments catch crops are found to increase N supply for the succeeding crop and sometimes quite large effects can be observed (e.g., Elers and Hartmann, 1987; Thorup-Kristensen, 1994b), but in other experiments consistent and sometimes large reductions in N availability have been observed (e.g., Muller et al., 1989; Francis et al., 1998). Already Mann (1959), one of the first who studied the N effect of catch crops in a systematic way, concluded that “the effect is short-lived, unpredictable and not related to factors such as the amount of catch crop dry matter or catch crop N incorporated.” The increases and decreases in N supply after catch crops have normally been explained simply as the effect of net N mineralisation or immobilisation during decomposition of the catch crop residues (e.g., Ditsch and Alley, 1991). However, examples of clearly decreased N supply have also been found in experiments where the data such as C/N ratio of the catch crops, inorganic N content in the topsoil (Schro¨der et al., 1997) or 15N uptake by a succeeding crop clearly indicate that a net N mineralisation from the catch crop residues has occurred (Francis et al., 1998; Martinez and Guiraud, 1990; Torstensson and Aronsson, 2000; Jensen, 1991; Muller et al., 1989; Torstensson and Aronsson, 2000; Fig. 6). Clearly, trying to explain the effect of catch crops by the N

242

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Figure 6 Amount and depth distribution of soil nitrate-N distribution in May as affected by catch crops. Fodder radish died off naturally during the winter, whereas ryegrass survived the winter. Both catch crops were incorporated around 1 April [data from the experiment of Thorup-Kristensen (2001)].

mineralisation or immobilisation caused by the catch crop residues have not lead to an understanding of the strongly variable effects of catch crops on N supply for succeeding crops observed in experiments. The main problem is that mineralisation and immobilisation only includes the effect of catch crop on soil inorganic N content from the time of their incorporation and onwards. This is similar to the way in which the effect of addition of organic manures is understood. But contrary to added manures, catch crops affect the soil also before they are “added to the soil.” During their growth catch crops remove mineral N from the soil, and before the catch crop is killed, soil mineral N content is therefore normally lower than if no catch had been grown. In short, catch crops do not add N to the soil, they take N from the soil and subsequently return it to the soil. The effect of a catch crop on N supply for the succeeding crop (Neff) is the combined effect of the N depletion made before catch crop incorporation and the N release due to mineralisation after incorporation (Thorup-Kristensen, 1993b; Thorup-Kristensen and Nielsen, 1998). Whether this combined effect of a catch

CATCH CROPS AND GREEN MANURES

243

crop will lead to increased or decreased N supply for a succeeding crop is the result of complex interactions between the characteristics of the catch crop, the soil, the climate and the succeeding crop. These interactions are complex, and may lead to counter-intuitive results; e.g., results where a negative Neff (see Eq. 1 below) is found even though a significant net N mineralisation from the catch crop residues is observed. As an example, Torstensson and Aronsson (2000) estimated that before incorporation Ninorg was 25 kg N ha21 lower below catch crops than in control plots, whereas the subsequent mineralisation from the catch crops was only 14 kg N ha21. Thus the mineralisation was not high enough to compensate for the effect of the catch crop N uptake. This negative effect of the catch crop soil N depletion on N supply for the succeeding crop has been termed pre-emptive competition (Thorup-Kristensen, 1993b). In the experiments of Francis et al. (1998) catch crops which had taken up more than 200 kg N ha21 and had C/N ratios below 15 clearly reduced N uptake of the succeeding crop. Soil analysis showed that the content of inorganic N in the soil was reduced with approximately 100 kg N ha21 by the catch crops before their incorporation, and the mineralisation from the catch crops was not high enough to compensate for this. Though the mechanisms behind such results are complex, they are the predictable result of well-known processes. This is illustrated by the fact that plant and soil models such as the Danish DAISY model (Hansen et al., 1991) was found to be able to simulate these interactions (Thorup-Kristensen and Nielsen, 1998) based on its simulation of N mineralisation, water and nitrate movement in the soil and root growth of crops and catch crops. Probably any other model simulating these factors based on generally accepted principles would be able to simulate the same phenomena. In spite of these more complex effects of catch crops, a number of papers discuss the effects of catch crops on the supply of N and other nutrients for succeeding crops as if they were added manures. No distinction is made between experiments where catch crops were grown and incorporated at the same plot and experiments where the catch crop material were added to plots where no catch crop had been grown (e.g., Ditsch and Alley, 1991; Thomsen, 1993; Yadvinder et al., 1992). This can lead to serious misinterpretations of the experimental results. In the example mentioned earlier (Torstensson and Aronsson, 2000) measurements including only the effect of N mineralisation after catch crop incorporation would have predicted a positive effect of 14 kg N ha21, whereas the effect observed in the field was a negative effect of 11 kg N ha21. Catch crops also affect N availability through other processes (e.g., Jackson, 2000) such as denitrification and ammonia volatilisation, but though these processes may be environmentally important (see Section “Gaseous losses of N”), they do not seem to be important for the effect of catch crops on N supply for succeeding crops.

244

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

A. CALCULATING THE N EFFECT, NEFF As discussed earlier, two of the main effects determining Neff of a catch crop are its effects through N depletion of the soil (pre-emptive competition) and the effect of the subsequent N mineralisation. Both of these effects are related to the N uptake of the catch crop. The mineralisation from a catch crop depends on the amount of N it has taken up and the fraction of this, which is mineralised to become available for the succeeding crop. The effect of the soil N depletion, i.e., the pre-emptive competition also relates to the amount of N taken up by the catch crop, and the fraction of this N which would otherwise have been retained in the soil and available for the succeeding crop. Based on this, Thorup-Kristensen (1993a,b) found that the effect of growing a catch crop on the N supply for the succeeding crop (Neff) can be calculated as: Neff ¼ Nupt · m 2 Nupt · r

ð1Þ

where Nupt is the amount of N taken up by the catch crop, m is the fraction of this N that is mineralised to become available for the succeeding crop, and r is the fraction of this N which would have been retained in the root zone and directly available for the succeeding crop if it had not been taken up by the catch crop. The fact that Neff is determined by these two main effects has important implications not only for the total amount of available N as described by Eq. (1), it also has important effects on other factors such as depth distribution of the available N (see Section “Depth distribution of inorganic N”).

B. WHAT DETERMINES PRE-EMPTIVE COMPETITION High pre-emptive competition is observed where most of the N taken up by a catch crop would have been retained within the rooting zone if no catch crop had been grown. In Eq. (1), the factor r will be close to 1.0 under low leaching conditions and close to 0 under high leaching conditions. A number of factors determine this retention. Soil type and the precipitation surplus during the cool season strongly affects how much of the mineral N is leached or retained (Burns, 1984), so that under conditions with low precipitation and high soil water holding capacity the N retention is high. In the data of Francis et al. (1998) as well as from the dry year of the experiment of Willumsen and Thorup-Kristensen (2001), the content of inorganic N in the soil in the control plots actually increases during the winter season, indicating very small leaching losses and r values close to 1. These are factors that the farmer cannot affect, but as soil type and general climate are known, the catch crop strategy can be adapted to the local conditions.

CATCH CROPS AND GREEN MANURES

245

The rooting depth of the succeeding crop also affects pre-emptive competition, as less precipitation is needed to leach all of the mineral N from the root zone of a shallow rooted crop than from the root zone of a deep rooted crop. Thus preemptive competition is lower and Neff of a catch crop is higher when it is followed by a shallow-rooted crop than by a deep-rooted crop (Thorup-Kristensen and Nielsen, 1998; Willumsen and Thorup-Kristensen, 2001). Finally, the alternative retention of N is not a constant for all N taken up by a catch crop. N taken early during catch crop growth or from deep soil layers has lower r values than N taken late or from upper soil layers. This also has consequences for optimal management of catch crops (see Section “incorporation time”).

C. WHAT DETERMINES N MINERALISATION Measurements of N mineralisation from catch crops have shown mineralisation varying from more than 50% mineralisation of catch crop N during the first few months (Breland, 1994a; Dou et al., 1994) to immobilisation during early stages of catch crop decomposition (Thorup-Kristensen, 1994a). Generally, N mineralisation from catch crops and green manures are fast, with much of the mineralisation occurring within 1– 2 months. In extreme examples, Breland (1994a) and Dou et al. (1994) found that as much as 80% of catch crop N was mineralised during the 4 weeks after incorporation. From our own experiments we have seen substantial transfers of N to the subsequent cash crop (Table II; Thorup-Kristensen, 1994b). In one trial, above ground cover crops were removed from the field, and then reintroduced in other plots in increasing amounts, where carrots were subsequently grown. This allows direct estimation of the N mineralisation from the green manure (Fig. 7), and it was estimated that 34% of the N added with the green manures were present as plant N or inorganic N in the soil at the end of the growing season. Table II Nitrate-N to Different Soil Depths in May after Catch Crops, and N Uptake in the Subsequent Barley Crop Soil nitrate-N (kg N ha21)

0–0.25 m 0–0.5 m 0–1.0 m 0–1.5 m Barley N uptake

Estimated Neff (kg N ha21)

Control

Italian ryegrass

Fodder radish

Italian ryegrass

Fodder radish

23 35 65 136 107

48 62 69 77 85

58 96 128 139 153

25 27 4 259 222

35 61 59 3 46

246

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Figure 7 Carrot N uptake (closed symbols) and residual soil inorganic N content at carrot harvest (open symbols) as a function of N content in green manure added to the plots in November (circles) or April (rectangles). The N recovery (carrot N þ soil inorganic N at harvest, shown with dashed lines) are plotted with regression line estimates [data from the experiment of Thorup-Kristensen and Van den Boogaard (1999)].

There are a number of reasons for the variation in N mineralisation. Quality of the catch crop plant material is a major determinant for the mineralisation of catch crop N, but also other factors such as temperature, humidity, soil type and the degree of mixing of soil and catch crop at incorporation have been found to be of importance. One of the most important parameters determining N mineralisation from catch crops in the field is the C/N ratio (Kuo and Jellum, 2000; Jensen, 1992; Ranells and Wagger, 1996; Thorup-Kristensen, 1994a; Wagger, 1989; Frankenberger and Abdelmagid, 1985). In most examples the C/N ratio of catch crop materials is in the range of 10 –30. At low C/N ratios very fast N release may occur, and more than 50% of the plant N may become mineralised within a couple of months (e.g., Marstorp and Kirchmann, 1991). At higher C/N ratios immobilisation of N is found during the early stages of plant matter decomposition. The estimated balance point between mineralisation and immobilisation varies among studies, from around 15 (Marstorp and Kirchmann, 1991; Thorup-Kristensen, 1994a) in some studies to more than 20 in others (Frankenberger and Abdelmagid, 1985).

CATCH CROPS AND GREEN MANURES

247

Though the C/N ratio is important, it does not in any precise way predict N mineralisation under field conditions. Even with low C/N ratio legume materials mineralisation of only 10– 20% of the plant N content can be observed (e.g., Ladd et al., 1981). Other quality parameters of the catch crop material may also be important. Studies have shown contents of lignin or polyphenols to affect N mineralisation from plant material (Fox et al., 1990). However, catch crops normally consist of young plant material and high contents of such compounds or very high C/N ratios are rarely found. Others have studied water solubility of the C and N compounds of the plant material as a quality parameter, which may affect turnover and N mineralisation (Magid et al., 1997; Mueller et al., 1998a). The water-soluble fraction was assumed to be the easily decomposable part and the non-soluble fraction assumed to be more recalcitrant parts such as cell wall components. Thus, the fraction of water-soluble C and N in plant materials could be an important input for model simulations of N mineralisation (Mueller et al., 1998b), but further studies have revealed that parts of the insoluble fractions are also decomposed very rapidly (Mueller et al., 1998b; Neergaard et al., 2002). Contents of soluble compounds may also be important in another way, as they can quickly be released and removed from the remaining more recalcitrant fractions of the plant material. When catch crops are decomposing aboveground (Quemada and Cabrera, 1996; Sanderson et al., 1999) the soluble compounds may be removed by rain, when it is decomposing in the soil it may be removed by diffusion or mass flow due to water movement. When N is released in this way and removed from the remaining residues, it will reduce the N availability for the turnover of the more recalcitrant residues. In this context a distance of a few centimetres is enough to ensure that immobilisation will be effectively hampered. An example of this effect is shown by Wang and Bakken (1997), though in their experiment it was plant uptake rather than mass flow which removed released N from the remaining plant residues. Some catch crops may even contain much of their N as nitrate. The content of nitrate-N may constitute from practically nothing to more than 25% of the total N content depending on catch crop species (Laine´ et al., 1993; Thorup-Kristensen, 1994a; Van Dam and Lantinga, 1998). Thus in some catch crops, a large fraction of the catch crop N was in mineral and highly mobile form even before incubation. Soil humidity and temperature strongly affects the conditions for the soil organisms, and thereby also decomposition of plant material. Temperature fluctuations may make the plant matter more susceptible to decomposition through freeze-thaw cycles (Breland, 1994a) and humidity through drying – rewetting cycles (Van Gestel et al., 1993). A general assumption has been that N mineralisation will be low when the soil temperature is below 58C, and simulation models are built reflecting this

248

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

perception. However, field data with catch crops have indicated that N is released fast also when they are incorporated into cold soil in the winter (Breland, 1994a; Thorup-Kristensen, 1994b; Mu¨ller and Sundman, 1988). This has led to more detailed studies of N mineralisation in cold soil (Andersen and Jensen, 2001; Magid et al., 2001; Van Scho¨ll et al., 1997; Vigil and Kissel, 1995). Magid et al. (2001) examined the decomposition of Medicago lupulina, Melilotus alba and Poa pratensis at 3, 9, and 258C. No retardation of N mineralisation was observed at low temperatures, and in the analysis of variance of mineralised N the residue type contributed 10 times as much to the regression as temperature did. Contrary to this, the evolution of CO2 was sensitive to temperature, and residue type and temperature were found to be equally important. The least retardation of carbon mineralisation at low temperature was found with M. alba that had a relatively low cellulose content, and a higher content of low molecular compounds. A decrease in the bioavailability of C-rich polymers at low temperatures (Nicolardot et al., 1994), and thus a preferential utilisation of N-rich low molecular substances is one possible explanation for the difference in temperature sensitivity for C and N mineralisation. Andersen and Jensen (2001) studied gross N mineralisation during decomposition of catch crops at 3 and 98C. These studies confirmed that while immobilisation processes were much retarded at low temperatures the mineralisation of N was only slightly affected. In order to further explore decomposition at low temperatures (Magid, personal communication) examined changes in plant residue quality during decomposition at 3 and 98C over a 140 day period (see Fig. 8). The N loss from rye was practically unaffected by temperature (3 or 98C) and it lost 60 –65% of its N content during the first 35 days of decomposition (Fig. 8a). N loss from ryegrass was strongly affected by temperature, and the loss was faster at 38C than at 98C (Fig. 8b). During the early stages N loss was observed at 38C while N immobilisation was observed at 9 8C, but after the first 35 days there was little difference in the N loss from ryegrass at the two temperatures. For both materials the C/N ratio of the remaining residues was higher at 38 than at 98C during most of the decomposition period (Fig. 8c and d). This indicates that the difference in temperature affected the decomposition process as such, and not only the rate at which C and N loss occurred. These results confirm the field observations that N mineralisation from easily decomposable plant material can occur rapidly even in cold soil as also found by Van Scho¨ll et al. (1997). This has a number of consequences for the optimal management of green manure crops and catch crops in cool temperate climate regions. As substantial nitrification has also been shown even at 38C (Magid et al., 2000; Van Scho¨ll et al., 1997), this mineralisation at low temperatures can allow early leaching losses as also indicated by field data. In the biological literature there is little general recognition of differences in temperature sensitivity in the decomposition of various substrates. Katterer et al. (1998) reviewed more than 20 experiments on temperature dependence of plant

CATCH CROPS AND GREEN MANURES

249

Figure 8 Decomposition of barley and ryegrass at 38C (X) and 98C (W): [(a) and (b)] N-loss (%) of initial added N versus time, [(c) and (d)] C/N ratios of remaining plant residues versus time (Magid, unpublished data).

residue decomposition and found that modelling of CO2 evolution data could best be achieved by a two compartment parameterisation of the organic resource, assuming that the rate constants for each compartment would be similarly affected by temperature. In accordance with Katterer et al. (1998) the generally recognised soil organic matter models used in integrated analysis of management and environmental issues are build on the assumption that organic resources can be conceptually divided in homogenous compartments characterised by first order rate decay and constant nutrient ratios (Parton et al., 1988). Therefore, the temperature dependency of N mineralisation in such models is the same as that of C mineralisation. To be able to simulate the effects of incorporation of catch crops or other green plant materials into cold soil, simulation models must be changed to simulate the fast N mineralisation at low temperatures.

250

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

It may be a matter of concern, that the conditions in laboratory studies may be too far removed from the field situation. Typically, the residues to be studied in the laboratory are dried and ground and mixed thoroughly with the soil. Further, temperature is kept constant, and often high compared to the normal conditions at the time of catch crop incorporation in the field. Soil humidity is also kept constant, and mass-flow of water in the soil does not occur. In the field, larger pieces of living plant material are mixed inhomogenously with the soil by incorporation. Further, much of the decomposition may occur even before incorporation, as it will clearly happen with catch crops which are winter killed in the field before incorporation (Thorup-Kristensen, 1994b). But it also happens with catch crops which are incorporated as living plants, as they may loose considerable amounts of leaf litter during growth. Most often the materials used in incubation studies are mixtures of all the aboveground plant parts. Only few have studied the N mineralisation from the plant roots (Franzleubbers et al., 1994a,b; Thomsen, 1993). Roots have a different morphology than stems and leaves, which often make them less prone to microbial invasion and decomposition (Neergaard et al., 2002). Some of these differences between the laboratory studies and the conditions in the field have been investigated (Esala, 1995; Breland, 1994a,b, 1996a), and have been found to affect the results. Breland (1994a) and Jensen (1994) studied effects of particle size and distribution. Jensen (1994) found that the normal practice of grinding straw material prior to incubation led to a much higher N immobilisation compared to the same straw materials cut in pieces. The extent of errors which may occur when using laboratory incubation data for modelling catch crop N release under field conditions is not clear, but they may be large in some situations (Thorup-Kristensen and Nielsen, 1998). It is therefore very important that the models are carefully tested and calibrated against field data before their simulations of the field effects can be trusted.

D. DEPTH DISTRIBUTION OF INORGANIC N As discussed above, the Neff of a catch crop in the field is determined by the two main effects pre-emptive competition and mineralisation, and this combined effect can be described with Eq. (1). However, not only the total amount of inorganic N available from the soil is affected; but also its depth distribution is strongly affected (Fig. 9b and c). This occurs as the normally positive effect of N mineralisation is reflected mainly as an increased content of inorganic N in the upper soil layers where the mineralisation occurs. Contrary to this, the negative effect of pre-emptive competition is mainly reflected as a reduced content of inorganic N in deeper soil layers (Thorup-Kristensen, 1993b). The reduced content of inorganic N in the deeper soil layers is found as catch crops prevent

CATCH CROPS AND GREEN MANURES Figure 9 Nitrate-N distributions in the soil profile at five dates during and after catch crop growth. The catch crops followed a green pea crop, and were sown in early August. The soil nitrate measurements were made in November under the still growing catch crops (a), in March just before incorporation of the catch crops (b), in mid-May approximately 6 weeks after catch crop incorporation and approximately 3 weeks after sowing of a barley crop (c), in August just after harvest of the barley crop (d), and in October under a new catch crop (e) consisting of a grass clover mixture which had been undersown in the barley crop [data from the experiment of Thorup-Kristensen (2001)].

251

252

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

N leaching to deeper soil layers, and to some extent due to direct catch crop N uptake from deeper soil layers. This altered depth distribution of inorganic N is shown in Fig. 9 and Table II. The difference in nitrate-N between control plots and catch crop plots in May is an estimate of Neff. Estimated in this way for the full 1.5 m soil layer, the results indicate that Neff of the fodder radish catch crop was close to zero, whereas the Neff of ryegrass was strongly negative (Table II). But this consideration depends strongly on the soil layer considered. If a soil layer of only 0.25 m or 0.5 m is considered, the data indicate a clearly positive Neff by both of the catch crops. If a soil layer of 1.0 m is considered, ryegrass had almost no effect, whereas fodder radish still had a clearly positive Neff. The nitrate-N data from August (Fig. 9d) indicate that the barley crop had an effective rooting depth of between 1.0 and 1.5 m. In accordance with this, ryegrass reduced the N uptake by the barley crop with approximately 20 kg N ha21 whereas fodder radish increased the N uptake of barley with approximately 45 kg N ha21. It is obvious from these data why Neff depends on the rooting depth of the succeeding crop. If the succeeding crop had a rooting depth of 0.5 m or less, both catch crops are estimated to have a positive Neff (Table II), but with deeper rooting, the estimated Neff is reduced, and with rooting depths of 1.5 m, ryegrass is estimated to have a clearly negative Neff. Thus the conclusion is that the same catch crop may have a positive Neff if it is followed by a shallow rooted main crop, and a negative Neff if it is followed by a deep rooted catch crop. In accordance with this, Willumsen and Thorup-Kristensen (2001) found that a winter rye catch crop increased the N uptake by shallow rooted onion crops, but reduced the N uptake by deep rooted white cabbage crops. Basically, legume catch crops have the same effect of concentrating available N in the topsoil. But as they take their N not only from the soil, but also adds further N to the system by biological N fixation, the risk of an overall negative Neff is much less than with non-legumes. Figure 9 may also be used to illustrate why the Neff of catch crops depends on soil type and precipitation. It is clear from the data from March and May how some of the nitrate present in upper soil layers in November has been leached to between 1.0 and 1.5 m during the winter season. On more sandy soils, or with higher winter precipitation the nitrate will leach even deeper, and the negative effect of catch crops observed in the deep soil layers in the spring will be reduced. This will lead to a more positive Neff especially for deep rooted crops. With less precipitation on the other hand, more of the autumn N will be retained in the soil, and it may be found closer to the surface, i.e., the risk of negative Neff will be higher. The importance of the winter precipitation on N availability for subsequent crops is further shown in Fig. 10. Here it becomes apparent that while the use of cover crops can ensure a rather stable delivery of N to the subsequent crops regardless of winter precipitation, soil inorganic N content varies strongly

CATCH CROPS AND GREEN MANURES

253

Figure 10 Spring soil nitrate-N content versus winter season precipitation with or without catch crops. The data shows spring soil nitrate N to be less sensitive to precipitation with than without catch crops. After dry winters catch crops tended to reduce soil nitrate-N content, whereas after wet winters they tended to increase it. Data from various experiments made at the Danish Institute of Agricultural Science, Department of Horticulture during the period 1985–2001.

depending on winter precipitation when no catch crop is grown. Without a catch crop, most of the available N is lost from the soil in the wet winters but most is retained in the dry winters. Thus, after dry winters a negative Neff was estimated but after wet winters a positive Neff was estimated.

E. GASEOUS LOSSES OF N During their decomposition green materials have been shown to loose NH3-N through volatilisation due to high NH4 concentrations in the decomposing tissues (Harper et al., 1995; Larsson et al., 1998). After mineralisation as NH4-N the nitrification –denitrification cycle may result in losses of N2O (Aulakh et al., 1991; Quemada and Cabrera, 1995; Larsson et al., 1998; Rosecrance et al., 2000).

254

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Larsson et al. (1998) found substantial ammonia (17–39% of applied N) emissions from decomposing grass and alfalfa mulches, and nitrous oxide emissions were seen to exceed those from bare soil (1% of added N). Harper et al. (1995) found minimal ammonia emissions from clover mulches, and Whitehead et al. (1988) found ammonia emission to be strongly temperature dependent. It would seem likely that ammonia emissions are small when a catch crop or green manure are incorporated in a cool moist soil, compared to the mulching situation. Baggs et al. (2000) studied the fate of nitrogen from incorporated cover crops and green manure residues in field trials in NE Scotland. Under the prevalent cool conditions they found very small differences in N2O emissions between plots with or without cover crops. Generally, emissions were lower on the planted plots when they effectively lowered NO3 concentrations in the soil system, but somewhat higher during their decomposition phase.

F. FIELD EFFECTS Various measures have been used to estimate the field effect of catch crops. The effect has also been measured as the effects on spring soil inorganic N content or on crop N uptake. Though these methods have their drawbacks, e.g., in deciding measurement depth for soil inorganic N content, they may be the most direct measures of Neff, which we can make. However, they do not tell the farmer how much the N fertilisation of the crops may be reduced due to a catch crop, as the uptake of N from chemical fertilisers is also well below 100%. Therefore, a number of experiments have been designed to determine a “fertiliser replacement value” (e.g., Stute and Posner, 1995a). Apart from the effect on the first succeeding crop, catch crops also increase the N supply for later crops. Finally, in a number of experiments the N mineralisation from catch crops residues have been studied by 15N techniques or by difference methods, but as discussed above, a measure of N mineralisation is not an estimate of the Neff of a catch crop. Though exceptions exist, the various field estimates of Neff are rarely high. Many examples can be found where soil tests or measurements of N uptake in subsequent crops indicate that only between 0 and 25% of catch crop N can be used in the next year (Andersen and Olsen, 1993; Ladd et al., 1981, 1983; Vyn et al., 1999, 2000). A number of results even show a clearly negative Neff (e.g., Francis et al., 1998; Muller et al., 1989; Sørensen, 1992), and in some experiments Neff varies strongly with catch crop species, with negative Neff of some species and positive Neff of others (Vyn et al., 1999, 2000; Willumsen and Thorup-Kristensen, 2001; Table IV).

CATCH CROPS AND GREEN MANURES

1.

255

Recovery of Catch Crop N

An issue of considerable agronomic interest is the efficacy of various fertilisers, whether organic or inorganic. For the commonly used compound N fertilisers a transfer of approximately 50% is typically found during the first season, and only a small residual effect can be measured in later years, indicating that considerable losses must have occurred. Even under the most carefully observed addition and subsequent management of the crops in research experiments, using 15N as a tracer, the recovery of 15N from compound fertiliser was most often less than 60%, which was comparable to recovery from 15N in stored urine, while recovery of 15N labelled faeces was lower, in the range of 9– 17% (Parton et al., 1988; Sørensen et al., 1994; Sørensen and Jensen, 1996, 1998; Theunissen, 1995; Thomsen et al., 1997). From our own experiments we have seen substantial transfers of N to the subsequent cash crop, although we have only one trial that seems appropriate for quantification (Fig. 7). In this trial above ground cover crops were removed from the field, and then reintroduced in other plots in increasing amounts, where carrots were subsequently grown. The resulting response curve indicates an efficacy of transfer of 34% during the first season after cover crop incorporation. This result cannot be generalised, but is consistent with a number of experiments where the succeeding crop has shown a very substantial positive response to green manures. Thus it would seem that the efficacy of cover crops might be on the same level as that commonly found for animal slurries (mixtures of urine and faeces). Very little data on efficacy of cover crops in temperate agro-ecosystems are available. Paustian et al. (1992) inferred from experiments covering a 30 year period that the long-term efficacy of transfer of N from green materials to subsequent crops would be near 40%. After a 6-year experiment Schro¨der et al. (1996) estimated that 58, 50, 115 and 73% of the N added in pig slurry, fertiliser N, rye catch crop and ryegrass catch crop, respectively, were recovered in the maize crops or in the soil by the end of the experiment. The effect of the catch crops is overestimated as only the aboveground plant parts were considered (Schro¨der et al., 1996), but anyhow, the results indicate that the utilisation of catch crop N in the longer term can be as high as that of inorganic fertiliser. Paustian et al. (1992) reported yield and soil organic matter data covering a 30 year period in long-term plots covering a number of treatments among which were green manure, farm yard manure, compound fertiliser and no fertiliser treatment. Like the farmyard manure and the compound fertiliser, the green manure was added to the plots and not grown there, thus precluding pre-emptive competition effects. A calculation indicates that losses mainly due to leaching were no different between the two treatments (Table III), and that the lower recovery of N in

256

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Table III Complete N Balances of a 30-year Long-term Organic Amendment and Fertiliser Experiment (Paustian et al., 1992)

Green manure (g N m22) Input Deposition Fertiliser total N Sum of input

% of total input

Compound fertiliser (g N m22)

49 313 362

Output Harvested N Measured net change in total soil N Lossesa % Fertiliser total N harvested relative to no fertilisation treatment (nitrogen use efficiency)

% of total input

49 256 305

269 38

74.3 10.5

291 225

95.4 28.2

55

15.2

39

12.8

39.6

57.0

a Losses ¼ Sum of input N 2 Harvested N 2 Measured net change in total soil N over a 30-year period.

harvested crops in the green manure treatment (74 versus 95%) was balanced completely by a build up of soil organic matter. The calculations of harvested N and losses both show that while the compound fertiliser treatment was most efficient in terms of transfer to the crop, the green manure treatment came quite close in efficiency, and that green manure carried on to a plot may be comparable to, e.g., pig slurry in terms of fertiliser effects. However, it is important to bear in mind that since the green manures were not grown in the amended plots themselves effects such as pre-emptive competition are not included in the overall picture in this experiment.

2.

Fertiliser Replacement Value

In experiments where the effect of catch crops is compared to the effect of added fertilisers, a mineral fertiliser equivalent (FE) can be estimated. This is a rather indirect measure of catch crop Neff, and will include other effects than the N effects (Janzen and Schaalje, 1992). An example of this was shown by Decker et al. (1994). They observed a negative Neff of a wheat catch crop, and positive Neff of legume catch crops, but they found little difference in the

CATCH CROPS AND GREEN MANURES

257

optimal N rate for the succeeding maize crop. Though the legumes increased maize yield at zero fertilisation to almost the level of optimally fertilised maize where no catch crop had been grown, they also increased the optimal yield, and thus a response to fertiliser application was observed. Though the measurements of fertiliser value are very indirect measurements of catch crop effects, they may still give our best estimate of the value of catch crops for the farmers. When catch crops are found to have a positive effect, FE is sometimes found to be very high as compared to more direct measurements of Neff. Decker et al. (1994) found that legumes increased the N uptake by maize with 50 – 70 kg N ha21, but to achieve the same N uptake without a catch crop, 135 kg N ha21 had to be added. Dou et al. (1994) compared the effect of hairy vetch and red clover to the addition of 200 kg N ha21. The effect of the legumes on maize N uptake was around 50 and 100% of the effect of adding 200 kg N ha21 in the no-till and conventional till systems, respectively. Schro¨der et al. (1996) found that an unfertilised ryegrass catch crop could not replace any N fertiliser, but when adding 100 or 200 kg N ha21 to the ryegrass, fertiliser values of almost 50 and almost 100 kg N ha21 was found. Red clover, which in the same experiment took up only 57 kg N ha21 in aboveground plant material was found to have a FE of almost 100 kg N ha21. Stute and Posner (1995a) found fertiliser replacements of various legume crops ranging from around 20 kg N ha21 to many examples of values between 100 and 200 kg N ha21. The effect of the green manures on maize N uptake was much smaller, and on average only about 20 kg ha21. During a 6-year experiment Schro¨der et al. (1996) found more moderate FE of rye and ryegrass catch crops in the order of 20– 40 kg N ha21. FE is not only a product of the catch crop itself, but also of the succeeding crop. In the experiment of Sørensen and Thorup-Kristensen (1993), catch crops of ryegrass and phacelia were found to increase the N uptake of both onion and white cabbage (Thorup-Kristensen, 1993b) with 35 kg N ha21, but as onion took up 32% of added fertiliser N and white cabbage took up 61% of added N, the estimated FE was 111 and 60 kg N ha21, respectively, for the two vegetable crops. As maybe the best measure of agricultural value of catch crop N effects, sometimes very high FE demonstrates a high potential for using catch crops. In systems with cheap and unlimited access to fertiliser N this may not be enough to make them attractive to farmers (Stute and Posner, 1995a). In agricultural systems where access to fertiliser N is limited, such as in organic farming, water protection areas or in less intensive agriculture the potential for using catch crops and green manure is high. The strongly variable results, varying from substantially negative effects to effects equivalent to 200 kg N ha21 of fertiliser N, show that it is very important to know how to manage catch crops to obtain the

258

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

better effects. Catch crop management to optimise catch crop effects will be discussed in detail in Chapter 5.

3. Second Year and Longer Term Effect As discussed in earlier sections, the immediate effects of catch crops and green manures on soil fertility are determined by a multitude of factors. As pre-emptive competition will rarely affect the N supply for the second succeeding crop, the effect of catch crops in the second year and onwards is likely to depend primarily on continued N mineralisation. One important factor for the short-term effects on soil C and N content is the relationship between residue quality and decomposition. In the longer term, this effect of plant matter quality on soil fertility is more likely to be overshadowed by the effect of total amount of organic matter added to the soil system. Mann (1959) concluded from the first year effects of catch crops that they were unpredictable and not related to catch crop biomass production or N content, but found clear relationships between the N content of catch crops and their effect on the second succeeding crop. Studies of the N effect of catch crops during several years after incorporation of the catch crop have indicated quite low mineralisation rates (Ladd et al., 1983; Mann, 1959; Thomsen, 1993). Approximately 4– 10% of the catch crop N is mineralised in the second year, and very little is mineralised in the third year (Jensen, 1992). Data on N uptake in a barley crop (Andersen and Olsen, 1993) showed no second year effect of catch crops, whereas Schro¨der et al. (1996) found that barley yields in the second year were slightly increased with 0.1– 0.2 Mg ha21. These studies considered the longer-term effect of one catch crop, but repeated growing of catch crops can be expected to show more clear longterm effects. A number of contributions give evidence of an appreciable increase in labile soil organic N after relatively few years (3 –7 years) with catch crops (Thomsen and Christensen, 1999; Lewan, 1994; McCracken et al., 1989; Schro¨der et al., 1996; Torstensson and Aronsson, 2000; Kuo and Jellum, 2000; Kuo et al., 1997; Sainju and Singh, 1997; Garwood et al., 1999; Hansen et al., 2000). Paustian et al. (1992) estimated a net change in soil organic N content equivalent to approximately 30% of the N added with green manure (average C/N ratio 16.8, average lignin content 6%). In this example a calculation shows that catch crop N was utilised almost as effectively as fertiliser N, and that the difference was due to higher soil N storage when catch crops were used (Table III). Sainju and Singh (1997) and Garwood et al. (1999) reported a consistent yield reduction by winter rye used as a catch crop over 7 years, but overall positive

CATCH CROPS AND GREEN MANURES

259

effects on the system N balance. Suction cups yielded increasing NO2 3 concentrations under catch cropped plots, but this was countered by an estimated decrease in drainage. These experiments have been conducted in a relatively dry environment where pre-emptive competition between winter rye and the following spring crops can have a substantial yield suppressive effect. Schro¨der et al. (1996) reported an increased dry matter production in N deficient maize systems, stating that the effect corresponded to fertiliser N rates of 105% of aboveground N in rye and 40% of aboveground N in perennial ryegrass. Averaged over 6 years 115 and 73% of the aboveground N in rye and ryegrass were recovered in the crop-soil system. Kuo and Jellum (2000) observed that while having vetch was the only cover crop that significantly increased N deficient maize yield, both winter rye and annual ryegrass benefited soil organic N and gradually improved maize biomass production compared with the control over the longer term (8 years). The contribution of catch crops to the long-term mineralisation rate of the soil is a contribution to the general soil fertility (Kuo and Jellum, 2000; Schro¨der et al., 1996; Hansen and Djurhuus, 1997), but it may also lead to increased nitrate leaching losses in later years. The results of Thomsen and Christensen (1999) indicate that increased leaching loss on the long-term reduced the overall catch crop effect with as much as 30% compared to the effect on nitrate leaching measured in the years when the catch crops were grown. However, the soil type and crop choice in this experiment favoured leaching, and in general such losses can be expected to be lower. The extent of such losses will depend on crop rotation and fertilisation level (Thomsen and Christensen, 1999) as well as soil type and climate. Gustafson et al. (2000) estimated that continued catch cropping reduced N leaching losses with an average of 50% over successive years. The issue of long-term efficiency of systematic use of catch crops and green manures may be more affected by the managers ability to choose appropriate crops suited for the local hydrological regime, and especially the avoidance of excessive pre-emptive competition. This may be formulated as a question: is it possible to continually loose less and sustain a higher production level with the use of catch crops and green manures?

IV. OTHER EFFECTS OF CATCH CROPS Although the major argument for incorporating catch crops and green manures into crop rotations has been to increase the N supply for succeeding crops, many other positive as well as some negative effects of catch crops and green manures have been documented (Biederbeck et al., 1998; Janzen and Schaalje, 1992; Sainju and Singh, 1997; Yadvinder et al., 1992). Some of these

260

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

effects may deserve equal attention as the pure N effects, since they may not only substitute other resources, but actually enhance the yield or quality potential of the crop. As an example, there may be a positive interaction between, e.g., soil structural effects of the catch crops and the yield response of the main crop to supplied N. Furthermore, as described earlier, conventional farmers are not likely to use catch crops and green manures for their value as N source alone, due to the low price of fertilizer N. However, if other significantly positive effects exist, the sum of these may prove sufficiently profitable to encourage their use by conventional farmers, even if their positive environmental effects are disregarded. Non-nitrogen effects may be divided into (i) effects on nutrient availability other than N, (ii) effects on soil biological activity, (iii) effects on the soil as a growth medium for the crop, (iv) effects on soil water content, and (v) effects on crop pests, pathogens and weeds.

A. EFFECTS ON OTHER NUTRIENTS THAN N For phosphorous availability, two principal effects of catch crops and green manures may be hypothesised. Firstly, catch crops and green manures take up soil P and thus convert it from inorganic to organic form. Some species may have especially high P uptake capability, e.g., by forming particularly long root hairs (Gahoonia and Nielsen, 1997; Gahoonia et al., 2000), by acidifying the rhizosphere (e.g., most leguminous species), root-exudation of organic acids (Jones, 1998; e.g., in buckwheat and lupines) or high affinity for mycorrhiza. Upon incorporation of the residues into the soil the plant P is released slowly and is not as susceptible to adsorption and precipitation as inorganic P fertilisers. Moreover, as catch crop residues are typically heterogeneously distributed in the soil after incorporation, the immobilisation of released residue P is further hampered by the reduced contact with the soil matrix. This is likely to improve the crop competitive advantage, as has been shown for crop residue N with increasing heterogeneity of the incorporated residues (Wang and Bakken, 1997). Secondly, it may be hypothesised that during decomposition, catch crop and green manure residues may mobilise some of the otherwise unavailable soil P, e.g., by release of organic acids, analogous to the release of organic exudates from roots. However, recent attempts to verify this hypothesis have not been able to identify significant amounts of organic acids in the residue-sphere, and other studies have questioned the presumption that organic acid exudation is the reason for the high P uptake capability of some species on low P soil (Jones, 1998). Catch crops and green manure effects on phosphorous availability may thus most likely be simply due to their P uptake, and the availability to the subsequent crop determined by the residue C/P ratio and the soil P status (Thibaud et al., 1988).

CATCH CROPS AND GREEN MANURES

261

Not many studies have quantified catch crop P uptake under temperate conditions. In a 6-year experimental period Vos and Van der Putten (2000) found that the catch crops winter rye, oil radish and forage rape catch crops in a conventional cropping system contained between 4 and 9 kg P ha21 in aboveground biomass prior to incorporation in spring. However, catch crop dry matter production in this study was low and ranged between 0.4 and 0.9 mg ha21; if higher catch crop biomass production occurs, larger P uptakes must be anticipated. For potassium, catch crops and green manures effects on K availability may be rather straightforward. Potassium taken up by the catch crop is usually very quickly released upon incorporation (e.g., Lupwayi and Haque, 1998) and thus likely to be equally available to fertiliser K. As for P, amounts of K taken up by catch crops may vary, Vos and van der Putten (2000) found between 21 and 45 kg K ha21 in aboveground biomass in spring. Catch crops may also affect the content of K and other cations in the soil water directly by their uptake of these ions, or indirectly as they deplete the soil water for anions such as nitrate and sulphate (Yanai et al., 1996). Ja¨ggli (1978) showed an example of how losses of K, Ca, and Mg were reduced by catch crops, and Scott et al (1919) showed how the soil content of a number of plant nutrients was affected by continuous use of catch cropping with vetch or rye/legume mixtures. Sulphur deficiency has become an important feature of most North European arable cropping systems, due to the greatly reduced sulphur emissions from fossil fuels. Sulphur behaves very similar to nitrogen in the soil system, and it can easily be lost by leaching in the form of sulphate. Very few studies have focussed specifically on the effects of catch crops on sulphur retention and availability. Eriksen and Thorup-Kristensen (2002) have shown that catch crops may exert a similar influence on the distribution of soil sulphate as for soil mineral N, although the depletion of the soil profile was not as efficient for sulphate. They also showed that the uptake capability can be quite high, and that the variation among catch crop species may be substantially larger than for N. In particular cruciferous species, which usually have a high plant S concentration, showed high uptakes of 22– 36 kg S ha21, compared to only 8 kg S ha21 taken up Italian ryegrass. This was also manifested in the S availability effect on the subsequent crop. Spring barley after an oil radish catch crop thus contained about 50% more S than barley in the control plots where no catch crop had been grown, whereas a ryegrass catch crop actually decreased the subsequent crop S uptake compared to the control. For micronutrients, catch crops do not constitute a significant source or sink. However, they may exert a significant influence on their availability, through their influence on redox potential of the soil or through increased chelation capacity (Hinsinger, 1998), and either may be caused by the increased biological activity and organic matter turnover (Yadvinder et al., 1992). However, no specific studies on catch crop and green manure effects on these properties have been conducted under temperate conditions.

262

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

B. EFFECTS ON SOIL MICROBIOLOGICAL AND FAUNAL ACTIVITY Effects of catch crops and green manures on soil biological activity are evident in both the short- and long-term, as also discussed in Section III.C “What Determines N Mineralisation” Depending on the magnitude of catch crop biomass inputs, several years with frequent catch crops may increase soil microbial biomass by up to 60% and cellulytic enzyme activities by 90% (Debosz et al., 1999; Mendes et al., 1999). Effects on total soil C are small, however, Kuo et al. (1997) demonstrated how 7 consecutive years of cover cropping in continuous maize only increased the soil C content with 3– 5% in the top 15 cm, even though the yearly biomass input from the most productive cover crops was in the order of 4 Mg ha21 aboveground and 4– 5 Mg ha21 below ground. Catch crops will add substrate to the soil microorganisms not only at the time of their incorporation, but all through their growth period as well. Through root exudation, root turnover, symbiosis with mycorrhiza and through leaf litter loss from aboveground plant parts, substrate will be added to the soil all through the growth period of the catch crop. Catch crops and green manures may also exert a significant effect on soil vesicular – arbuscular mycorrhiza (VAM). VAM fungal infection potential has been shown to increase in maize after cover cropping (Boswell et al., 1998; Kabir and Koide, 2000), whereas no effect or even slight negative effects were shown in small grain cereals (Baltruschat and Dehne, 1989). Only few studies exist on the specific effects of catch crops and green manures on soil fauna. However, as the application of organic matter will increase the densities of most members of the detritus food web, this is likely to occur also in systems where catch crops and green manures are frequently applied. In a study of microarthropods, significantly higher abundances of collembola and mites were found after incorporation of catch crops and green manures (Axelsen and Kristensen, 2001), with fodder radish promoting densities of up 120,000 collembola and 90,000 mites m22, compared to 30,000 and 50,000 m22 in the control without catch crop. Filser (1995) and Scholte and Lootsma (1998) found much higher densities of collembola in green manured compared to mineral fertilised fields, albeit at much lower total densities than (Axelsen and Kristensen, 2001). Recently, it has also been speculated that cover crops may influence the degradation potential of the soil for pesticides. Bottomley et al. (1999) studied the degradation of the herbicide 2,4-D, and consistently found increased degradative capabilities of both surface and subsoil layers after a rye cover crop compared to no cover crop in a vegetable cropping system. The increased capability persisted longer than the cropping season immediately following the cover crop incorporation, and thus cover crops may possibly play a role in preventing pesticide leaching.

CATCH CROPS AND GREEN MANURES

263

C. EFFECTS ON SOIL PHYSICAL PROPERTIES Organic matter is known to improve a range of soil physical properties, including both soil structure and water retention. Catch crop and green manure effects may be divided into effects on soil tilth (soil surface strength, friability, aggregate stability) and soil porosity (air permeability, hydraulic conductivity, water retention). These may in turn affect crop establishment (seedling emergence), root development and soil losses by runoff and erosion (Meyer et al., 1999). Cover crops have been shown to positively influence soil tilth by reducing the soil surface strength, created by crusting of the soil upon the impact of rainfall (Folorunso et al., 1992), which disrupts and slakes soil aggregates at the surface. They showed that an oat-vetch cover crop reduced soil surface strength by 24 – 41% depending on soil type. Winter cover crops have also been shown to increase aggregate stability, protecting against aggregate breakdown during winter and resulting in better friability and structure after spring tillage (Hermawan and Bomke, 1997). If catch crop residues are left on the soil surface or only lightly incorporated, this may also affect the soil moisture regime dramatically by altering evaporation and soil temperature (Teasdale and Mohler, 1993). However, during active growth, catch crops and green manures deplete the available soil water and may thus impact the water supply for establishment of a subsequent crop, at least in climates with a low winter precipitation (Mitchell et al., 1999). Soil porosity (Scott et al., 1919) and subsequent root growth have been shown to be positively influenced by a clover green manure prior to a lettuce crop (Stirzaker and White, 1995). However, effects on root development may not only be due to green manure effects on soil porosity, but also to effects on soil mineral N distribution (Thorup-Kristensen and Van den Boogaard, 1999) or possibly to allelopathic effects (Burgos et al., 1999; Creamer et al., 1996).

D. EFFECTS ON SOIL WATER CONTENT The use of water by catch crops may be a serious problem when catch crops are grown in areas with limited rainfall (Harper et al., 1995). The effects on soil water content are very different from the effects on plant nutrients. Whereas catch crops do not remove nutrients from the field, and may even prevent losses, a growing catch crop will remove water from the field. The use of water by a catch crop may occur in pre-emptive competition with the succeeding crop, as previously discussed with N. The effects on water are simpler than the effect on N though, as there is no positive effect resembling the N mineralisation from catch crop residues obscuring the observation of the preemptive competition. Thus only the negative effects of water use by the catch

264

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

crops are observed. In California, McGuire et al. (1998) and Mitchell et al. (1999) observed reductions in spring soil water content of up to 80 mm due to catch crops. The water use by catch crops depends strongly on the growing conditions and growth period of the catch crops and under more northern conditions the water use will generally be small. However, not all water use by catch crops cause pre-emptive competition. In areas with sufficient surplus precipitation the water used by catch crops will be replaced, and only water percolation will be reduced, whereas spring soil water storage is not affected, as also McGuire et al. (1998) observed in one year with high winter precipitation. This is also one reason why spring water use by catch crops become a special problem; it not only increases the total water use by the catch crop, and whereas the water used in the autumn has a good chance of being replenished by subsequent precipitation this is much less likely with water used in the spring. Thus as with pre-emptive competition for N, the pre-emptive competition for water may be strongly increased by allowing active catch crop growth in the spring.

E. EFFECTS ON PESTS, PATHOGENS AND WEEDS The majority of studies on catch crop and green manure effects on crop pests have dealt with possible effects on plant parasitic nematodes (Abawi and Widmer, 2000). On the one hand, a catch crop may increase the risk of infestation with certain parasitic nematodes that may propagate on a susceptible catch crop; on the other hand, catch crops may exert a nematicidal effect through their decomposition products. The effect seems to be largely dependant on the catch crop species. Abawi and Widmer (2000) tested the effect of a range of incorporated catch crops on infestation of bean with lesion nematodes and found effects to vary by a factor of 35, the most efficient repressors being ryegrass and rapeseed, the most promoting being Hairy vetch. McBride et al. (1999) demonstrated that rye catch crop foliage significantly reduced cotton root-knot nematode populations for at least 21 days. They hypothesised that the effect was due to the production of low molecular weight acids during decomposition, but they were only able to detect low concentrations of such acids in the soil solution, and found that these acids were being degraded extremely rapidly in the soil. Thus the mechanisms behind the nematicidal effects are not known. A very common concern amongst farmers is the possibility that introducing catch crops or green manures may propagate certain soil-borne pathogens, which may render the soil inappropriate for certain crops for decades. This has been a particular concern for a number of root diseases of different crops. However, several studies have shown that catch crops and green manures can be used as break crops, actually having a suppressive effect, reducing the soil-borne

CATCH CROPS AND GREEN MANURES

265

pathogen intensity (Davis et al., 1996; Muehlchen et al., 1990; Sumner et al., 1995; Theunissen and Schelling, 2000; Tu, 1988; Williams-Woodward et al., 1997; Yamagishi et al., 1986). Cunningham (1983) on the other hand found no effect of ryegrass or white mustard catch crops on take-all and eyespot on barley in a long-term (14 years) experiment. The mechanisms involved in disease suppression or propagation are numerous and often unknown. Some of the mechanisms may be common with effects of organic matter amendments in general (Linderman, 1989), such as release of certain organic compounds with toxic effects or the stimulation of microbial activity resulting in increasing competitive or antagonistic suppression of pathogenic organisms. Many root disease studies indicate positive effects only of certain cover crops. Yamagishi et al. (1986) found that cruciferous cover crops reduced club root, Theunissen and Schelling (2000) found undersown clover to reduce cavity spot in carrots, and Davis et al. (1996) found sudangrass or corn green manures to reduce Verticillium wilt of potato. These studies also underline that the cover crop species greatly influences repression effectivity, Abawi and Widmer (2000) found anything from a 10% increase (after white clover) to a 40% decrease (after rapeseed) in the effect of a range of incorporated cover on root rot of beans. A number of studies have been concerned with root rot of pea, WilliamsWoodward et al. (1997), Tu and Findlay (1986) and Muehlchen et al. (1990) have found cover crops, in particular oats and crucifers (white mustard and rape), to significantly reduce pea root rot caused by Aphanomyces eutiches. However, Bødker and Thorup-Kristensen (1999) found the effect to be highly variable between years, with oil radish reducing incidence of A. eutides one year and increasing it another year. The effects of catch crops and green manures on weeds are often more variable than those observed for pests and diseases. Again, the mechanisms may be multiple (Liebman and Dyck, 1993). On the one hand, weeds may be suppressed by either direct competition and allellopathy or by phytotoxic effects on germinating weeds upon incorporation of the green manure (Creamer et al., 1996; Hoffmann et al., 1996; Teasdale & Daughtry, 1993). On the other hand, catch crops may promote weed infestation by hindering chemical or mechanical weed control, or for persistent catch crops and green manures actually acting as weeds themselves later in the cropping sequence. Based on such very different mechanisms it is not surprising that greatly variable and inconsistent results have often been found, e.g., in pea (Khatib et al., 1997), in soybean (Moore et al., 1994; Leiebl et al., 1992) and in corn (Hoffmann et al., 1993). However, Boydston and Hang (1995) found that a rapeseed catch crop was very effective in controlling weed density (73 – 85% reduction) in potato production, greatly enhancing the yield (17 – 25%). As mentioned above, decomposing cover crop material may produce phytotoxic substance that retard weed germination, but this may affect the main crop establishment and growth as

266

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

well, especially in sensitive cultures as many vegetables (Stirzaker and Bunn, 1996). This underline the fact that also in this respect, correct cover crop management strategies are crucial for success.

V. MAKING THE MOST OF CATCH CROPS IN CROPPING SEQUENCES AND WHOLE CROP ROTATIONS Before deciding on a strategy for optimal effects of catch crops, it is important to define the precise goals. It is clear that the goals may differ between farmers and the policy makers. Where the policy makers aim at reducing leaching losses and environmental problems, the farmers must try to optimise profitability through low costs of establishment and positive effects on succeeding crops (Lu et al., 1999). Also policy makers must try to find cost effective methods to reduce leaching losses (Hasler, 1998; Lu et al., 1999; Gustafson et al., 2000). As the goals are different, the methods may also be different, but the strategies which can be adopted to optimise Neff of a catch crop will often also improve its environmental effect. Where the farmers prioritise other goals, such as effects on soilborne diseases, soil structure, erosion control, minimising the cost of growing a catch crop, or N fixation by legumes, the choices made by the farmer may conflict with the interests of the policy makers. To reach the goals at the farm level, the farmer has several agronomic management tools to optimise the Neff of catch crops: † The placement of the catch crops within the crop rotation, which determines both how much N will be available for the catch crops to take up, and the subsequent Neff of the catch crop due to interactions with factors such as rooting depth of the succeeding crop. † The choice of time and method of catch crop establishment, which determines cost and efficiency of catch crop establishment. † The choice of catch crop incorporation time which may strongly affect Neff through effects on mineralisation, leaching, pre-emptive competition and the depth distribution of the inorganic N in the soil. † The choice of catch crop species, which affects practically all other factors through effects on N uptake capacity, rooting depth, kill off time and C/N ratios. In the text below, we will discuss these main management tools available to the farmers for optimising the effects of catch crops. The management of catch crops is complex, and tools and effects should of course not be considered in isolation but in the context of whole crop rotations.

CATCH CROPS AND GREEN MANURES

267

In order to integrate the various processes and their interactive effects, we have carried out model scenario simulations, which will be presented to illustrate effects on both crop production and environmental protection. At the policy level, the goals are different, and other management tools are available. Regulations can be made to favour catch crops on specific farm types or soil types to reduce overall leaching losses or in specific geographical regions to protect vulnerable environments or aquifers. The management of catch crops at the policy level will be discussed at the end of this chapter.

A. PLACING CATCH CROPS IN THE CROP ROTATION One of the most important management tools which can be used to improve the results of catch crops, is choosing the right place in the crop rotation to grow the catch crops. To achieve a high Neff, the farmer must consider (1) whether much leachable N is present in the field, (2) whether an efficient catch crop can be established, and (3) whether the next years crop will respond well. The amount of N available to a catch crop depends on the previous main crop and the cropping history (Janzen and Radder, 1989; Aufhammer et al., 1992; Jensen, 1991; Shepherd et al., 1993; Kessavalou and Walters, 1999). Much N can be available in the soil after intensively fertilised crops, especially if organic fertilizer has been used (Jackson et al., 1993; Kessavalou and Walters, 1999; Sainju et al., 1999; Thorup-Kristensen and Van den Boogaard, 1999), after crops leaving high amounts of N rich residues in the field or after plough down of perennial crops (Francis et al., 1995; Shepherd, 1993). Much N can also be available if shallow-rooted crops have been grown, as they have only exploited the uppermost parts of the soil (Jackson and Stivers, 1993; Thorup-Kristensen and Sørensen, 1999), and much of this will then be available in deeper soil layers. Even with relatively deep-rooted crops increased fertilisation within one year may, though the fertiliser is added to upper soil layers, lead to increased subsoil nitrate content at harvest (Kessavalou and Walters, 1999; Sainju et al., 1999; Thorup-Kristensen, 1993b). Such deep N can only be taken up if deep-rooted catch crops are grown. Furthermore, much N can be present in the autumn as a result of what happened in previous years. If perennial crops as a grass-clover ley have been ploughed under 1 or 2 years earlier, or if much farmyard manure has been used, mineralisation rates may still be high. Further, if much N is left in the soil in one autumn, some of this can still be available in the next autumn. This will typically be found in deeper soil layers, as there has been considerable time for downwards leaching, and a main crop have taken up what was available in upper soil layers (Fig. 9d; Dick and Christ, 1995; Izaurralde et al., 1995; Kessavalou and Walters, 1999; Sainju et al., 1999).

268

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Another important point is to grow catch crops where they have a good chance to develop well. If the main crop is harvested late, there may be no point in growing a catch crop afterwards even if much leachable N is left in the soil after harvest. This dilemma is illustrated by the many studies made on catch crops in silage maize production in Europe (Schro¨der et al., 1992; Ballcoelho and Roy, 1997; Schro¨der et al., 1996; Shipley et al., 1992). Silage maize is grown extensively in many parts of Europe, and is a crop which typically leaves much inorganic N in the soil after harvest. Still, the crop is harvested late, and the duration of the growing season for the catch crop after maize harvest will often be the factor limiting catch crop efficiency (Schro¨der et al., 1996). Therefore, experiments with undersown catch crops in silage maize production have been made to secure a longer growing season for the catch crop (e.g., Ballcoelho and Roy, 1997). Generally, early harvested crops, or crops which allow effective establishment of undersown catch crops (see below) can give very good opportunities for growing catch crops. Legume catch crops are generally not as effective as non-legumes to deplete soil inorganic N, and should therefore only be grown where little inorganic N is left in the soil (see Chapter 2), and where non-legumes would give little effect. However, the need for early establishment may be more important than with nonlegume catch crops, as none of the legumes establish as rapidly as the fastest non-legumes, and it takes some time for the N2 fixation to start. In some systems legume catch crops are expected to grow and fix N2 mainly in the spring, e.g., before maize crops (Wagger, 1989), and then the autumn period should just allow efficient establishment, but not necessarily much growth of the legume catch crop. Finally, the effect of a catch crop may depend strongly on crop species which are grown afterwards. As discussed in Section “Depth distribution of inorganic N” catch crops generally increase N availability in the uppermost soil layers and reduce it in the subsoil (Fig. 9c). This is advantageous to shallow rooted crops, but less so, or even disadvantageous to deeper rooted crops (Table II). As discussed in Section III.F.2 “Fertiliser Replacement Value” the estimated fertiliser replacement value of catch crops (Sørensen and Thorup-Kristensen, 1993) was 111 kg N ha21 when it was followed by onions which have a very shallow root system, whereas it was 60 kg N ha21 when it was followed by white cabbage. A similar result was found by Willumsen and Thorup-Kristensen (2001). Schro¨der et al. (1997) compared catch crop effects on sugar beets and potatoes, and found less clear results. Sugar beet took up added fertiliser N more effectively than potatoes (70 versus 40%) and also took up catch crop N more effectively when the catch crop was ryegrass. Therefore, the fertiliser replacement value was not much affected by succeeding crop. However, potatoes recovered more of the N from a red clover catch crop than sugar beet, and a calculated fertiliser replacement value for the red clover catch crop was therefore

CATCH CROPS AND GREEN MANURES

269

approximately twice as high when it was followed by potatoes as when it was followed by sugar beet. (Shepherd, 1999) found that catch crops increased the yield and N uptake of succeeding potatoes but not that of sugar beet. However, as the potatoes were well fertilised they speculate that the effect on potato yield was not due to N effects of catch crops. These results show that the farmer can strongly improve the effect of catch crops by choosing the optimal position in the crop rotation to grow catch crops. This includes N availability for the catch crop, the possible duration of the growing season, and the crop to be grown in the following year.

B. ESTABLISHING CATCH CROPS Catch crops are grown in periods of the year, which are not suitable for commercial crop production. Thus the climatic conditions are often cold and the growing season short. To grow an effective catch crop, it is important to make the most of this period, and therefore it is important that the catch crop growth and N uptake starts as fast as possible after harvest of the main crop. In most catch crop experiments catch crops have been established either during main crop growth or after main crop harvest, but in some experiments both methods have been used (Stute and Posner, 1995a; Jensen, 1991; Aufhammer et al., 1992). Undersowing allows the catch crop to be established already before harvest, but it may cost a yield loss in the main crop due to competition (Breland, 1996b). Growth of the main crop will influence the establishment of an undersown catch crop, and a reduced seeding rate of the main crop have sometimes been recommended when establishing catch crops in small cereals. Also the N supply to the main crop can be important. Breland (1996b) found that increased N fertilisation of a barley crop increased the subsequent growth and N uptake of a ryegrass catch crop, whereas it had the opposite effect on white clover and subterranean clover. Catch crops sown after harvest will not interfere with the main crop, but may be more expensive to establish due to the extra field operations needed, and they will have shorter time for growth and N uptake. Based on this, Karlsson-Strese et al. (1998) worked to identify plant species and genotypes which do not compete strongly with the main crop. In maize, vegetables, or other row crops the competition problem with undersown catch crops is even worse (Mu¨ller-Scha¨rer, 1996; Lotz et al., 1997), as the row structure tends to reduce the competitiveness of the Main crop. However, as many row crops are late harvested and leave much available N in the soil, the potential of undersowing catch crops is large. Thus, experiments have been performed to develop systems where catch crops are undersown in row crops at a later stage, where the main crop is well established and has a higher competitive

270

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

ability (Mu¨ller-Scha¨rer, 1996; Lotz et al., 1997; Schro¨der et al., 1996; Aufhammer et al., 1992; Ballcoelho and Roy, 1997). It is not possible to draw a general conclusion from these experiments, but the results do show that in many situations such systems can be developed and the potential advantages of undersowing can thus be used. The specific solutions will depend on the main crop, catch crop species and local conditions. When catch crops are sown after main crop harvest, attempts to establish them with reduced soil tillage is one way to reduce the cost of establishment (Shepherd, 1999), but it may also reduce the effect of the catch crop. After harvest of a main crop the soil may often be too dry to allow germination of a catch crop. Choosing catch crop species is also important, as the amount and price of seed to be used differs greatly, and some species may be more tolerant to sub optimal seedbed preparation. This is the case with some of the crucifer species, where it has even been found possible to establish a catch crop just by spreading the seeds into a cereal crop approximately 2 weeks before its harvest. Choosing species, which germinate rapidly and establish a plant cover quickly will improve the utilisation of the short growing season. One way to reduce the cost of catch crop establishment is to reduce the seeding rate. Few results have been published on this subject, but the results of Clark et al. (1994) show an example where reduced seeding rate reduced catch crop N uptake and the estimated Neff.

C. INCORPORATION TIME In a number of experiments the effect of incorporation time or kill date have been studied, and sometimes quite large effects have been found, indicating that incorporation time can be an important management tool. Many of the experiments have simply compared autumn incorporation to spring incorporation (Vyn et al., 2000; Hansen et al., 1997), but in some experiments two or three incorporation dates in the spring (Clark et al., 1994, 1997a; Garwood et al., 1999; Wagger, 1989) have been compared and in a few studies two incorporation dates in the autumn (Torstensson, 1998; Wallgren and Linden, 1994) have been compared. Thorup-Kristensen (1996) compared the effect of catch crops incorporated at five dates during the winter season, two in the autumn and three in the spring. The results from these experiments demonstrate that changing incorporation date with a few weeks during the autumn or during the spring can affect leaching losses and Neff strongly (Fig. 11). Postponing incorporation during the autumn will allow catch crops to take up more N. As N mineralisation will start later, it will also reduce the risk of leaching loss of N mineralised from the catch crop residues. The results from the published experiments (Thorup-Kristensen, 1996;

CATCH CROPS AND GREEN MANURES 271

Figure 11 Effects of incorporation time on catch crops Neff as dependent on precipitation regime.

272

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Wallgren and Linden, 1994; Sanderson et al., 1999) show that postponing incorporation during the autumn can significantly increase Neff. Postponing incorporation during the spring will lead to changes in plant residue quality parameters such as C/N ratio (Clark et al., 1997a; Wagger, 1989) and lignification (Wagger, 1989) and thereby lead to slower N mineralisation (Wagger, 1989) or even immobilisation (Wyland et al., 1995) from the incorporated residues. Postponing incorporation in the spring will also allow the catch crops to take up spring mineralised N from the soil. As this N would normally be directly available for the succeeding crop this uptake leads to a strong pre-emptive competition, and several results have shown reduced Neff of non-legume catch crops when the incorporation is postponed during spring (Thorup-Kristensen, 1996). With legume species postponing incorporation will allow more time for biological N fixation, and changes in C/N ratio will be smaller than with nonlegumes, thereby making Neff of legume catch crops less sensitive to incorporation time than that of non-legumes (Vyn et al., 2000; Wagger, 1989; Clark et al., 1994, 1997a; Wallgren and Linden, 1994). Vyn et al. (2000) found that a red clover catch crop increased maize yield with approximately 1.5 mg ha21 irrespective of incorporation time, whereas rye reduced corn yield with 1.4 mg ha21 when killed in the spring, but only 0.5 t ha21 when killed in the autumn. When legumes can be allowed to grow for a substantial period in the spring before incorporation this can actually be used to obtain increased N fixation (Clark et al., 1997a) and increased mineralisation after incorporation. As an example, Clark et al. (1994) found that vetch increased subsequent corn yield with approximately 2 mg ha21 when incorporated in April, but with approximately 3 mg ha21 when incorporated in May. The effect of postponing incorporation of rye was opposite, as it had no effect on corn yield when incorporated in April, but reduced corn yield with approximately 1.5 mg ha21 when incorporated in May. After catch crops the available soil N is normally more strongly concentrated in the topsoil layers, with higher amounts of inorganic N in the uppermost soil layers and less in the deeper soil layers, an effect which becomes more marked with later incorporation (Sanderson et al., 1999; Thorup-Kristensen, 1993b; Thorup-Kristensen and Van den Boogaard, 1999). It is also seen in experiments where species which survive the winter are compared to species which are naturally winter killed (Fig. 9c; Thorup-Kristensen, 1994b). This concentration of available N in the topsoil is an advantage for shallow rooted crops, but not necessarily for all crops (Table II). Different incorporation dates of catch crops will also have other effects than effects on Neff for the first succeeding crop. One example is that results from one long-term experiment indicate that repeated spring incorporation leads to higher soil C content than continued autumn incorporation (Hansen and Djurhuus, 1997).

CATCH CROPS AND GREEN MANURES

273

Postponing incorporation will also allow more water use by the catch crop (Bollero and Bullock, 1994), though Clark et al. (1997b) found no effect. In areas where water availability is a main limitation to crop growth, increased water use can make late incorporation very disadvantageous for subsequent crop growth. Some results show that catch crops can contain phytotoxic compounds, which may reduce the subsequent germination of other plant species, both crop and weed species (Dabney et al., 1996). Incorporation of a catch crop very shortly before the establishment of the next crop will increase such effects, and can lead to problems of making an optimal seedbed preparation. In conclusion, the highly significant effects on Neff often found by changing incorporation date with a few weeks emphasise incorporation date as an important management tool in optimising catch crop effects. Incorporation time or kill date can be decided not only by choosing tillage date, but also by choosing catch crops which are winter-hardy or winter-killed as will be discussed in Section “Kill date.” The highly significant differences also emphasise the importance of the choice of incorporation date in catch crop experiments. Especially in experiments where more species are compared it should be kept in mind that the differences observed in Neff may be mainly due to differences in kill date if some are winter killed and others not.

D. CHOOSING CATCH CROP SPECIES A recurring theme in most catch crop experiments is comparison of different plant species. In most published research on catch crops several species have been included in the experiments, even where the objective of the experiments was not to study species effects but other factors such as incorporation time, mineralisation rates, etc. One reason for this continued interest in comparing species is that virtually any aspect of catch cropping is affected by the choice of species. Many results show that the effects of catch crops can depend strongly on the species choice. Just regarding the N effects of catch crops, at least the following important factors can be controlled by the choice of catch crop species: † Speed of establishment, growth rate and rooting depth, as well as cold tolerance. All these factors affect N uptake capacity during the autumn. † Kill date can be controlled by choice of more or less winter hardy species. † Nitrogen fixing capacity, by choosing legumes and choosing among legume species. † Quality of the catch crop plant material, e.g., C/N ratio, contents of lignin, nitrate-N and other water-soluble compounds affecting subsequent N mineralisation.

274

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Apart from all this, the non-N effects of catch crops discussed in Chapter 4 will also often depend on species choice. The fact that all these factors depend on the choice of species makes the choice of plant species a powerful management tool in catch cropping. In the following the significance of some of the specific characters which vary among catch crops will be discussed. At the end of the section, effects of different plant species or “species types” on Neff and yield of the succeeding crop will be discussed. Apart from choosing among species, many experiments with species mixtures have been made. There can be several reasons for wanting to mix different species, the most obvious is mixing non-legumes which can effectively deplete soil N with legumes with their biological N fixation (Ranells and Wagger, 1997a,b; Clark et al., 1997b).

1.

N Uptake Capacity

As the growing season of catch crops is normally short, it is important to look for species with fast establishment, fast growth and root growth. Among the tested species several crucifer species fulfil this, and in many experiments they are found to have superior N uptake capacity compared to other non-legumes (see Section II.A “Growth and N Uptake Potential”). Though some of the crucifer species used as catch crops have fast growth and root growth they may not be the best choice when catch crops are sown very late. This is illustrated in the results of Thorup-Kristensen (2001) who found that during the early growth stages the cereals tended to have the deepest rooting, as they were faster to initiate their root growth. However, due to a clearly faster rooting depth penetration, the crucifers developed the deepest rooting at later growth stages but it took approximately 600 day degrees from sowing before the crucifers had a clearly deeper rooting than the cereals. Thus with a short growing season these data suggest that cereals were superior to crucifers, only with longer growing seasons were the advantages of crucifers realised. One reason for the faster initial rooting of cereals may be the larger seeds and sowing rates of cereals, which will give them a head start compared to crucifers or small seeded grasses. The results of Power and Zachariassen (1993) and Ilgen and Stamp (1993) show how seed size determines the ranking of plant N uptake at early growth stages, whereas at later stages this relationship disappears as other characteristics of the plant species become more important. Differences in initial biomass due to seed weight and seeding rates can be quite dramatic. Crucifer crops are often sown at a rate of 5 or 10 kg seed ha21 compared to typical rates of 100 – 200 kg seed ha21 of cereals. Using these sowing rates, a crucifer crop will typically have to increase its weight 20-fold, just to reach the biomass a cereal catch crop has right from the time it is sown.

CATCH CROPS AND GREEN MANURES

275

Cold tolerance will also become increasingly important when catch crops are sown later, as their effect then depend more on their ability to continue growth also into the colder periods. A number of studies have dealt with the response of catch crop N uptake and N metabolism to low temperatures (Laine et al., 1994; Macduff et al., 1994; Clarkson et al., 1992; Power and Zachariassen, 1993; Van Dam and Lantinga, 1998). Laine et al. (1994) found the N uptake of crucifer species to be less sensitive to low temperatures than the N uptake by monocot species. Power and Zachariassen (1993) found faba bean and especially hairy vetch to have higher N uptake than other legumes at low temperatures, also when compared to legumes normally grown in cool climates such as white clover, sweet clover and field pea. When undersown, catch crops should preferably not grow too fast in the beginning, as this will cause competition against the main crop, but still they should be able to utilise the growing season after main crop harvest effectively (Karlsson-Strese et al., 1996). Karlsson-Strese et al. (1998) tested many species and genotypes of potential catch crops, and found that the ryegrass species, which appear to be the most commonly used species for undersown catch crops, were not optimal, as they competed too strongly with the main crop. Within each of the plant types grasses, legumes and non-legume dicots, species were identified which combined good autumn growth with less competition against the main crop.

2. N Fixing Capacity Though large differences in N fixation among legume catch crops are observed, it is hard to draw general conclusions. With short growing periods, species such as hairy vetch, crimson clover, and red clover are often found to give good results. As an example, Stute and Posner (1995a) compared red clover, sweet clover, hairy vetch and two types of alfalfa, and found N uptake of the different legumes ranging from 42 to 128 kg N ha21 on an average of 4 years. When the catch crops were undersown in oats the best results were obtained with red clover yielding 128 kg N ha21, compared to 78 kg N ha21 by hairy vetch and between 47 and 56 kg N ha21 for the three other crops. When the catch crops were sown after oat harvest hairy vetch showed the best result with 108 kg N ha21 compared to between 42 and 67 kg N ha21 for the three other crops. However, many other species may give good results depending on the growing conditions. The conclusion seems to be that we know species which will normally give good results and which can be used in most cases, but also that much may be gained by identifying species especially suited to the relevant conditions.

276

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

CATCH CROPS AND GREEN MANURES

3.

277

Kill Date

As discussed in Section V.C “Incorporation time” the choice of incorporation date is important for optimising Neff. Kill date may be controlled by mechanical incorporation of catch crops, but can also be controlled by choice of catch crop species. Species that are killed off naturally may be an advantage as the soil may often be wet and unsuited for field operations at the optimal time of catch crop incorporation. With a catch crop, which dies of during winter, incorporation may be postponed until the spring when the soil is suitable for tillage again. Whereas winter killed species may be an advantage where catch crops should optimally be incorporated in the late autumn or during winter, they are not well suited where catch crops should be kept alive during winter and not incorporated until the spring (Figs. 11 and 12). Thorup-Kristensen (1994b) compared the effect of winter killed and winter hardy catch crops, and showed how differences in kill date affected the timing of mineralisation and the depth distribution of inorganic N in the spring soil. Even among winter killed species, the kill date may vary significantly. In the data of Thorup-Kristensen (1994b) white mustard and fodder radish was still fully viable in mid November, whereas oats and phacelia had apparently already started to release their N content. Such differences among winter killed species may be significant, as even a few weeks delay in incorporation time of catch crops during the autumn may strongly affect Neff (Section “Incorporation time”).

4.

Plant Material Quality and N Mineralisation

The plant matter produced by different catch crop species may have very different quality, leading to different N mineralisation rates after incorporation (Kuo et al., 1996; Thorup-Kristensen, 1994a; Wagger, 1989; Wivstad, 1997). Plant matter quality varies due to growing conditions, but systematic differences between species are also found. Some of the commonly used winter hardy cover crops (e.g., Italian ryegrass and perennial ryegrass) have been bred as fodder crops with the view to ensure a high energy delivery to ruminants, and have the ability to assimilate substantial amounts of carbon even at winter time. This in turn can result in a rather high C/N ratio (. 30) in the cover crop at the time of incorporation—and thus diminish the return of N to the subsequent crop. Accordingly, ryegrass is repeatedly found to have a higher C/N ratio than winter rye (Kuo et al., 1996; Thorup-Kristensen, 1994b).

Figure 12 Diagram showing a proposed practical guidance for the use of catch crops. By adding up the score from the three boxes on the left side of the diagram, an estimate of “leaching intensity from the root zone” is made. This score is used in the right side of the diagram to find the relevant advise.

278

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Among legumes hairy vetch has repeatedly been found to have a very low C/N ratio around 10, which is lower than that of most other legume species (e.g., Kuo et al., 1996).

5. Species Differences in Neff As mentioned above, the different characteristics of catch crop species regarding growth, root growth, N uptake, winter hardiness, etc. leads to very different effects on the yield and N uptake of the succeeding crop. It is not surprising that legumes with their capacity for biological N fixation often have a better effect on succeeding crops than non-legumes. What may be more surprising is that systematic and sometimes quite large differences are observed among legume species as well as among non-legume species. This is found even among species which all seem to be efficient catch crops. Some of the more striking examples have been reported by Breland (1996b), Thorup-Kristensen (1994b) and Vyn et al. (1999, 2000). Thorup-Kristensen (1994b) found that while barley grown without a preceding catch crop took up 45 kg N ha21, barley grown after catch crops took up anything from 58 kg N ha21 after ryegrass to 112 kg N ha21 after fodder radish. A similar difference is shown in Table II, though in this example the N uptake without a catch crop was intermediate to the two catch crops. On average across the results of Vyn et al. (1999), the yield of maize without a preceding catch crop was 8.1 Mg ha21, whereas the maize yield was 6.5, 8.6 and 10.1 Mg ha21 after catch crops of ryegrass, fodder radish and red clover, respectively. Similarly, across the results of Vyn et al. (2000), yields of maize was reduced by approximately 1 Mg ha21 after rye or oat catch crops, increased with 0.2 Mg ha21 after fodder radish and increased with 2.5 Mg ha21 after red clover. Schro¨der et al. (1997) found that the fertiliser N replacement value of undersown red clover was almost 100 kg N ha21, practically the same as found with ryegrass catch crops fertilised with 200 kg N ha21, whereas unfertilised ryegrass had practically no effect. Several other examples can be found. The differences are not only dependent on the choice of “plant types,” such as monocots, crucifers or legumes, but clear differences can also be found within these groups. In comparisons between rye and ryegrass catch crops, rye is found to have a better effect on the succeeding crop. When more crucifer crops are compared, fodder radish is often found to have the best effect. In experiments where legumes are compared, hairy vetch is normally superior (e.g., Stute and Posner, 1995b). At least some of the differences observed among plant types can be understood based on the knowledge we have about root growth, N fixation, C/N ratios and other factors. However, the apparently systematic differences

CATCH CROPS AND GREEN MANURES

279

among species within the plant groups indicate that there are still important mechanisms we do not understand. The comparisons mentioned above are only those which involve differences which are so obvious, that their effect can be clearly observed across a number of very different experiments. Many lesser differences, but differences, which can still be agriculturally important, could be found in experiments designed for that purpose. It should also be kept in mind, that only a small fraction of the plant species, which could be used as catch crops, has been tested as such. Generally most catch crop experiments include only species which are also grown for commercial purposes, but many other species could be used and might offer considerable advantages. As an example, Kabir and Koide (2000) performed an experiment to compare the effect of a wheat catch crop with the effect of a weed covered soil. They established dandelion as an “artificial weed,” but they found dandelion to have better effects on the succeeding maize crop than the wheat catch crop, whether this was measured as yield, P uptake or mycorrhiza infection of the maize root system.

6. Research Perspectives in New Catch Crop Species Based on the large differences observed among plant species grown as catch crops, it is obvious to continue the work in studying plant species with potential as catch crops, and to study the mechanisms behind the differences among species. Karlsson-Strese et al. (1996) screened 518 accessions belonging to 134 plant species for their suitability as undersown catch crops in cereals. Based on this, Karlsson-Strese et al. (1998) tested 118 accessions belonging to 39 plant species, and measured their competition effect against a barley main crop, and their ability to grown and cover the soil in the autumn after barley harvest. Their results clearly show that within the groups of grasses and legumes a number of species could be used as catch crops. They also identified a number of interesting nonlegume dicot species, which generally showed less competition against the main crop than the grasses or even the legumes. Among the non-legume dicots, especially chicory seemed to combine low competition against the main crop with an efficient plant cover after harvest. Based partly on the work of Karlsson-Strese et al. (1998), we are currently testing a number of plant species as undersown catch crops, measuring also their root growth, soil N depletion and their Neff. Initial results show that the roots of chicory reach approximately twice as deep as the roots of ryegrass, and that the N uptake by chicory is higher than that of ryegrass. As ryegrass undersown in cereals is presently the most common catch crop in Denmark, chicory may have the potential to improve catch crop effects significantly.

280

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

E. MODEL SIMULATION OF CATCH CROPS IN THE CROP ROTATION In this section, we will try to illustrate some of the complex interactions between management tools, climatic and environmental factors described above, by presenting the results of model simulations of catch crop scenarios (Text box 1). By using a well tested and validated dynamic simulation model for this (DAISY), we can achieve realistic simulations with regard to most processes, while at the same time avoiding the experimental “noise” often caused by atypical climatic conditions. We have set up a mixed six-course rotation, with 50% spring barley, a legume and two root crops. In this rotation we introduce 33 or 66% catch crops, either shallow-rooted (ryegrass) or deep-rooted (fodder radish). The rotation is subjected to two different N input levels (corresponding to a typical organic farm and a conventional intensive farm with animal production) and two climatic regimes (high or moderate winter precipitation). For more details on the scenarios and model, please consult Text box 1. Overall, the catch crop scenarios reduce N leaching compared to no catch crops (Table IV). The data have been calculated as the average over the last 18 years (three rotations) of the simulated 24-year period, and the result can thus be considered the medium to long-term effect of catch crops. Defining the correct leaching depth is very important for the estimated leaching loss. As seen in Table IV, somewhat different conclusions may be reached if the lower flux boundary is set at either 1.0 or 2.5 m. This is most prominent for the rotations including deep-rooted catch crops, but underlines the errors, which may often be made in measurements of nitrate leaching based on suction cups placed at 1.0 m depth. Actually, if deep-rooted crops are included in the rotation, N measured to be leached to below 1.0 m may be taken up again, and after mineralisation actually contribute to estimated leaching loss once again! When looking at the leaching loss at 2.5 m, it should be noted that almost the same reduction in N leaching loss was achieved with 33% deep-rooted as with 66% mixed catch crop in the moderate precipitation regime. Table V shows that especially in this combination of deep rooted catch crops and moderate precipitation, measuring N leaching loss at 1.0 m may lead to an underestimation of the catch crop effect. This indicates that by growing deep-rooted catch crops, they need to be used less frequently under moderate or low precipitation. This may be an advantage, as strong pre-emptive competition and negative Neff is found especially under moderate or low precipitation conditions, and also negative effects on the soil water balance are most likely under such conditions as indicated in Table VI. In the low N input scenario, leaching is increased significantly (ca. 30%) by the high compared to the moderate precipitation, whereas at high N input the simulated leaching loss was comparable in the two climates.

CATCH CROPS AND GREEN MANURES

281

Text box 1. Description of model simulations of catch crop scenarios

Model Simulations were carried out with the latest version of the Daisy model (Hansen et al., 1991a; Abrahamsen and Hansen, 2000), which has been extensively validated and used for crop rotation modelling and N leaching (Willigen, 1991; Diekkru¨ger et al., 1995; Hansen et al., 2001; Jensen et al., 1994, 1997, 2001; Magid and Kølster, 1995; Smith et al., 1997). The model was parameterised according to earlier studies (Jensen et al., 1999; Mueller et al., 1997, 1998b). Crop modules were either standard (Abrahamsen and Hansen, 2000) or in the case of Oil radish catch crop, the one developed by Jensen et al. (1999). Soils and climate Simulations were carried out with a sandy loam soil parameterised from Aarslev Experimental Station (Jensen et al., 1999). The climate used in the simulations was an average Danish climate (mean annual temperature of 7.88C, warmest and coldest months 15.8 and 20.28C, respectively), but with two different precipitation regimes, either 991 or 661 mm annual precipitation, corresponding to a simulated percolation at 2.5 m depth of 508 or 198 mm, respectively, with the chosen crop rotation. The climate file contained natural day-today variations in temperature and precipitation. Crop rotation and catch crop scenarios An overview of the crop rotation and catch crop scenarios simulated is given below. The scenarios include catch crops in 33% of the six-course rotation with either a deep (Oil radish, Raphanus sativus, sown after harvest of main crop) or a shallow (Ryegrass, Lolium multiflorum, undersown) rooted catch crop. A scenario with the maximum frequency of catch crops in the rotation (66%) was also applied with a combination of the catch crops. Finally, a control scenario without catch crops was included.

Major crops in rotation:

Catch crop strategy: Deep-rooted (Oil radish) Shallow-rooted (Ryegrass) Combination (both) No catch crops

Green peas

Spring barley

OR RG OR

OR RG RG

Potatoes (late)

Spring barley

RG

Sugar beets

Spring barley

RG

N input scenarios For each of these catch crop scenarios, a low and a high N input regime were adopted. An overview of the fertilizer and manure inputs is given below for the whole crop rotation. Low N input corresponds to application of animal manure containing 70 kg total N ha21 year21, but no inorganic fertilizer application. This is a typical N level for many organic farms. High N input corresponds to application of animal manure containing 140 kg total N ha21 year21), supplemented with inorganic fertilizer N up to Danish statutory N norms for crop N fertilisation, corresponding to an average of 106 kg available N ha21 year21 (176 kg total N ha21 year21) over the crop rotation. This corresponds to a moderately intensive N input level, typical for many conventional farms in northern Europe.

282

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Major crops in rotation: Green peas

Spring barley

Potatoes (late)

Spring barley

Sugar beets

Spring barley

Average (kg N ha21

Low N input: Pig slurry (tot-N) Effective-N 50% Fertilizer Sum available N

0 0 0 0

50 25 0 25

150 75 0 75

50 25 0 25

100 50 0 50

70 35 0 35

70 35 0 35

High N input: Pig slurry (tot-N) Effective-N 50% Fertilizer Sum available N

0 0 0 0

170 85 30 115

120 60 110 170

170 85 30 115

210 105 20 125

170 85 30 115

140 70 36 106

Simulations Scenarios were simulated by carrying out the rotation four times, i.e., for a period of 24 years. The initial rotation, the first 6 years, were considered a ‘warm up’ period, and all data were calculated as average figures for the last three rotations only.

In the low precipitation regime this effect continues for several years, whereas in the high precipitation regime, the effect is only observed in the first year and only when using a leaching depth of 2.5 m. In subsequent years, leaching is actually increased, e.g., after sugar beet and spring barley in years 5 and 6, due to N mineralisation from the previous catch crops. An example is seen after spring barley in the fourth year of the rotation (Table IV), where N leaching is increased from 44 to 69 kg N ha21 year21 in the high N, high precipitation regime. As described earlier, frequent use of catch crops make years with a bare soil more vulnerable to N leaching due to the build up in soil organic matter and mineralisation capacity. As discussed earlier nitrate concentration of the percolate, rather than total N loss may be of greater concern for groundwater quality. In this case, the conclusions are somewhat different from those on quantitative N loss (Table V). Without a catch crop, both N input levels cause the percolate to exceed the drinking water limit of 11.3 mg NO3-N l21 in the moderate precipitation regime, whereas in the high precipitation regime the limit was not exceeded at any of the scenarios. The large differences in percolation volume, where only 198 mm is lost in the low precipitation regime compared to 509 mm in the high precipitation regime (Table VI), allows much higher N leaching losses in the high precipitation regime without exceeding the limit. Catch crops may clearly improve percolate quality, but it is evident that in the long-term, the shallow-rooted catch crop will only have little effect, if only grown in 33% of the rotation. The reason for this is shown in Fig. 13, where the shallow-rooted catch crop lowers the nitrate

Table IV Simulated N Leaching (kg N ha21 year21, Average over Three Crop Rotations) as a Function of Precipitation Regime, N Input Level and the Effect of Catch Crop Rooting Depth, Frequency in the Rotation and Whether the Leaching Depth is Defined at 1.0 or 2.5 m. The Leaching Loss has been Cumulated in Each Crop Year until the Following Spring Leaching depth

Climate Moderatea precipitation

N input regime

Rotation

Year

100 cm

250 cm

Nil (0%)

Deep rooted (33%)

Shallow rooted (33%)

Combination (66%)

Nil (0%)

Deep rooted (33%)

Shallow rooted (33%)

Combination (66%)

16p 14p 14 16 17 20 16

6p 1p 2 9p 14 13p 7

Lowb

Pea Sp. barley Potato Sp. barley Sugar beet Sp. barley Average

1 2 3 4 5 6

45 20 31 21 9 17 24

2p 0p 33 28 12 20 16

15p 2p 29 24 10 19 17

0p 3p 36 9p 8 4p 10

21 24 26 26 21 22 23

6p 0p 1 10 16 21 9

Highb

Pea Sp. barley Potato Sp. barley Sugar beet Sp. barley Average

1 2 3 4 5 6

55 45 48 48 23 43 43

6p 1p 43 68 36 76 38

27p 5p 40 53 26 55 34

2p 10p 50 29p 33 37p 27

39 39 46 44 39 42 41

20p 1p 2 14 29 45 19

34p 11p p 31 0p 31 4 27 14p 31 25 39 29p 32 14 (continued on next page)

CATCH CROPS AND GREEN MANURES

Catch crop scenario (freq.)

283

284

Table IV (continued) Leaching depth

Climate Higha precipitation

N input regime

Rotation

Year

100 cm

250 cm

Nil (0%)

Deep rooted (33%)

Shallow rooted (33%)

Combination (66%)

Nil (0%)

Deep rooted (33%)

Shallow rooted (33%)

Combination (66%)

Lowb

Pea Sp. barley Potato Sp. barley Sugar beet Sp. barley Average

1 2 3 4 5 6

60 26 42 29 8 25 31

1p 1p 47 35 10 29 20

11p 2p 42 33 9 28 21

1p 2p 50 6p 8 3p 11

35 43 31 37 26 16 31

6p 0p 8 39 33 21 18

22p 11p 11 35 29 18 21

2p 1p 10 34p 14 9p 12

Highb

Pea Sp. barley Potato Sp. barley Sugar beet Sp. barley Average

1 2 3 4 5 6

66 39 56 44 17 49 45

2p 1p 60 69 21 78 38

17p 3p 56 56 19 66 36

1p 3p 65 16p 26 31p 24

47 53 44 54 38 32 45

14p 0p 10 59 55 49 31

41p 18p 15 52 45 40 35

6p 1p 14 47p 25 32p 21

Values given are leached N (kg N ha21 year21). p A catch crop was grown after this main crop in the rotation. a Moderate and high precipitation was 661 and 991 mm year21 respectively. b With low and high N input regime 35 and 106 kg N ha21 year21 was added respectively.

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Catch crop scenario (freq.)

Leaching depth Catch crop scenario (freq.) Precipitation

Moderate High

100 cm N input

Low High Low High

250 cm

Nil (0%)

Deep rooted (33%)

Shallow rooted (33%)

Combination (66%)

Nil (0%)

Deep rooted (33%)

Shallow rooted (33%)

Combination (66%)

13 23 6 9

8 18 4 7

9 23 4 8

5 10 2 4

13 24 6 9

5 12 4 7

9 23 4 9

4 6 2 4

Values given are soil solution nitrate N concentration (average mg NO3-N l21).

CATCH CROPS AND GREEN MANURES

Table V Simulated Average Soil Solution Nitrate-N Concentration (mg NO3-N l21) Over Time at 1.0 and 2.5 m Depth (Average Over Three Rotations) as a Function of Catch Crop Rooting Depth and Frequency in the Rotation, Climate and N Input. The Recommended Limit for Nitrate in Drinking Water is 11.3 mg Nitrate-N l21 (50 mg nitrate l21), Figures Exceeding This has been Marked in Bold in the Table

285

286

Water balance Actual evapotranspiration (mm year21)

Precipitation

Catch crop scenario (freq.)

Moderate

Nil (0%)

Deep rooted (33%)

Shallow rooted (33%)

Combination (66%)

Nil (0%)

Deep rooted (33%)

Shallow rooted (33%)

Combination (66%)

461

471 2% 488 2%

476 3% 497 3%

493 7% 515 7%

197

186 26% 502 21%

182 28% 493 23%

159 220% 476 27%

Diff.a High

481 Diff.a

a

Difference to Nil treatment in %.

Percolation (at 250 cm; mm year21)

509

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Table VI Simulated Differences in Water Balance Components (Actual Evapotranspiration and Percolation at 2.5 m) as Affected by Catch Crop Rooting Depth, Frequency in the Rotation, N Input and Precipitation Regime. Figures are Averages Over High and Low N Input (Only Small Differences)

CATCH CROPS AND GREEN MANURES

287

concentration in half the years, but increases the concentration in the subsequent years, where no catch crop is grown. This again underlines that by incorporating catch crops in rotations, years without catch crops can become more vulnerable. In the high precipitation regime increased leaching is often seen already in the next year, whereas in the moderate precipitation regime the effect of catch crops continue for two or more years. Together these results indicate that different catch crop strategies should be used depending on precipitation regime. Furthermore, the inclusion of catch crops affects the water balance by increasing evapotranspiration and thereby decreasing the percolation. The relative effect on percolate quantity is more dramatic in moderate precipitation regime, where percolate volume is reduced by up to 20% compared with 7% in the high precipitation regime (Table VI). This means, that in order to decrease percolate nitrate concentrations, the catch crop has to be even more efficient. An important question can then be raised concerning the N retained from leaching by the catch crops: How efficiently is it then used for increased crop production? The results from the simulations show that the catch crops reduce annual N leaching by 7 –19 kg N ha21 in the low input and 9 –27 kg N ha21 in the high N input scenarios (Table IV). The increase in harvested N was as little as 0– 7 kg N ha21 (data not shown) and thus the utilisation of the retained N is low, ranging from 0 to 50%, with the highest utilisation generally found in the low N scenarios, confirming the conclusions in the literature reported in Section III.F.3 “Second year and longer term effect.” What happens to the remainder? The only important component of the N balance remaining is the change in soil organic N stocks. Figure 14 shows a clear relationship between the amount of N accumulated by the catch crops and the increase in soil organic matter N over the nil treatment (which has a continuous annual decrease in soil organic N of 0.14 and 0.09% in the low and high N input scenarios, respectively). From Fig. 14 it seems that a steady state in soil organic matter is not approaching within this 24 year period, more likely it will take many decades or even a century before a new near equilibrium is reached. In this example catch crops are grown in 33 or 66% of the years. These scenarios in most cases exceed what is common practice with respect to fraction of catch crops in the rotation. In Denmark, farmers are only required by legislation to grow catch crops on 6% of the arable area; from the above it is evident that such a low proportion is very unlikely to have any significant effect on soil C and N storage in the medium to long-term. The output from the simulations illustrate how catch crops can contribute to solving problems with N leaching losses from agriculture, by reducing the amount of N lost as well as the nitrate concentration in the water lost from the soil. The simulations also illustrate how different catch crop strategies may be optimal depending on the soil and climate conditions, as illustrated by the differences between the climate scenarios.

288

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

CATCH CROPS AND GREEN MANURES

289

F. PLACING CATCH CROPS AT THE “POLICY SCALE” Catch crops should be encouraged in areas where nitrate leaching affects sensitive aquifers. The use of catch crops is just one of several possible ways to try to reduce the nitrate load of the environment, and it should be compared to other methods to find the most cost effective way of reducing losses (Gustafson et al., 2000; Hasler, 1998; Lu et al., 1999). Few studies have attempted to compare the effectiveness of different available methods, but Gustafson et al. (2000) compared reduced fertilisation with catch crops as methods to reduce leaching losses. In their comparison catch crops were clearly the most effective method. Catch crops reduced nitrate leaching slightly more than when no fertiliser N was added to the crops, but with much less effect on main crop yield. Catch crops can also be the part of a more complex approach to reduction of nitrate leaching losses. Hasler (1998) suggested levies on chemical N fertiliser use, combined with a demand for using catch crops if the farmers shift to certain crops such as grain legumes, which have a lower N demand but which can still lead to relatively high N leaching losses. The problems can be nitrate contamination of drinking water resources or eutrofication of streams, lakes or coastal waters. It is often assumed that the worst problems are found where the highest amounts of nitrate is leached, but that is not always so. Especially when considering water resources for human use, the nitrate concentration rather than the amount of nitrate is important. This means that in dry areas where water percolation is limited, even low amounts of nitrate lost can lead to unacceptably high nitrate concentrations in the aquifers. As discussed below, such situations may require different catch crop strategies compared to situations where larger absolute amounts of nitrate leaching is to be handled. Apart from placing catch crops in areas affecting sensitive aquifers, it can be relevant to try to place the catch crops at specific types of farms. Typically, the largest nitrate leaching losses occur from farms with intensive animal husbandry (Hall et al., 2001) or intensive horticulture (Jackson et al., 1993). Catch crops grown at such farms will prevent much more leaching loss than catch crops grown at arable farms with cereal rotations. Also within the farm it is important where the catch crops are grown. Depending on previous crop and cropping history, and on the type of crop that is to be grown in the next season, the effect will be very different as discussed in chapter III “Catch Crop Effect on N Supply for Succeeding Crops.” It may be difficult to make regulations that go into this sort of detail on within-farm management of catch cropping. Still, regulations on factors as latest time of

Figure 13 Simulated nitrate-N conc. (mg NO3-N l21) over time at 1.0 and 2.5 m depth (average over three crop rotations) as a function of catch crop rooting depth, climate and N input. The recommended limit for nitrate in drinking water is 11.3 mg nitrate-N l21 (50 mg nitrate l21).

290

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Figure 14 Simulated change in soil organic matter N content to 1 m depth over the four rotations (4 £ 6 ¼ 24y), as affected by catch crop rooting depth, frequency and N input regime. Data from the low precipitation regime (see Text box 1).

establishment, earliest date of incorporation and of species choice can be made as it is already the case with, e.g., Danish regulations on catch cropping. A special case is the drier areas where nitrate leaching becomes a problem because of a low volume of drainage water rather than because of high amounts of nitrate lost by leaching. Low precipitation areas are exactly the areas where catch crops will normally cause strong pre-emptive competition, as much of the N they take up would not otherwise have been lost by leaching. This will often lead to a negative Neff. If this further leads to depressed crop yields catch crops may become a very expensive way to control the nitrate problem. If the negative Neff leads to increased N fertiliser use, it is uncertain whether any environmental effect is achieved in the long-term. Still, catch crops may be a valuable part of the solution also under such conditions, if used correctly. With low water percolation, the nitrate will only move slowly downwards in the soil profile, thus there is longer time to catch it before it leaves the root zone, especially if deep rooted catch crops are grown. This offers the possibility of growing catch crops at some year’s interval to take up the N left by the crops during the previous years. Such a possibility is also indicated by the model simulations presented in Section “Model simulation of catch crops in the crop rotation,” where deep rooted catch crops in 33% of the years had almost the same effect on N leaching loss as 66% catch crops in the low precipitation scenario.

CATCH CROPS AND GREEN MANURES

291

Using deep rooted catch crops grown at some years interval to “clean up” after the previous years nitrate loss will induce much less pre-emptive competition and negative agricultural effects over the rotation over the rotation than if they are grown more often. Further, if the catch crops can then take up substantial amounts of nitrate from deeper soil layers, this will involve little pre-emptive competition and the catch crops may lead to positive Neff even under such relatively dry conditions.

VI. PERSPECTIVES Catch crops can be one of the important tools in trying to reduce nitrate leaching losses from agriculture. They also have the potential to supply significant amounts of available N to the main crops, which can be valuable especially in low input agriculture such as organic farming. Placing catch crops in the crop rotation, selection of species or species mixtures and other aspects of catch crop management will have large effects on both environmental and agronomic effects of catch crops. A main conclusion is that the solution is not just “grow catch crops” but that catch cropping should be adapted to the relevant soil, climate and cropping situation to make sure that the desired effects are achieved. Considering the large differences in Neff and other characteristics found among the relatively few plant species which have been tested, it is highly likely that some of the many other species which could be used will offer important advantages as catch crops. Thus, looking for new species may be one of the most important lines of research in order to improve their practical applicability both agronomically and environmentally. At the same time, studying the different effects of catch crop species, and the mechanisms behind them, may be one of the most promising ways to improve our understanding of the basic effects of catch crops, and thereby to improve the results of catch cropping on the long-term. Apart from the knowledge we have about catch crop effects on N losses and Neff, there is still a number of other effects of which we presently know little, and which may differ strongly among catch crop species. Factors such as the degree and timing of winter dormancy, and plant growth reactions to N limitations may be important. N limitation may affect plant production, plant quality, or both, and thereby determine Neff. These and many other factors may affect catch crop Neff and may be the reasons for the differences observed among catch crop species, but many other “non-N” effects may also vary strongly among plant species. A central conclusion must also be that it is important that the N effect of catch crops is utilised by subsequent crops. In low input systems where the crops are more or less N limited this may not be a problem. In high input systems where the

292

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

crops are already optimally supplied with fertiliser N, the effect of a catch crop may lead to excessive supply and thereby to increased losses at a later time. This will to some extent reduce the overall effect of the catch crops on N leaching loss compared to the reduced leaching during the period when the catch crop was actually grown. Methods to predict how much the fertiliser application to the main crops can be reduced will therefore be important. Apart from the effects on N losses and N availability, catch crops can have a number of other effects both beneficial and potentially negative effects. Catch crops may lead to increased problems with weeds, pests and diseases. However, it seems that most of these problems can be handled relatively easily; as an example, disease problems can probably be handled through the choice of catch crop species. Beneficial effects can be achieved on many areas, regarding weeds, pests and diseases, but also on soil structure, erosion control, soil organic matter content, mycorrhiza and effects on other plant nutrients. The scientific literature on these subjects is still scattered, and few conclusions can be drawn, but generally the results indicate that catch crops can be used actively to handle several problems apart from the desired effects on N. The new understanding of catch crop effects gained through recent research lead to a more detailed understanding of when catch crops can be a useful tool for environmental protection and farm N management. It has also increased our understanding of the mechanisms of catch crop effects, and to manage catch crops to achieve the optimal results. Continued research on the mechanisms of catch crop effects, and on the reasons for the variable results are likely to lead to further progress, as several of the new but important topics have only been subjected to a limited research effort.

REFERENCES Abawi, G. S., and Widmer, T. L. (2000). Impact of soil health management practices on soilborne pathogens, nematodes and root diseases of vegetable crops. Appl. Soil Ecol. 15, 37–47. Abrahamsen, P., and Hansen, S. (2000). Daisy: an open soil–crop–atmosphere system model. Environ. Modell. Soft. 15, 313 –330. Al-Khatib, K., Libbey, C., and Boydston, R. (1997). Weed suppression with Brassica green manure crops in green pea. Weed Sci. 45, 439–445. Andersen, A., and Olsen, C. C. (1993). Rye grass as a catch crop in spring barley. Acta Agric. Scand., Sect. B—Soil Plant Sci. 43, 218–230. Andersen, M. K., and Jensen, L. S. (2001). Low soil temperature effects on short-term gross N mineralisation–immobilisation turnover after incorporation of a green manure. Soil Biol. Biochem. 33, 511–521. Aufhammer, W., Stu¨tzel, H., Ku¨bler, E., and Figenbaum, A. (1992). Interplanting of non-legumes into faba bean (Vicia faba L.) to reduce the risk of nitrate contamination of the ground water. Eur. J. Agron. 1, 59–69. Aulakh, M. S., Doran, J. W., Walters, D. T., and Power, J. F. (1991). Legume residue and soil water effects on denitrification in soils of different textures. Soil Biol. Biochem. 23, 1161–1167.

CATCH CROPS AND GREEN MANURES

293

Axelsen, J. A., and Kristensen, K. T. (2001). Collembola and mites in plots fertilised with different types if green manure. Pedobiologia 44, 556–566. Baggs, E. M., Watson, C. A., and Rees, R. M. (2000). The fate of nitrogen from incorporated cover crop and green manure residues. Nutr. Cycl. Agroecosyst. 56, 153 –163. Ballcoelho, B. R., and Roy, R. C. (1997). Overseeding rye into corn reduces NO3 leaching and increases yields. Can. J. Soil Sci. 77, 443– 451. Baltruschat, H., and Dehne, H. W. (1989). The occurrence of vesicular–arbuscular mycorrhiza in agro-ecosystems. II. Influence of nitrogen fertilization and green manure in continuous monoculture and in crop rotation on the inoculum potential of winter barley. Plant Soil 113, 251–256. Biederbeck, V. O., Campbell, C. A., Rasiah, V., Zentner, R. P., and Wen, G. (1998). Soil quality attributes as influenced by annual legumes used as green manure. Soil Biol. Biochem. 30, 1177–1185. Bollero, G. A., and Bullock, D. G. (1994). Cover cropping systems for the central corn belt. J. Prod. Agric. 7, 55 –58. Boswell, E. P., Koide, R. T., Shumway, D. L., and Addy, H. D. (1998). Winter wheat cover cropping, VA mycorrhizal fungi and maize growth and yield. Agric. Ecosyst. Environ. 67, 55–65. Bottomley, P. J., Sawyer, T. E., Boersma, L., Dick, R. P., and Hemphill, D. D. (1999). Winter cover crop enhances 2,4-D mineralization potential of surface and subsurface soil. Soil Biol. Biochem. 31, 849–857. Boydston, R. A., and Hang, A. (1995). Rapeseed (brassica napus) green manure crop suppresses weeds in potato (solanum tuberosum), Weed Technology 9, 669 –675. Bødker, L., and Thorup-Kristensen, K. (1999). Effect of green manure crops on root rot and arbuscular mycorrhizal fungi in pea roots. DARCORF Report no. 1. In “Designing and testing crop rotations for organic farming”. (J. E. Olesen, R. Eltun, M. J. Goulding, E. S. Jensen and U. Ko¨pke, Eds.), pp. 337–345. Breland, T. A. (1994a). Enhanced mineralization and denitrification as a result of heterogeneous distribution of clover residues in soil. Plant Soil 166, 1–12. Breland, T. A. (1994b). Measured and predicted mineralization of clover green manure at low temperatures at different depths in two soils. Plant Soil 166, 13–20. Breland, T. A. (1996a). Green manuring with clover and ryegrass catch crops undersown in small grains—effects on soil mineral nitrogen in field and laboratory experiments. Acta Agric. Scand., Sect. B–Soil Plant Sci. 46, 178–185. Breland, T. A. (1996b). Green manuring with clover and ryegrass catch crops undersown in small grains: crop development and yields. Acta Agric. Scand., Sect. B—Soil Plant Sci. 46, 30 –40. Bulson, H. A. J., Welsh, J. P., Stopes, C. E., and Woodward, L. (1996). In “Stockless organic farming in the UK: potential and limits, New Research in Organic Agriculture [Proceedings from the 11th International Scientific IFOAM Conference]”, (N. H. Kristensen and H. Høeg-Jensen, Eds.), pp. 17 –22. Burgos, N. R., Talbert, R. E., and Mattice, J. D. (1999). Cultivar and age differences in the production of allelochemicals by Secale cereale. Weed Sci. 47, 481– 485. Burns, I. G. (1984). Estimation of annual variations in leaching of residual nitrate under deep-rooted crops during winter. In “The Nitrogen Requirements of Cereals, MAFF/ADAS Reference book 385”, pp. 95–101. Her Majesty’s Stationary Office, London. Clark, A. J., Decker, A. M., and Meisinger, J. J. (1994). Seeding rate and kill date effects on hairy vetch –cereal rye cover crop mixtures for corn production. Agron. J. 86, 1065–1070. Clark, A. J., Decker, A. M., Meisinger, J. J., and Mcintosh, M. S. (1997a). Kill date of vetch, rye, and a vetch –rye mixture. 1. Cover crop and corn nitrogen. Agron. J. 89, 427 –434. Clark, A. J., Decker, A. M., Meisinger, J. J., and Mcintosh, M. S. (1997b). Kill date of vetch, rye, and a vetch –rye mixture. II. Soil moisture and corn yield. Agron. J. 89, 434 –441. Clarkson, D. T., Jones, L. H. P., and Purves, J. V. (1992). Absorption of nitrate and ammonium ions by Lolium perenne from flowing solution cultures at low root temperatures. Plant Cell Environ. 15, 99–106.

294

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Creamer, N. G., Bennett, M. A., Stinner, B. R., Cardina, J., and Regnier, E. E. (1996). Mechanisms of weed suppression in cover crop-based production systems. HortScience 31, 410–413. Cunningham, P. C. (1985). Characteristics of Trends in Disease Caused by Soilborne Pathogens with Spring Barely Monoculture. (C.A. Parker, A.D. Rovira, K. J. Moore, P.T.W. Wong, editors, In “Ecol. Manage. Soilborne Plant Pathogens” St. Paul, Minnesota, USA pp. 9–13. Dabney, S. M., Schreiber, J. D., Rothrock, C. S., and Johnson, J. R. (1996). Cover crops affect sorghum seedling growth. Agron. J. 88, 961– 970. Davis, J. R., Huisman, O. C., Westermann, D. T., Hafez, S. L., Everson, D. O., Sorensen, L. H., and Schneider, A. T. (1996). Effects of green manures on verticillium wilt of potato. Phytopathology 86, 444 –453. Debosz, K., Rasmussen, P. H., and Pedersen, A. R. (1999). Temporal variations in microbial biomass C and cellulolytic enzyme activity in arable soils: effects of organic matter input. Appl. Soil Ecol. 13, 209 –218. Decker, A. M., Clark, A. J., Meisinger, J. J., Mulford, F. R., and Mcintosh, M. S. (1994). Legume cover crop contributions to no-tillage corn production. Agron. J. 86, 126–135. Dick, R. P., and Christ, R. A. (1995). Effects of long-term residue management and nitrogen fertilization on availability and profile distribution of nitrogen. Soil Sci. 159, 402–408. Diekkru¨ger, B., So¨ndergarth, D., Kersebaum, K. C., and McVoy, C. W. (1995). Validity of agroecosystem models: a comparison of results of different models applied to the same data set. Ecol. Model. 81, 3 –29. Ditsch, D. C., and Alley, M. M. (1991). Nonleguminous cover crop management for residual N recovery and subsequent crop yields. J. Fert. Effects 8, 6–13. Dou, Z., Fox, R. H., and Toth, J. D. (1994). Tillage effect on seasonal nitrogen availability in corn supplied with legume green manures. Plant Soil 162, 203–210. Elers, B., and Hartmann, H. D. (1987). Biologische Konservierung von Nitrat. Gemu¨se 4, 210– 214. Eriksen, J., and Thorup-Kristensen, K. (2002). The effect of catch crops on sulphate leaching and availability of S in the succeeding crop. Agric. Ecosyst. Environ. 90, 247–254. Esala, M. J. (1995). Changes in the extractable ammonium- and nitrate-nitrogen contents of soil samples during freezing and thawing. Commun. Soil Sci. Plant Anal. 26, 61–68. Filser, J. (1995). The effect of green manure on the distribution of collembola in a permanent row crop. Biol. Fert. Soil. 19, 303 –308. Folorunso, O. A., Rolston, D. E., Prichard, T., and Louie, D. T. (1992). Soil surface strength and infiltration rate as affected by winter cover crops. Soil Technol. 5, 189–197. Fox, R. H., Myers, R. J. K., and Vallis, I. (1990). The nitrogen mineralization rate of legume residues in soil as influenced by their polyphenol, lignin, and nitrogen contents. Plant Soil 129, 251– 259. Francis, G. S. (1995). Management practices for minimising nitrate leaching after ploughing temporary leguminous pastures in Canterbury, New Zealand. J. Contam. Hydrol. 20, 313–327. Francis, G. S., Bartley, K. M., and Tabley, F. J. (1998). The effect of winter cover crop management on nitrate leaching losses and crop growth. J. Agric. Sci., Cambridge 131, 299 –308. Francis, G. S., Haynes, R. J., and Williams, P. H. (1995). Effects of the timing of ploughing-in temporary leguminous pastures and two winter cover crops on nitrogen mineralization, nitrate leaching and spring wheat growth. J. Agric. Sci. 124, 1 –9. Frankenberger, W. T., and Abdelmagid, H. M. (1985). Kinetic parameters of nitrogen mineralization rates of leguminous crops incorporated into soil. Plant Soil 87, 257– 271. Franzleubbers, K., Weaver, A. S., Juo, A. S. R., and Franzleubbers, A. J. (1994a). Carbon and nitrogen mineralization from cowpea plant parts decomposing in moist and repeatedly dried and wetted soil. Soil Biol. Biochem. 26, 1379–1387. Franzleubbers, K., Weaver, A. S., Juo, A. S. R., and Franzleubbers, A. J. (1994b). Carbon and nitrogen mineralization from cowpea plant parts decomposing in moist and repeatedly dried and wetted soil. Soil Biol. Biochem. 26, 1379–1387.

CATCH CROPS AND GREEN MANURES

295

Gahoonia, T. S., Asmar, F., Gise, H., Gissel-Nielsen, G., and Nielsen, N. E. (2000). Root-released organic acids and phosphorus uptake of two barley cultivars in laboratory and field experiments. Eur. J. Agron. 12, 281–289. Gahoonia, T. S., and Nielsen, N. E. (1997). Variation in root hairs of barley cultivars doubled soil phosphorus uptake. Euphytica 98, 177–182. Garwood, T. W. D., Davies, D. B., and Hartley, A. R. (1999). The effect of winter cover crops on yield of the following spring crops and nitrogen balance in a calcareous loam. J. Agric. Sci., Cambridge 132, 1–11. Grindlay, D. J. C (1995). Rooting depth in cover crops and its relation to shoot size. From PhD thesis: Principles governing the ability of cover crop species to trap nitrate. 230 –278, University of Nottingham. Gustafson, A., Fleischer, S., and Joelsson, A. (2000). A catchment-oriented and cost-effective policy for water protection. Ecol. Engng. 14, 419–427. Hall, M. D., Shaffer, M. J., Waskom, R. M., and Delgado, J. A. (2001). Regional nitrate leaching variability: What makes a difference in northeastern Colorado. J. Am. Water Resour. Assoc. 37, 139 –150. Hansen, E. M., and Djurhuus, J. (1997). Yield and N uptake as affected by soil tillage and catch crop. Soil Till. Res. 42, 241– 252. Hansen, E. M., Djurhuus, J., and Dick, W. A. (1997). Nitrate leaching as influenced by soil tillage and catch crop. Soil Biol. Till. 41, 203– 219. Hansen, E. M., Djurhuus, J., and Kristensen, K. (2000). Nitrate leaching as affected by introduction or discontinuation of cover crop use. J. Environ. Qual. 29, 1110–1116. Hansen, S., Jensen, H. E., Nielsen, N. E., and Svendsen, H. (1991). Simulation of nitrogen dynamics and biomass production in winter wheat using the Danish simulation model DAISY. Fert. Res. 27, 245–259. Hansen, S., Thirup, C., Refsgaard, J. C., and Jensen, L. S. (2001). Modelling of nitrate leaching at different scales—application of the Daisy model. In “Modeling Carbon and Nitrogen Dynamics for Soil Management”. (M. Shaffer, M. Liwang and S. Hansen, Eds.), pp. 511–547. Lewis Publishers, Boca Raton. Harper, L. A., Hendrix, P. F., Langdale, G. W., and Coleman, D. C. (1995). Clover management to provide optimum nitrogen and soil water conservation. Crop Sci. 35, 176 –182. Hasler, B. (1998). Analysis of environmental policy measures aimed at reducing nitrogen leaching at the farm level. Environ. Pollut. 102, 749– 754. Hendrix, P. F., Coleman, D. C., and Crossley, D. A. (1992). Using Knowledge of Soil Nutrient Cycling Processes to Design Sustainable agriculture. In “Integrating sustainable agriculture, ecology, and environmental policy”. (Olson R.K.,), 63– 82.. Hermawan, B., and Bomke, A. A. (1997). Effects of winter cover crops and successive spring tillage on soil aggregation. Soil Till. Res. 44, 109 –120. Hinsinger, P. (1998). How do plant roots acquire mineral nutrients? Chemical processes involved in the rhizosphere. Adv. Soil Sci. 64, 225– 265. Hoffman, M. L., Regnier, E.E., and Cardina, J., (1993). Weed and Corn (Zea mays) Responses to a Hairy Vetch (Vicia villosa) Cover Crop, Weed Technology 7, 594–599. Hoffman, M. L., Weston, L.A., Snyder, J. C., and Regnier, E.E. (1996). Allelopathic influence of germinating seeds and seedlings of cover crops on weed species, Weed Science 44, 579–584. Ilgen, B., and Stamp, P. (1993). Root development in seedlings of oilradish, white mustard and pea. J. Agron. Crop Sci. 169, 122–127. Izaurralde, R. C., Feng, Y., Robertson, J. A., McGill, W. B., Juma, N. G., and Olson, B. M. (1995). Long-term influence of cropping systems, tillage methods, and N sources on nitrate leaching. Can. J. Soil Sci. 75, 497 –505. Jackson, L. E. (2000). Fates and losses of nitrogen from a nitrogen-15-labeled cover crop in an intensively managed vegetable system. Soil Sci. Soc. Am. J. 64, 1404–1412.

296

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Jackson, L. E., and Stivers, L. J. (1993). Root distribution of lettuce under commercial production: Implications for crop uptake of nitrogen. Biol. Agric. Hort. 9, 273–293. Jackson, L. E., Wyland, L. J., and Stivers, L. J. (1993). Winter cover crops to minimize nitrate losses in intensive lettuce production. J. Agric. Sci. 121, 55–62. Janzen, H. H., and Radder, G. D. (1989). Nitrogen mineralization in a green manure-amended soil as influenced by cropping history and subsequent crop. Plant Soil 120, 125 –131. Janzen, H. H., and Schaalje, G. B. (1992). Barley response to nitrogen and non-nutritional benefits of legume green manure. Plant Soil 142, 19–30. Ja¨ggli, F. (1978). Sickerverluste an mineralstoffen. Mitteilungen fu¨r die Schweizerische Landwirtschaft 26, 130–136. Jensen, E. S. (1991). Nitrogen accumulation and residual effects of nitrogen catch crops. Acta Agric. Scand. 41, 333–344. Jensen, E. S. (1992). The release and fate of nitrogen from catch-crop materials decomposing under field conditions. J. Soil Sci. 43, 335–345. Jensen, E. S. (1994). Mineralization immobilization of nitrogen in soil amended with low C:N ratio plant residues with different particle sizes. Soil Biol. Biochem. 26, 519–521. Jensen, L. S., Mueller, T., Bruun, S., and Hansen, S. (2001). Application of the Daisy model for short and long-term simulation of soil carbon and nitrogen dynamics. In “Modeling carbon and nitrogen dynamics for soil management”. (M. Shaffer, M. Liwang and S. Hansen, Eds.), Lewis Publishers, Boca Raton. Jensen, L. S., Mueller, T., Eriksen, J., Thorup-Kristensen, K., and Magid, J. (1999). Simulation of plant production and N fluxes in organic farming systems with the soil–plant–atmosphere model DAISY. In “Designing and Testing Crop Rotations for Organic Farming. Proceedings from an International Workshop”. (J. E. Olesen, R. Eltun, M. J. Gooding, E. S. Jensen and U. Ko¨pke, Eds.), pp. 235–247. Danish Research Centre for Organic Farming, Tjele, Denmark. Jensen, L. S., Mueller, T., Nielsen, N. E., Hansen, S., Crocker, G. J., Grace, P. R., Klir, J., Korschens, M., and Poulton, P. R. (1997). Simulating trends in soil organic carbon in long-term experiments using the soil–plant–atmosphere model DAISY. Geoderma 81, 5–28. Jensen, L. S., Stougaard, B., and Olsen, P. (1994). Simulation of nitrogen dynamics at three Danish locations by use of the DAISY model. Acta Agric. Scand., Sect. B—Soil Plant Sci. 44, 75 –83. Jones, D. L. (1998). Organic acids in the rhizosphere–a critical review. Plant Soil 205, 25–44. Kabir, Z., and Koide, R. T. (2000). The effect of dandelion or a cover crop on mycorrhiza inoculum potential, soil aggregation and yield of maize. Agric. Ecosyst. Environ. 78, 167–174. Karlsson-Strese, E. M., Rydberg, I., Becker, H. C., and Umaerus, M. (1998). Strategy for catch crop development. II. Screening of species undersown in spring barley (Hordeum vulgare L.) with respect to catch crop growth and grain yield. Acta Agric. Scand. 48, 26– 33. Karlsson-Strese, E. M., Umaerus, M., and Rydberg, I. (1996). Strategy for catch crop development. I. Hypothetical ideotype and screening of species. Acta Agric. Scand., Sect. B—Soil Plant Sci. 46, 106–111. Katterer, T., Reichstein, M., Andre´n, O., and Lomander, A. (1998). Temperature dependence of organic matter decomposition: a critical review using literature data analyzed with different models. Biol. Fert. Soil 27, 258–262. Kessavalou, A., and Walters, D. T. (1999). Winter rye cover crop following soybean under conservation tillage: Residual soil nitrate. Agron. J. 91, 643–649. Kuo, S., and Jellum, E. J. (2000). Long-term winter cover cropping effects on corn (Zea mays L.) production and soil nitrogen availability. Biol. Fert. Soil 31, 470– 477. Kuo, S., Sainju, U. M., and Jellum, E. (1996). Winter cover cropping influence on nitrogen mineralization, presidedress soil nitrate test, and corn yields. Biol. Fert. Soil 22, 310– 317. Kuo, S., Sainju, U. M., and Jellum, E. J. (1997). Winter cover crop effects on soil organic carbon and carbohydrate in soil. Soil Sci. Soc. Am. J. 61, 145–152.

CATCH CROPS AND GREEN MANURES

297

Ladd, J. N., Amato, M., Jackson, R. B., and Butler, H. A. (1983). Utilization by wheat crops of nitrogen from legume residues decomposing in soils in the field. Soil Biol. Biochem. 15, 231 –238. Ladd, J. N., Oades, J. M., and Amato, M. (1981). Distribution and recovery of nitrogen from legume residues decomposing in soils sown to wheat in the field. Soil Biol. Biochem. 13, 251 –256. Laine, P., Bigot, J., Ourry, A., and Boucaud, J. (1994). Effects of low temperature on nitrate uptake, and xylem and phloem flows of nitrogen, in Secale cereale L. and Brassica napus L. New Phytol. 127, 675–683. Laine´, P., Ourry, A., Macduff, J., Boucaud, J., and Salette, J. (1993). Kinetic parameters of nitrate uptake by different catch crop species: Effects of low temperatures or previous nitrate starvation. Physiol. Plant. 88, 85 –92. Larsson, L., Ferm, M., Kasimirklemedtsson, A., and Klemedtsson, L. (1998). Ammonia and nitrous oxide emissions from grass and alfalfa mulches. Nutr. Cycl. Agroecosyst. 51, 41 –46. Lewan, E. (1994). Effects of a catch crop on leaching of nitrogen from a sandy soil: Simulations and measurements. Plant Soil 166, 137 –152. Liebl, R., Simmons, F.W., Wax, L.M., and Stoller, E.W., (1992). Effect of Rye (Secale cereale) Mulch on Weed Control and Soil Moisture in Soybean (Glycine max), Weed Tech 6, 838 –846. Liebman, M., and Dyck, E., (1993). Crop rotation and intercropping strategies for weed management, Ecol. Appl. 3, 92 –122. Linderman, R. G. (1989). Organic amendments and soil-borne diseases. Can. J. Plant Phatol. 11, 180 –183. Lotz, L. A. P., Groeneveld, R. M. W., Theunissen, J., and Vandenbroek, R. C. F. M. (1997). Yield losses of white cabbage caused by competition with clovers grown as cover crop. Neth. J. Agric. Sci. 45, 393 –405. Lu, Y. C., Watkins, B., and Teasdale, J. (1999). Economic analysis of sustainable agricultural cropping systems for mid-Atlantic states. J. Sust. Agric. 15, 77 –93. Lupwayi, N. Z., and Haque, I. (1998). Mineralization of N, P, K, Ca and Mg from Sesbania and Leucaena leaves varying in chemical composition. Soil Biol. Biochem. 30, 337 –343. Macduff, J. H., Jarvis, S. C., and Cockburn, J. E. (1994). Acclimation of NO3—fluxes to low root temperature by brassica napus in relation to NO3—supply. J. Expl. Bot. 45, 1045–1056. Magid, J., Henriksen, O., Thorup-Kristensen, K., and Mueller, T. (2001). Decoupled C and N cycling at low temperatures—implications for green manure efficiency in temperate agro-ecosytems. Plant Soil 228, 73 –82. Magid, J., and Kølster, P. (1995). Modelling nitrogen cycling in an ecological crop rotation—an explorative trial. Biol. Agric. Hort. 11, 77– 87. Magid, J., Mueller, T., Jensen, L. S., and Nielsen, N. E. (1997). Modelling the measurable— interpretation of field-scale CO2 and N-mineralization, soil microbial biomass and light fractions as indicators of oilseed rape, maize and barley straw decomposition. In “Plant Litter Quality and Decomposition”. (G. Cadisch and K. E. Giller, Eds.), pp. 349 –632. CAB International, Wallington, UK. Mann, H. H. (1959). Field studies in green manuring. II. Empire J. Expp. Agric. 27, 243 –251. Marstorp, H., and Kirchmann, H. (1991). Carbon and nitrogen mineralization and crop uptake of nitrogen from six green manure legumes decomposing in soil. Acta Agric. Scand. 41, 243 –252. Martinez, J., and Guiraud, G. (1990). A lysimeter study of the effects of a ryegrass catch crop, during a winter wheat/maize rotation, on nitrate leaching and on the following crop. J. Soil Sci. 41, 5 –16. McBride, R. G., Mikkelsen, R. L., and Barker, K. R. (1999). Survival and infection of root-knot nematodes added to soil amended with rye at different stages of decomposition and cropped with cotton. Appl. Soil Ecol. 13, 231–235.

298

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

McCracken, D. V., Corak, S. J., Frye, W. W., and Blevins, R. L. (1989). Residual effects of nitrogen fertilization and winter cover cropping on nitrogen availability. Soil. Sci. Soc. Am. J. 53, 1459–1464. McGuire, A. M., Bryant, D. C., and Denison, R. F. (1998). Wheat yields nitrogen uptake, and soil moisture following winter legume cover crop vs. fallow. Agron. J. 90, 404– 410. Mendes, I. C., Bandick, A. K., Dick, R. P., and Bottomley, P. J. (1999). Microbial biomass and activities in soil aggregates affected by winter cover crops. Soil Sci. Soc. Am. J. 63, 873–881. Meyer, L. D., Dabney, S. M., Murphree, C. E., Harmon, W. C., and Grissinger, E. H. (1999). Crop production systems to control erosion and reduce runoff from upland silty soils. Trans. ASAE 42, 1645–1652. Mitchell, J. P., Peters, D. W., and Shennan, C. (1999). Changes in soil water storage in winter fallowed and cover cropped soils. J. Sust. Agric. 15, 19–31. Moore, M.J., Gillespie, T.J., and Swanton, C.J., (1994). Effect of cover crop mulches on weed emergence, weed biomass, and soybean (glycine max) development, Weed Tech. 8, 512–518. Muehlchen, A. M., Rand, R. E., and Parke, J. L. (1990). Evaluation of crucifer green manures for controlling aphanomyces root rot of peas. Plant Dis. 74, 651– 654. Mueller, T., Jensen, L. S., Magid, J., and Nielsen, N. E. (1997). Temporal variation of C and N turnover in soil after oilseed rape straw incorporation in the field: Simulations with the soil– plant–atmosphere model DAISY. Ecol. Model. 99, 247–262. Mueller, T., Jensen, L. S., Magid, J., and Nielsen, N. E. (1998). Turnover of carbon and nitrogen in a sandy loam soil after incorporation of chopped maize, barley straw and blue grass in the field. Soil Biol. Biochem. 30, 561–571. Mueller, T., Magid, J., Jensen, L. S., and Nielsen, N. E. (1998). Soil C and N turnover after the incorporation of chopped maize, barley straw and blue grass in the field: Evaluation of the DAISY soil-organic-matter submodule. Ecol. Model. 111, 1–15. Mueller, T., and Thorup-Kristensen, K. (2001). N-fixation of selected green manure plants in an organic crop rotation. Biol. Agric. Hort. 18, 345– 363. Muller, J. C.D., Denys, D., Morlet, G., and Mariotti, A. (1989). Influence of catch crops on mineral nitrogen leaching and its subsequent plant use, pp. 85–98. Elsevier, London, New York. Mu¨ller-Scha¨rer, H. (1996). Interplanting ryegrass in winter leek: Effect on weed control, crop yield and allocation of N-fertilizer. Crop Prot. 15, 641–648. Mu¨ller, M. M., and Sundman, V. (1988). The fate of nitrogen (15N) released from different plant materials during decomposition under field conditions. Plant Soil 105, 133 –139. Neergaard, T. A. F. D., Nielsen, H. H., Jensen, L. S., and Magid, J. (2002). Decomposition of white clover (Trifolium repens ) and ryegrass (Lolium perenne ) components: C and N dynamics simulated with the DAISY soil organic mattter submodule. Eur. J. Agron. 16, 43–55. Nicolardot, B., Fauvet, G., and Cheneby, D. (1994). Carbon and nitrogen cycling through soil microbial biomass at various temperatures. Soil Biol. Biochem. 26, 253–261. Olesen, J. E., Askegaard, M., and Rasmussen, I. (2000). Design of an organic farming crop rotation experiment. Acta Agric. Scand., Sect. B—Soil Plant Sci. 50, 13–21. Parton, W. J., Stewart, J. W.B., and Cole, C. V. (1988). Dynamics of C, N, P and S in grassland soils— a model. Biogeochemistry 5, 109 –131. Paustian, K., Parton, W. J., and Persson, J. (1992). Modeling soil organic matter in organic-amended and nitrogen-fertilized long-term plots. Soil Sci. Soc. Am. J. 56, 476–488. Pound, B., Anderson, S., and Gundel, S. (1999). Species for niches: When and for whom are cover crops appropriate? Mount. Res. Developp. 19, 307– 312. Power, J. F., and Zachariassen, J. A. (1993). Relative nitrogen utilization by legume cover crop species at three soil temperatures. Agron. J. 85, 134–140. Quemada, M., and Cabrera, M. L. (1995). Carbon and nitrogen mineralized from leaves and stems of four cover crops. Soil Sci. Soc. Am. J. 59, 471 –477.

CATCH CROPS AND GREEN MANURES

299

Quemada, M., and Cabrera, M. L. (1996). Temperature and moisture effects on C and N mineralization from surface applied clover residue. Plant Soil 189, 127–137. Ranells, N. N., and Wagger, M. G. (1996). Nitrogen release from grass and legume cover crop monocultures and bicultures. Agron. J. 88, 777– 782. Ranells, N. N., and Wagger, M. G. (1997). Grass –legume bicultures as winter annual cover crops. Agron. J. 89, 659 –665. Ranells, N. N., and Wagger, M. G. (1997). Nitrogen-15 recovery and release by rye and crimson clover cover crops. Soil Sci. Soc. Am. J. 61, 943–948. Ranells, N. N., and Wagger, M. G. (1997). Winter annual grass–legume bicultures for efficient nitrogen management in no-till corn. Agric. Ecosyst. Environ. 65, 23– 32. Rask, N., Pedersen, S. E., and Jensen, M. H. (2002). Response to lowered nutrient discharges in the coastal waters around the island of Funen. Hydrobiologia 393, 69–81. Richards, I.R., Wallace, P.A., and Turner, I.D.S., (1996). A comparison of six cover crop types in terms of nitrogen uptake and effect on response to nitrogen by a subsequent spring barley crop. J. Agric Sci 127, 441–449. Renius, W., and Entrup, E. L. (2002). In “Zwischenfruchtbau Zur Futtergewinnung und Gru¨ndu¨ngung”, pp. 206. DLF Verlag, Frankfurt/Main. Robinson, D., Linehan, D. J., and Caul, S. (1991). What limits nitrate uptake from soil? Plant, Cell Environ. 14, 77–85. Robinson, D., Linehan, D. J., and Gordon, D. C. (1994). Capture of nitrate from soil by wheat in relation to root length, nitrogen inflow and availability. New Phytol. 128, 297–305. Rosecrance, R. C., McCarty, G. W., Shelton, D. R., and Teasdale, J. R. (2000). Denitrification and N mineralization from hairy vetch (Vicia villosa Roth) and rye (Secale cereale L.) cover crop monocultures and bicultures. Plant Soil 227, 283–290. Sainju, U. M., and Singh, B. P. (1997). Winter cover crops for sustainable agricultural systems—influence on soil properties, water quality, and crop yields [review]. HortScience 32, 21 –28. Sainju, U. M., Singh, B. P., Rahman, S., and Reddy, V. R. (1999). Soil nitrate-nitrogen under tomato following tillage, cover cropping, and nitrogen fertilization. J. Environ. Qual. 28, 1837–1844. Sainju, U. M., Singh, B. P., and Whitehead, W. F. (1998). Cover crop root distribution and its effects on soil nitrogen cycling. Agron. J. 90, 511–518. Sanderson, J. B., MacLeod, J. A., and Kimpinski, J. (1999). Glyphosate application and timing of tillage of red clover affects potato response to N, soil N profile, and root and soil nematodes. Can. J. Soil Sci. 79, 65–72. Scholte, K., and Lootsma, M. (1998). Effect of farmyard manure and green manure crops on populations of mychophagous soil fauna and Rhizoctonia stem cracker of potato, Pedobiologia. 42, 223–231. Schro¨der, J. J., Groot, W. M., and Dijk, W. (1992). Nitrogen losses from continuous maize as affected by cover crops. Aspects Appl. Biol. 30, 317–326. Schro¨der, J. J., Holte, L. T., Janssen, B. H., and Ten-Holte, L. (1997). Non-overwintering cover crops: a significant source of N. Neth. J. Agric. Sci. 45, 231 –248. Schro¨der, J. J., Vandijk, W., and Degroot, W. J.M. (1996). Effects of cover crops on the nitrogen fluxes in a silage maize production system. Neth. J. Agric. Sci. 44, 293– 315. Scott, H. D., Keisling, T. C., Waddle, B. A., Williams, R. W., and Frans, R. E. (1919). Effects of winter cover crops on yield of cotton and soil properties. Arkansas Expp. Station Bull. 924, 1 –21. Shepherd, M. A. (1993). Measurement of soil mineral nitrogen to predict the response of winter wheat to fertilizer nitrogen after applications of organic manures or after ploughed-out grass. J. Agric. Sci. Camb. 121, 223–231. Shepherd, M. A. (1999). The effectiveness of cover crops during eight years of a UK sandland rotation. Soil Use Manag. 15, 41 –48.

300

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Shepherd, M. A., Davies, D. B., and Johnson, P. A. (1993). Minimizing nitrate losses from arable soils. Soil Use Manag. 9, 94–99. Shipley, P. R., Meisinger, J. J., and Decker, A. M. (1992). Conserving residual corn fertilizer nitrogen with winter cover crops. Agron. J. 84, 869–876. Smith, P., Smith, J. U., Powlson, D. S., McGill, W. B., Arah, J. R.M., Chertov, O. G., Coleman, K., Franco, U., Frolking, S., Jenkinson, D. S., Jensen, L. S., Kelly, R. H., KleinGunneweik, H., Komarov, A. S., Li, C., Molina, J. A.E., Mueller, T., Parton, W. J., Thornley, J. H.M., and Whitmore, A. J. (1997). A comparison of the performance of nine soil organic matter models using datasets from seven long-term experiments. Geoderma 81, 153–225. Sørensen, J. N. (1992). Effects of catch crops on the content of soil mineral nitrogen before and after winter leaching. Z. Pflanzenerna¨hr. Bodenk. 155, 61–66. Sørensen, J. N., and Thorup-Kristensen, K. (1993). Nitrogen effects of non-legume catch crops. Z. Pflanzenerna¨hr. Bodenk. 156, 55–59. Sørensen, P., and Jensen, E. S. (1996). 15N-labelling of animal manure: problems, solutions and use. In “Trans. 9th Nitrogen Workshop”, pp. 67 – 70. Technische Universitet Braunschweig, Braunschweig, Germany. Sørensen, P., and Jensen, E. S. (1998). The use of N-15 labelling to study the turnover and utilization of ruminant manure N. Biol. Fert. Soil 28, 56–63. Sørensen, P., Jensen, E. S., and Nielsen, N. E. (1994). The fate of N-15 labeled organic nitrogen in sheep manure applied to soils of different texture under field conditions. Plant Soil 162, 39 –47. Strizaker, R.J., and Bunn, D.G. (1996). Phytotoxicity of ryegrass and clover cover crops, and a lucerne alley crop for no-till vegetable production, Biol. Agric. Hort. 13, 83–101. Stirzaker, R. J., and White, I. (1995). Amelioration of soil compaction by a cover-crop for no-tillage lettuce production. Aust. J. Agric. Res. 46, 553– 568. Stute, J. K., and Posner, J. L. (1995). Legume cover crops as a nitrogen source for corn in an oat corn rotation. J. Prod. Agric. 8, 385–390. Stute, J. K., and Posner, J. L. (1995). Synchrony between legume nitrogen release and corn demand in the upper midwest. Agron. J. 87, 1063–1069. Sumner, D. R., Phatak, S. C., Gay, J. D., Chalfant, R. B., Brunson, K. E., and Bugg, R. L. (1995). Soilborne pathogens in a vegetable double-crop with conservation tillage following winter cover crops. Crop Prot. 14, 495–500. Teasdale, J.R., and Daughtry, G.S.T. (1993). Weed suppression by live and desiccated hairy vetch (Vicia villosa), Weed Sci. 41, 207–212. Teasdale, J. R., and Mohler, C. L. (1993). Light transmittance, soil temperature, and soil moisture under residue of hairy vetch and rye. Agron. J. 85, 673–680. Theunissen, J. (1995). Application of intercropping in organic agriculture. Entomological research in organic agriculture Austrian Federal Ministry of Science and Research, Vienna, March 1995. Biol. Agric. Hortic. 15, 251–259. Theunissen, J., and Schelling, G. (2000). Undersowing carrots with clover: suppression of carrot rust fly (Psila rosae ) and cavity spot (Pythium spp.) infestation. Biol. Agric. Hort. 18, 67–76. Thibaud, M., Morel, C., and Fardeau, J.-C. (1988). Contribution of phosphorus issued from crop residues to plant nutrition. Soil Sci. Plant Nutr. 34, 481 –491. Thomsen, I. K. (1993). Nitrogen uptake in barley after spring incorporation of 15N labelled Italian ryegrass into sandy soils. Plant Soil 150, 193– 201. Thomsen, I. K., and Christensen, B. T. (1999). Nitrogen conserving potential of successive ryegrass catch crops in continuous spring barley. Soil Use Manag. 15, 195 –200. Thomsen, I. K., Kjellerup, V., and Jensen, B. (1997). Crop uptake and leaching of N-15 applied in ruminant slurry with selectively labelled faeces and urine fractions. Plant Soil 197, 233–239. Thorup-Kristensen, K. (1993a). Root development of nitrogen catch crops and of a succeeding crop of broccoli. Acta Agric. Scand., Sect. B—Soil Plant Sci. 43, 58 –64.

CATCH CROPS AND GREEN MANURES

301

Thorup-Kristensen, K. (1993b). The effect of nitrogen catch crops on the nitrogen nutrition of a succeeding crop I. Effects through mineralization and pre-emptive competition. Acta Agric. Scand., Sect. B—Soil Plant Sci. 43, 74–81. Thorup-Kristensen, K. (1994). An easy pot incubation method for measuring nitrogen mineralization from easily decomposable organic material under well defined conditions. Fert. Res. 38, 239 –247. Thorup-Kristensen, K. (1994). The effect of nitrogen catch crop species on the nitrogen nutrition of succeeding crops. Fert. Res. 37, 227–234. Thorup-Kristensen, K. (1996). Effect of catch crop incorporation time on N availability for a succeeding crop. In “Long Term Reduction of Nitrate Leaching by Cover Crops, First progress report of EU concerted action (AIR3) 2108”. (J. J. Schro¨der, Ed.), pp. 49–54. Thorup-Kristensen, K. (2001). Are differences in root growth of nitrogen catch crops important for their ability to reduce soil nitrate-N content, and how can this be measured? Plant Soil 230, 185 –195. Thorup-Kristensen, K., and Nielsen, N. E. (1998). Modelling and measuring the effect of nitrogen catch crops on the nitrogen supply for succeeding crops. Plant Soil 203, 79–89. Thorup-Kristensen, K., and Sørensen, J. N. (1999). Soil nitrogen depletion by vegetable crops with variable root growth. Acta Agric. Scand., Sect. B—Soil Plant Sci. 49, 92 –97. Thorup-Kristensen, K., and Van den Boogaard, R. (1999). Vertical and horizontal development of the root system of carrots following green manure. Plant Soil 212, 145 –153. Torbert, H. A., Reeves, D. W., and Mulvaney, R. L. (1996). Winter legume cover crop benefits to corn: rotation vs fixed-nitrogen effects. Agron. J. 88, 527– 535. Torstensson, G. (1998). Nitrogen delivery and utilization by subsequent crops after incorporation of leys with different plant composition. Biol. Agric. Hort. 16, 129–143. Torstensson, G., and Aronsson, H. (2000). Nitrogen leaching and crop availability in manured catch crop systems in Sweden. Nutr. Cycl. Agroecosyst. 56, 139–152. Tu, J. C. (1988). The impact of interseason green manuring on root diseases and soil environments. Med. Fac. Landbouww. Rijksuniv. Gent 53, 2–321. Tu, J. C., and Findlay, W. I. (1986). The effect of different green manure crops and tillage practices on pea root rots. Br. Crop Protect. Conf.—Pest Dis. 3C-1, 229 –236. Van Dam, A. M., and Lantinga, E. A. (1998). Leaf CO2 assimilation and leaf dynamics in catch crops during autumn and winter at two levels of nitrogen supply. Neth. J. Agric. Sci. 46, 249–265. Van Dam, A. M., and Leffelaar, P. A. (1998). Root, soil water and nitrogen dynamics in a catch crop— soil system in the Wageningen Rhizolab. Neth. J. Agric. Sci. 46, 267– 284. Van Gestel, M., Merckx, R., and Vlassak, K. (1993). Soil drying and rewetting and the turnover of 14C-labelled plant residues: First order decay rates of biomass and non-biomass 14C. Soil Biol. Biochem. 25, 125– 134. Van scho¨ll, L., Van Dam, A. M., and Leffelaar, P. A. (1997). Mineralisation of nitrogen from an incorporated catch crop at low temperatures: experiment and simulation. Plant Soil 188, 211 –219. Vigil, M. F., and Kissel, D. E. (1995). Rate of nitrogen mineralized from incorporated crop residues as influenced by temperature. Soil. Sci. Soc. Am. J. 59, 1636–1644. Vos, J., van der Putten, P. E. L., Hussein, M. H., Van Dam, A. M., and Leffelaar, P. A. (1998). Field observations on nitrogen catch crops. II. Root length and root length distribution in relation to species and nitrogen supply. Plant Soil 201, 149–155. Vos, J., and van der Putten, P. E. L. (2000). Nutrient cycling in a cropping system with potato, spring wheat, sugar beet, oats and nitrogen catch crops. I. Input and offtake of nitrogen, phosphorus and potassium. Nutr. Cycl. Agroecosyst. 56, 87–97. Vos, J., and van der Putten, P. E. L. (1997). Field observations on nitrogen catch crops. 1. Potential and actual growth and nitrogen accumulation in relation to sewing date and crop species. Plant Soil 195, 299–309.

302

K. THORUP-KRISTENSEN, J. MAGID AND L. S. JENSEN

Vyn, T. J., Faber, J. G., Janovicek, K. J., and Beauchamp, E. G. (2000). Cover crop effects on nitrogen availability to corn following wheat. Agron. J. 92, 915 –924. Vyn, T. J., Janovicek, K. J., Miller, M. H., and Beauchamp, E. G. (1999). Soil nitrate accumulation and corn response to preceding small-grain fertilization and cover crops. Agron. J. 91, 17–24. Wagger, M. G. (1989). Time of desiccation effects on plant composition and subsequent nitrogen release from several winter annual cover crops. Agron. J. 81, 236–241. Wallgren, B., and Linden, B. (1994). Effects of catch crops and ploughing times on soil mineral nitrogen. Sweed. J. Agric. Res. 24, 67–75. Wander, M. M., Traina, S. J., Stinner, B. R., and Peters, S. E. (1994). Organic and conventional management effects on biologically active soil organic matter pools. Soil Sci. Soc. Am. J. 58, 1130–1139. Wang, J., and Bakken, L. R. (1997). Competition for nitrogen during decomposition of plant residues in soil: Effect of spatial placement of N-rich and N-poor plant residues. Soil Biol. Biochem. 29, 153–162. Whitehead, D. C., Lockyer, D. R., and Raistrick, N. (1988). The volatilization of ammonia from perennial ryegrass during decomposition, drying and induced senescence. Ann. Bot. 61, 567–571. Williams-Woodward, J. L., Pfleger, F. L., Fritz, V. A., and Allmaras, R. R. (1997). Green manures of oat, rape and sweet corn for reducing common root rot in pea (Pisum sativum ) caused by Aphanomyced euteiches. Plant Soil 188, 43 –48. Willigen, P. D. (1991). Nitrogen turnover in the soil–crop system; comparison of fourteen simulation models. Fert. Res. 27, 141 –149. Willumsen, J., and Thorup-Kristensen, K. (2001). Effects of green manure crops on soil mineral nitrogen available for organic production of onion and white cabbage in two contrasting years. Biol. Agric. Hort. 18, 365–384. Wivstad, M. (1997). Plant morphology and content of nitrogen, cell wall and lignin at different phenological stages of red clover and yellow sweetclover. Sweed J. Agric. Res. 27, 3–14. Wyland, L. J., Jackson, L. E., and Schulbach, K. F. (1995). Soil –plant nitrogen dynamics following incorporation of a mature rye cover crop in a lettuce production system. J. Agric. Sci. 124, 17 –25. Yadvinder, S., Bijay, S., and Khind, C. S. (1992). Nutrient transformations in soils amended with green manures. Adv. Soil Sci. 20, 237– 309. Yamagishi, H., Yoshikawa, H., and Ashizawa, M. (1986). Effects of resistant plants as a catch crop on the reduction of resting spores of clubroot (Plasmodiophora brassicae Worn.) in soil. J. Jpn Soc. Hortic. Sci. 54, 460–466. Yanai, J., Linehan, D. J., Robinson, D., Young, I. M., Hackett, C. A., Kyuma, K., and Kosaki, T. (1996). Effects of inorganic nitrogen application on the dynamics of the soil solution composition in the root zone of maize. Plant Soil 180, 1–9.

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.): ITS ECOLOGY, AND USE AS A PASTURE LEGUME IN AUSTRALASIA p Michael L. Smetham1* Department of Plant Science, Lincoln University, P.O. Box 94, Lincoln, near Christchurch, New Zealand

I. Characteristics of the Species Trifolium Subterraneum L. A. Description B. Distribution C. Adaptation To Soil Conditions II. Physiological Response to the Environment A. Vegetative Responses B. Reproductive Responses C. Influence On Germination III. Management of Subterranean Clover A. History Of Use B. Choice Of Variety C. Establishment D. Seed Production E. Impact Of Hardseeded On Performance F. Hardseededness Carryover IV. Productivity A. Herbage Production B. Herbage Quality C. Use As A Pure Sward Or In Mixture With Grass V. Pests and Diseases A. Viruses B. Foliar Fungal Diseases C. Root Fungal Diseases D. Insects E. Oestrogenicity Acknowledgments References

p

Previously published by the Spanish Society for the Study of Pastures, in “PASTOS”, vol. XXIV(1), pp. 5–30. 1

Now retired. Present address: 75A, Aikmans Road, Merivale, Christchurch 8001, New Zealand. 303 Advances in Agronomy, Volume 79 Copyright q 2003 by Academic Press. All rights of reproduction in any form reserved 0065-2113/02$35.00

304

M. L. SMETHAM

Australian and New Zealand research investigating the ecology and use of the annual pasture legume subterranean clover (Trifolium subterraneum L.) is reviewed. The species and its distribution are described, together with the soil conditions and soil reaction to which it is adapted. Vegetative responses to light and temperature are briefly considered. Environmental control of reproduction is described in relation to vernalisation, seed dormancy, hard-seed development and decline, and burr burial. The history and manner of use of subterranean clover in Australasia is detailed. The management of this legume is discussed in relation to choice of variety, establishment, and seed production. Problems associated with hardseededness are listed together with possible solutions. The usefulness of hardseed carryover is debated. The level of herbage productivity obtained in various localities is catalogued, and quality for animal performance is discussed. Use of the species as a pure sward or in mixture with grass is debated. Pests and diseases are catalogued, together with an assessment of relative importance. Symptoms associated with the occurrence of oestrogenic substances in this species are detailed. q 2003 Academic Press.

I. CHARACTERISTICS OF THE SPECIES TRIFOLIUM SUBTERRANEUM L. A. DESCRIPTION Subterranean clover is an annual legume divided into several, but mainly three subspecies—T. subterraneum L. subsp. subterraneum var. subterraneum; T. subterraneum L. subsp. brachycalycinum (Katz. and Morley) var. brachycalycinum; and T. subterraneum L. var. yanninicum (Katz. and Morley) Zohary and Heller differentiated mainly (Katznelson (1974) in Rossiter (1978)) but not entirely (Zohary and Heller, 1984) on the basis of edaphic adaptation. Flowering occurs early in spring so that the plant can mature seed before moisture stress curtails plant growth. The species exists over the dry part of the year as seed, and germinates with the advent of rain in the autumn. After germination in the late autumn the seedling rapidly grows a taproot, to be followed in mid-winter by 1 – 20 shoots (runners) depending on competition, which closely hug the ground and spread out from the crown in all directions. With an increase in daylength and temperature, from three to five florets appear on the peduncle in each leaf axil (Fig. 1). Normally one or two fertilised ovules develop in each inflorescence. On fertilisation the florets reflex on the peduncle, and at the same time sterile florets develop (Fig. 1) into 30 – 40 hooked structures which envelope

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)305

Figure 1 A runner of subterranean clover (T. subterraneum L.) with a newly formed inflorescence on the left, and on the right, one where the fertilised florets have reflexed upon the peduncle, which is growing downwards to the ground surface. The start of growth of some of the sterile flower parts can be seen below the base of the reflexed florets.

the developing seeds to form the “burr” (Fig. 2). During this period the peduncle exhibits negative geotropism and grows towards the ground, and a proportion of the burrs become buried. This is facilitated by the fact that each time the hook structures absorb moisture they relax, whereas when they dry they contract, so helping to draw the burr into the ground. Subterranean clover is self-fertile, and fertilisation occurs before florets open (Morley, 1961). This is considered to have facilitated the retention in the wild of many discreet true-breeding populations, each occupying a landform, edaphic or climatic niche with subtle and unique differences. However, Cocks (1992) found 67% of a population originally sown in 1938 had diverged genetically, and he considered hybridisation to have been the main cause of this. Gladstones (1966) describes 89 ecotypes collected in the wild from Western Australia. These mainly true breeding populations differ visually in leaflet shape, presence of anthocyanin pigmentation patterns, or chlorophyll free marks on the lamina, presence or absence and prolificacy of hairs on stem, peduncle and pedicel; colour of florets, and presence and number of coloured rings on the calyx. There are also differences in the timing of, and duration of flowering, and the degree of hardness of the seeds.

306

M. L. SMETHAM

Figure 2 A ripe burr of subterranean clover (T. subterraneum L.) is illustrated, containing two seeds enclosed in the remains of the flower parts, and surrounded by many wiry structures which have developed from sterile flower parts.

B. DISTRIBUTION Subterranean clover occurs naturally as a component of herbaceous cover in areas of the old world having a Mediterranean climate (Katznelson (1974) in Rossiter (1978)). This is a regime with moderate rainfall of 250– 600 mm per annum falling mostly in late autumn, winter, and spring. Cold season temperatures are moderate with day means of 7 –158C. Summer and autumn temperatures are hot, with day mean temperatures ranging from 18 to 248C (Gentilli, 1971). Soils are below wilting point from late spring – early summer until late autumn, a period ranging from 4.5 to 6.5 months. Moderate frosts of up to 2 108C are experienced. The main areas where subterranean clover occurs naturally are Spain, Portugal, Italy, Greece, Turkey; in the Baltic around the Caspian Sea; Israel, Syria, Ethiopia, Afghanistan and neighbouring far-eastern countries; in the islands of the Mediterranean; and in North Africa. It is also found in areas with a marginal Mediterranean climate and occurs in sandy heath lands in southern England, southern France and country bordering areas with a true Mediterranean climate. Coincident with the utilisation of lands in the southern hemisphere (SH) as whaling bases, and later during settlement in the 18th century, subterranean clover became widespread in areas of

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)307

temperate Australia (Gladstones, 1966). It was first reported in Australia in 1896 (Davies, 1952) and in New Zealand the species was recognised growing north of Auckland in the early 1900s where it was known as Mangere clover (Saxby, 1956). In these instances it is thought that spread was facilitated by the occurrence of subterranean clover in the hays used as fodder for animals during passage to the new colonies. A proportion of legume seed can pass undamaged through the ruminant digestive system—for white clover this is about 8% (Suckling, 1952), and dung disposal at the port of entry would therefore provide a colonising source. By 1970 there was some 20 million ha sown to subterranean clover in Australia (Donald, 1970). In New Zealand there is little if any intensive use of subterranean clover as a pure sward, although it is widespread on dry hill country (Suckling et al., 1983). Commercial plantings are in use in parts of the western USA, as well as in temperate South Africa and several South American countries including Chile, Uruguay, and Argentina.

C. ADAPTATION 1.

TO

SOIL CONDITIONS

Species Adaptation

The natural distribution of the three main species of subterranean clover is to a large degree determined by edaphic factors. T. sub. subsp. sub. var. subterraneum and T. sub. var. yanninicum prefer acid to neutral soils, while T. sub. subsp. brachy. var. brachycalycinum tolerates higher pH: even alkaline soils (Katznelson (1974) quoted by Rossiter (1978)). Preferred soil conditions for these subspecies are: well-drained and alluvial soils for var. subterraneum; stony or ruderal soils for var. brachycalycinum and poorly drained alluvial soils for var. yanninicum (Katznelson, 1970). Var. brachycalycinum is also found on heavier clay soils. The greater tolerance of flooding by var. yanninicum (Marshall and Millington, 1967) is a result of several adaptive features of this species including a greater tolerance of root-rotting pathogens (Flett et al., 1993), a smaller build-up of ethanol following immersion, and a greater ability of the root system physiology to achieve higher levels of oxidation when waterlogged (Katznelson, 1970) probably due to the shallow root system.

2.

Soil Fertility

It is not proposed to cover in detail here the aspect of subterranean clover agronomy. It has been reviewed in depth by Williams and Andrew (1970).

308

M. L. SMETHAM

II.

PHYSIOLOGICAL RESPONSE TO THE ENVIRONMENT A. VEGETATIVE RESPONSES 1.

Response to Temperature

Growth cabinet studies (Mitchell, 1956a) have shown that subterranean clover has an optimum temperature for growth near 178C, although the response curve is fairly flat. This is in contrast to that for white clover (248C) and 278C for lucerne (Smith, 1970). At a population of 2000 plants/m2 Silsbury and Hancock (1990) obtained twice as much growth at 208C than at 108C, but no response at higher densities. Similarly, Fukai and Silsbury (1976) found that growth rate was maximised by temperatures between 20 and 258C although differences were small and occurred only until a closed canopy was reached at LAI 3.0 or 2000 kg DM ha. Although further increases in herbage mass increased leaf area, it also increased mutual shading, and the photosynthate required for increased respiration progressively absorbed any additional photosynthate produced (Fukai and Silsbury, 1976). The results of Mitchell (1956a,b) also demonstrated the ability of subterranean clover to grow faster than white clover at temperatures below 158C, and down to 78C. This explains the ability of the autumn germinating subterranean clover to utilize cooler autumn – winter conditions for growth, and enables this annual to establish competitive bulk in the limited growth season available to it. At latitude 438380 S in southern New Zealand, Smetham and Jack (1995) recorded substantial subterranean clover growth rates of 16 – 19 kg DM ha d in autumn, 12 kg DM ha d over June and July, and 17 –54 kg DM ha d from spring onwards. Monthly mean day temperatures for these periods were 9– 12, 5 – 6, and 8– 108C, respectively (Smetham and Jack, 1995).

2.

Response to Light

Artificial shading to one-third of daylight caused an increase in petiole length and leaf area together with a reduction of leaf thickness (Mitchell, 1956b). These changes also occured with mutual shading in a pure sward (Smetham and Dear, unpublished results). This response allows the species to survive considerable competition in a mixed grass legume sward although the maximum petiole elongation appears to be 25 cm, which can be exceeded by grasses if there is a long interval between defoliations. In quasi-Mediterranean climates with some rainfall occurring in the summer, perennial grass ingress following soil N

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)309

increase can provide severe competition for light, most particularly at the time of establishment following autumn germination (Dear et al., 1998).

B. REPRODUCTIVE RESPONSES The distribution of subterranean clover is determined mainly by features of climate which affect seed production and therefore survival. Donald (1960, 1970) suggests that in Australia, three climatic boundaries can be recognised for subterranean clover. A rainfall boundary, which if too low results in a season of growth too short for the production of sufficient seed. A high temperature boundary above which vernalisation and floral induction fail to occur, and a low temperature boundary below which flowering is inhibited. While most subterranean clovers can survive frosts of 2 7 to 2 108C during vegetative growth (Smetham, 1977), at altitude temperatures may be too low to allow the induction of flowering (Evans, 1959), and frosts can severely damage flowers and seed setting (Donald, 1960). 1.

Vernalisation and the Timing of Flowering

Subterranean clover flowers in response to low temperatures, followed by long days and warm temperatures (Evans, 1959). By mid-winter ground-hugging branches (runners) start growing out from the rosette of subterranean clover, and flowers eventually appear in spring in the axils of leaves at nodes on these runners. Flower initiation in subterranean clover is dependant on both temperature and daylength. Aitken and Drake (1941) found that the very early flowering variety “Dwalganup” initiated floral structures at a low node (number 4– 7 on a runner) when this cultivar was exposed to a mean temperature of 128C for 1– 2 weeks during the month after sowing, and when the photoperiod was 10 h. If temperatures were higher in the month after sowing, flowering was delayed until nodes 8 –14 appeared, and above 258C few flowers were formed. The very late flowering cultivar “Tallarook”, however, needed exposure to a temperature of 48C or lower before flowers appeared at a low node. Flowering occurred at nodes 11 –18 at higher temperatures, and above 138C no flowers appeared. Exposure to a longer daylength decreased the requirement for cold. Although delayed, “Dwalganup” managed to flower with initiation temperatures up to 258C and a daylength of 14 h, but “Tallarook” remained vegetative above a vernalisation temperature of 188C regardless of daylength (Aitken, 1955). In addition to exposure to low temperature, Aitken (1955) demonstrated that the requirement for cold was quantitative. Although the early flowering “Dwalganup” did flower without exposure to cold, flowering occured at a lower node

310

M. L. SMETHAM

after 2.5 weeks at 108C, but with no further response beyond that period. “Tallarook” on the other hand flowered at a high node after 6 weeks at 108C but needed 8– 10 weeks at 108C to flower at the lowest node. Morley and Davern (1956) have demonstrated that cultivars flowering at a time between very early and very late have an intermediate vernalisation requirement. From these studies, it can be seen that the temperature requirement for initiation constitutes a warm boundary for the species and its ecotypes. This can be determined geographically by reference to climatic data. Donald (1960) quotes field experience which finds that the warm boundary for adequate seed production with the mid-season variety “Mount Barker” coincides closely with the 138C isotherm for July on lowland close to the NSW – Queensland border, and the boundary moves further north on the higher altitude, and therefore colder tablelands.

2.

Soil Moisture and Seed Production

Adequate soil moisture for long enough to allow full maturation of sufficient seed is critical to the success of subterranean clover. The findings of Aitken (1955) explain how early flowering subterranean clovers flower in late winter – early spring—July or August in the SH (Smetham, 1977)—early enough to produce viable seed in situations where moisture becomes limiting for plant growth 6– 8 weeks later in spring (September in SH) They have little or no requirement for cold to trigger floral initiation. However, at the other extreme, late flowering ecotypes flower late because they need some 10 weeks of exposure to temperatures below 108C before initiation occurs. These commence flowering in late spring –early summer (mid-October in SH) and need adequate moisture into summer (until late December in SH) for a good seed set. There is of course a continuum of ecotypes which lie between the extremes of flowering time, having an increasing requirement for cold which leads to a progressively later period of flowering. The early flowering cultivars like “Dwalganup” and “Geraldton” have been successful in areas of Australia with rainfall as low as 280– 300 mm year, whereas the late flowering “Mount Barker” has been used in areas with 800 – 1500 mm year (Rossiter and Ozanne, 1970). The inland or arid boundary for these cultivars is determined by the length of time during which moisture conditions satisfactory for growth and seed maturation are maintained at any location (Trumble, 1957) combined with local experience of the actual growth period required. For a mid-season flowering variety this is estimated to be 7.5 months and for an early type like “Dwalganup” 6 months (Donald, 1970). On this basis for example the boundary for “Mt Barker”

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)311

is not far inland from the southwestern tip of Western Australia, whereas for “Dwalganup” the boundary is some 380 km inland and runs north– east from Hopetown to near Geraldton. Since germination of subterranean clovers occurs with the advent of autumn rains, all varieties commence growing at much the same time regardless of whether they are early or late flowering. As the latter have to grow for a longer period of time in order to acquire the cold requirement for vernalisation, it follows that they grow vegetatively for longer before flowering. Thus Aitken (1955) found that early ecotypes produced a first flower in the axil of the third, fourth, or fifth leaf on any stem, whereas late varieties had produced 12 –15 leaves before flowers appeared. In addition, and for the same reason, the later flowering varieties develop more inflorescences and hence have a greater potential for seed production. This is only realised, however, if the moisture environment allows this to happen (Rossiter, 1959). Strains also differ in the rate and pattern of production of flowers (Francis and Gladstones, 1974). The duration of flowering also differs widely between ecotypes. Ru and Fortune (1999) found that some varieties flowered for twice as long as others, even in the same maturity grouping. For instance Green Range; a mid-season flowering cultivar flowered for 66 d but York flowered for only 26 d. Maximum seed production, and hence success in a given area, will be given by a variety which just manages to mature adequate seed, rather than an earlier flowering type, especially since excess moisture during and after seed maturation can lead to large losses following inhibition and rotting of the seed (Archer, 1990).

C. INFLUENCE

ON

GERMINATION

There appear to be several mechanisms which operate to prevent the germination of subterranean clover seed until conditions are likely to be suitable for good survival of seedlings.

1.

Embryo Dormancy

Embryo dormancy is promoted by high temperature, inhibitory substances in the seed coat, and by an absence of an high concentration of carbon dioxide; all of which have been shown to prevent the germination of fully imbibed subterranean clover seeds. Unlike hardseededness, embryo dormancy is a behaviour which only protects seeds immediately after maturity. It is a mechanism which is of advantage to a plant growing in a summer-dry environment, particularly in areas where there is sometimes likely to be enough summer rain to promote

312

M. L. SMETHAM

germination, but not enough to sustain the seedlings. Thus fully imbibed seeds of subterranean clover mostly fail to germinate at temperatures above 208C although they do so readily at 108C (Loftus Hills, 1944). Strains vary in their response, which is under genetic control. Some strains, for example, “Geraldton” and “Dinninup” have greater dormancy than others, for example, “Woogenellup” and “Mt Barker” (Taylor, 1970). Dormancy is most strongly expressed at high temperatures although some cultivars, e.g., Clare, can germinate at 308C (Gladstones, 1987). Dormancy diminishes in storage, the rate being faster with higher storage temperature. Protection against widespread loss of seedlings is therefore assured in summer, but by autumn (late March – April in the SH) day temperatures have declined enough to allow germination of seeds at a time when survival is almost guaranteed to occur. Germination in the autumn has also been shown to be promoted by exposure to the high levels of CO2 likely to be experienced in the soil but not on the soil surface, and to leaching with water (Taylor, 1970). In February, continuous leaching with running water, or confinement in a sealed container with water, enabled substantial germination even at 308C though with varietal differences, whereas merely soaking in water gave only 8% germination. At 208C leaching resulted in complete germination. This requirement for leaching, equivalent to about 10 mm of rain before full removal of the inhibitor occurs ensures at least some continuity of soil moisture to help seedling survival. Walker (1971) has some evidence that the purple anthocyanin in the seed coat may act as an inhibitor of germination, and it could be assumed that this would be removed by any leaching process.

2.

Hardseededness

(a) Development of hardseededness. Subterranean clover has the ability, like most legumes, to produce so-called “hard” seed which remains unable to absorb water and thus germinate, for a period of weeks to years after maturation. A comprehensive review of seed coat impermeability was made by Quinliven (1971). The facility to produce hardseed is a mechanism which allows the species to survive seasons where seedlings are lost due to drought or pest attack or where seed production has failed. It particularly protects against premature germination following summer rains in less than true Mediterranean climates. Studies by Taylor et al. (1984) of seed losses from drought after so-called “false strikes”, showed that over a 12-year period these averaged 24% of all seed set. In New Zealand, Dodd et al. (1995a) has measured losses from this cause of 4 –55%. If the bulk of the current seed set is lost as a result of false strike, softening of a proportion of hardseed in the next season then allows the species to re-establish. For instance, Donald (1959) found that 6% of “Dwalganup” seed

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)313

survived and germinated in the next season after formation; while 2% still germinated after 4 years in the ground. Hardseed is a condition obtained when the moisture content of the seed has dropped to a critical low level. As the seed matures and dries the testa shrinks and becomes impermeable at a moisture content of around 14% of dry weight. Thereafter structures below the hilum, or point of attachment of the seed (Bell, 1991), and surrounding the longitudinal groove seen in the pit of the hilum (Fig. 3), act as an hygroscopic valve, preventing water from entering but allowing moisture out. The moisture content of the seed eventually reaches equilibrium with the driest conditions to which it is exposed (Hyde, 1954). Seeds need to reach a low level of moisture to attain an irreversible impermeability (Quinliven, 1971) which for subterranean clover appears to be 5– 7%. Hardseededness is maximised when the seed matures slowly over a long growing season and in the absence of moisture stress (Aitken, 1939; Quinliven, 1965). Seeds matured under these conditions subsequently soften more slowly. An additional mechanism contributing to impermeability is the presence of a continuous layer of suberin, laid down over the coloured part of the seed coat (Aitken, 1939) which further prevents ingress of moisture into the seed. However, if moisture stress occurs during seed maturation, then this suberin layer may be thinner and even discontinuous. It is evident therefore that if soil moisture deficit occurs soon after flowering has started, subsequent seed may lack the protection of hardseededness. If excessive moisture is present during seed maturation and subsequently, hardness of seed fails to develop, and considerable amounts of seed can be lost to fungal rots (Collins and Quinliven, 1980; Archer, 1990).

Figure 3 The seed of subterranean clover (T. subterraneum L. subspp. subterraneum var. subterraneum in elevation (left), and plan view (middle). The drawing on the right shows a muchenlarged hilum H, with its deep longitudinal groove, surrounded by the raised rim-aril structure RA (see Corner (1951)). The position of the tip of the radicle within the seed is indicated by R. The position of the strophiole, just under the seed coat, is indicated by S.

314

M. L. SMETHAM

Maximum levels of hardseededness occur in late spring –early summer, with a significant difference between strains and between maturity groups, although all groups exhibit a relatively high percentage of hardseed at this time. In the Mediterranean climate of Western Australia Quinliven and Millington (1962) measured hardseededness in the first month after cessation of growth in mid-November of 95 –97% for early varieties through to 59% for late flowering cultivars (Table I). (b) Decline of hardseededness. With exposure to sufficient variation between day and night temperatures, hardseededness declines steadily in any seed lot between maturation and the following autumn. Softening to allow germination is influenced positively by the amplitude of the diurnal temperature variation and the magnitude of the high temperatures reached. Quinliven (1961, 1966) found that softening was significantly faster under a 608C day and a 158C night regime than a 468C/158C diurnal variation. There was, however, no increase in rate of softening when the day temperature was 738C. The regime causing the most rapid softening does in fact approximate soil surface temperatures in many parts of southern Australia in mid-summer. For example, measurements at Griffith, New South Wales recorded a mean January (SH) day maximum temperature at 1 p.m. of 518C and a 3 a.m. mean minimum night temperature of 258C (Quinliven, 1961). Exposure to cooler conditions during the summer will slow up the softening process and Evans and Hall (1995) found that 12 –31% of the original seed was still hard after exposure to the cooler Tasmanian summer monthly maximums of 20 – 238C and minimums of 10 –118C. Hagon (1971) found that the diurnal variation of temperature responsible for softening caused alternate expansion and contraction of the tissues comprising

Table I The Proportion of Hardseed in Nine Cultivars of Subterranean Clover

Cultivar Geraldton Dwalganup Burerang Carnamah Morocco Woogenellup Palestine Bacchus Marsh Mount Barker

In early summer in Western Australia (% hard)

After 4 months exposure to summer temperatures in Western Australia (% hard)

After 6 months exposure to alternating 158C night/608C day temperatures (% hard)

97.0 96.5 82.4 95.5 80.7 86.0 89.2 71.0 58.8

64.7 62.4 52.4 51.2 35.3 34.4 29.6 16.5 11.1

29.3 10.3 28.8 8.5 26.4 2.8 3.1 0.2 2.1

From Quinliven and Millington (1962).

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)315

the “strophiole,” resulting in the mechanical rupture and chemical degradation of this region of the testa (Young, 1985), allowing moisture to enter the seed, which did so at this point only. Bell (1991) describes the strophiole as a conspicuous swollen fleshy addition on that part of the ovule stalk fused with the outer integument known as the raphe. Diagrams in Corner (1951) and photographs in Hagon and Ballard (1970) where the alternative name “lens” is mentioned, indicate that this structure is in line with the tip of the radicle, and the hilum, but on the other side of the latter (Fig. 3). It was at this point that percussion had the effect of softening hardseeds of T. subterraneum L. (Hagon and Ballard, 1970) and Melilotus alba (Hamly, 1932) by splitting palisade cells in the testa, allowing water to enter. Significant variation occurs between strains of subterranean clover for the rate and pattern of decline of hardness, and Quinliven and Millington (1962) record that while the initial levels of hardseed in the early flowering “Geraldton” and late flowering “Mount Barker” were 97 and 59%, respectively, after 4 months in the field, these had dropped to 67 and 11%, respectively, with similar behaviour over the range of intermediate varieties (Table I). Enough variation exists in the rate of autumn softening to allow selection for this trait to be beneficial (Smith et al., 1996). Hot summers were noticed by Quinliven (1965) to cause more rapid softening in the autumn, and Taylor (1981) has proposed that softening is a two stage process, being influenced initially by the value and duration of constant temperatures experienced in the summer of seed maturation. All temperatures from 15 to 808C accelerated softening in the following autumn, but 608C and above had most effect, with seed softening in the autumn after only one to three cycles of alternating 608C/158C temperatures in the second stage of the process. Taylor (1981) suggests that the initial steady temperatures cause the thermal degradation of the strophiolar region, while the fluctuating temperatures of the second stage physically disrupt the strophiole allowing the entry of moisture. Burial in soil reduces the amplitude and extremes of temperature experienced by the seed and so slows the softening process. Ninety-five percent of “Geraldton” seed, initially hard, germinated during the first year on the surface, but only 25% had softened during 4 years buried at 10 cm (Taylor and Ewing, 1988). Conditions of storage and burial also affect the rate of hardseed breakdown and Taylor and Ewing (1996) found that softening was accelerated by up to four times after burial for very long rather than short periods in soil, and he postulated a pre-conditioning effect of storage on the rate of softening. Differences in rate of decline of hardseededness also occur because of different amounts of herbage top cover. This progressively reduces the magnitude of the temperature fluctuations experienced at the soil surface (Quinliven and Millington, 1962) and consequently reduces the rate of softening. Grazing to reduce herbage cover can clearly be used as a management tool to accelerate softening, particularly in areas with cool summers.

316

M. L. SMETHAM

3.

Burr Burial

The ability of subterranean clover to bury seed in a burr below ground level results in heavier and more viable seeds, and in more seeds per burr, giving a higher seed yield particularly in early flowering varieties (Yates, 1958). The burial of burrs, which occurs as the seed is maturing, protects the developing seed from the adverse effects of high temperature and low humidity and provides a more favourable moisture regime (Yates, 1957). Burial in the field resulted in a fivefold increase in seeds per burr, and a variable but always positive increase in the weight of seed produced. In addition, it tends to aid the development of hardseededness (Aitken, 1939). When burial was artificially prevented, the number of burrs produced, and the weight per seed was reduced in all three subspecies of subterranean clover, leading to a drastic fall in weight of seed produced, often to less than 50% of the weight obtained when natural burial was allowed (Bolland and Collins, 1986). Defoliation significantly increases burial (Walton, 1975) and Collins et al. (1983) found burial increased with defoliation during flowering to include 95% of all seed formed. There are genetic differences in the ability to bury burrs, with early types burying a high proportion and late types a low proportion of burrs (Francis et al., 1972). Burial is to some extent dependant on soil surface roughness and soil texture and is facilitated by the presence of cracks and clods, and is greater in sands than in clays. The species T. sub. subsp. sub. var. subterraneum has a greater ability to bury burrs than T. sub. var. yanninicum, which in turn is better than T. sub. subsp. brachy. var. brachycalycinum (Francis et al., 1972), which only manages to bury burrs in cracks formed in cracking clays. Burial also aids survival of seed from year-to-year (Taylor and Ewing, 1988). For instance, only 5% of hardseeds of “Geraldton” survived after 1 year on the soil surface but 75% were still present 4 years after formation, when buried at 10 cm.

III. MANAGEMENT OF SUBTERRANEAN CLOVER A. HISTORY

OF

USE

From the early 1920s, use was made of subterranean clover in New Zealand mainly hill country where soil moisture severely limited plant growth for 1 to 4 months over summer (Saxby, 1956). This can be caused by low rainfall, shallow soils, and high summer temperatures of 26 –328C or a combination of any or all of these. Suitable areas have a mild winter with mean day temperatures of 5– 108C, but with the possibility of some frosts of up to 108C, allowing cool-season growth but without

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)317

causing the death of plants. From the early 1930s subterranean clover use was extended to the Canterbury Plains and other flat to rolling areas of shallow, stony or sandy soils in New Zealand receiving 600 –750 mm annual rainfall (RF), and drying below wilting point for most of the period between November and March –April (Saxby, 1956). This necessitated the development of what became a particularly successful farming system involving very early lambing starting in July, so that a proportion of lambs reached killable weight fat off their mothers, before the subterranean clover-based swards died off in early summer—late November in the SH. Previously, short-lived annual types of white clover were the legume base for much of this country, but the use of subterranean clover together with a perennial grass, increased carrying capacity more than threefold to nearly six ewe equivalents per ha (Calder, 1954). From the late 1940s, subterranean clover was increasingly sown on dry, sunny face hill country of the North Island after bush clearance and burning, being broadcast by hand initially (Saxby, 1956), but later using aircraft. Although the annual rainfall on most of this land is 1000 –2000 mm year, effectiveness is markedly reduced by wind; the high summer temperatures, and the steepness of the terrain, leading to summer drought, particularly on country with a sunny aspect. Although Levy and Gorman (1936) recommended from experimental results that more productive cultivars should be used, the late season flowering “Tallarook” and the mid-season flowering “Mount Barker” still constitute the majority of stands in New Zealand (Suckling et al., 1983; Macfarlane and Sheath, 1984) mainly because these were the only varieties originally available. In New Zealand, subterranean clover has been used almost exclusively in mixed stands, generally with a perennial ryegrass. This plant has probably never expressed its full potential in New Zealand (Smetham, 1977) partly because it has been sown at too low a seed rate, and but mainly because it is rarely sown as a pure sward. In Australia, subterranean clovers arrived with the first whalers and settlers and their sheep early in the 19th century (Gladstones, 1966). Plants spread into areas with annual rainfall of 280– 1400 mm, and a true Mediterranean or near Mediterranean climate. These are areas of mild winters with a mean day temperature of 7– 158C; hot summers with mean day temperatures of 18 –248C, and relatively little rain for the 4 –6 months of summer (Gentilli, 1971). Different varieties have been recognised since 1907, when “Mount Barker” was first commercialised (Symon, 1961) with most cultivars being named after districts or properties in Australia where they were first identified (Donald, 1960). Cocks and Phillips (1979) listed 18 cultivars in commerce by 1976, 15 of which were strains naturalised since settlement; two were interstrain hybrids, and one was an overseas selection. These included the early flowering “Dwalganup”; commercialised in 1929; and “Yarloop” (1947), mid-season-flowering “Bacchus Marsh”

318

M. L. SMETHAM

(1940) and “Nangeela” (1961), and mid- to late-season flowering “Mount Barker” (1907)and “Tallarook” (1935). In Western Australia only 10 –30% of rainfall occurs in summer (Gentilli, 1971), and apart from the extreme tip of the land area, the climate is too dry and hot in summer for perennial grasses to survive. Annual grasses are sometimes sown with the subterranean clover, but pastures on acid soils are mainly sown to pure subterranean clover to build fertility for wheat, and for grazing (Rossiter and Ozanne, 1970). Between the 450 mm rainfall isohyet and the inland arid boundary for subterranean clover which is around 280 mm, early varieties are used. Between 450 and 750 mm annual rainfall early mid-season varieties are used, while late types are sown in the extreme western tip where rainfall is greater than 750 mm (Gillespie and Nicholas, 1991). In eastern Australia summer rainfall is a significant proportion of the total, increasing from 30% of total in the western plains to around 40% on the slopes and 60% on the tablelands of the main divide (Gentilli, 1971). Perennial grasses such as phalaris are used in mixture with subterranean clovers where annual rainfall exceeds 560 mm, while perennial ryegrass survives above 750 mm and in irrigated pastures with subterranean clover on the tablelands and in parts of Tasmania (Moore, 1970). Subterranean clovers are sown pure on acid soils in areas of eastern and southern Australia with rainfall of less than 550 mm, down to around 350 mm, mostly as a fertility building break in a cropping regime but also to feed sheep and cattle. In both the west and east of Australia, free draining soils which are not too acid, with rainfall above 350 mm and a significant proportion of this occurring in summer, are often sown to a mixture of subterranean clover and lucerne. Although annual grasses like Lolium rigidum Gaud. are sometimes encouraged in mixture with subterranean clover, grass species including annuals are not favoured in the restorative years of a cropping rotation because some, including the weed species in the genus Vulpia, are carry-over hosts of Gaeumannomyces graminis var. tritici, which causes take-all disease in cereal crops (Leys et al., 1993). This can severely reduce wheat yield.(Dear, personal communication). Subterranean clover is used in these areas as a pure sward down for 3– 5 years in rotation with one to four crops of wheat. The relatively low numbers of sheep on these farms lamb in the autumn in early April; and graze on wheat stubble to the end of April, or until the subterranean clover on the rest of the property has re-established and is strong enough to be grazed in early June. Grazing of the subterranean clover continues over spring – early summer while the clover is flowering. Weaned ewes utilise crop residues from wheat harvest in December (mid-summer) through the end of April. By mid-November, the subterranean clover top growth has died off, leaving considerable bulk of leaf, stem, and some burr, which is progressively utilised by

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)319

sheep until subterranean clover germination occurs in early April (Dear, personal communication).

B. CHOICE

OF

VARIETY

In 1950, a project to select more appropriate ecotypes of subterranean clover for the wheat belt was started by Dr Millington in Western Australia. This evolved over the years to become the (Australian) National Subterranean Clover Improvement Programme (Collins, 1987). The programme was administered and run jointly by the Western Australian Department of Agriculture and the University of Western Australia, with interstate collaboration from all states except Northern Territory. Associated was the establishment at Perth, of the Australian Trifolium Genetic Resource Centre, following a number of collecting expeditions to the Mediterranean region between 1951 and 1967 (Donald, 1970). The ongoing evaluation procedure for subterranean clover involved the crossing of accessions and selection for up to 8 years; screening for isoflavones, disease and pest tolerance, and then evaluation in each collaborating state. In 1997, after evolving through many intermediate bodies, the programme became the National Annual Pasture Legume Improvement Programme, and was broadened to include other alternative legumes (Dear, personal communication). The culmination of this work has been the approval and release of new cultivars with superior attributes for all the areas of use in Australia. This is accomplished by approval from the CSIRO Australian Herbage Plant Registration Authority, and documented by publication in the Australian Journal of Experimental Agriculture, which gives full details of parentage, screening results, productivity, special attributes and recommended areas of use of any new cultivar. For instance, volume 32, 1992, pp 539 –542 documents the release of one T. sub. var. yanninicum cv. Gosse; and three T. sub. subsp. subterraneum var. subterraneum cv. Denmark, cv. Leura, and cv. Goulburn. Full details of Australian pasture plant cultivar releases, including all subterranean clovers prior to 1990, can also be found in Oram (1990) and Barnard (1972). This facilitates the publication for farmers of extension publications such as “Subterranean clover in New South Wales-identification and use” by Dear and Sandral (1997). This gives details of a wide range of recommended subterranean clover cultivars, many of which are recent releases from the NAPLIP; and all of which embody significant improvements in performance, disease and pest tolerance, as well as low levels of antiquality agents. Table II summarises some of the information given in this publication. The choice of variety for a given area will hinge initially on the date at which moisture stress can be expected to occur and the time of flowering required to allow the maturation of adequate seed before this occurs. Theoretical

320

M. L. SMETHAM

Table II Characteristics of Subterranean Clover Varieties Recommended for Use in New South Wales

Variety

Flowering startsa

Days from sowing in mid-May to flowering

Nungarin Dalkeith Seaton Park (L) York Trikkala Riverina Rosedale Gosse Junee Woogenellup Clare Goulburn Denmark Leura Nuba

Early August Late August Early September Early September Early September Mid-September Mid-September Late September Mid-September Mid-September Late September Late September Early October Early October Early October

110 120 125 125 122 128 120 136 138 140 142 145 149 156 152

Minimum rainfall for persistenceb (mm)

Hardseed in autumn (0 ¼ nil; 5 ¼ high)

375 400 475 475 525 500 500 650 500 525 650 525 600 750 700

5 5 3 5 2 3 3 3 3 2 2 3 2 1 3

After Dear and Sandral (1997). a The flowering data is based on observations at Wagga Wagga, New South Wales. b Rainfall figures are a guide only and will vary with aspect, slope, and soil type.

considerations can be applied to arrive at an appropriate time of first flower appearance and as a result choose a variety for a given situation. Between 38 d (“Geraldton”) and 54 d (“Dwalganup”) were required from anthesis to allow maximum weight per seed to develop, both for buried and surface burrs (Tennant, 1965) although viable seed was formed earlier at 30 and 38 d, respectively. Consequently to set some seed and mature this, a variety needs to start flowering 4 or 5 weeks before soils fall below wilting point. However, to set high yields of seed Dear and Sandral (1997) consider that flowering needs to continue for at least 6 weeks, with a further period of 4 weeks to allow full maturation of the seed formed—a total of 70 d. Rossiter (1978) suggests that a slightly longer period of about 80 d from start of flowering is needed to maximise seed production. This means, therefore, that for success in a given situation, a cultivar needs to be chosen that will start flowering 70 –80 d before severe moisture stress sets in. Rossiter (1959) introduced the idea of describing the time of first flower appearance of varieties by using the number of days which had elapsed between the flowering of “Dwalganup”; one of the earliest varieties, and the variety in question. He called this figure the maturity grading (MG). Thus “Mount Barker” had an MG of 98 since this variety commenced flowering 98 d after Dwalganup. This system eliminates from varietal descriptions the fact that the time between

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)321

germination and first flower appearance is shorter, and particularly so for later varieties, when sowings, or the natural time of germination, are progressively later (Dear and Loveland, 1984), and that this period differs between sowings made in western and eastern Australia (Dear and Sandral, 1997). On average in Canterbury, New Zealand, in the SH, soils reach wilting point by mid-November (Garnier, 1958) in early summer. Thus for a variety to be agriculturally successful from the seed production point of view, it needs to start flowering at least 10 weeks before this date i.e., by 1 September in early spring. Observations by Smetham (1977) record that in 1969 “Dwsalganup” flowered on 1 August (MG 0), “Geraldton” 14 August (MG 14), “Yarloop” 3 September (MG 34), “Woogenellup” 14 September (MG 45), “Mount Barker” 5 October (MG 66) and “Tallarook” 13 October (MG 74). From these data one can predict that Yarloop or Woogenellup will set adequate seed for success in Canterbury. Under similar conditions of rainfall and climate, Scott (1971) did obtain adequate yields (499 – 739 kg ha) of seed from “Geraldton” and “Woogenellup”, but cultivars flowering later than “Woogenellup” set insufficient seed (70 – 239 kg ha) for success. On the other hand, at a high rainfall site with no moisture stress until late December, the highest seed yields (1100 – 1400 kg ha) were given by the late varieties “Mt Barker” and “Tallarook”. Further support for the aforementioned theoretical approach to choice of cultivar is given by the results of an evaluation of cultivars and numbered accessions from the Australian Trifolium Genetic Resource Centre, Perth, in Canterbury, New Zealand (Latitude 438500 S) on shallow stony soils with an evenly spread rainfall of 650 mm year (Smetham et al., 1993). Out of 47 lines, only 22 produced more than 240 kg ha of seed and these flowered 99 – 138 d after sowing at or before 18 September. As before, these successful lines were all in the early-mid to mid-season flowering group, equivalent to a MG of 9 – 48. However, in this region rainfall can be erratic, and an evaluation conducted during the next year (Smetham et al., 1994), when an atypical drought occurred in mid-spring, showed that early and mid-season flowering varieties were severely disadvantaged by this. Successful lines were all numbered accessions in the mid-season and late mid-season flowering group, flowering up to 146 d after sowing, with MG 46 – 56. They all possessed an ability to recommence flowering after a check to growth—a characteristic termed “kick-on” by Australian agronomists. This is known to be possessed most strongly by annual medics, but also by subterranean clover lines with stout stems and peduncles Gladstones (1985) in Nichols (1987). These results suggest that in areas of variable erratic spring rainfall patterns, it is wise to include both early and later flowering varieties of subterranean clover in the sown mixture to provide some insurance should one fail; a recommendation also made by both Gillespie and Nicholas (1991) and Dear and Sandral (1997) for Western Australia and New South Wales, respectively. In Western Australia in a 550– 600 mm rainfall zone Rossiter (1966) examined the performance of 51 ecoptypes with MG from 0 (“Dwalganup”) to

322

M. L. SMETHAM

62 (“Wenigup”), and found that seed production was successful only where MG was less than 25, this again being a reflection of time of flowering relative to the early onset of water stress in this environment. It will be evident that whilst seed production is a first criterion to be used in selection of a variety for a district, herbage production, hardseededness and other attributes including disease and pest tolerance must be taken into account. The results of Dear et al. (1993) show that while early varieties were successful in terms of seed set in their evaluation, the use of these early varieties was not justified in areas of New South Wales with a rainfall of greater than 500 mm, since they were unable to take advantage of the extended season for growth. Later varieties set sufficient seed too, but yielded around 10 t DM ha, while early varieties produced only 4 t DM ha. Similarly Smetham and Jack (1995) record that in a 625 mm annual rainfall area in New Zealand although from a seed production point of view early mid-season flowering varieties were superior, one mid-season, and two late mid-season flowering accessions produced significantly more herbage. The recognition of characteristics needed in a cultivar for a given district, can form the basis for the definition of an ideotype. An example of this is given by Evans (1993) and Dear et al. (1992), which resulted in the selection of cv. Goulburn, a replacement for cv. Woogenellup, but with the greatly improved disease resistance and a higher level of hardseededness as defined in the ideotype. In a comprehensive study of the performance of nine subterranean clover cultivars at eight hill country sites in New Zealand (Williams et al., 1990) it was found that the ability of a cultivar to survive and set seed under the close and continuous grazing; often to within 10 –20 mm of the ground surface, which is a feature of New Zealand hill country management (Sheath and Macfarlane, 1990b) was of overriding importance for success when this was measured as herbage productivity. This ability was more important than time of flowering and seed production, however, all but two of the eight sites had rainfall between 1200 and 1500 mm, with appreciable rain falling in summer. Such a climate meant therefore that most varieties, regardless of date of flowering, could survive and produce some seed. Overall Tallarook was the most successful cultivar, being consistently high yielding with good regeneration (Chapman et al., 1986), and conforming to the ideotype proposed for North Island hill country by Sheath and Macfarlane (1990a) of a variety flowering in late October– November, with prostrate crown and runner habit, and a low level of oestrogenic activity. Results show that the cultivars Woogenellup, Clare, and Nangeela (Sheath and Macfarlane, 1990a) and these together with Seaton Park (Sheath and Macfarlane, 1990b) were unable to persist and regenerate under continuous close grazing. These cultivars have an open habit of growth, and the runners do not hug the ground as closely as the more persistent Mount Barker and Tallarook, with the result that portions of runners with flowers and seed attached were often removed during grazing. The more persistent Larissa, Mt Barker

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)323

and Tallarook also had smaller leaves contained in a dense growth habit and were very prostrate (Chapman and Williams, 1990). The amount of seed set at the eight hill sites initially ranged from 300 to 3000 seeds m2 (Macfarlane et al., 1990) and occasionally up to 7500 seeds m2 (Sheath et al., 1990), but at most sites this declined over the years to low levels, as a result of increasing competition from browntop (Agrostis spp.) and ryegrass (Lolium spp) (Sheath and Macfarlane, 1990a). It is unfortunate that at no sites was hardseed assessed. At most sites Tallarook gave the best seed set and seedling numbers. However, at only 3 of 63 site £ cultivar records reported by Chapman et al. (1986) did seedling numbers reach the 1000 m2 suggested by several Australian authors, for example, Taylor et al. (1984) as being a minimum for success. Woogenellup, Nangeela and Howard were frequently reported to lose considerable numbers of seedlings killed by desiccation (Sheath and Macfarlane, 1990b). Rains would have been experienced at most sites during seed maturation, and this moisture would have tended to minimise the full development of the hardseeded condition (Collins and Quinliven, 1980). It is significant that at only two sites; Porangihau, with a rainfall of 700 –900 mm, and Carvossa with 657 mm, were the numbers of seeds set; 1800 – 3000 and 1200– 4000 m2, and seedling numbers of 330 and 500– 700 m2, respectively (Chapman et al., 1986), maintained or even increased during the period of evaluation. This is not surprising since at both of these sites the climatic conditions were closer to those for which subterranean clover is adapted. Many of the subterranean clover cultivars named to date in this review have been superseded for various reasons including high levels of oestrogen, susceptability to diseases and lack of productivity. Whilst details of date of first flower, herbage production and other attributes of the latest releases for Australian conditions can be obtained from sources quoted at the start of this section, performance data for these under New Zealand conditions are scarce. An evaluation of some recent Australian releases, Australian breeding material, and some North Island, New Zealand hill country selections of subterranean clover (Dodd et al., 1995a) showed that in typical North Island hill country with 1200 – 1500 mm annual rainfall and summer drought from relatively late in the summer, Tallarook was superior, although six other lines were similar, with the ability to consistently generate adequate (Sheath and Macfarlane, 1990b) winter populations of 200 plants/m2 in a mixed sward, and give more than 1000 kg DM ha in spring. At a somewhat drier site (Dodd et al., 1995b) the late mid-season flowering Karridale—a 1984 release, was superior to Tallarook but with five other lines produced more than 1000 seeds/m2 resulting in more than 300 plants/m2. The performance of a wide range of Australian cultivars, including the latest releases, have been examined by Widdup and Pennell (2000) growing in the South Island of New Zealand in a somewhat drought prone soil and a moderate rainfall of 500 –640 mm with sheep grazing. The new Australian cultivars, particularly the late flowering Denmark and Leura and also a North Island

324

M. L. SMETHAM

selection Ak 948 had improved seed set, seedling regeneration and herbage yield compared to older cultivars. The amount of seed set in the first season by recent releases and by material selected in New Zealand was extremely high, and was often much higher than that set by the older named cultivars. Denmark and Leura were both derived from Sardinian ecotypes, and it is significant that all successful cultivars were small leaved and dense, previously identified as a pre-requisite for success under the hard grazing practised in New Zealand. A low percentage of hardseed was produced in the year when this was measured, probably a result of the moist conditions during seed maturation.

C. ESTABLISHMENT 1.

Initial Rate of Sowing

Recent extension advice for pure stands in the eastern Australian wheat belt (Dear and Sandral, 1997) recommends 7 kg ha on dry land and 10 kg ha for irrigated country. In Western Australia the recommended sowing rate after wheat has been to sow 9– 13 kg ha of seed (Rossiter and Ozanne, 1970). In New Zealand 2– 4 kg ha of subterranean clover has been used for oversowing onto hill country with an existing grass cover, or for sowing into cultivated ground together with a grass or grasses.(Levy, 1970). In the absence of data concerning optimum seed rates of subterranean clover, it is appropriate to consider the problem from a theoretical viewpoint. As already mentioned, the minimum desirable seedling population for pure swards after germination is at least 1000 seedlings/m2 (Taylor et al., 1991; Cocks, 1974). Machine harvesting of legume seeds normally substantially reduces hardseededness due to abrasion in the thrashing process. As a result germination is likely to be at least 65% and perhaps as high as 85%. With seed weights between 6 and 9 g/1000 and a germination of 85%, the theoretical seed rate for a pure stand to achieve minimum seedling density in the year of sowing is therefore between 70 and 100 kg ha. However, in practise this is too expensive, although dairy farmers on the east coast of Australia, and in Western Australia sometimes sow 20– 25 kg ha. The time taken by a pure stand of subterranean clover to achieve minimum seedling density for success from a low initial seed rate is, however, likely to be quite short because at low density plants set more seed per plant than at high density (Donald, 1954). Thus a sowing rate of 25 kg ha produced 149– 305 seedlings/m2 and set 247 –345 kg ha seed in the first season (Dear et al., 1993). The initial seedling numbers are well below requirements, but seed production was quite adequate for the next year. There is no record for the second season, but between 2000 and 6000 seedlings/m2 were counted in the third. When the aim is to establish a pure stand of subterranean clover, it would therefore seem desirable to sow at least 25 kg ha of seed, while recognising that

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)325

the sowing of a lesser amount will lead to a delay in the achievement of threshold seed production and seedling numbers for success. Populations of subterranean clover obtained by aerial oversowing in New Zealand, traditionally sown at 2 – 4 kg ha, have been considered too low to be effective by Smetham (1980). He proposed an arbitrary optimum population for this purpose of 35 plants/m2 in a sward with grass. This was obtained by sowing 9 kg ha of seed, although population did increase in linear fashion up to the maximum seed rate used of 36 kg ha. Since the 1980s, problems have been experienced with establishment of subterranean clover by under sowing in the last wheat crop of the rotation, particularly in New South Wales. Sheep have proved to be less profitable than wheat, and in an attempt to maximise profitability farmers have been reluctant to reduce the seeding rate of the last wheat crop in the cropping phase of the rotation. In addition, clover seed sown directly behind the wheat coulters tends to be sown too deeply for good emergence. These conditions have led to a slower than usual build up in clover plant numbers (Dear, 1988). Alterations to equipment and a halving of the crop sowing rate or a doubling of clover sowing rate to 7 kg ha, can correct this problem (Dear, 1988). However, in addition the period in wheat has been extended from two or three crops to four or five, which has exacerbated the problem, since the varieties of subterranean clover used (“Woogenellup” and “Seaton Park”) do not have particularly high levels of hardseededness to allow them to persist through a series of crops without resowing. Selection of varieties for use in the wheatbelt has now placed more emphasis on hardseededness (Nichols, 1987).

2. Management to Ensure Re-establishment Severe competition from perennial grasses or lucerne can have a major effect on seedling survival during natural autumn re-establishment. Dear and Cocks (1997) found that by the end of March only 57% of subterranean clover seedlings had survived when growing with phalaris (summer inactive), 13% with a danthonia, and only 1% with cocksfoot—both summer active. By mid-May, only 2 seedlings/m2 survived in any of the perennial companion plant plots, compared with 964 seedlings/m2 in the pure subterranean clover. A decline in the moisture content of the surface soil horizons indicated that competition for moisture was involved. Subsequent work (Dear et al., 1998) indicated that competition for light and nitrate as well as water, is involved when perennials are growing with establishing subterranean clover seedlings. It is therefore vital for the agricultural success of this plant that competition is minimised during natural seedling re-establishment so that an adequate population survives. In practise, this has been achieved in New Zealand by thinning out about

326

M. L. SMETHAM

one-third of the perennial grass tillers using light cultivation with grubber, discs or heavy harrows, followed by rolling, in mid-summer every 2 – 3 years (Calder, 1954), or by very heavy grazing in the weeks just prior to the commencement of germination. Future productivity is assisted by spelling the germinating stand from grazing for some 3 –5 weeks, so allowing leaf area to build up (Silsbury and Fukai, 1977). This helps the clover compete with other vegetation, and also enhances production due to the compounding effect of leaf area on cool season growth (Hoglund and Pennell, 1989). Each year of subterranean clover growth increases soil N (Watson and Lapins, 1964; Dear et al., 1999) and therefore associated grass competitiveness. This can be minimised by 2 or 3 years of cropping to reduce N levels (Watson and Lapins, 1964). However, even if soil N levels are only moderate, but the grazing of a mixed subterranean clover is lax and infrequent, then seed production, and the reestablishment of subterranean clover is likely to be low (Dear et al., 1998, 2000) and it will eventually disappear. This problem is particularly hard for hill country farmers to deal with. Typically farmers on such country (Emmersen, 1980) obtain the necessary degree of top growth control by rotating hard, clean-up grazing around a small number of paddocks in summer and autumn in any 1 year, so that adequate cover removal is achieved at least once in 3 years. It is possible to use a selective grass-killing herbicide to achieve control, and Smetham (1980) obtained a sixfold increase of seedlings establishing in a herbage mass of 3100 kg DM ha after spraying. Subsequent seed production showed a fourfold advantage over no spraying.

D. SEED PRODUCTION 1.

Threshold Values for Seed Production

Carter and Cochrane (1985) propose that for the continued agricultural success of subterranean clover, a minimum reserve of 200 kg ha seed should be maintained in the soil. Rossiter (1966) quotes a considerably higher figure of 600 kg ha but this was to enable a variety to survive in competition with nonleguminous sward components. More recently Dear et al. (1993) maintain that seed reserves of early mid-season flowering cultivars in areas receiving 500– 600 mm annual rainfall with rain in summer, should be at least 500 kg ha. Seedling densities of at least 1000 plants/m2 (Taylor et al., 1991; Silsbury and Fukai, 1977) and up to 5000 plants/m2 after full germination in the autumn (Taylor et al., 1984) are suggested as being necessary for highest winter and total DM production. If germination is 100%, then with a seed weight of 7 g/1000, 70 kg ha of seed will be required to give 1000 seedlings/m2. However, in Australia most subterranean clover cultivars still retain a degree of hardseededness at the break in the autumn. This ranges from up to 45% in early

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)327

cultivars, e.g., Dalkeith; 35% in early mid-season types, e.g., Seaton Park; and 15% in late varieties like Mount Barker (Collins et al., 1984). When allowance is made for this, the minimum amounts of seed required just before germination in the autumn are therefore 127, 108, and 82 kg ha for early, mid- and late-season flowering cultivars, respectively. Substantial loss of seed reserves can occur when seedlings are killed by drought in a false strike (Rossiter, 1966; Taylor, 1972; Taylor et al., 1984). The latter authors recorded a mean loss of 24% over 12 periods of regeneration. In New Zealand hill country receiving some summer rain, losses from false strike ranged from 4 to 55% over three seasons (Dodd et al., 1995a). If such losses are assumed to be at least 25%, the minimum amount of seed required needs to be increased to 168, 144, and 109 kg ha. Up to 83% of seed can be consumed when sheep are grazed on pastures after the herbage has dried off (de Koning and Carter, 1989). If it is assumed that 50% of seed is lost in this way, then the theoretical minimum amounts of seed which need to be set become 336 kg ha for early, 288 kg ha for mid-season, and 218 kg ha for late-flowering cultivars. It should be noted that the above calculations take no account of hardseed which may carry over from one season to the next, and no allowance has been made for seed quality, diseases or pests. Since insect predation has not been widely reported, it is therefore not allowed for in the amendment of the amount of seed required for success.

2.

Effect of Grazing on Seed Production

Over the last 20 years in Australia there has been a growing awareness that many subterranean clover pastures were not as productive as they once were. Among the several factors blamed for the decline of productivity has been the increase of stock pressure on pastures, particularly those grown as part of a cropping rotation (Carter et al., 1982). Sheep have been shown to progressively deplete by ingestion seed reserves of subterranean clover during the grazing of dried-off herbage in summer and autumn (de Koning and Carter, 1989), but grazing during flowering and seed maturation has a major impact also. Several workers (Collins, 1978, 1981; Collins et al., 1983; Rossiter, 1972) have recorded an increase in seed yield in response to defoliation, or defoliation and grazing (Rossiter, 1961) or grazing alone (Rossiter and Park, 1972; Bolland, 1987) up to first flower appearance. However, grazing (Rossiter, 1961) or defoliation (Collins, 1978; Collins et al., 1983) during flowering decreased seed yield. Increases in yield with defoliation before flowering were not, however, obtained with spaced plants (Rossiter, 1972) or sparse swards (Walton, 1975) because subsequent flower numbers were reduced. Collins et al. (1978) obtained a large reduction in seed yield correlating with a reduction in inflorescence numbers

328

M. L. SMETHAM

where swards were shaded during flowering. Insufficient light penetrating the canopy of dense swards to allow maximum floral initiation to occur is the probable explanation. Rossiter (1972) counted 17% more flowers as a result of defoliation during the early part of the flowering period, but more flowers still, and 27% more seed where grazing and defoliation were stopped before flowering commenced. Subterranean clover seems to respond to grazing in similar fashion to white clover, the stolons of which branch to give more sites for inflorescence initiation (Beinhart, 1962) and more flowers at axilliary sites (Zaleski, 1964) when light intensity at stolon, i.e., ground level, is high. This hypothesis tends to be confirmed by Hagon (1973) who obtained more seed from defoliation, but only where the LAI of the sward before defoliation exceeded that for full light interception. There is little information on the effect of grazing throughout the whole period of flowering. Conlan et al. (1994) obtained a reduction of seed yield with some varieties, particularly “Clare”, but in others there was no effect. The grazing imposed was, however, intermittent and in the most severe treatments still left a leaf area index (LAI) of 2.7– 3.6 which would still have enabled full light interception (Fukai and Silsbury, 1976), and is unlikely to have unduly stressed plants. Reductions of 30 – 50% in seed yield were obtained by Young et al. (1994) when grazing was continued through to pod formation, but here again grazing was periodic and at long intervals. The most significant finding of this study was that spring DM yields accounted for 90% of the variation in subsequent seed yields. One of the few investigations where grazing was continuous during the flowering period (Smetham and Dear, unpublished results) demonstrated a large reduction in seed following continuous severe grazing. Stocking so that 1405 kg DM ha of herbage mass was sustained during grazing (equivalent to an LAI of 0.5, or roughly half a layer of leaf per unit ground area) representing very severe defoliation, allowed only 71 kg ha seed to be produced. Maintaining a herbage mass of 1613 kg DM ha (or an LAI of 1.5, equivalent to leaving roughly one and one-half layers of leaf per unit ground area), resulted in significantly more seed being produced (324 kg ha), while no grazing gave a seed yield of 1254 kg ha. More florets were produced when the sward was grazed, but at the severe level of defoliation relatively more flowers were consumed by grazing than at the intermediate level. Nevertheless, the study showed that moderately severe continuous grazing to leave a sustained residual herbage mass of around 1600 kg DM ha can be practised during flowering, and that this will still allow sufficient seed to be produced for the stand to be agriculturally successful.

3.

Other Factors Affecting Seed Production

Seed production is reduced by competition with both grass and legume perennials when these are grown with subterranean clover. Although water in

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)329

spring was not a factor affecting seed yields, these were strongly correlated with the amount of light reaching the clover component, resulting in a 50 – 75% reduction of seed yield at the highest density of 35– 40 perennial plants/m2 (Dear et al., 2000). Even so, seed yields were with one exception all at or above 400 kg ha and so above the minimum for success proposed above. In addition, although seedling numbers were depressed by competition, they generally exceeded 750 seedlings/m2 counted at the highest perennial density, and were often above 1000 seedlings/m2, so exceeding the minimum suggested for maximum DM production. Under the moderately dry (580 mm year annual rainfall) conditions at Wagga Wagga, New South Wales, the use of herbicides has been shown to increase stomatal resistance and reduce leaf area (Dear et al., 1996) causing a watersaving effect and resulting in up to twice as much seed being produced. However, at wetter sites seed production was either not affected or depressed by up to 66% (Sandral et al., 1995). Although trace element deficiencies are widely recognised in Australia (Williams and Andrews, 1970), at that date there was little mention of B deficiency. However, Dear and Lipsett (1987) have subsequently obtained 5 – 21fold increases in seed production on acid cropping soils deficient in B, although foliage analysis did not indicate a shortage.

E. IMPACT

OF

HARDSEEDEDNESS

ON

PERFORMANCE

The high summer temperatures and considerable diurnal amplitude of temperature variation of a Mediterranean climate causes the hardseededness developed during seed maturation to decline to a lower level by autumn (Quinliven and Millington, 1962; Collins et al., 1984; and Table I.), leading to substantial germination at this time. Such conditions exist in Western Australia with February mean daily temperatures of 18 –278C and a diurnal range of up to 178C (Rossiter and Ozanne, 1970), and mean extremes up to 488C (Gentilli, 1971). Summer temperatures are much the same in South Australia, and southeastern Australia west of the main divide on the slopes and plains. However, on the eastern tablelands, and in much of Tasmania, temperatures are not as high, with a mean daily maximum around 208C; and a diurnal variation of about 138C (Gentilli, 1971). As a result the rate of hardseed breakdown is reduced, so that in the autumn hardseededness is still quite high and populations of seedlings are very much below optimum (Fitzgerald, 1987; Evans and Hall, 1995). This is further compromised by the grass often grown as a companion species with subterranean clover in these areas, due to the extra insulating effect of this additional herbage cover. The problem is even more acute in New Zealand with mean daily summer maximum temperatures of 16 –198C and a diurnal range of 9 –108C although

330

M. L. SMETHAM

Central Otago, possessing a more continental climate, has an increased range of nearly 138C (Gentilli, 1971). Smetham and Wu Ying (1991) consider this to be a major impediment to the success of subterranean clover in New Zealand, although the protection against loss of seedlings frequently recorded in summer (Sheath and Macfarlane, 1990a) afforded by hardseed is still very much required. In a major study of subterranean clover cultivar performance on North Island hill country (Williams et al., 1990) a typical result was a population of 107 seedlings/ m2 in March from 1540 seeds/m2 counted in January (Sheath et al., 1990). Unfortunately no measurements of hardseed were made, so that the losses from false strike cannot be separated from the lack of germination due to hardseededness, although in another study 30– 50% of seedlings appeared to die from drought (Sheath and Macfarlane, 1990b). Smetham (1980) has recorded 76% hardseededness in the autumn in early varieties, and 62 –72% in mid-season varieties from dry hill country, and calculations using the figures of Hoglund (1990) for cultivars grown in the same area gave much the same range of hardseededness (Smetham and Wu Ying, 1991). The area experiences monthly mean maximum of 20 – 258C in mid- to latesummer, with minimum of 9 – 138C. These temperatures are well below those promoting rapid loss of hardseededness (Quinliven and Millington, 1962) and resulted in low populations of 285 seedlings/m2 from 403 seeds (31% hard), to only 120 seedlings from 1494 seeds (92 % hard; Smetham and Wu Ying, 1991). Seed from the same field experiment, subjected to a range of day/night temperatures in growth cabinets gave up to 57% germination after only 1 month of 508C day/58C night temperatures (Smetham and Jack, unpublished results), thus reinforcing the hypothesis that low field germination of subterranean clover in New Zealand is frequently due to a slow rate of hardseed breakdown as a result of low summer temperatures and a small diurnal range. An evaluation of 47 numbered accessions of subterranean clover in a dry South Island, New Zealand environment (Smetham et al., 1993) also showed that high levels of hardseed were still extant at the end of the summer. In mid-autumn twenty-two lines were 75 – 100% hard, while 37 had hardseededness above 50%. However, whilst these levels were high, the amounts of seed formed by some lines were also exceptionally high. Consequently, 14 lines germinated to give more than 1000 seedlings/m2 and could therefore be judged to be successful in the agricultural sense.

F. HARDSEEDEDNESS CARRYOVER A commonly held view is that hardseededness in annual legumes ensures that there will always be seed for next year if the majority of this year’s seed production is lost. However, whilst this may ensure biological survival, it does

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)331

not guarantee agricultural success, when this is defined as the production of at least 1000 seedlings/m2 in autumn. There are two reasons why this may be so. Firstly, only small percentages of seed soften to germinate in later years, particularly if they are buried, since this slows the rate of softening, although this does vary with variety. Taylor (1984) found that most hardseeds of “Yarloop” germinated within 3 years at any depth of burial, while few seeds of “Northam” did so. Whereas 92% of the seed set by four subterranean clover varieties germinated in the first autumn, only 6% (or 12 kg ha seed) did so in the second autumn, giving only 193 seedlings/m2 (Donald, 1959): quite insufficient for agricultural success, but more than enough for biological survival. Secondly, there is some evidence (Chapman and Anderson, 1987) that the viability of hardseed does not remain high for long. Using white clover, they found that of seed which was originally 75% hard, only 25% remained 7 months after burial. 25% was recovered and 22% had germinated, but 53% was lost to unknown causes. A half-life of 1 year was postulated, and few seeds were found to have survived for more than 2 years. Taylor (1984) found some evidence for the microbial decomposition of hardseed buried in soil, with a mean loss of 3%, although in one batch, 49% of seeds died after 1 year buried at 6 cm. Considerable loss of seed was also recorded by Rossiter (1966), who was able to account for only 35– 44% of the seed sown, as having germinated or was still hard after 5 years in the field. This constitutes considerable mortality. On purely theoretical grounds the environment to which an embryo is subjected inside a hardseed is not one which is likely to ensure a high rate of survival. The seed coat has limited permiability for oxygen; no permeability for water, and may, if on the surface of the ground, be subjected to near, or above temperatures normally lethal for any plants (Levitt, 1956). At Griffith, New South Wales the bare soil surface mean maximum temperature at 1 p.m. for 1 week’s observations in mid-January was 51.28C (Quinliven, 1961), while mean weekly maximum temperatures at the soil surface were 55 – 598C during February and March for a site near Perth, Australia (Smith et al., 1996). With increasing depth of burial, however, survival of hardseed is progressively greater, and whilst Taylor (1984) found that on the surface little hardseed remained hard for long, some varieties buried at 10 cm survived for 3 years or more. Nevertheless, the apparent losses of seed from a hill site in Canterbury, New Zealand tend to support the hypothesis that hardseed may not survive as well as has been supposed. For instance, less than 50% of seeds were present in February 1985 when compared to the 1984 figures (Hoglund, 1990) yet seed harvested in 1983 and 1984 had a range of hardseededness from 80 to 98% which should have ensured a higher survival. Such apparent losses of hardseed, possibly due to disease or degradation need investigation.

332

M. L. SMETHAM

IV. PRODUCTIVITY A. HERBAGE PRODUCTION 1.

Total Production

Differences between cultivars for date of first flower appearance has implications for herbage production since leaf production slows dramatically once flowering starts, and means that late flowering varieties produce more herbage dry matter than those flowering earlier (Rossiter, 1959) although this can only occur if the moisture regime allows potential growth to be realised. Hence, in an environment which is moist enough to support late season flowering varieties, herbage production is directly related to time of first flower appearance. Evans (1993) for instance found that yield increased by 43 kg DM ha for every day later flowering occurred. In eastern Australia and with a long-term average 530 mm annual rainfall, but extended season, two early cultivars—“Nungarin” (110 d from germination to first flower at Wagga Wagga, New South Wales) and “Dalkeith” (120 d) gave 4.1 and 4.6 t DM ha, respectively, whereas the later varieties “Seaton Park” (125 d) and “Mount Barker” (156 d) gave considerably higher yields of 11.5 and 10.5 t DM ha (Dear et al., 1993, Dear and Sandral, 1997). In this situation early varieties were clearly unable to take advantage of the extended season. Scott (1971) found the same in New Zealand with a progressive increase in yield from the earliest to the latest flowering varieties at the site where rainfall allowed this to occur. Likewise with the latest cultivars from Australia Widdup and Pennell (2000) obtained June to November herbage production ranging from 2500 kg DM ha for early flowering types to 7300 kg DM ha for the late flowering Denmark, with a steady gradation between these, since at this site rainfall and moisture were adequate even for the late varieties. In the same way, the potential for yield is strongly influenced by the length of the growing season (Silsbury and Fukai, 1978), which necessarily equates with autumn, winter and spring – early summer rainfall. Blumenthal and Ison (1993) recorded an average over three seasons of 6000 kg DM ha at Forbes with rainfall of 280 –301 mm and additional water to prolong the growth. These authors obtained a very strong correlation (r ¼ 0.99) between water use and DM yield. Increased use of water gave increases of production up to 11 t DM ha from 600 mm water used. At Wagga Wagga with a long-term average 530 mm rainfall, Dear and Loveland (1984) obtained yields up to 10.3 t DM ha from a mid-season variety. In Western Australia irrigation during the growing season with 440 mm of water after the first 30 mm gave a linear relationship of 30 kg DM ha for each millimeter of water applied with an upper limit to yield of 12000 kg DM ha (Bolger and Turner, 1999). In New Zealand, total production of subterranean clover-grass

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)333

pasture is closely related to spring rainfall, and to spring herbage growth which generally constitutes two-thirds of total yield (Iversen, 1957). Herbage yield is also affected by seedling population (Taylor et al., 1991). Cocks (1974) has recorded a maximum total yield of 16.4 t DM ha, finding yield strongly dependant on plant density and leaf area, but Silsbury and Fukai (1978) measured growth rate increasing with LAI up to a herbage mass of 2000 kg ha but decreasing beyond 6000 kg ha as respiratory load increased. Calculated yields (Cocks, 1974) rose from 7.7 t DM ha with a sustained LAI of 2, to 13.2 t DM ha where LAI was maintained at LAI 5. Herbage yield is also affected by the date of germination in the autumn (Silsbury and Fukai, 1977) but while winter growth was maximised by early germination, total yield was highest from a May sowing. Almost without exception the mid-season flowering “Mount Barker”, and the late season flowering “Tallarook” have been the varieties extant in New Zealand hill country (Suckling et al., 1983) and on flat to rolling terrain (Saxby, 1956). On steep North Island (NZ) hill country Ledgard et al. (1987) have measured total sward productivity of 5– 6400 kg DM ha with subterranean clover contributing 1200– 1500 kg DM ha. Use of “Mount Barker” and “Tallarook” on stony free draining soils of the Canterbury Plains in New Zealand has resulted in long term average total sward yields of 6000 kg DM ha, with subterranean clover contributing 1300 kg (23%) of this (Rickard and Radcliffe, 1976). While this legume is very seldom used in New Zealand as a pure sward, various authors have shown pure swards to be highly productive, and possibly more so than where associated with a grass or grasses. Two years of results under periodic cutting and grazing (Harris et al., 1973) gave annual yields from 1000 kg DM ha for early flowering varieties to 4400 kg DM ha for mid- and late-season types, however, there are no New Zealand comparisons of a pure sward with a mixed grass clover sward. Under a climate and on soils almost identical to those used by Rickard and Radcliffe (1976) and Smetham and Jack (1995) obtained production from a large number of lines sown as pure swards, and rotationally grazed by sheep, the best three of which exceeded total production of 5000 kg DM ha under hard periodic grazing. Scott (1971) harvested 7900– 8900 kg DM ha from two early midseason cultivars cut three times in a 550 mm rainfall area of the South Island, New Zealand, while Suckling (1960) recorded 11,000 – 13,000 kg DM ha from late cultivars cut monthly in a 1500 mm rainfall area of the North Island, NZ. Overall, on the evidence presented earlier, subterranean clover is often a more productive option used as a pure sward, but more comparative data is needed to confirm this. Clearly, unless herbicides are used, this is not an issue on steep hill country where grasses will be an existing basic component of the sward. Under hard monthly grazing Smetham and Jack (1995) found total production from the three best subterranean clover lines averaged 5100 kg DM ha compared to 3310 kg DM ha from a 2-year-old stand of lucerne cv. “Grasslands Otaio”. However, in this particular year a very dry late spring – early summer

334

M. L. SMETHAM

disadvantaged dry land lucerne productivity. Normally such lucerne could be expected to yield around 8500 kg DM ha per annum (Vartha and Fraser, 1978). Nevertheless the substantial production from subterranean clover makes this species a viable alternative for use in dry land without the more demanding fertility requirements of lucerne. In areas of Australia with soils of pH 5.2– 7.5, having good drainage, and a rainfall of more than 350– 400 mm year, lucerne can be a highly productive pasture option (McDonald and Waterhouse, 1989) except that little herbage is produced over winter. The sowing of subterranean clover with the lucerne can correct this imbalance (Fitzgerald, 1979; Wolfe and Sutherland, 1980). Where suitable conditions occur a proportion of the wheat belt in both eastern and western Australia is sown to this mixture, the two species of which give additive and complimentary production (Wolfe and Sutherland, 1980) with good persistence as long as the lucerne is sown at a wide row spacing.

2. Seasonal Production Being an autumn germinating annual, the amount of herbage produced by subterranean clover in each of autumn, winter and spring is dependant both on the timing of germination, and the numbers of seedlings per unit area. Silsbury and Fukai (1977) found that winter growth was maximised by early germination. Early autumn applications of N to pasture, have been shown to increase leaf area, leading to increased winter and spring DM yields (Hoglund and Pennell, 1989). Similarly the early germination of subterranean clover results in leaf area which builds up early and quickly, before the cooler and shorter days of early winter, and this leads to higher winter growth rates than where sowing or germination is later (Blumenthal and Ison, 1993). Swards sown in mid-April or late May give little winter production compared to those sown in early March (Dear and Loveland, 1984). At latitude 438S in southern New Zealand, Smetham and Jack (1995) recorded substantial subterranean clover growth rates from early April germination and seedling numbers in excess of 1000 m22 of 16 – 19 kg DM ha d in autumn, 12 kg DM ha d over June and July, and 17 – 54 kg DM ha d from spring onwards. Monthly mean day temperatures for these periods were 9 – 12, 5 –6, and 8 –108C, respectively (Smetham and Jack, 1995). These growth rates were substantially higher than those from lucerne over the cool season in the same experiment, and equal to or better than those for winter active perennial grasses.

B. HERBAGE QUALITY The mature dry herbage of subterranean clover can be quite adequate as a maintenance feed for sheep, having a digestibility of DM of 35– 60% depending

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)335

on variety and exposure to rain (Rossiter et al., 1994), and the dried off herbage and associated burr can feed up to 6 dry sheep equivalents per ha (de Koning and Carter, 1989) depending on the previous green herbage productivity of the stand. Summer rain decreased dry matter digestibility by 5 –16%. Unlike most legumes, the digestibility of leaf is equal to or less than that of petiole and stem (Rossiter et al., 1994; Stockdale, 1992) due to a higher ratio of lignin to cell wall material in the leaf (Mulholland et al., 1996). In addition petiole and stem fractions had a higher level of water soluble carbohydrate than did leaf. These attributes were highly correlated (r ¼ 0.70) with animal performance in a grazing experiment, although it was not until the second year that significant differences in digestibility and liveweight gain between cultivars developed. Performance by cows grazed leniently was not as good as it was where hard grazing meant that a greater proportion of their intake was stem and petiole (Stockdale, 1992). These findings are questioned by Ru and Fortune (2000) who found no difference in the DMD between the plant parts of 26 cultivars in August, although significant differences were measured after the cessation of flowering when leaves were of higher quality than stem or petiole. However, close examination of the results shows that three cultivars did have stem and petiole fractions with significantly higher DMD than leaf in August, while 16 of the 26 cultivars had non-significantly higher DMD for stem and petiole than leaf. There are also differences in the energy required to shear the dried off herbage which are related to voluntary intake (Baker et al., 1993) and while this is to some extent dependant on the ratios of stem and petiole to leaf, there are marked varietal differences too. Differences in biomechanical properties are considered by Ridsill-Smith et al. (1994) to have contributed significantly to differences in the eating rate of sheep on six cultivars of subterranean clover. Stockdale (1992) reports that irrigated pure subterranean clover is an excellent feed for dairy cows, which produced up to 28 kg milk/cow in early lactation although he produced no comparative data for other legumes.

C. USE AS A PURE SWARD OR IN MIXTURE WITH GRASS The question of whether subterranean clover confers most benefit as a pure sward or mixed with a grass—either annual or perennial, needs to take account of many considerations. However, in most of temperate Western Australia, and the drier parts of the wheatbelt in eastern Australia the summer climate is too severe for perennials to survive (Moore, 1970). Elsewhere in Australia subterranean clover is used in mixture with perennial grasses where the climate allows these to survive. In eastern Australia where 35 –60% of rain falls in summer (Gentilli, 1971), grasses in mixture with subterranean clover are used, depending on total rainfall

336

M. L. SMETHAM

(Moore, 1970). Phalaris, and more recent selections of Dactylis can be used where this is 450– 630 mm year. However, the survival of some of these dryland adapted cultivars for example, Phalaris aquatica cv. Sirocco is dependant on a degree of summer dormancy, hence productivity may not be significant over this period (Cooper and Tainton, 1968) Below 450 mm in Victoria and South Australia, the annual ryegrass L. rigidum Gaud. is used in combination with subterranean clover (Donald, 1970). Above 630 mm perennial ryegrasses can survive for use in mixed pastures (Moore, 1970). In New Zealand subterranean clover is almost invariably used in a mixture, which includes perennial ryegrass (Lolium spp.) and sometimes cocksfoot (Dactylis spp.), The contribution of subterranean clover to the yield of these mixed pastures in New Zealand; up to 1500 kg DM ha in hill pasture (Ledgard et al., 1987) or 1300 kg DM ha in flat dryland pasture on stony soils (Rickard and Radcliffe, 1976), is not particularly impressive, whereas experimental sowings of pure subterranean clover have given substantial yields of 8000 – 13,000 kg DM ha (Suckling, 1960; Scott, 1971). In addition, a pure sward of subterranean clover will give cool-season growth rates as high as any perennial grass (Smetham and Jack, 1995). However, a pure sward of subterranean clover will give little production of green herbage from mid-summer until rains occur in autumn. Whilst stock can exist over summer at a maintenance feeding level on dried off herbage and burr at stocking rates of up to 6 dry sheep equivalents per ha (de Koning and Carter, 1989), provision of a higher quality crop or greenfeeds is needed to allow adequate flushing of ewes before joining with the ram. The inclusion of dryland persistent perennial grasses with summer production capability in mixture with subterranean clover in New Zealand, for example “Grasslands Wana” cocksfoot (Barker et al., 1985), or “Grasslands Maru” (Stevens et al., 1989), or as already noted as being practiced in Australia (Moore, 1970), will provide grazing in summer, but overall herbage quality will be lower. Although the digestibility of grasses and clovers under grazing may be similar, the greater intake of clover herbage when grazed as a pure sward will lead to higher growth rates of stock (Ulyatt et al., 1977). Consequently animal growth rates are likely to be greater on pure, than on mixed grass-clover swards. However, pure subterranean clover swards are seen in Australia as responsible for the acidification associated with legume nitrogen fixation (Ridley et al., 1990), and, being shallow-rooted, as doing little to solve problems of rising water tables and salinisation (Williams, 1991). It is now widely recognised (Dear, personal communication) that by growing deeper rooting perennials, including lucerne and trees, with subterranean clover can minimise or prevent acidification (Porter, 1981), and salinisation (Schofield, 1989) by minimising the downward movement of water through the soil profile. Although subterranean clover may be sown as a pure sward, in practice these are rapidly invaded by volunteer annual broadleaved weeds like capeweed (Arctotheca spp.), and annual grasses, such as Vulpia spp. Lolium spp. and Bromus spp. Such stands are impossible to maintain

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)337

in a pure state without the extensive use of herbicides (Dear, personal communication). However, this is especially so when an absence of competition for establishing weeds occurs as a result of a low subterranean clover population. This can occur in mixed grass and subterranean clover pastures as a result of the insulating effect of continuing herbage cover over the summer period (Quinliven and Millington, 1962), and or as a result of lower summer temperatures in some locations. Both can cause hardseededness to be lost only slowly (Fitzgerald, 1987), resulting in clover germination levels often marginal for success. In addition, the purposeful use of grass with subterranean clover or its association with weed grasses like Vulpia spp. can lead to problems with allellopathy. Residues from Phalaris have been shown to significantly reduce germination and aspects of root growth in subterranean clover seedlings, with many of the Australian cultivars being severely affected, compared to lines from grassy environments in the Mediterranean (Leigh et al., 1995). Breeding strategies to capture resistance are predicted to lead to an improvement in new releases. Where climate allows perennial grasses to survive, the use of these with subterranean clover is likely to severely restrict its ability to perform to potential. In the absence of special circumstances which demand the use of a grass with subterranean clover, the strategy with greatest benefit to a pastoral system may on balance be to use subterranean clover as a pure stand.

V. PESTS AND DISEASES A. VIRUSES Viruses affecting subterranean clover have been reviewed by Johnstone and Barbetti (1987) and only brief comment about them will be made here. In summarising the situation, Johnstone (1987) made the following points. Virus diseases considered important at one time or another in Australia have been alfalfa mosaic virus, bean yellow mosaic virus, clover yellow vein virus, cucumber mosaic virus, soybean dwarf virus (also known as subterranean clover red leaf virus), subterranean mottle virus and subterranean clover stunt virus. Since the mid-1950s when virus suddenly emerged as a problem in Australia, the relative prevalence of the various viruses has continually changed. Occurrence of virus epidemics has been sporadic and is generally localised in importance. Nevertheless losses of herbage can be 50% and more. Two main classes of virus can be recognised. The first group are persistent viruses spread by aphid vectors from host plant species. These include subterranean clover stunt virus and subterranean clover red leaf virus. All 1987 registered subterranean clover varieties were susceptable. The second group of viruses are spread in a non-persistent manner, and are

338

M. L. SMETHAM

mostly transmitted through the seed of at least some of their hosts. These are alfalfa mosaic virus, cucumber mosaic virus, and bean yellow mosaic virus. These have increased in incidence since the arrival in Australia in the late 1970s of the pea aphid Acyrthosiphon pisum. Losses from the two most important virus diseases of subterranean clover in Australia: cucumber mosaic and bean yellow mosaic viruses, spread by aphid vectors, depend very much on the survival of infected seedlings and the rate of spread of infection, and were found to differ greatly from season to season (Jones, 1993). Herbage yield reduction ranged from 7 to 49% while seed yields were reduced by 14 –64%. There is wide variation in subterranean clovers in reaction to the viruses; the vectors, and to transmission via seed, and there is evidence that good tolerance exists in some lines held in the subterranean clover section of the Trifolium Gene Bank operated by the Department of Agriculture, Western Australia, Perth, which could allow resistant lines to be selected. In New Zealand Ashby (1980) has surveyed viruses of annual legumes, and notes stunting of subterranean clover from bean yellow mosaic virus; a wide incidence of alfalfa mosaic virus with blue green aphid as vector; and the prevalence of subterranean clover red leaf virus and aphid vector Aulacorthum solani (syn. Acyrthosiphon solani; Ferro, 1978) overwintering on white clover spreading to cause infection of subterranean clover in spring.

B. FOLIAR FUNGAL DISEASES In Australia the most prevalent and damaging fungous disease is Clover Scorch (Kabatiella caulivora ), which causes necrosis mainly at the junction of petiole and lamina followed by collapse of the leaf and the browning off of the whole canopy. The disease is worst, and spread most rapid, in pure swards closed for seed or hay, and growing on heavy soils. It can be so devastating that screening of introductions and crossbreds for susceptability or otherwise has been a priority in the Australian Subterranean Clover Improvement Programme. There are wide differences in tolerance or susceptability to disease amongst varieties, which allows selection of resistant lines (Barbetti and Gillespie, 1987). Clover scorch occurs on red clover hay crops in New Zealand at wet sites, but is not recorded on subterranean clover (Close, 1990), probably because this species is almost always grown on free-draining soils, and seldom if ever saved as a pure sward for hay. Rust disease (Uromyces trifolii ) is sporadic and can be severe locally in Australia, and has been found on New Zealand stands of subterranean clover (Close, 1990). Screening for this problem has been routinely carried out in subterranean clover improvement programmes in Australia (Barbetti and Gillespie, 1987). These authors also list a number of other foliar diseases, including Cercospora zebrina which can be serious. In a review of pasture diseases occurring in New Zealand, Close (1990) likewise mentions a number of fungus diseases affecting clovers but few were noted on subterranean clover with

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)339

the exception of C. zebrina, which causes reddish leaf spotting. Close (1990) cautions that many of the leaf spotting diseases cause the host to produce oestrogenic substances potentially detrimental to animals.

C. ROOT FUNGAL DISEASES Several organisms can be responsible for root rots in subterranean clover including Pythium, Fusarium, Rhizoctonia, Aphanomyces, and Phytophthora clandestina. However, it is mostly the latter that is involved in serious field infections (Greenhalgh and Flett, 1987). Losses can range from 30 to 70% and are worst when moisture stress with warm temperatures follows an early break, or after irrigation. The disease is sufficiently severe in some seasons in Australia for screening for resistance to this organism to be another essential preliminary in the selection of new varieties of subterranean clover (Flett et al., 1993). Lines have shown great variation in susceptability, but with 22% of the 800 tested up until 1988 being resistant. More than 95% of T. sub. var. yanninicum lines tested were resistant, with the exception of “Yarloop”. Whilst “Larisa” is highly resistant, the older cultivars “Woogenellup”, “Mt Barker”, “Northam” and “Tallarook” are highly susceptable. In New Zealand root rots do not seem to cause problems with subterranean clover, although the crown rot Sclerotinia trifoliorum has been recorded on this species (Close, 1990).

D. INSECTS Commenting on insect pests of subterranean clover pastures in Western Australia, Sandow and Gillespie (1987) consider that although there are a variety of moth larvae, beetles, weevils, and springtails which do attack clover, the most damaging pests are the red legged-earth mite (Halotydeus destructor ), and the blue-green aphid (Acyrthosiphon kondoi ). The red-legged earth mite cause mechanical damage resulting from feeding lesions on established plants but more important is the very high mortality of seedlings which can result if attack occurs shortly after germination. This pest is only of concern in areas with winter rainfall and an absence of spring and summer rain (Wallace and Mahon, 1971). In areas of summer rain, the very similar blue oat mite is of equal importance in causing damage (Dear and Sandral, 1997). Screening for resistance to red-legged earth mite, which is widespread amongst lines, has been conducted as part of subterranean clover breeding in Australia since 1985. Henry et al. (1997) have shown that resistant cultivars have a significantly higher biomechanical strength than susceptable lines. The blue – green aphid debilitates plants by feeding activity which, if aphid populations are high enough, can cause plants to die off

340

M. L. SMETHAM

prematurely. However, more serious is the transmission of virus diseases which can occur, even with the low population of aphids which frequently overwinter on clover stands. To date screening trials have not been conclusive. Since blue –green aphids also arrived in New Zealand in the late 1970s, it is likely that they may be involved in virus spread and reduction of plant vigour in that country also (Ferro, 1978). Pasture cockchafer (Aphodius tasmaniae ), known as Tasmanian grassgrub in New Zealand, can in some seasons cause localised sporadic damage to subterranean clover in winter (Chapman, 1990), although in New Zealand it appears to be restricted to the driest shallow, stony, and sandy soils. Both adult beetles and larvae feed on the aerial parts of clover plants but larvae are more damaging since they feed until pupation in July (Ferro, 1978). Whitefringed weevil (Graphognathus leucoloma ) larvae, feeding on roots and nodules are reported as causing localised and sporadic damage in New Zealand (Chapman, 1990), while the lucerne flea (Sminthurus viridis ) is classified as localised and persistent.

E. OESTROGENICITY Depending on variety, subterranean clover can cause major problems with lambing and calving. The ingestion of phyto-oestrogenic substances by breeding sheep or cattle over the weeks prior to mating as a result of grazing subterranean clover leaf, or to a lesser extent by consuming hay, can cause temporary infertility, reducing lambing and calving to as little as 8%. There is however no danger from naturally dried off herbage (Davies and Dudzinski, 1965). The symptoms are a general disturbance of reproductive function, uterine prolapse, and dystokia (Bennetts et al., 1946). Thinning of the mucous strands leads to the inability of sperm to migrate through the cervix to achieve fertilisation of the ovum (Bennetts et al., 1946). Prolonged exposure causes permanent infertility following encystment of the uterine wall (Collins and Cox, 1985). The problem is caused by the isoflavones—formononetin, biochanin A, and genistein, of which formononetin is considered to be the principal causal compound (Millington et al., 1964a,b). The level of isoflavones in the herbage is increased by low soil fertility (Rossiter and Beck, 1966), and by leaf diseases (Jagusch, 1982). The level of oestrogen in subterranean clovers is a varietal characteristic (Davies and Dudzinski, 1965). The early-flowering variety “Dwalganup”, widely used in the 300– 500 mm rainfall areas of Australia since the early 1930s, particularly in Western Australia, possesses high levels of oestrogens. Although associated crop yields were good, animal problems were serious. “Yarloop” and “Dinninup” are also classified as having high levels of isoflavones. “Northam”, “Geraldton”, “Woogenellup” and “Clare” have intermediate levels, while “Daliak”, “Bacchus Marsh” and “Mount Barker” have low levels. In Australia testing for oestrogen level was an established first priority in the selection process during the evaluation of new introductions of subterranean

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)341

clovers (Francis et al., 1970). Those chosen for further evaluation were allowed no more than 0.2% DM formononetin in the leaf (Nichols, 1987). On the basis of formononetin content, rather than total isoflavenes, lines tend to segregate on the basis of subspecies (Piano et al., 1993).

ACKNOWLEDGMENTS The author wishes to thank Dr Warwick Harris, of Lincoln, Canterbury, New Zealand, and Dr Brian S. Dear, Senior Research Scientist, New South Wales Agriculture, Wagga Wagga, Australia, for their valuable constructive criticism, and for supplying reference material.

REFERENCES Aitken, Y. (1939). The problem of hard seeds in subterranean clover. Proc. R Soc. Victoria 51, 187 –210. Aitken, Y. (1955). Factors affecting flower initiation in Trifolium subterraneum L. Aust. J. Agric. Res. 6, 212–244. Aitken, Y., and Drake, F. R. (1941). Studies on the varieties of subterranean clover. Proc. R Soc. Victoria 53, 342 –393. Archer, K. A. (1990). The effects of moisture supply and defoliation during flowering on seed production and hardseededness of T. subterraneum L. Aust. J. Expt. Agric. 30, 515– 522. Ashby, J. W. (1980). Virus diseases of annual legume crops. Proc. Agron. Soc. New Zealand 10, 77 –80. Baker, S. K., Klein, L., de Boer, E. S., and de Purser, D. B. (1993). Genotypes of dry mature subterranean clover differ in shear energy. In “Proceedings of the XVII International Grassland Congress. Palmerston North, New Zealand February 8–21, 1993”. (M. J. Baker, J. R. Crush and L. R. Humphreys, Eds.), pp. 592 –593. New Zealand Grassland Association, Palmerston North. Barbetti, M. J., and Gillespie, D. J. (1987). Fungal foliar diseases of subterranean clover. In “Proceedings of the 3rd National Workshop: National Subterranean Clover Improvement Programme. Wagga Wagga, New South Wales, 31 August–2 September 1987”. (B. S. Dear and W. J. Collins, Eds.), pp. 26–31. Department of Agriculture, New South Wales. Barker, D. J., Lancashire, J. A., and Meurk, C. (1985). “Grasslands Wana” cocksfoot—an improved grass suitable for hill country. Proc. NZ Grassland Assoc. 46, 167– 172. Barnard, C. (1972). In “Register of Australian Herbage Plant Cultivars”. CSIRO Division of Plant Industry, Canberra. Beinhart, G. (1962). Effects of temperature and light intensity on CO2 uptake, respiration and growth of white clover. Plant Physiol. 37, 709– 715. Bell, A. D. (1991). Seed morphology. In “Plant Form. An Illustrated Guide to Flowering Plant Morphology”, pp. 158–159. Oxford University Press, London. Bennetts, H. W., Underwood, E. J., and Shier, L. F. (1946). The effect of oestrogenic subterranean clovers on reproduction in sheep. Aust. Vet. J. 22, 2–8. Blumenthal, M. J., and Ison, R. L. (1993). Water use and productivity in subterranean clover and murex medic swards. 1. Dry matter production. Aust. J. Agric. Res. 44, 89 –107.

342

M. L. SMETHAM

Bolger, T. P., and Turner, N. C. (1999). Water use efficiency and water use of Mediterranean annual pastures in Southern Australia. Aust. J. Agric. Res. 50, 1035–1046. Bolland, M. D. A. (1987). Seed production of Trifolium subterraneum spp. subterraneum and brachycalycinum as influenced by soil type and grazing. Aust. J. Expt. Agric. 27, 539–559. Bolland, M. D. A., and Collins, W. J. (1986). Effect of burr burial on seed production of Trifolium subterraneum subspp. brachycalycinum and other annual legumes. Aust. J. Expt. Agric. 26, 59–64. Calder, J. W. (1954). Agricultural Bulletin 301. In “Subterranean Clover Pastures on Light Land”. Canterbury Chamber of Commerce, Christchurch, New Zealand. Carter, E. D., and Cochrane, M. J. (1985). The poor subterranean clover status of dairy pastures in the Adelaide hills. In “Proceedings of the 3rd Australian Agronomy Conference, University of Tasmania, Hobart Jan/Feb 1985”. (J. J. Yates, Ed.), pp. 217–219. Australian Society of Agronomy, Parkdale, Victoria. Carter, E. D., Wolfe, E. C., and Francis, C. M. (1982). Problems of maintaining pastures in the cereallivestock areas of southern Australia. Proceedings of the 2nd Australian Agronomy Conference, Wagga Wagga, July 1982. (M. J. T. Norman, Ed.), pp. 68–82. Australian Society of Agronomy, Parkdale, Victoria. Chapman, R. B. (1990). Insect Pests. In “Pastures; Their Ecology and Management”. (R. H. M. Langer, Ed.), pp. 448 –467. Oxford University Press, Auckland. Chapman, D. F., and Anderson, C. B. (1987). Natural reseeding and Trifolium repens demography in grazed hill pastures. 1. Flowerhead appearance and fate, and seed dynamics. J. Appl. Ecol. 24, 1025–1035. Chapman, D. F., and Williams, W. M. (1990). Evaluation of clovers in dry hill country 8. Subterranean clover at Ballantrae. NZ J. Agric. Res. 33, 569–576. Chapman, D. F., Sheath, G. W., Macfarlane, M. J., Rumball, P. J., Cooper, B. M., Crouchley, G., Hoglund, J. H., and Widdup, K. C. (1986). Performance of subterranean and white clover varieties in dry hill country. Proc. NZ Grassland Assoc. 48, 53– 62. Close, R. C (1990). Pasture plant diseases. In “Pastures; Their Ecology and Management” (R. H. M. Langer, Ed.), pp. 468– 490. Oxford University Press, Auckland. Cocks, P. S. (1994). Potential production of grass and clover monocultures in a Mediterranean-type environment—an experimental approach. Aust. J. Agric. Res. 25, 835 –846. Cocks, P. S. (1992). Evolution in sown populations of subterranean clover (Trifolium subterraneum L.) in South Australia. Aust. J. Agric. Res. 43, 1383–1395. Cocks, P. S., and Phillips, J. R. (1979). Evolution of subterranean clover in South Australia. I. The strains and their distribution. Aust. J. Agric. Res. 30, 1035–1052. Collins, W. J. (1978). The effect of defoliation on inflorescence production, seed yield, and hardseededness in swards of subterranean clover. Aust. J. Agric. Res. 29, 789 –801. Collins, W. J. (1981). The effects of length of growing season with and without defoliation on seed yield and hardseededness in swards of subterranean clover. Aust. J. Agric. Res. 32, 783–792. Collins, W. J. (1987). Structure and organisation of the National Subteranaean Clover Improvement Programme. Proceedings of the 3rd National Workshop: National Subterranean Clover Improvement Programme. Wagga Wagga, New South Wales, 31 August–2 September 1987. (B. S. Dear and W. J. Collins, Eds.), pp. 1 –4. Department of Agriculture, New South Wales. Collins, W. J., and Cox, R. I. (1985). Oestrogenic activity in forage legumes. Proceedings of a Trilateral Workshop Forage Legumes for Energy Efficient Annual Production. 30 April–4 May 1984. Palmerston North, New Zealand. (R. F. Barnes, P. R. Ball, R. W. Brougham, G. T. Marten, and E. J. Minson, Eds.), pp. 268 –276. USDA/ARS, Palmerston North, New Zealand. Collins, W. J., and Quinliven, B. J. (1980). The effects of a continuous water supply during and beyond seed development on seed production and losses in subterranean clover swards. Aust. J. Agric. Res. 31, 287 –296.

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)343 Collins, W. J., Rossiter, R. C., Monreal, R. C., and Ramos, A. (1978). The influence of shading on seed yield in subterranean clover. Aust. J. Agric. Res. 29, 1167– 1175. Collins, W. J., Rhodes, I., Rossiter, R. C., and Palmer, M. J. (1983). The effect of defoliation on seed yield of two strains of subterranean clover grown in monoculture and in binary mixtures. Aust. J. Agric. Res. 34, 671–680. Collins, W. J., Francis, C. M., and Quinliven, B. J (1984). Registered cultivars of subterranean clover—their origin, identification and potential use in Western Australia. In “Western Australian Department of Agriculture Bulletin 4083.” Perth, Western Australia. Conlan, D. J., Dear, B. S., and Coombes, N. E. (1994). Effect of grazing intensity and number of grazing on herbage production and seed yield of Trifolium subterraneum, Medicago murex and Ornithopus compressus. Aust. J. Expt. Agric. 34, 181 –188. Cooper, J. P., and Tainton, N. M. (1968). Light and temperature requirements for growth of tropical and temperate grasses. Herb. Abstr. 38, 167–176. Corner, E. J. H. (1951). The leguminous seed. Phytomorphology 1, 117–150. Davies, J. G. (1952). Pastures in the Australian economy. J. Aust. Inst. Agric. Sci. 18, 60– 67. Davies, H. L., and Dudzinski, M. L. (1965). Further studies of oestrogenic activity in strains of subterranean clover (T. subterraneum L.) in south-western Australia. Aust. J. Agric. Res. 16, 937 –950. Dear, B. S. (1988). Establishing sub.clover and lucerne under cover crops. In “Information from NSW Agriculture and Fisheries 12.10.88”. (J. Sykes, Ed.), pp. 1–3. Government Printer, New South Wales. Dear, B. S., and Cocks, P. S. (1997). Effect of perennial pasture species on surface soil moisture and early growth and survival of subterranean clover (Trifolium subterraneum L.) seedlings. Aust. J. Agric. Res. 48, 683–693. Dear, B. S., and Lipsett, J. (1987). The effect of boron supply on the growth and seed production of subterranean clover (Trifolium subterraneum L.). Aust. J. Agric. Res. 38, 537–546. Dear, B. S., and Loveland, B. (1984). Herbage production and seed yield of defoliated subterranean clover swards in a cool temperate environment. Aust. J. Expt. Agric. 24, 543–549. Dear, B. S., and Sandral, G. A. (1997). In “Subterranean Clover in NSW—Identification and Use. Agfact P 2.5.16” (2nd edition). New South Wales Agriculture, Sydney. Dear, B. S., Conlan, D. J., Richards, M., and Coombes, N. E. (1992). Development of a new midseason cultivar of subterranean clover conforming to a pre-defined ideotype. In “Proceedings of the 6th Australian Society of Agronomy Conference, University of New England, Armidale 10 –14 February, 1992”. (K. J. Hutchinson and P. J. Vickery, Eds.), pp. 452–455. Australian Society of Agronomy, Parkville, Victoria. Dear, B. S., Cregan, P. D., and Murray, G. M. (1993). Comparison of the performance of subterranean clover cultivars in southern New South Wales. I. Persistence, productivity and seed yields. Aust. J. Agric. Res. 33, 581–590. Dear, B. S., Sandral, G. A., and Coombes, N. E. (1996). Changes in stomatal resistance and water use in subterranean clover in response to broadleaf herbicides. Aust. J. Agric. Res. 47, 625 –635. Dear, B. S., Cocks, P. S., Wolfe, E. C., and Collins, W. J. (1998). Established perennial grasses reduce growth of emerging subterranean clover seedlings through competition for water, light, and nutrients. Aust. J. Agric. Res. 49, 41–51. Dear, B. S., Cocks, P. S., Peoples, M. B., Swan, A. D., and Smith, A. B. (1999). Nitrogen fixation by subterranean clover (Trifolium subterraneum L.) growing in pure culture and in mixtures with varying densities of lucerne (Medicago sativa L.) or phalaris (Phalaris aquatica L.). Aust. J. Agric. Res. 50, 1047–1058. Dear, B. S., Cocks, P. S., Swan, A. D., Wolfe, E. C., and Ayre, L. M. (2000). Effect of phalaris (Phalaris aquatica L.) and lucerne (Medicago sativa L.) density on seed yield and regeneration of subterranean clover (Trifolium subterraneum L.). Aust. J. Agric. Res. 51, 267–278.

344

M. L. SMETHAM

Dodd, M. B., Sheath, G. W., and Richardson, S. (1995a). Development of subterranean clover (Trifolium subterraneum L.) genotypes for New Zealand pastures. 1. Whatawhata persistence evaluation. NZ J. Agric. Res. 38, 33 –47. Dodd, M. B., Sheath, G. W., and Richardson, S. (1995b). Development of subterranean clover (Trifolium subterraneum L.) genotypes for New Zealand pastures. 2. Wairakei persistence evaluation. J. Agric. Res. 38, 49 –56. Donald, C. M. (1954). Competition among pasture plants. II. Influence of density on the flowering and seed production of annual pasture plants. Aust. J. Agric. Res. 5, 585–597. Donald, C. M. (1959). The production and life span of seed of subterranean clover. Aust. J. Agric. Res. 10, 771 –787. Donald, C. M. (1960). The influence of climatic factors on the distribution of subterranean clover in Australia. Herb. Abstr. 30, 81–90. Donald, C. M. (1970). Temperate pasture species. In “Australian Grasslands”. (R. M. Moore, Ed.), pp. 303 –320. Australian National University Press, Canberra. Emmersen, R. S. (1980). Managing and developing dry hill country. Proc. Lincoln College Farmers’ Conf. 29, 347–357. Evans, L. T. (1959). Flower initiation in Trifolium subterraneum L. 1. Analysis of the partial processes involved. Aust. J. Agric. Res. 10, 1 –16. Evans, P. M. (1993). A subterranean clover ideotype for a cool temperate environment. In “Proceedings of the XVII International Grassland Congress. Palmerston North, New Zealand February 8–21, 1993”. (M. J. Baker, J. R. Crush and L. R. Humphreys, Eds.), pp. 254 –255. New Zealand Grassland Association, Palmerston North. Evans, P. M., and Hall, E. J. (1995). Seed softening patterns for single seed crops of subterranean clover (Trifolium subterraneum L.) in a cool temperate environment. Aust. J. Expt. Agric. 35, 1117–1121. Ferro, D. N. (1978). Pasture pests in New Zealand. In “Insect Pests”. (D. N. Ferro, Ed.), (2nd edition), pp. 105 –127. Lincoln University College of Agriculture, Canterbury, New Zealand. Fitzgerald, R. D. (1979). A comparison of four pasture types for the wheat belt of southern New South Wales. Aust. J. Expt. Agric. Anim. Husb. 19, 216–224. Fitzgerald, R. D. (1987). Subterranean clover for Northern New South Wales. 1. Special requirements for a summer dominant rainfall. In “Proceedings of the 3rd National Workshop: National Subterranean Clover Improvement Programme. Wagga Wagga, New South Wales, 31 August–2 September 1987”. (B. S. Dear and W. J. Collins, Eds.), pp. 116–119. Department of Agriculture, New South Wales. Flett, S. P., Taylor, P. A., and Wakley, V. (1993). A method for the glasshouse evaluation of subterranean clover breeding lines for Phytophthora clandestina root rot. In “Proceedings of the XVII International Grassland Congress. Palmerston North, New Zealand February 8–21, 1993”. (M. J. Baker, J. R. Crush and L. R. Humphreys, Eds.), pp. 961–962. New Zealand Grassland Association, Palmerston North. Francis, C. M., and Gladstones, J. S. (1974). Relationships among rate and duration of flowering and seed yield components in subterranean clover (Trifolium subterraneum L.). Aust. J. Agric. Res. 25, 435 –442. Francis, C. M., Gladstones, J. S., and Stern, W. R. (1970). Selection of new subterranean clover cultivars in south-western Australia. In “Proceedings of the XI International Grassland Congress. Surfers Paradise, Queensland, Australia 13 – 23 April 1970”. (M. J. T. Norman, Ed.), pp. 214 –218. International Grassland Congress, Brisbane, Australia. Francis, C. M., Quinliven, B. J., and Nicol, H. I. (1972). Variation in burr burial in subterranean clover. Aust. J. Agric. Res. 23, 605 –610. Fukai, S., and Silsbury, J. H. (1976). Responses of subterranean clover communities to temperature. I. DM production and plant morphogenesis. Aust. J. Plant Physiol. 3, 527–543. Garnier, B. J. (1958). In “The Climate of New Zealand”. Arnold, London.

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)345 Gentilli, J. (1971). The main climatological elements. Climates of Australia and New Zealand. World Survey of Climatology 13. (J. Gentilli, Ed.), pp. 119– 188. Elsevier, New York. Gillespie, D. J., and Nicholas, D. A. (1991). In “Pasture legume recommendations for sowing in 1991. Bulletin 4204”. Western Australian Department of Agriculture, Perth. Gladstones, J. S. (1966). Naturalised subterranean clover (Trifolium subterraneum L.) in Western Australia: The strains, their distributions, characteristics, and possible origins. Aust. J. Bot. 14, 329 –354. Gladstones, J. S. (1987). Measurement of high temperature seed dormancy in subterranean clover. J. Aust. Inst. Agric. Sci. 53, 99 –101. Greenhalgh, F. C., and Flett, S. (1987). Recent research on root diseases of subterranean clover in Victoria and its implications. In “Proceedings of the 3rd National Workshop: National Subterranean Clover Improvement Programme. Wagga Wagga, New South Wales, 31 August– 2 September 1987”. (B. S. Dear and W. J. Collins, Eds.), pp. 40–42. Department of Agriculture, New South Wales. Hagon, M. W. (1971). The action of temperature fluctuations on hard seeds of subterranean clover. Aust. J. Expt. Agric. Anim. Husb. 11, 440–443. Hagon, M. W. (1973). The effect of defoliation on flowering and seed yield of subterranean clover. Aust. J. Agric. Res. 24, 211–217. Hagon, M. W., and Ballard, L. A. T. (1970). Reversibility of strophiolar permiability to water in seeds of subterranean clover (Trifolium subterraneum L.). Aust. J. Biol. Sci. 23, 519–528. Hamly, D. H. (1932). Softening of seeds of Melilotus alba. Bot. Gazette 4, 345–375. Harris, A. J., Brown, K. R., Turner, J. D., Johnston, J. M., Ryan, D. L., and Hickey, M. J. (1973). Some factors affecting pasture growth in Southland. NZ J. Expt. Agric. 1, 139 –163. Henry, D.A., Baker, S.K., and Ridsill-Smith, T.J. (1997). Implications of biomechanical characters of subterranean clover on feeding by red-legged earth mite and intake by sheep. In “Proceedings of the XVIII International Grassland Congress, Winnipeg, Manitoba, Saskatoon, Saskatchewan, Canada, 8 –19 June 1997”, pp. 879–881. International Grasslands Congress, Winnipeg, Canada. Hoglund, J. H. (1990). Evaluation of clovers in dry hill country. 1. Subterranean clover in North Canterbury, New Zealand. NZ J. Agric. Res. 3, 585–590. Hoglund, J. A., and Pennell, C. G. L. (1989). Autumn nitrogen fertilizer in a dryland sheep system. Proc. NZ Grassland Assoc. 50, 135 –138. Hyde, E. O. C. (1954). The function of the hilum in some Papilionaceae in relation to the ripening of the seed and the permiability of the testa. Ann. Bot (Lond.) 18, 241–256. Iversen, C. E. (1957). Agricultural Bulletin 332. In “Pastures for Light Land”. Canterbury Chamber of Commerce, Christchurch, New Zealand. Jagusch, K. T. (1982). Nutrition of ruminants grazing lucerne. In “Lucerne for the 80’s: Agronomy Society of New Zealand Special Publication No. 1”. (R. B. Wynn-Williams, Ed.), pp. 73–78. Agronomy Society of New Zealand, Christchurch, NZ. Johnstone, G. R. (1987). Overview of virus disease problems affecting subterranean clover in Australia. In “Proceedings of the 3rd National Workshop: National Subterranean Clover Improvement Programme. Wagga Wagga, New South Wales, 31 August–2 September 1987”. (B. S. Dear and W. J. Collins, Eds.), pp. 16 –19. Department of Agriculture, New South Wales. Johnstone, G. R., and Barbetti, M. J. (1987). Impact of fungal and virus diseases on pastures. In “Production and Utilisation of Pasture in Southern Australia”. (J. L. Wheeler, C. J. Pearson and G. E. Robards, Eds.), pp. 25–39. CSIRO, Melbourne. Jones, R. A. C. (1993). Effects of three viruses on the productivity of annual pasture legume swards. In “Proceedings of the XVII International Grassland Congress. Palmerston North, New Zealand February 8–21 1993”. (M. J. Baker, J. R. Crush and L. R. Humphreys, Eds.), pp. 937–938. New Zealand Grassland Association, Palmerston North. Katznelson, J. (1970). Edaphic factors in the distribution of subterranean clover in the Mediterranean region. In “Proceedings of the XI International Grassland Congress, Surfers Paradise,

346

M. L. SMETHAM

Queensland, Australia 13–23 April, 1970”. (M. J. T. Norman, Ed.), pp. 192–196. International Grasslands Congress, Brisbane, Queensland, Australia. de Koning, C. T., and Carter, E. D. (1989). Losses of subterranean clover seed from dry pasture residues during grazing by sheep in summer–autumn. In “Proceedings of the 5th Australian Agronomy Conference, Perth, 24 –29 September 1989”. (C. M. Francis, Ed.), pp. 408– 410. Australian Society of Agronomy, Parkville, Victoria. Ledgard, S. F., Brier, G. J., and Littler, R. A. (1987). Legume production and nitrogen fixation in hill pasture communities. NZ J. Agric. Res. 30, 413–424. Leigh, J. H., Halsall, D. M., Muller, W. J., Oram, R. N., Edlington, J. P., and Greissler, A. E. (1995). The role of allellopathy in legume decline in pastures. III. Tolerance of subterranean clover lines to allellopathic compounds released from Phalaris residues. Aust. J. Agric. Res. 46, 1039– 1049. Levitt, J. (1956). In “The Hardiness of Plants”. Academic Press, New York. Levy, E. B. (1970). In “Grasslands of New Zealand”. Government Printer, Wellington, NZ. Levy, E. B., and Gorman, L. W. (1936). Strain in subterranean clover. Proc. NZ Grassland Assoc. 5, 19–33. Leys, A. R., Dowling, P. M., Plater, B., and Smetham, M. L. (1993). Long term management of vulpia in pastures. In “Proceedings of the XVII International Grassland Congress. Palmerston North, New Zealand February 8– 21 1993”. (M. J. Baker, J. R. Crush and L. R. Humphreys, Eds.), pp. 879 –880. New Zealand Grassland Association, Palmerston North. Loftus Hills, K. (1944). Dormancy and hardseededness in Trifolium subterraneum L. 2. The progress of after harvest ripening. J. Council Ind. Sci Res. 17, 186 –190. Macfarlane, M. J., and Sheath, G. W. (1984). Clovers-what type for dry hill country? Proc. NZ Grassland Assoc. 45, 140 –151. Macfarlane, M. J., Sheath, G. W., and Tucker, M. A. (1990). Clovers for dry hill country. 6. Subterranean and white clovers at Wairakei, New Zealand. NZ J. Agric. Res. 33, 557 –564. Marshall, T., and Millington, A. J. (1967). Flooding tolerance of some Western Australian pasture legumes. Aust. J. Expt. Agric. Anim. Husb. 7, 367 –371. McDonald, W. J., and Waterhouse, D. B. (1989). In “Lucerne for Pasture and Fodder. Agfact P2.2.25” (1st edition). New South Wales Agriculture, Tamworth. Millington, A. J., Francis, C. M., and McKeown, N. R. (1964a). Weather bioassay of annual pasture legumes. 1. Oestrogenic activity in Medicago tribuloides DESR. var. Cyprus relative to four strains of Trifolium subterraneum L. Aust. J. Agric. Res. 15, 520–526. Millington, A. J., Francis, C. M., and McKeown, N. R. (1964b). Weather bioassay of annual pasture legumes. II. The oestrogenic activity of nine strains of Trifolium subterraneum L. Aust. J. Agric. Res. 15, 527 –536. Mitchell, K. J. (1956a). Growth of pasture species under a controlled environment. 1. Growth at various levels of constant temperature. NZ J. Sci. Tech. 38A, 203 –216. Mitchell, K. J. (1956b). Growth of pasture species in a controlled environment. III. White clover (Trifolium repens L.), subterranean clover (Trifolium subterraneum L.) and Lotus major (Lotus uliginosus ). NZ J. Sci. Tech. 38A, 395–413. Moore, R. M. (1970). South eastern temperate woodlands and grasslands. In “Australian Grasslands”. (R. M. Moore, Ed.), pp. 169 –190. Australian National University Press, Canberra. Morley, F. H. W. (1961). Subterranean clover. Adv. Agron. 13, 57–123. Morley, F. H. W., and Davern, C. I. (1956). Flowering time in subterranean clover. Aust. J. Agric. Res. 7, 388– 400. Mulholland, J. G., Nandra, K. S., Scott, G. B., Jones, A. W., and Coombes, N. E. (1996). Nutritive value of subterranean clover in a temperate environment. Aust. J. Agric. Res. 36, 803–814. Nichols, P. G. H. (1987). Early generation selection criteria. In “Proceedings of the 3rd National Workshop: National Subterranean Clover Improvement Programme. Wagga Wagga, New South Wales, 31 August– 2 September 1987”. (B. S. Dear and W. J. Collins, Eds.), pp. 5– 11. Department of Agriculture, New South Wales.

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)347 Oram, R. N. (1990). In “Register of Australian Herbage Plant Cultivars”. CSIRO Australian Herbage Plant Registration Authority, Division of Plant Industry, Canberra. Piano, E., Tava, A., Spanu, F., and Pecetti, L. (1993). Variability for oestrogen content in large collections of subterranean clover and its implications for selection. In “Proceedings of the XVII International Grassland Congress. Palmerston North, New Zealand 8–21 February 1993”. (M. J. Baker, J. R. Crush and L. R. Humphreys, Eds.), pp. 1383–1384. New Zealand Grassland Association, Palmerston North. Porter, W. M. (1981). Soil acidification—the cause. In “The Acid Soils Affair: 10th Riverina Outlook Conference Riverina College of Advanced Education, Wagga Wagga, New South Wales”, pp. 31–46. Quinliven, B. J. (1961). The effect of constant and fluctuating temperatures on the permeability of hard seeds of some legume species. Aust. J. Agric. Res. 12, 1009–1022. Quinliven, B. J. (1965). The influence of the growing season and the following dry season on the hardseededness of subterranean clover in different environments. Aust. J. Agric. Res. 16, 277 –291. Quinliven, B. J. (1966). The relationship between temperature fluctuation and the softening of hard seeds of some legume species. Aust. J. Agric. Res. 17, 625– 631. Quinliven, B. J. (1971). Seed coat impermiability in legumes. J. Aust. Inst. Agric. Sci. 37, 283–295. Quinliven, B. J., and Millington, A. J. (1962). The effect of a Mediterranean summer environment on the permeability of hardseeds of subterranean clover. Aust. J. Agric. Res. 13, 377– 387. Rickard, D. S., and Radcliffe, J. E. (1976). Seasonal distribution of pasture production in New Zealand. 12. Winchmore, Canterbury Plains dryland and irrigated pastures. NZ J. Expt. Agric. 4, 329 –335. Ridley, A. M., Helyar, K. R., and Slattery, W. J. (1990). Soil acidification under subterranean clover (Trifolium subterraneum L.) pastures in north-eastern Victoria. Aust. J. Expt. Agric. 30, 195–201. Ridsill-Smith, T. J., Baker, S. K., and Dynes, R. A. (1994). Feeding by redlegged earth mite and sheep on six subterranean clover cultivars. Proc. Aust. Soc. Anim. Prod. 20, 417– 419. Rossiter, R. C. (1959). The influence of MG on total yield and seed production in strains of Trifolium subterraneum L. grown as single plants or in swards. Aust. J. Agric. Res. 10, 305–321. Rossiter, R. C. (1961). The influence of defoliation on the components of seed yield in swards of subterranean clover. Aust. J. Agric. Res. 12, 821 –833. Rossiter, R. C. (1966). The success or failure of strains of Trifolium subterraneum L. in a Mediterranean environment. Aust. J. Agric. Res. 17, 425–446. Rossiter, R. C. (1972). The effects of defoliation on flower production in subterranean clover. Aust. J. Agric. Res. 23, 427–435. Rossiter, R. C. (1978). The ecology of subterranean clover-based pastures. In “Plant Relations in Pastures”. (J. R. Wilson, Ed.), pp. 325– 339. CSIRO, Melbourne. Rossiter, R. C., and Beck, A. B. (1966). Physiological and ecological studies on the oestrogenic isoflavones in subterranean clover (Trifolium subterraneum L.). II. Effects of phosphate supply. Aust. J. Agric. Res. 17, 447–456. Rossiter, R. C., and Ozanne, P. G. (1970). South western temperate forests, woodlands and heaths. In “Australian Grasslands”. (R. M. Moore, Ed.), pp. 199 –218. Australian National University Press, Canberra. Rossiter, R. C., and Park, R. J. (1972). The effect of stocking rate and defoliation on relative seed production of a mix of two subterranean clover varieties. J. Aust. Inst. Agric. Sci. 38, 209–211. Rossiter, R. C., Taylor, G. B., and Klein, L. (1994). Environmental effects in particular of rainfall on the digestibility of dry mature subterranean clover. Aust. J. Expt. Agric. 34, 25– 32. Ru, Y. J., and Fortune, J. A. (1999). Effect of grazing intensity and cultivar on morphology, phenology and nutritive value of subterranean clover. 1. Morphology and phenology of subterranean clover during the growing season. Aust. J. Agric. Res. 50, 1191–1201.

348

M. L. SMETHAM

Ru, Y. J., and Fortune, J. A. (2000). Variation in the nutritive value of plant parts of subterranean clover (Trifolium subterraneum L.). Aust. J. Expp. Agric. 40, 397–403. Sandow, J. D., and Gillespie, D. J. (1987). Insects in legume pastures in Western Australia. In “Proceedings of the 3rd National Workshop: National Subterranean Clover Improvement Programme. Wagga Wagga, New South Wales, 31 August–2 September 1987”. (B. S. Dear and W. J. Collins, Eds.), pp. 43–46. Department of Agriculture, New South Wales. Sandral, G. A., Dear, B. S., and Coombes, N. E. (1995). Differential tolerance of Trifolium subterraneum L (subterranean clover) cultivars to broadleaf herbicides. 2. Seed yield and quality. Aust. J. Expt. Agric. 35, 475–482. Saxby, S. H. (1956). The history of subterranean clover in New Zealand. NZ J. Agric. 92, 518 –527. Schofield, N. J., Loh, I. C., Scott, P. R., Bartle, J. R., Ritson, P., Bell, R. W., Borg, H., Anson, B., and Moore, R. C. (1989). Agricultural strategies. In “Vegetation Strategies to Reduce Stream Salinities of Water Resource Catchments in South-west Western Australia”, pp. 50–56. Water Authority of Western Australia, Perth. Scott, W. R. (1971). An agronomic evaluation of subterranean clover cultivars. Proc. NZ Grassland Assoc. 33, 134–140. Sheath, G. W., and Macfarlane, M. J. (1990a). Evaluation of clovers in dry hill country. 3. Regeneration and production of subterranean clover at Whatawhata, New Zealand. J. Agric. Res. 33, 533 –539. Sheath, G. W., and Macfarlane, M. J. (1990b). Components of subterranean clover regeneration at Whatawhata, New Zealand. NZ J. Agric. Res. 33, 541– 547. Sheath, G. W., Macfarlane, M. J., and Crouchley, G. (1990). Evaluation of clovers in dry hill country. 7. Subterranean and white clovers at Poragihau, Hawkes Bay, New Zealand. NZ J. Agric. Res. 33, 565–568. Silsbury, J. H., and Fukai, S. (1977). Effects of sowing time and sowing density on the growth of subterranean clover at Adelaide. Aust. J. Agric. Res. 28, 427–440. Silsbury, J. H., and Fukai, S. (1978). A growth model for Trifolium subterraneum L. Aust. J. Agric. Res. 29, 51 –65. Silsbury, J. H., and Hancock, T. W. (1990). Growth responses of cultivars of subterranean clover to temperature, plant density and nitrate supply. Aust. J. Agric. Res. 41, 101–114. Smetham, M. L. (1977). Pasture legume species and strains. In “Pastures and Pasture Plants”. (R. H. M. Langer, Ed.), pp. 85 –127. A.H.&A.W. Reed, Wellington. Smetham, M. L. (1980). The establishment and management of subterranean clover and other annual legumes on the dry hill country of the South Island. Proc. Lincoln College Farmers’ Conf. 30, 326–346. Smetham, M. L., and Jack, D. W. (1995). Herbage production under grazing of some subterranean lines compared to lucerne. Proc. Agron. Soc. NZ 25, 69–76. Smetham, M. L., and Wu Ying, C. C. K. (1991). Establishment of subterranean clover (Trifolium subterraneum L.) in New Zealand. 1. Hardseededness and autumn germination. NZ J. Agric. Res. 34, 31– 44. Smetham, M. L., Hines, S., and Jack, D. W. (1993). Seed production and autumn germination as determinants of the success of subterranean clover in a cool temperate environment. In “Proceedings of the XVII International Grassland Congress. Palmerston North, New Zealand 8– 21 February 1993”. (M. J. Baker, J. R. Crush and L. R. Humphreys, Eds.), pp. 307– 309. New Zealand Grassland Association, Palmerston North. Smetham, M. L., Jack, D. W., and Hammond, S. E. H. (1994). The influence of patterns of flowering of some subterranean clover (Trifolium subterraneum L.) accessions and cultivars on total seed set and autumn germination in a cool temperate environment with sporadic summer rain. Proc. NZ Grassland Assoc. 56, 127 –131. Smith, D. (1970). Influence of temperature on yield and chemical composition of five forage legume species. Agron. J. 62, 520– 523.

A REVIEW OF SUBTERRANEAN CLOVER (TRIFOLIUM SUBTERRANEUM L.)349 Smith, F. P., Cocks, P. S., and Ewing, M. A. (1996). Short term patterns of seed softening in Trifolium subterraneum, Trifolium glomeratum and Medicago polymorpha. Aust. J. Agric. Res. 47, 775 –785. Stevens, D. R., Turner, J. D., Barker, D. J., and Moloney, S. (1989). “Grasslands Maru” phalaris: Productive and persistent in hill country. Proc. NZ Grassland Assoc. 50, 231– 236. Stockdale, C. R. (1992). The productivity of dairy cows fed irrigated subterranean clover herbage. Aust. J. Agric. Res. 43, 1281–1295. Suckling, F. E. T. (1952). Dissemination of white clover (Trifolium repens L.) by sheep. NZ. J. Sci. Tech. 35, 64–77. Suckling, F. E. T. (1960). Production of pure legume swards at Te Awa. NZ J. Agric. Res. 3, 579–591. Suckling, F. E. T., Forde, M. B., and Williams, W. M. (1983). Naturalised subterranean clovers in New Zealand. NZ J. Agric. Res. 26, 35–43. Symon, D. E. (1961). Commonwealth Bureau of Pastures and Field Crops, Hurley, Berkshire. Mimeograph Publication Number 1/1961. In “A Bibliography of Subterranean Clover Together with a Descriptive Introduction”. Commonwealth Bureau of Pastures and Crops, Hurley. Taylor, G. B. (1970). The germinability of soft seed of a number of strains of subterranean clover. Aust. J. Expt. Agric. Anim. Husb. 10, 293–297. Taylor, G. B. (1972). Time distribution of seedlings emergence from single seed crops of several annual legumes. Aust. J. Expt. Agric. Anim. Husb. 12, 628 –633. Taylor, G. B. (1981). Effect of constant temperature treatments followed by fluctuating temperatures on the softening of seeds of Trifolium subterraneum L. Aust. J. Plant Physiol. 8, 547–558. Taylor, G. B. (1984). Effect of burial on the softening of the hard seeds of subterranean clover. Aust. J. Agric. Res. 35, 201–210. Taylor, G. B., and Ewing, M. A. (1988). The effect of depth of burial on the longevity of hard seeds of subterranean clover and annual medics. Aust. J. Expt. Agric. 28, 77– 81. Taylor, G. B., and Ewing, M. A. (1996). Effects of extended (4–12 year) burial on seed softening in subterranean clover and annual medics. Aust. J. Expt. Agric. 36, 145 –150. Taylor, G. B., Rossiter, R. C., and Palmer, M. J. (1984). Long term patterns of seed softening and seedling establishment from single seed crops of subterranean clover. Aust. J. Expt. Agric. 24, 200 –212. Taylor, G. B., Maller, R. A., and Rossiter, R. C. (1991). A model describing the influence of hardseededness on the persistence of an annual legume, in a ley farming system, in a Mediterranean-type environment. Agric. Ecosyst. Environ. 37, 275– 301. Tennant, D. (1965). The differential rate of seed development in Dwalganup and Geraldton varieties of subterranean clover. Aust. J. Expt. Agric. Anim. Husb. 5, 46–48. Trumble, H. C. (1957). The climatic control of agriculture in South Australia. Trans. R. Soc. Aust. 61, 41 –62. Ulyatt, M. J., Lancashire, J. A., and Jones, W. T. (1977). The nutritive value of legumes. Proc. NZ Grassland Assoc. 38, 107–118. Vartha, E. W., and Fraser, T. J. (1978). Lucerne-Tama ryegrass systems for fat lamb production. NZ J. Expt. Agric. 6, 195– 200. Walker, M. G. (1971). Changes in germination promotion and inhibition in seed extracts of subterranean clover (Trifolium subterraneum L.) related to dormancy and germination. Aust. J. Biol. Sci. 24, 897–903. Wallace, M. M. H., and Mahon, J. A. (1971). The ecology of Sminthurus viridis Collembola. IV. The influence of climate and land use on its distribution and that of an important predator Bdellodes lapidaria (Acari: bdellidae). Aust. J. Zool. 19, 177–188. Walton, G. H. (1975). Response of burr burial in Subterranean clover (Trifolium subterraneum L.) to defoliation. Aust. J. Expt. Agric. Anim. Husb. 15, 69 –73. Watson, E. R., and Lapins, P. (1964). The influence of subterranean clover pastures on soil fertility. II. The effects of certain management systems. Aust. J. Agric. Res. 15, 885 –894.

350

M. L. SMETHAM

Widdup, K., and Pennell, C. (2000). Suitability of new subterranean clovers in the Canterbury region. Proc. NZ Grassland Assoc. 62, 161–165. Williams, B. (1991). Dryland salinity: An underview. In “Trees—the essential farm ingredient. 20th Riverina Outlook Conference, Wagga Wagga, 15 August 1991”. (J. E. Prattley, Ed.), pp. 57 –66. Charles Sturt University, Riverina. Williams, C. H., and Andrews, C. S. (1970). Mineral nutrition of pastures. In “Australian Grasslands”. (R. M. Moore, Ed.), pp. 321 –338. Australian National University Press, Canberra. Williams, W. M., Sheath, G. W., and Chapman, D. F. (1990). Evaluation of clovers in dry hill country. 1. General objectives and descriptions of sites and plant material. NZ J. Agric. Res. 33, 521–526. Wolfe, E. C., and Sutherland, O. R. (1980). Plant production and persistence in mixed pastures containing lucerne at a range of densities with subterranean clover and phalaris. Aust. J. Expt. Agric. Anim. Husb. 20, 189–196. Yates, J. J. (1957). Seed setting in subterranean clover (Trifolium subterraneum L.). 1. The importance of the microenvironment. Aust. J. Agric. Res. 8, 433 –443. Yates, J. J. (1958). Seed setting in subterranean clover. II. Strain environment interactions in single plants. Aust. J. Agric. Res. 9, 754– 766. Young, K. A. (1985). Seed dormancy. In “Annual Report: Official Seed Testing Station, New Zealand Ministry of Agriculture and Fisheries”, pp. 41–48. Government Printer, Wellington. Young, R. R., Morthorpe, K. J., Nicol, H. I., and Croft, P. H. (1994). Effect of sowing time and grazing on the dry matter yield, phenology, seed yield and hardseed levels of annual pasture legumes in western New South Wales. Aust. J. Expt. Agric. 34, 189 –204. Zaleski, A. (1964). The effect of density of plant population, photoperiod, temperature and light intensity on inflorescence formation in white clover. J. Br. Grassland Soc. 19, 239 –247. Zohary, M., and Heller, D. (1984). In “The Genus Trifolium”. Israel Academy of Sciences and Humanities, Jerusalem.

BREEDING HEVEA BRASILIENSIS FOR ENVIRONMENTAL CONSTRAINTS P. M. Priyadarshan Rubber Research Institute of India, Regional Station, Agartala 799 006, India

I. Introduction II. Growing Conditions A. Ideal Environments B. Marginal Areas III. Constraints A. Geo-climatic Stresses B. Biotic Stresses IV. Hevea Under Marginal Conditions A. Immature Phase B. Yield Depression, Patterns, Regimes and Specific Adaptation V. Breeding Programs A. Polyclonal Seedlings B. Recombination Breeding C. Increasing Genetic Diversity D. Molecular Breeding VI. Conclusions Acknowledgments References

Breeding research on Hevea brasiliensis under marginal areas worldwide is reviewed. Ideal and marginal environments are described together with geo-climatic and biotic stresses. The performance of rubber in immature and mature phases is presented with due emphasis on factors affecting yield depression and specific adaptation. The use of various breeding programs like evaluation of polyclonal seedlings, recombination breeding and integration of molecular diversity from both nuclear and cytoplasmic sources is presented. A special mention is made on allied species and their utility in evolving clones for areas with environmental constraints. The usefulness of molecular diversity, tissue specific gene expression and their categorization along with importance of molecular markers to breed Hevea for marginal areas is debated. Molecular linkage maps and their utility in mapping QTLs especially towards horizontal resistance to diseases are explained. The utility of direct gene transfer to increase genetic 351 Advances in Agronomy, Volume 79 Copyright q 2003 by Academic Press. All rights of reproduction in any form reserved 0065-2113/02$35.00

352

P. M. PRIYADARSHAN variation and expression of foreign genes in Hevea latex is briefly presented. q 2003 Academic Press.

I. INTRODUCTION Crop yield is a multichannel end point, influenced by several resources in the environment, wherein a fraction of the resources is captured by the crop, converted in to dry matter and partitioned to harvestable yield. In a new environment, the limitations imposed by both biological and physical hazards of the environment over the growth and yield of the crop will be significant and substantial, but varies with degree of tolerance/susceptibility of the crop. The detection, measurement and interpretation of this differential performance of genotypes in an environment and over the environments are challenging. Many genetic and physical attributes of the crop, viz., shoot and canopy architecture, stomatal resistance, turgour pressure, vascular structure, translocation, root structure, permeability and distribution, soil moisture content and depletion, diseases, insects, disasters, flowering and fruiting, vapour pressure, relative humidity, wind, temperature and photoperiod are few factors that influence phenotypic expression of yield (Fig. 1). The aforesaid factors control biological

Figure 1 A system analysis of phenotypic expression of yield in H. brasiliensis.

BREEDING HEVEA BRASILIENSIS

353

yield accumulation further bifurcated to economic yield and residual biological yield (Wallace and Yan, 1998). Varietal/clonal differences are evident in every plant species in relation to environment. In majority of the cases, the interpretation offered for this differential performance of genotypes is based on GE interaction studies reflected through stability analysis and adaptability. In a perennial crop species like rubber, where yield is retrieved throughout the year, the factors governing yield are intricate due to intrinsic attribute of latex production, since latex is the end product of several biochemical steps. Breeding for yield and secondary attributes in such species become challenging especially under areas with environmental constraints. Rubber is synthesised in over 2000 plant species confined to 300 genera of seven families, viz., Euphorbiaceae, Apocyanaceae, Asclepiadaceae, Asteraceae, Moraceae, Papaveraceae and Sapotaceae (Backhaus, 1985; Cornish et al., 1993; Heywood, 1978; John, 1992; Lewinsohn, 1991). At least two fungal species are also known to make natural rubber (Stewart et al., 1955). The para rubber tree [Hevea brasiliensis Willd. ex Adr. Juss. (Muell. Arg.)] of Euphorbiaceae is the chief contributor towards the natural rubber produced worldwide (Greek, 1991). Rubber is a hydrocarbon polymer constructed of isoprene units and natural rubber is a secondary metabolite (cis-1,4-polyisoprene) chiefly originating in the secondary phloem of the tree. No other synthetic substitute has comparable elasticity, resilience and resistance to high temperature (Davies, 1997). The genus Hevea has 10 species. An elaborate description of the taxonomical and botanical aspects of Hevea is out of scope of this article. Wycherley (1992) refers the readers to an excellent narration of the subject. However, an account of the salient features of different species of Hevea is given in Table I. Hevea species occur in Bolivia, Brazil, Colombia, French Guyana, Guyana, Peru, Surinam and Venezuela in its natural habitat. These countries need a special mention since they are around the centre of origin. All species except H. microphylla occur in Brazil; five species have been found in Colombia; four occur in Peru and Venezuela and two occur in Bolivia and Guayanas. H. guianensis is the widely adapted species. An alternate source of natural rubber, Guayule (Parthenium argentatum—Asteraceae), a shrub native to Chihuahuan desert of Texas provides 10% of the world’s natural rubber (George and Panikkar, 2000). Guayule can withstand temperature range of 2 18 to 498C and can grow in well drained soils with an annual rainfall of 230– 400 mm. The yield potential of guayule is only 600– 900 kg/ha (Estilai and Ray, 1991). However, guayule latex is useful in developing hypoallergenic latex products (Cornish and Siler, 1996). The first description of rubber was given by Columbus in 15th Century and the astronomer de la Condamine was the first to send samples of the elastic substance “caoutchouc” from Peru to France in 1736 with full details of habit and habitat of the trees and procedures for processing (Dijkman, 1951). History recapitulates names of five distinguished men: Clement Markham (of British India Office), Joseph Hooker (Director, Kew Botanic Gardens), Henry Wickham (Naturalist),

354

Table I Allied Species of the Genus Hevea—Occurrence and Features Species H. benthamiana Muell. Arg. H. brasiliensis (Willd. ex. A. de. Juss.) Muell. Arg.

Notable featuresa

North and west of Amazon forest basin, upper Orinoco basin (Brazil) South of Amazon river (Brazil, Bolivia, Ecuador, Peru)

Complete defoliation of leaves. Medium size tree. Habitat: swamp forests Complete defoliation of leaves. From medium to big tree size. Habitat: well drained soils Possibility of natural hybridisation. H. brasiliensis from 2 to 25 m tree height. Habitat: seasonally flooded swamps Retain old leaves until new leaves appear. Maximum 2 m tall. Habitat: dry savannahs

H. camargoana Pires

Restricted to Marajo island of Amazon river delta (Brazil)

H. camporum Ducke

South of Amazon between Marmelos and Manicore´ rivers tributaries of Madeira river Throughout the geographic range of the genus (Brazil, Venezuela, Bolivia, French Guyana, Peru, Colombia, Surinam, Ecuador)

H. guianensis Aublet

H. microphylla Ule

H. nitida Mart. ex. Muell. Arg.

H. pauciflora (Spr. ex Bth.) Muell. Arg.

Upper reaches of Negro river in Venezuela. It is not found in other region of geographic range of the genus Between the rivers Uaupes and Icana tributaries of the upper Negro river (Brazil, Peru, Colombia) North and west of Amazon river (Brazil, Guyana, Peru). Distribution discontinuous due to habitat preferences

Retain old leaves until new leaves and inflorescences appear. Grows at higher altitudes (1100 m MSL); medium size tree. Habitat: well drained soils Complete defoliation of leaves. Small trees. They live on flooded area (igapo´s). Habitat: sandy or lateritic soils Inflorescences appear when leaves are mature. Small to medium size trees (2 m). Habitat: quartzitic soils Retain old leaves until new leaves and inflorescences appear. No wintering. Small to big size trees. Habitat: well drained soils, rocky hill sides

P. M. PRIYADARSHAN

Occurrence

Table I (continued) Species

Occurrence Among Negro river and its affluents. Uaupes and Ic¸ana rivers (Brazil, Colombia, Venezuela)

H. spruceana (Bth.) Muell. Arg

Banks of Amazon, Rio Negro and lower Madeira (Brazil)

H. paludosa Uleb

Marshy areas of Iquitos, Peru

Retain old leaves even after inflorescences appear. Small tree from savannas. Sometime tall, with small crown on the topp. Habitat: well drained soils Retain old leaves until new leaves and inflorescences appear. Flowers reddish purple. Medium size tree. Habitat: muddy soils of islands Small leaflets, narrow and thin in the fertile branches;up to 30 m height. Habitat: marshy areas

After Wycherley (1992), Schultes (1970, 1977), Goncalves et al. (1990), Pires (1973) and Brazil (1971). a Wintering characteristics mentioned here has a bearing on the incidence of fungal diseases especially secondary leaf fall (Oidium ) since retention of older leaves may make the tree “oidium escape.” Dwarf types are desirable of the possible wind fastness. All species are diploid (2n ¼ 36) (Majumder, 1964), and are crossable among themselves (Clement-Demange et al., 2000). b Pires (1973) considered 11 species including H. paludosa; Brazil considers 11 species.

BREEDING HEVEA BRASILIENSIS

H. rigidifolia (Spr. ex Bth.) Muell. Arg.

Notable featuresa

355

356

P. M. PRIYADARSHAN

Henry Ridley (Director, Singapore Botanic Gardens), and R.M. Cross (Kew Gardner), with Kew Botanic Gardens playing the nucleus of rubber procurements and distribution. As per directions of Markham, Wickham collected 70,000 seeds from Rio Tapajoz region of Upper Amazon (Boim district) and transported it to Kew Botanic Gardens during June 1876 (Wycherly, 1968; Schultes, 1977; Baulkwill, 1989). Of the 2700 seeds germinated, 1911 were sent to Botanical Gardens, Ceylon during 1876 where 90% of them survived. During September 1877, 100 rubber plants specified as “Cross material” were sent to Ceylon. Earlier (during June 1877), 22 seedlings, not specified either as Wickham or Cross sent from Kew to Singapore, were distributed in Malaya which formed the prime source of 1000 tappable trees found by Ridley during 1888 (Baulkwill, 1989). Seedlings from Wickham collection of Ceylon were also distributed worldwide. Some how, the modern planting materials are believed to be derived from “Wickham genetic base.” There are reasons to believe that an admixture of Cross and Wickham materials were likely since 22 seedlings sent to Singapore during 1877 were unspecified (Baulkwill, 1989). The first large rubber estates came in to being in 1902 in Sumatra’s East Coast (Dijkman, 1951). At present, Thailand leads in rubber production followed by Indonesia, Malaysia, India, China, Sri Lanka, Vietnam, Nigeria, Cote d’Ivoire, Philippines, Cameroon, Cambodia, Liberia, Brazil, Myanmar, Bangladesh, Papua New Guinea, Ghana, Gabon, Guatemala and Zaire (Barlow, 1997). The Southeast Asian countries continue to dominate rubber production and trade accounting for more than 90% of the 6.74 million ton produced annually worldwide, most of which comes from Thailand, Indonesia, Malaysia and India. South America, the centre of origin, accounts for only 2% of world production primarily due to increased infestation of South American leaf blight (SALB—Microcyclus ulei P. Henn. von Arx. (Dean, 1987; Clement-Demange et al., 2000). World production was expected to exceed 7 million ton by 2001 (Cain, 2001).

II. GROWING CONDITIONS A. IDEAL ENVIRONMENTS H. brasiliensis is native to the rain forests of the Tropical region of the Great Amazonian basin of South America. Its flat land distinctly characterizes this area, between equator and 158 south with altitudes not exceeding 200 m with a wet equatorial climate (Strahler, 1969). The climate is characterised by a mean monthly temperature of 25 –288C and abundant rainfall of more than 2000 mm/ year (Pushparajah, 2001). The attributes ideal for rubber cultivation are (a) 2000 – 4000 mm rainfall distributed over 100 – 150 rainy days/annum

BREEDING HEVEA BRASILIENSIS

357

(Watson, 1989), (b) mean annual temperature of around ^ 288C with a diurnal variation of about 78C (Barry and Chorley, 1976), and (c) sunshine hours of about 2000 h/year at the rate of 6 h/day in all months (Ong et al., 1998). In a study with hydrothermal index, Rao et al. (1993) rationalised Senai of Malaysia (18360 N; 1038390 E) to be the most suitable for rubber cultivation and production. Amazon Basin is the largest area in the world with a typical equatorial climate with the rainfall exceeding 2000 mm, without any real dry season. Tropical temperatures (27 –328C) make the environment in Brazilian plateau a different one, where some of the areas are found to be ideal for rubber. However, in southern states, rubber is not a regular species. The increased global demand for rubber as also the extension in cultivation of other agricultural crops prompted the countries out side the hitherto traditional zone to focus their attention on the cultivation of rubber. Such a tendency often extended rubber to marginal soil and environmental conditions.

B. MARGINAL AREAS The mean annual temperature decreases when moved away from the equator with more prominent winter conditions, either during November to January (towards north) or June to August (towards south). Northeastern states of India, south China, north and northeast Thailand, North Vietnam, north Coˆte d’Ivoire and southern plateau of Brazil are well recognised as inhospitable for the crop, experiencing stress situations like low temperature, typhoons, dry periods and altitude (Priyadarshan et al., 2001; Zongdao and Yanqing, 1992; Hoa et al., 1998; Dea et al., 1997). It may also be worthwhile to note that rubber areas of China and Tripura fall under same latitude range, though climatic conditions in vivid pockets of China shall vary since its tropical and sub-tropical regions are undulating and diversified (Priyadarshan et al., 1998a). Similarly, southern plateau of Brazil (450 –500 m MSL) especially Sa˜o Paulo (238S) is being experimented for rubber cultivation (Costa et al., 2000). Brazil, being on the west of the Greenwich Meridian, offers entirely different climate for rubber inflicting considerable phenological changes. A geo-climatic comparison of various environments with India, China, Brazil, Malaysia, Vietnam, Indonesia, French Guiana, Thailand and Coˆte d’Ivoire would amply reveal a spectrum of climatic conditions over which rubber is being grown (Tables II and III) (Fig. 2). In India, marginal areas (non-traditional) are delineated as non-traditional zones spread over to the states of Maharashtra, Madhya Pradesh, Orissa, Tripura, Assam, West Bengal, Meghalaya and Mizoram. Similarly, east and northeast provinces of Thailand, central highlands of Vietnam and north Coˆte d’Ivoire are counted as non-traditional. Multitude of hazards, viz., moisture stress, low temperature, wind, high altitude and disease epidemics apart from altered soil physical factors

358

Table II Spectrum of Weather Variables Under Different Geo-Climates

Attributes

Longitude Altitude (m)

Pindorama (Sa˜o Paulo, Brazil)b

Kourou (French Guiana)a

Odienne (Cote d’Ivoire)b

Nong Khai (Thailand)b

Hainan (China)b

Agartala (Tripura, India)b

Senai (Malaysia)a

Dak Lak (Vietnam)b

27.4

22.9

26.3

25.6

26.8

22.6

25.4

26.9

21.5

9.1

11.8

7.8

12.7

10.2

7.8

9.9

7.2

7.9

67 59.2 1.3 1297.9 119 0.67

74 58.1 1.2 1455.96 128 0.7

79.9 46.8 2.7 1431.29 151 0.6

76.8 50.8 1.38 1960.1 93 1.1

82.3 47.8 2.1 2282.2 182 1.2

75.7 48.8 2.5 1669.31 163 0.8

3.97

3.48

3.39

3.9

3.57

79 61 2.4 1791.5 159 0.78

67 55.1 1.6 1117.6 117 0.49

81.5 49.9 1.35 2573.53 193 1.4

3.87

3.78

4.4 5890 S 0

106858 E 16

208250 S 0

49859 W 505

5870 N 0

52856 W 48

4.3 98300 N 0

7834 W 451

178510 N 0

102844 E 164

19820 N 0

109830 E 671

238490 N 0

91816 E 31

Source: International Water Management Institute. Senai (Malaysia) is considered as the area offering optimum environment. a Traditional. b Non-traditional.

18360 N 0

103839 E 13

148550 N 1088100 E 655

P. M. PRIYADARSHAN

Temperature (8C; mean) Daily tempera ture range (8C) Relative humidity (%) Sunshine (% h) Wind run (m/s) Rain fall (mm/annum) No. of rainy days Moisture availability index Penman ET0 (mm/day) Latitude

Bogor (Indonesia)a

BREEDING HEVEA BRASILIENSIS

359

Table III Spectrum of Climatic Features of Rubber Growing Countries Country

General climatic features

Malaysia

Tropical, annual southwest (April– October) and northeast (October–February) monsoons Tropical; rainy, warm, cloudy southwest monsoon (mid-May–September); dry, cool northeast monsoon (November–mid-March); southern isthmus always hot and humid. North and northeast areas are non-traditional for rubber Tropical monsoon type with winter (November –January), summer (March–May), southwest monsoon season (June–September) and postmonsoon or northeast monsoon season (October –December). Most of the rainfall brought by southwest monsoon. Because of the geographical diversity of India, regional climate conditions in the extreme north, east and west varies from the general conditions given here. Specific areas of west, east and northeast are nonraditional for rubber Tropical monsoon; northeast monsoon (December–March); southwest monsoon (June–October) Tropical, climate even all year around. Heavy rainfall usually between December and January. The equatorial position of the country makes opposite climates in the north and the south. Extremely diverse, tropical in south to subarctic in the north, with great climatic differences resulting from the monsoon, the expanse of the land mass, and the considerable differences in altitude. Typhoons are prudent in southeast China between July and September. China is a non-traditional zone for rubber Tropical in south; monsoonal in north with hot, rainy season (mid-May–midSeptember) and warm, dry season (mid-October–mid-March). Diverse range of latitude, altitude and weather patterns produces enormous climatic variation. North Vietnam like China has two basic seasons: a cold humid winter from November to April, and warm, wet summer for the reminder of the year. The northern provinces share the climate of the north, while the southern provinces share the tropical weather of the south. South Vietnam is relatively warm. Central highlands and the coastal regions are non-traditional areas for rubber Tropical along coast, semi-arid in far north; three seasons—warm and dry (November–March), hot and dry (March–May), hot and wet (June–October); three main climatic regions: the coast, the forest and the savannah. Low rainfall areas in north (less than 1300 mm) are non-traditional experimental zones for rubber Varies; equatorial in south, tropical in center, arid in north. Two principal wind currents affect Nigeria; the harmattan, from the northeast, is hot and dry and carries reddish dust from the desert and causes high temperatures during the day and cool nights. The southwest wind brings cloudy rainy weather Tropical; hot, humid; dry winters with hot days and cool to cold nights; wet, cloudy summers with frequent heavy showers Range: equatorial, tropical, semi-arid, highland tropical and subtropical. Annual average temperature in the Amazon region is 22– 268C. Brazil is in the south of the equator, seasonal changes are vice versa compared to north of the equator. Plateau of Sa˜o Paulo is non-traditional area for rubber

Thailand

India

Sri Lanka Indonesia

China

Vietnam

Coˇte d’Ivoire

Nigeria

Liberia Brazil

After Priyadarshan and Gonealves, 2002.

360 P. M. PRIYADARSHAN

Figure 2 Latitudinal distribution of Hevea species.

BREEDING HEVEA BRASILIENSIS

361

make these areas moderate/marginal. The aforesaid range of geo-climatic attributes are noteworthy and deserve special attention while deriving adaptive clones, evolving agro management strategies and rescheduling exploitation systems. Most of these areas are non-traditional, since they are away from the equator, where a higher annual input of radiation energy would facilitate greater potential for dry matter production (Oldeman and Frere, 1982). However, this is not reflected in yielding potential of clones under non-traditional environments. Constraints prevailing in these areas are discussed in some detail in Section “Constraints”.

III. CONSTRAINTS A. GEO-CLIMATIC STRESSES 1.

Regions of India, Thailand and Vietnam

Climatologically India has five main zones, viz., tropical rain, tropical wet and dry, sub tropical rain, temperate and desert. Of these, former three are identified to be suitable for rubber cultivation. Several locations of these zones are counted as non-traditional due to latitude and altitude changes. In northeast India (23 – 258N and 90– 958E), such potential areas experience low temperature period during November to January (as low as 3.88C), complete defoliated period during February to March, brief moisture stress during March, tropical storms during monsoon (June to August) and infestation of powdery mildew (Oidium heveae Stein) during refoliation (March to April) are the constraints in these states. Rubber is a prominent species in the states of Tripura, Assam, Meghalaya, Mizoram and Arunachal Pradesh. Tripura (228560 and 248320 N and 918100 and 928210 E) is a representative environment of these states and owes maximum area under rubber. The climate is sub tropical (Mediocre) with moderate temperature (summer: 17.9 – 36.68C; winter: 7.17– 28.98C) and high humid atmosphere. The areas between 15 and 208N of western and eastern India have also been identified as non-traditional zones for rubber cultivation. For instance, the Konkan region of western India experience long dry periods, high temperatures, low atmospheric humidity and zero rainfall between September and May. Daytime temperatures range from 38 to 418C during summer months with occasionally days getting as hot as 478C. The region gets rainfall of 2430 mm, but with an uneven distribution (Devakumar et al., 1998). High solar radiation coupled with high temperature and low relative humidity results in high vapour pressure deficit between the leaf and the surrounding atmosphere, and this subsequently increases the evapotranspirative demand of the atmosphere. Thus, rubber trees in this region are subjected to prolonged periods of both soil and

362

P. M. PRIYADARSHAN

atmospheric drought stress. Irrigated plants showed 32% increment in leaf area index (LAI) leading to 52% more shoot biomass/tree (Devakumar et al., 1998). Water deficit in the dry period is 1070 mm, whereas in traditional areas it is 350 mm (Jacob et al., 1999). Reduction in girth of trees (0.2 – 0.5 mm) was observed during summer months (Chandrashekar et al., 1996). Towards the end of summer, moisture level falls below permanent wilting point (PWP; 17.5%). The high intensities of sunlight, more than what is required to saturate photosynthesis can aggravate the harmful effects on Hevea leaves (Devakumar et al., 1998; Table IV). Almost an analogous situation prevails in the eastern part of India also. Similarly, the non-traditional areas of Thailand (13 –188N), viz., Chachoengsao (east), Nong Khai and Chiang Mai provinces (northeast) experience marked dry season for six months, severe moisture deficit (temperature 14 –398C) with a minimum temperature of 58C during January (Saengruksowong et al., 1983). Rainfall (1110 –1550 mm) is confined to mainly June to September (Watson, 1989). The rubber areas of Vietnam are scattered between 12 and 218N. The research and development of rubber in non-traditional areas are streamlined depending on altitude, viz., high lands of 450 –600 MSL, high lands of 600– 700 MSL and coastal regions (Hoa et al., 2002a). Southeast area is the traditional region for rubber where nearly 3 million hectares are under rubber. While southeast region is with relatively flat terrain, highlands and coastal regions are , 550 m. The highlands and coastal regions that are non-traditional experience low temperatures (5.58C), regular strong winds, rain fall lasting for several days, lesser sunshine, higher number of misty days (Hoa et al., 1998; Tuy et al., 1998). The highlands are predominantly ferrallitic and belong to major family of red or yellowish-red soils. They are clayey, deep and basalt (Eschbach et al., 1998). Ever since rubber was introduced in 1897, Vietnam has taken steps to extend the area to 500,000 ha including expansion to marginal areas (Hoa et al., 2002b). Rubber is a second priority crop for Vietnam (Chapman, 2000).

2. Chinese Conditions China has been divided into six climatologic zones, viz., tropical wet and dry, sub-tropical wet, sub-tropical summer rain, temperate, desert and temperate continental. Of these, the former three are being experimented with rubber. The rubber growing areas of China fall under 18 –248N and 97 –1218E, spread over to five provinces of south China, viz., Hainan, Guangdong, Fujian, Yunnan and Guangxi. These areas are under tropics and sub-tropics having monsoonal climate. Pronounced monsoon and dry seasons prevail from May to November and December to April, respectively. Two types of cold regimes have been

BREEDING HEVEA BRASILIENSIS

363

identified, viz., radiative and advective (Zongdao and Xueqin, 1983). In radiative type, the night temperature falls sharply to 58C and the day temperature ranges from 15 to 208C or above; while in advective type, the daily mean temperature remains below 8 – 108C, with a daily minimum of 58C. In both these types, under extreme circumstances, complete death of the plant is the ultimate outcome. Reports from China point that clones GT 1 and Haiken 1 can withstand temperatures up to 08C for a short span, while SCATC 93-114 can endure temperature of even 2 18C. The cold wave conditions prevailing over other than China can be conveniently classified as radiative type (Priyadarshan et al., 2001). Wind is yet another abiotic stress influencing the establishment and growth of rubber. While an annual mean wind velocity of 1 m/s has favourable effect on the growth of rubber trees, wind speeds of 2.0– 2.9 m/s retards rubber growth and latex flow and that of 3.0 m/s and above severely inhibits normal growth (Table IV). Wind over Beaufort force 10 (more than 24.5 m/s) play havoc with branch breaks, trunk snaps and uprooting of trees, mainly confined in China, during June to October. During 1949 –1982, storms and typhoons lashed rubber-growing areas of China for at least 55 times (Zongdao and Yanqing, 1992). Most of these originate between 5 and 208N near Philippines and are influenced by lowpressure areas over Pacific ocean (Zongdao and Xueqin, 1983). Typhoons, which take westward track, lash south China during June, September and October. Weather data from Hainan shows an average wind velocity of 2.7 m/s which is higher among the rubber growing areas of the world, sufficient enough to retard growth (Tables II – IV).

3. Conditions in West Africa Countries in West Africa (Coˆte d’Ivoire, Liberia, Ghana, Nigeria, Guinea and Sierra Leone) are suitable for rubber. Rainfall is confined to April to October as southwest monsoon that winds over Gulf of Guinea, resulting in high rainfall in the coastal region that diminishes steadily northwards (Edingon, 1991). The presence of mount Cameroon acts as a great barrier for rain bearing winds to settle and to give the second highest rainfall in the world (1000 cm). These areas also experience average annual temperature of 258C with least diurnal temperature range. Northern parts of the rubber growing countries experience dry wind popularly known as Harmattan during November to April, originating in Sahara desert. Cameroon experiences tornadoes during rainy season. Soils are derived from sedimentary rocks, which have been weathered, leached, eroded and deposited. They are naturally deep and poorly supplied with nutrients. But soils of west Cameroon are more fertile and have a tendency to fix nutrients. The coastal areas are densely forested and suitable for rubber.

364

Table IV Geo-Climatic Factors Influencing Growth and Yield of Rubber Attribute Ambient temperature (8C) ,0 ,5 10 18

Manifestations

Reference

Diurnal variation (7–108C) Monthly temp. 208C

Optimum Negligible growth

Shamshuddin (1988) and Rao and Vijayakumar (1992) Jiang (1984) Jiang (1984)

Rainfall (mm) 1300–1500 1800–2000 9–11 mm/day

Optimum for growth and production Optimum for growth and production Congenial

Pushparajah (1983) Pakianathan et al. (1989) Liyanage et al. (1984)

Optimum

Ong et al. (1998)

,18 18–24 22–28 27–30 34–40

Rainy days 100–125 days/year at 125 mm/month

Jiang (1984) Zongdao and Xueqin (1983) Zongdao and Xueqin (1983) Zongdao and Xueqin (1983)

Zongdao and Xueqin (1983) Shuochang and Yagang (1990) Shangpu (1986) and Jiang (1984) Shangpu (1986) and Shamshuddin (1988) Lee and Tan (1979), Chandrashekar et al. (1990) and Ong et al. (1998)

P. M. PRIYADARSHAN

Annual temp. 20–288C

Severe cold damage Cold damage Mitosis occurs but photosynthesis discontinues Plant cells divide normally just for survival (crucial temperature for tissue differentiation) Yield decreases with late dripping Optimum for latex flow Favourable for latex flow Optimum range for photosynthesis Respiration exceeds photosynthesis; retardation of growth and scorching of young leaves Optimum for growth, latex production

Table IV (continued) Attribute Water requirement 3–5 mm/day

Manifestations

Reference

Monteny et al. (1985) and Haridas (1985) Zongdao and Xueqin (1983) Oldeman and Frere (1982) Yee et al. (1969) Zongdao and Xueqin (1983)

8–13.8 17.2 24.5

Favorable No evident hindrance Growth and latex flow retards Severe inhibition of growth and latex flow Leaf laceration Branch breaks, trunk snaps Uprooting

Sunshine 2000 h/year

Optimum

Ong et al. (1998)

Ambient vapour pressure deficit (mbar) .12 28 35

Decrease in latex flow Initiation of stomatal closure Stomata closes

Paardekooper and Sookmark (1969) Rao et al. (1990) Rao et al. (1990)

Wind (m/s) 1.0 1.0–1.9 2.0–2.9 .3.0

Zongdao and Yanqing (1992) Zongdao and Yanqing (1992) Zongdao and Yanqing (1992)

BREEDING HEVEA BRASILIENSIS

Optimum

365

366

P. M. PRIYADARSHAN

Coˆte d’Ivoire, a prominent rubber producer, is located between latitudes 5 and 68N and at longitudes 3 and 88W. Though the areas fall under the tropical belt, water is the limiting factor due to low rainfall. Considering isobar of 1500 mm, and dry season not exceeding five months (monthly rainfall below 100 mm), 20% of the area is suitable for rubber cultivation (Dea et al., 1997). Areas towards north are identified as marginal, where rainfall is below 1300 mm. Even under moderate conditions, in spite of favourable rainfall and short dry season, areas having gravelled elements in soil profile impose 20 –30% weak growth in rubber (Dea et al., 1997).

4.

Situation in South America

Brazil has four main climatic zones, viz., tropical rain, tropical wet and dry, subtropical rain and temperate. Though the former two are congenial for rubber, the southern plateau of Sa˜o Paulo (20 – 248S; 44– 528W) with tropical wet and dry climate is the main production area, due to absence of epidemic of SALB (M. ulei ). The most important production region is in the north west, where the climate is tropical of altitude type with a summer rainy season from October to March and a cold dry winter from June to August with temperature reaching 15– 208C. The yearly total rainfall ranges from 1000 to 1400 mm. The ideal altitude for rubber is 350 – 900 m above sea level. The undulating flat areas are with podzolic and latossolic soils, deep and well drained both with eutrophic and dystrophic types. A few plantations are located in volcanic red soils of high fertility. The low leaf wetness duration and relative low temperature in the winter reduces the epidemics of SALB (Goncalves et al., 2001).

B. BIOTIC STRESSES 1.

Diseases

Diseases, especially SALB that is singularly devastating is yet another stress limiting the yield of Hevea. It is noteworthy that viral diseases do not affect Hevea (Simmonds, 1989). Other diseases of economic importance are the Gloeosporium leaf disease (Colletotrichum gloeosporioides Penz. Sacc.), powdery mildew, and the Phytophthora leaf fall (Phytophthora sp.). Clonal specificity is evident towards resistance to these diseases (Wycherly, 1969). A study with Gloeosporium showed that clones from Malaysia and Indonesia are fairly resistant while clones from Sri Lanka and China are less resistant. But clones from South America are seen to be highly resistant indicating local adaptation rather than breeding is the cause for the resistance (Simmonds, 1989). Ho (1986) gives a good narration of the breeding implications of

BREEDING HEVEA BRASILIENSIS

367

diseases in Hevea. It is imperative that too much susceptible genotypes are rejected at the first instance and the survivors are seen to be moderately resistant. The phenomenon of local adaptation is more evident in the case of minor leaf spot (Corynespora cassiicola Berk et. Curt. Wei.). While Malaysian clones exhibited fairly good HR, clones from Thailand and Malaysia were susceptible. The case of SALB is evidently different. The resistance exhibited by wild relatives like H. benthamiana, H. pauciflora and H. spruceana has been exploited through crosses with H. brasiliensis but was turned to be VR, and was susceptible to newly evolved pathotypes (Ho, 1986). Since the wild relatives own only VR, the breeding programmes need to start from a very low level of genetic variability. On the other hand, achieving HR would imply several cycles of selections under epiphytotic conditions. Since the HR is polygenic, a fairly high h2 would be evident through additive inheritance, where advanced generations produce more resistant progenies. Only RRIM 600 and PR 107 are seen with nominal resistance (Chee, 1976). An immediate remedy to SALB is to practice crown budding (Tan, 1979). This is based on the assumption that a vigorous, wind-fast, disease-resistant crown would a provide good flow of photosynthate to a trunk capable of high partition (Simmonds, 1982). However, such exercises need to be done at the field level, where the infection of SALB largely depends on the climatic conditions of the location. M. ulei requires at least 10 consecutive hours of relative humidity above 95%, with optimum average daily temperatures of 24 –268C with intermittent rains are most favourable for germination and infection (Watson, 1989). Powdery mildew or secondary leaf fall is yet another disease of economic importance for the non-traditional areas. Weather towards the end of wintering is crucial and infection is increased if refoliation takes place at a time of low temperature, with overcast days and cool nights. Also, very light rains giving prolonged periods of high humidity are ideal for increased infection. Though the yield loss is difficult to assess, yield increase of over 100% is reported from traditional areas (Johnston, 1989). There must be resistant sources in allied species especially in types that defoliate partially. Infestation of powdery mildew has a profound effect on flowering and seed set in all growing areas and is a set back to the multiplication of clones in addition to yield depression.

2. Phenology under Differential Geo-Climates Phenology of a crop is vital that inflicts significant changes in the yielding behaviour, especially under a new environment. Hevea normally takes 3 –4 years

368

P. M. PRIYADARSHAN

to attain reproductive stage, and shows seasonal flowering in response to alteration in seasons. In the north of equator, March to April experience the main flowering season and during August to September, a short spell of secondary flowering occurs in most of the Asian countries. Defoliation is experienced during December to January and refoliation commences by February. It seems reasonable to presume that geographic location has a bearing on the trees to flower during the secondary season. While it flowers in southern parts of India (6 – 88N) only during March to April, Malaysia (3 – 68N) experiences flowering with viable seeds during both seasons. Tripura (22 – 248N) on the other hand, though experiences flowering and seed set during both seasons, the viability of seeds is largely less during secondary season. This prompts to confine hand pollination experiments during March to April only when substantial number of clones undergo flowering for a short span of 10 –15 days (Sowmyalatha et al., 1997). The situation in the south of the equator is in the opposite fashion. This phenomenon of phenological changes becomes more prudent in a comparison of areas towards north and south of the equator (e.g., Tripura, India and Sa˜o Paulo, Brazil). While Tripura lies at 22 –248N, Sa˜o Paulo is at 20 – 228S (400 – 500 m MSL) making these areas non-traditional (Priyadarshan et al., 2001; Costa et al., 2000; Ortolani et al., 1998). Flowering and fruit formation precede low yielding phase in rubber both in Tripura and Sa˜o Paulo. The environmental conditions inducing defoliation, flowering and low and high yielding periods are analogous. The peak yielding period in Sa˜o Paulo is January to May followed by winter and defoliation, while in Tripura May to September is the low yielding period (Table V). Apart from Brazil, Indonesia is another country where the equator bifurcates into north and south. The change in geo-climate ensures stabilised supply of rubber in the international market.

Table V Seasons and Phenological Attributes Expressed During Various Periods in Tripura and Sa˜o Paulo Phenology Defoliation Refoliation Flowering Lean yield Peak yield Rainy season Winter a b

After Priyadarshan et al. (2000a). After Ortolani et al. (1998).

Tripuraa

Sa˜o Paulob

December–January February–March March–April May–September October–December May–August November–January

August–September September–October October–November August–January February–July October–March June– August

BREEDING HEVEA BRASILIENSIS

369

IV. HEVEA UNDER MARGINAL CONDITIONS A. IMMATURE PHASE Clones multiplied through bud grafting unto seedlings that attain required girth (50 cm) early are preferred, since yield can be retrieved from them especially under a new environment. Accordingly, girth increment under immature phase becomes a crucial attribute in Hevea. In a comparison of girth increment of RRIM 600 in traditional and non-traditional areas of India, Sethuraj et al. (1989) reported 4.3 cm less girth in the northeastern region of India compared to traditional belt. While RRII 105 is counted as one of the best suitable clones for the traditional areas, PB 235, RRIM 600, RRII 208 and Chinese clone Haiken 1 are seen to be adaptable in the north east region of India (Priyadarshan et al., 2000a,b; Mondal et al., 1999). In a study with seven clones and five hybrids, Meenattoor et al. (2000) rationalised RRII429 to attain higher girth in nontraditional environments. Girth increment is seen to be minimum during winter months (November to January; Meenattoor et al., 1991; Priyadarshan et al., 1998a), which is over 20% of the total annual girth (Vinod et al., 1996). These preliminary evaluations amply rationalised that clones, which perform well under traditional areas, do not behave similarly under non-traditional environments. In the water-limiting environment of Konkan region, shrinkage of tree stems has been observed during moisture deficit period (March to June). The monsoon period (July to August) though experiences cloudy and low sunshine hours, girth increment indicated trees received adequate photosynthetically active radiation (Chandrashekar et al., 1998). Also, a full potential irrigation during dry period gave maximum growth that is 50% less than the growth observed in the preceding monsoon period (Mohanakrishna et al., 1991), presuming that Hevea prefers low vapour pressure deficits for growth. Clonal differences were evident in stomatal characteristics in trees grown under moisture stress (Chandrashekar, 1997). While Konkan region experiences active girth increment between July and September, in northeast India (Chandrashekar et al., 1998), May to August is the congenial period for better growth (Priyadarshan et al., 1998a). Both the regions require 8– 9 years for the trees to attain maturity. In a comparative study involving 15 clones of Indian, Malaysian, Srilankan and Indonesian origin, RRII 208, RRIC 52, RRII 6, RRIC 100 and RRIC 102 were seen to exhibit better growth in Konkan region of India. Even in low temperature affected northeast India, RRII 208 showed better growth in addition to PB 235, RRIM 600, RRII 118, and SCATC 93/114. Evidently, these clones were developed under hydrographic environments specific to each location. However, RRII 105, a potential clone for traditional region was not adjudged as drought/low temperature tolerant, and hence not adapted to these conditions (Chandrashekar et al., 1998; Meenattoor et al., 1991). However, Rao et al. (1990) reported that

370

P. M. PRIYADARSHAN

RRII 105 responded well to dry weather of traditional areas through higher values of stomatal resistance, leaf water potential and lower transpirational water loss. This differential performance needs to be studied in depth with physiological tools. In a comparative stability analysis of girth in Tripura, Haiken 1, PR 107 and SCATC 93/114 were seen to be more stable. However, higher contribution towards girth increment was seen in RRII 208 followed by Haiken1 and SCATC 93/114. Clones with higher stability were with wind endurance also (Priyadarshan et al., 1998a). In an analysis with clones of vivid geographic origin (GT1, AVROS 2037, RRIM 600, PB 217 and PB 235) under different locations in Coˆte d’Ivoire, Dea et al. (1997) demonstrated growth is influenced by availability and extent of rainfall (Fig. 3). Rainfall in these areas varied from 1090 to 1600 mm with 4– 6 dry months. Trees took 7– 8 years to attain maturity. A similar exercise was done in Vietnam, where non-traditional areas imposed immaturity period of 1.5– 2 years more compared to traditional zones (Tuy et al., 1998). Immaturity period increased with altitude. GT 1, RRIC 110, RRIC 121, PB 235 and VM 515 were seen to be with higher girth increment. Though genotype –environment interaction studies have been undertaken at several sites earlier (Jayasekara and Karunasekara, 1984), the environment had not been bifurcated into climatic and edaphic factors. Studies with seven clones of Indonesian (GT1, PR 261, PR 255), Malaysian (RRIM 701, PB 235, RRIM

Figure 3 Growth–pluviometry relationship (1984–1991) at different locations of Coˆte d’Ivoire.

BREEDING HEVEA BRASILIENSIS

371

600) and Brazilian (IAN 873) origin, Goncalves et al. (1998) could bifurcate the climatic and edaphic factors affecting the interactions. This was done by exercising clone x site interactions (four test locations) through calculation of estimated heritability (h 2b ) and genetic gains (GGs) that showed PB 235, IAN 873 and RRIM 600 with greater values under different sites. In yet another study with half-sib progenies of 22 Asian clones evaluated under three test sites demonstrated genotype –site interactions were significant for rubber production and girth increment (Costa et al., 2000). However, these studies never rationalised clones suitable for a specific location. The aforesaid discussion amply proves that growth trends of clones are location-specific and clones exhibiting better growth need to be evolved for a specific environment.

B. YIELD DEPRESSION, PATTERNS, REGIMES AND SPECIFIC ADAPTATION Like immature phase, the mature phase of rubber also exhibits differential performance of clones under various non-traditional environments with single or multitude of stresses. Yield depression during a specific period is the main set back when we examine the phenotypic expression of this attribute of Hevea under marginal conditions. This is evident when yield profiles are taken from Tripura (India), Sa˜ Paulo (Brazil) and highlands of Vietnam, where two yielding regimes are prudent in a year (Fig. 4). Months preceding the low temperature period experience depression in yield. In India, in the northeastern states, May to September used to experience a low yielding period. This is the carried-overeffect of stress periods that is not prudent in traditional areas. There are multitude of factors that induce a low yielding period, viz., low temperature (November to February), utilisation of carbohydrate reserves for refoliation (February to March), flowering and fruit development after refoliation (April to August), low moisture period (March), and incidence of leaf diseases during refoliation (February to March). These factors together impose an ensuing low yielding period (Priyadarshan et al., 2000a). An analogous situation prevails in the nontraditional areas of Brazil (southern plateau), but in a vice versa fashion (Ortolani et al., 1998; Priyadarshan et al., 2001; see Table V). However, fall in temperature during November stimulates yield. The daily temperature range in non-traditional areas of northeast India during winter is around 8– 128C, making the atmosphere most ideal for latex flow and production. Minimum temperature experienced in the early morning during tapping is 15 –188C and after 10 a.m., the temperature shoots to 27 –288C. While the former is congenial for latex flow, the latter is ideal for latex regeneration through accumulation of rubber particles (Ong et al., 1998). The rubber growing areas of Vietnam fall under the same latitude range experience and same trend. However, the areas of China are diversified and hence exhibit a trend depending on the temperature and altitude. Chinese clones Haiken

372

P. M. PRIYADARSHAN

Figure 4 Contribution towards yield in GT 1 and PB 235 over months in Vietnam (Highlands), India (Tripura) and Brazil (Sa˜o Paulo).

1, SCATC 88-13 and SCATC 93-114 are being evaluated in Tripura. Initial yielding pattern shows Haiken 1 to be a high yielder against RRIM 600 as a local check (Priyadarshan et al., 1998b). SCATC 93-114 is proclaimed as cold endurant under Chinese conditions (Zongdao and Xueqin, 1983), and shows the same trend in Tripura also (Priyadarshan et al., 1998b). There are clones that show consistency in yield over months, viz., PB 235, RRII 203, and RRII 208. Among these, PB 235 has been evaluated under differential conditions. PB 235 shows consistency under stressful conditions of Tripura (low temperature area), Coˆte d’Ivoire (high minimum temperature)

BREEDING HEVEA BRASILIENSIS

373

and Vietnam (high altitude; Priyadarshan et al., 2000b; Dea et al., 1997; Thanh et al., 1998). Its latex contains low sucrose concentrations implying rapid utilisation of the precursor. Its biosynthetic activity is also seen to be intense with higher values of latex yield, dry extract (dry rubber yield) and high inorganic phosphorus (Pi) with a rapid regeneration between two tapping (Serres et al., 1994; Jacob et al., 1995). PB 235 does not tend to increase yield significantly at longer tapping intervals. Such observations were made under warm climatic conditions of Coˆte d’Ivoire (Serres et al., 1994), which amply conform to the inferences drawn from our studies on yielding trends (Priyadarshan et al., 2000a,b). The aforesaid attributes of PB 235 amply suggest its utility for Tripura, Coˆte d’Ivoire and Vietnam that can be confirmed through on-farm trials. The low wind tolerance of PB 235 shall be circumvented through induction of branches at a lower height (2 m), high density planting and commencement of tapping upon attainment of 60 cm girth instead of the usual 50 cm (Clement-Demange et al., 1998). GT 1 is yet another clone that deserves special mention, since it is counted as a high yielding clone in China (Zongdao and Xueqin, 1983; Zongdao and Yanqing, 1992). GT 1 has not been counted as a high yielder in Tripura, though Tripura and rubber growing areas of South China fall under the same latitude range. This disparity in yielding potential could be attributed to diverse climatic and edaphic factors. A comparison of yield and secondary attributes of clones evaluated in Tripura and Sa˜o Paulo would reveal their differential performance (Table VI). In Vietnam, clones are being evaluated under different altitude ranges. While PB 312, PB 280, RRIC 101 and RRIC 130 gave 100 –146% more yield than GT 1 under altitudes . 650 m, PB 235, VM 515 and PB 255 exhibited 72 – 93.5% yield increase under altitudes of 450 –600 m (Tuy et al., 1998). This evidently indicated that the performances of clones are not complimentary under differential altitudinal climates (Table VII). In Thailand, nearly 2.6 million hectares are delineated in the north and northeast region that has been divided into three zones depending on soil profile and climatic information. GT1, PB 28/59, RRIM 600 and PB 5/51 are the prominent clones adapted to these regions (Krisanasap and Dolkit, 1989; Watson, 1989). An insight into the impact of climate would amply rationalise the role of certain attributes over the yielding ability of clones. Minimum temperature, wind velocity and evaporation are seen to have negative correlation with monthly mean yield (Priyadarshan et al., 2000a). The rationale is that, fall in temperature along with reduced evaporation and low wind speeds prevail upon the microenvironment to influence yield-stimulation during cold period. It is evident that turgour pressure in laticiferous system is vital for the flow of latex and yield. Turgour pressure as high as 10– 14 atmospheres is observed before sunrise and studies on diurnal variations in latex yield gave a correlated response between latex yield and variations in atmospheric vapour pressure (Moraes, 1977). The atmospheric vapour pressure is very high during cold months thus increasing

374

P. M. PRIYADARSHAN Table VI Yield and Secondary Attributes of Clones being Evaluated in Tripura and Sa˜o Paulo Yield (projected; Crop kg/ha) efficiencya

Clones

Stand (initial)

Girth (mature)

RRII 105T RRII 118T RRII 203T RRII 208T RRIM 600T RRIM 703T RRIC 105T PB 5/51T

Good Good Good Good Good Average Average Good

Moderateb Highb Moderateb Moderated Moderateb Moderateb Highb Lowb

1303c 1620c 1512c 1080e 1364c 1449c 896c 888c

1.0 1.07 1.14 0.93 0.99 1.21 0.59 0.74

PB 235T GT 1T PR 107T

Good Good Good

Highb Moderateb Goodb

1889c 1045c 305e

1.34 0.85 0.29

SCATC 88/13T SCATC 93/114 HIAKEN 1T IAC 35S IAC 40S IAC 301S IAN 3156S IAN 873S RO 45S FX 3864S

Good Good Good Average Good Good Average Good Average Good

Goodb Goodb Goodb Moderate High Moderate Low High High High

744e 279e 798e 1680c 1755e 1750e 2499e 1243c 1940c 1755c

0.67 0.24 0.68 1.4 1.84 1.85 1.99 1.82 1.55 0.85

Wind damage

TPD

Moderate High Low High Low Moderate High Low

Low Mild Low Very mild Moderate Low Low Mild

Oidium incidence

Severe Moderate Mild Severe Severe Mild Low Very severef Moderate Moderate Severe Low Mild Moderate Very low Mild Very severef Low Mild Severe Medium Very mild Low Medium Mild Moderate High Low Moderate Moderate Low Moderate High Mild Moderate Low Mild Mild Moderate Low Mild Moderate Low Mild High Low Mild

Projected yield ¼ g/tree/tap £ no. of tapping £ total stand (350); T ¼ Tripura; S ¼ Sa˜o Paulo. a g/cm of the tapping cut. b Over 7 years. c BO II panel. d Over 2 years. e BO I panel. f With secondary infection.

the latex flow. But there are clones like PB 235 and RRII 208 that show less stimulation towards the onset of cold period. Studies conducted in revealed clones, especially PB 235 and GT 1 as less responsive to ethrel stimulation (Gohet et al., 1995). From these observations, it can very well be presumed that PB 235 is less responsive to stimulation irrespective of the stimulant, which is a positive attribute. PB 235 owns a specific adaptive mechanism, whereby it yields more when ambient temperature ranges from 22 to 288C. When all clones continue with a higher yield in combination with descending temperature, PB 235

BREEDING HEVEA BRASILIENSIS

375

Table VII Performance of Rubber Clones under Marginal Areas of Vietnam (kg ¼ kg/ha)

Site

Latitude

1 Highlands 12–158N (450–600 m MSL; Gia Lai, Daklak, Kontum)

Clones

PB 235, RRIC 105, RRIC 110, RRIC 117, RRIC 121, VM 515, PB 255, PB 310, PB 324, RRIM 600, GT 1, PR 255 12–158N PB 235, GT 1, 2 Highlands (600–700 m MSL; RRIM 600, RRIM 712, Gia Lai, Daklak) VM 515, RRIC 121, PB 260, RRIV 1, RRIV 3, RRIV 4 3 Coastal region 16–198600 N PB 255, RRIM 600, RRIM 712, GT 1, (Quang Tri RRIV 1, RRIV 3, province) RRIV 4

High yielding clones

Reference

RRIC 121 (1522 kg), Hoa et al. PB 235 (1390 kg), (2001a) VM 515 (1387 kg), RRIM 600 (1232 kg), PB 255 (1226 kg) RRIV 1 (1041 kg), RRIM 712 (951 kg), RRIC 121 (940 kg), VM 515 (920 kg), PB 260 (964 kg) PB 235 (1427 kg), RRIM 600 (1420 kg)

Hoa et al. (2001a)

Hoa et al. (2001a)

recedes yield during January when the ambient temperature gets below 158C. Studies conducted in China with few other clones endorse the same trend in GT 1 (Zongdao and Xueqin, 1983). Ambient temperatures ranging from 18 to 248C is conducive for latex flow (Zongdao and Yanqing, 1992). Evidently, the existence of genetic homeostasis and their subsequent expression in the changed environment might be the reason for the near uniform yielding trend in these clones. Through homeostasis, perhaps, yield is reduced and the source– sink relations are brought to equilibrium to ensure the survival during cold/stimulated period. The trend shown by clones is in sharp contrast to that of traditional areas of India where RRII 105 and RRIM 600 are prominent yielders when evaluated separately (Nazeer et al., 1991; Mydin et al., 1994). A comparison of yielding trends of PB 235 and RRIM 600 rationalised that these clones under a specific environment expresses “cross-over” type of GE interactions, wherein 28 g represents the threshold level below which these clones are expected to experience stress (Priyadarshan et al., 2000a; Fig. 5). Presumably, a clone giving more than 28 g/tree/tap shall not experience any stress. Clones of varied geographical origin could be delineated into three groups, viz., high, moderate and low yielding clones. Also, in these environments, PB 235 has been adjudged as a high yielding clone. Performances of Hevea clones under immature and mature phases are different and the clone that attains maturity is not necessarily be the best yielding clone. This is due to lack of significant relationship between girth increment and yield (Tan, 1987).

376

P. M. PRIYADARSHAN

Figure 5 Regression of mean yield of PB 235 and RRIM 600 over environmental yield under two yielding regimes.

V. BREEDING PROGRAMS Evaluation of available genetic diversity and derivation of adaptable variability are the two strategies for evolving clones for a specific environment. Since the marginal areas are diversified, the breeding programs to be followed in Hevea can be categorised into: evaluation of polyclonal seedlings, recombination breeding and increasing genetic diversity. Evaluation of clones has already been dealt in Section “Yield depression, patterns, regimes and specific adaptation.”

A. POLYCLONAL SEEDLINGS Whitby (1919) was the first to report the considerable variability in productive capacity in routine seedlings. First clones released out of the seedlings were Cramer’s Cultuurtuin (Ct3, Ct9, Ct88) selected from 33 seedlings planted in Penang through Java in Indonesia (Dijkman, 1951). Mixed planting of these clones gave an yield of over 1700 kg/ha that was very much higher than that of the unselected seedlings (496 kg/ha; Tan et al., 1996). During 1924, Major Gough selected 618 seedlings from a population of about one million seedlings in

BREEDING HEVEA BRASILIENSIS

377

Kajang district of Malaysia, which yielded prominent primary clones like Pil A44, Pil B84, Pil B16, PB 23, PB 25, PB 86, PB 186 and Gl 1. By 1930s it was understood that the primary clones have reached a plateau of yield (Tan, 1987). Hence, the emphasis shifted from primary clones to recombinants derived through controlled pollination (see Section “Recombination breeding”). While recombination breeding was underway, polyclonal seed gardens were set up duly with improved clones to derive polycross seedlings for supplementary planting materials. Thus, the best seedlings came from Prang Besar Isolated Gardens (PBIG), Gough Gardens and Prang Further Proof trails (Tan et al., 1996). By 1970, polycross seedling areas extended to 7700 hectares with more than 2 million trees. Both yield and secondary attributes need to be given the deserving importance while selecting clones (Ho et al., 1979). Final selection was on the basis of 65 and 35% scores for yield and secondary attributes, respectively (Tan et al., 1996). The procedure involves field selection in the estates, nursery selection, small-scale selection (16 trees) and large scale testing (128 trees). After popularisation of clones in the 1980s, the potentiality of extending rubber to marginal areas was understood and the concept of producing polyclonal seedlings by constituting polyclonal seed gardens had emerged. There is a contention that yield and girth variation can be largely accounted by additive genetic variance (Gilbert et al., 1973; Nga and Subramaniam, 1974; Tan, 1981), suggesting that phenotypic selection would be effective. However, as per general genetic principles, selection based on genotypic values as reflected by general combining ability (GCA) will be more reliable and desirable. GCA can be estimated through evaluation of seedling progenies to choose parental clones. It is here that the Biotechnology can contribute significantly to assess molecular diversity of parents and the resultant seedlings (see Section “Molecular diversity”). The number of parents is very crucial in determining the constitution of polyclonal seed garden. Though gardens with more than four clones are possible, an optimum of nine clones had been suggested (Simmonds, 1986). Accordingly, a repeated three-step two-dimensional rubber polycross-design with nine clones can be envisaged that allows only heteroneighbours for a given clone, ensuring cross-pollination (Fig. 6). A polyclonal seed garden involving clones with high GCA that are panmictic, ensures seedlings with high genetic divergence. The extent of selfing may reduce the vigour of first generation (SYN1) population,

Figure 6

Two-dimensional design for the production of polycross progenies.

378

P. M. PRIYADARSHAN

since there is no evidence of self-incompatibility. However, it can be presumed that the seeds produced are of cross-pollination, given the argument that zygotic inability reduces germination due to inbreeding (Simmonds, 1986). Such SYN1 progenies are still considered as Class I planting material in Malaysia. Moreover, SYN1 progenies must be of better use in non-traditional/marginal areas, where attributes rendering resistance towards stresses also attain prominence. In a comparative evaluation of polyclonal seedlings and multiclonal population in Tripura, Sasikumar et al. (2001) rationalized the mean yields of both the populations as on par, indicating thereby that a highly heterogeneous poly clonalseedling population can be successful for marginal areas.

B. RECOMBINATION BREEDING Rubber breeding over the last century has made significant progress due to recombination breeding and selection. Yields have increased from 496 kg/ha in primary clones to more than 3000 kg/ha in RRIM 2000 (Rubber Research Institute of Malaysia) series. The RRIC 100 series (Rubber Research Institute of Sri Lanka) released in Sri Lanka during 1980s is yet another example. Much of the hybridisation work at Malaysia, Indonesia, India, Coˆte d’Ivoire Brazil, Thailand and Vietnam further strengthened the array of hybrid clones (Table VIII). These clones are known for their adaptability to specific hydrothermal/agroclimatic situations, since selection pressure was exerted to derive clones with local adaptation apart from yield, especially to stress factors like wind, low temperature, moisture stress and diseases. At least 16 primary clones can be considered as prime progenitors for modern clones, viz., PB 56, PB 24, PB 25, PB 28, PB 86, Tjir 1, Gl 1, PR 107, Mil 3/2, Hil 28, AVROS 255, RRIC 52, Pil B50, Pil B84, PB 28/59 and GT 1. It is presumed that families of crosses involving reasonably good clones will be of high average performance (Simmonds, 1989), provided if data on parental combining abilities are available. GCA estimates are especially valuable in focussing attention on good combinations. However, such a concerted effort has not been seen in Hevea breeding (Tan, 1987). Needless to say, this approach would consume more time in exploiting selective parents. Hence, it is always advisable to advance further with promising clones through small-scale clone trials (SSCTs) as parents Fig. 7). The approach must be either to involve clones of proven performance and breeding value or early cross between promising locally adapted imperfectly known clones (Simmonds, 1989). The major strategy followed was to use the best yielding genotype of one generation as the parent of the next generation. Many valuable recombinants must have been lost during the course of this assortative mating of primary/hybrid clones and subsequent directional selection for yield under varied climates. Also, most of the clones had cytoplasm of clones like PB 56 (through PB 5/51) or Tjir 1 (Table IX). It is presumable that the success of

Table VIII Profile of Prominent Clones Evaluated in Their Areas of Origin Resistance to

Clone

Yield (kg/ha)

Girth increment during tapping

Tjir 1 £ Gl 1 PB 86 £ Mil 3/2 Mil 3/2 £ AVROS 255 RRIC 52 £ PB 83 Tjir 1 £ PB 86 PB 49 £ PB 84 RRIM 605 £ RRIM 71 GT 1 £ PR 107 PB 5/51 £ RRIM 703 PB 5/51 £ IAN 873 PB 5/51 £ PB 6/9 PB 5/51 £ PB S/78 PB 5/51 £ PB 32/36 Primary clone Tjir 1 £ PR 107 Tjir 1 £ PR 107 Primary clone PB 5/51 £ RRIM 600 PB 5/51 £ GT 1 PB 5/51 £ RRIM 501 Primary clone GT 1 £ AVROS 1734

2210 1618 1587 1774 2199 1622 2264 2146 2483 2760 1778 2485 2283 2023 2018 1838 1475 1446 1807 2086 1500 1394

3 4 3 3 4 4 2 3 2 4 4 3 3 1 3 3 4 5 5 2 3 2

Wind damage

Panel dryness

3 3 3 5 4 2–3 5 4 5 NA 4 2 4 3 4 4 4 3 3 NA 4 3

5 2 3 3 4 3 4 3 3 NA 4 2 2 3 3–4 3–4 4 3 3 NA 3 3

Pink Disease 5 3 NA 3 1 2 –3 3 4 4 NA 2 3 2 2 3 3 4 NA NA NA 2 3

Oidium 3 3 3 4 3 1–2 3 3 3 4 2 2 2 2 1 1–2 2 NA NA 3 NA 3

Colletotrichum 5 NA NA 3 3 3–4 1 4 3 4 3 2 2 2 3 4 NA NA NA NA NA 2

Corynespora

Phytophthora

5 3 NA 5 1 4 3 4 5 4 4 4 4 4 4 3 3 NA NA 3 NA 3 –4

1 3 NA NA 1 1 3 2 3 3 1 3 2 2 3 3 3 NA NA NA NA 4

379

(1) poor; (2) below average; (3) average; (4) good; (5) very good; (NA) not available, since the disease is not prominent. Under conditions of (M) Malaysia; (I) India; (C) China; (CD) Cote d’Ivoire; (B) Brazil; (T) Thailand. Tapping system ¼ s/2 d/2 6d/7 86%; number of tapping days per year ¼ 158 þ 11 trees/ha ¼ 327 þ 34.

BREEDING HEVEA BRASILIENSIS

RRII 105I RRII 203I RRII 208I RRIC 100M RRIM 600M RRIM 623M RRIM 712M RRIM 936M RRIM 937M RRIM 2015M PB 217M PB 235M PB 255M PB 28/59M PB 255M PR 261M GT 1M IRCA 111CD IRCA 230CD RRIT 163T HAIKEN 1C BPM 24M

Parentage

380

P. M. PRIYADARSHAN

Figure 7 Various breeding schemes.

wider adaptive RRIM 600 with Tjir 1 cytoplasm paved the way for the production of many modern clones of Malaysia that are being experimented under various marginal areas. However, hand pollination experiments leading to recombinants need to be conducted under the environment in question since selection pressure either in favour or against gene combinations commences from the induction of embryo onwards.

BREEDING HEVEA BRASILIENSIS

381

Table IX Source of cytoplasm of prominent clones Tjir 1

PB 56 PB 5/51, PB 217, PB 235, PB 255, PB 260, PB 330, PB 355, IRCA 18, IRCA 111, IRCA 130, RRIM 901, RRIM 905, RRIM 908, RRIM 911, RRIM 921, RRIM 931, RRIM 2001, RRIM 2016, RRIM 2017, RRIM 2020, RRIT 163

RRII 105, RRIM 600, RRIM 605, RRIM 628, RRIM 703, RRIM 712, RRIM 722, RRIM 928, RRIM 929, RRIM 2001, SCATC88/13, RRIC 50, PR 255, PR 261, PB 311, PB 312, PB 314, PB 350, IAN 3457, IAN 3460

China has recently developed five wind fast clones that are recombinants of Haiken 1 or PR 107. Their cumulative percentage of wind damage is lower than the control Haiken 1. Such clones have been evolved through recombination breeding involving locally bred genotypes (Tianren 31-445, Haiken 1, SCATC 93-114). The clone Xuyu 141-2 could withstand winds of . 12 Beaufort scale (Huasun et al., 1998; Table X). In an evaluation with locally bred clones, Goncalves et al. (2001) rationalised IAN 3156 (Fx 516 £ PB 86) having 50% more yield than RRIM 600. It is noteworthy that Fx 516 owns the cytoplasm of H. benthamiana. Apart from China and Brazil, institutions in the other nontraditional areas are focussing attention on the production of hybrids, which are under experimental phase.

Table X Yield and Secondary Attributes of Chinese Clones Clone/attribute

Parentage

Yield (kg/ha)

High yielders Yunyan 277-5 SCATC 7-33-97 SCATC 8-333

PB 5/63 £ Tjir 1 RRIM 600 £ PR 107 SCATC 88-13 £ SCATC 217

2036 1977 2187

Yield and wind endurance Wenchang 217 Wenchang 193 Wenchang 33-24 Wenchang 11 Xuyu 141-2

Haiken 1 £ PR 107 PB 5/51 £ PR 107 Za 39 £ Haiken 1 RRIM 600 £ PR 107 Haiken 1 £ PR 107

1319 919 893 1356 1007

Cold endurance SCATC 88-13 SCATC 93-114 Haiken 1

RRIM 600 £ Pil B84 Tianren 31-45 £ HK 3-11 Primary clone

1592 750–900 1050–1500

Huasun et al. (1998) and Zongdao and Xueqin (1983).

382

P. M. PRIYADARSHAN

1. Latex Timber Clones Of late, a concept has been evolved to extract maximum quantity of rubber in a stipulated time and then use the trees as source of wood. An estimation from RRIM shows that a hectare of rubber plantation can yield 190 m3 of rubber wood. By 2000, 2.7 million m3 of Hevea wood would be available from Malaysia (Arshad et al., 1995). This is used for chip logs (for the production of cement board, chip board, band medium density fireboard) and saw logs (for plywood and veneer operations). Theoretical estimations indicate that India is expected to have 43 million m3 of growing stock from 518,000 ha (Anonymous, 1996). Hence, nearly 741 million m3 of wood must be available from 892,7000 ha worldwide. The demand is expected to increase by 2012 and RRIM, RRIT, and RRII have been making concerted efforts in deriving latex timber clones (Table XI). Clones PB 235, PB 260, RRIM 2008 and RRIM 2014 are promising because they are complimented with higher yield also. A few accessions of allied species like H. pauciflora, H. guianensis and H. nitida also yielded wood volume in the range of 1.19– 4.43 m3/tree. Nearly 20 clones of 1981 Amazonian collection were also selected for timber yield by the RRIM yielding at a range of 1.438 – 2.518 m3/tree at the age of 13 years. It is pertinent to increase production of Hevea wood due to constant decline in area both under smallholdings and estates. Among a number of genotypes tested for wood production, H. guianensis appeared to be the best with clear bole volume at 1.77 m3/tree. However, this attribute needs to be complimented with latex yield probably through intercrossing and selection.

C. INCREASING GENETIC DIVERSITY Since the introduction of Hevea during 1877 by Wickham and Cross, there have been a few attempts to collect the new material and increase genetic diversity. During 1951 – 1952, 1614 seedlings of five Hevea species (H. brasiliensis, H. guianensis, H. benthamiana, H. spruceana and H. pauciflora ) were introduced in Malaysia (Tan, 1987). In Sri Lanka, 11 clones of H. brasiliensis and H. benthamiana and 105 hybrid materials were imported during 1957 – 1959, through triangular collaboration of USDA, Instituto Agronomico do Norte (IAN), Brazil, and Liberia. Many of these clones were later given to Malaysia which were used for further breeding programmes at RRIM (Tan, 1987). Due to the initiatives taken up by the International Rubber Research and Development Board (IRRDB), 63,768 seeds, 1413 m of bud wood and 1160 seedlings were collected during 1981 from Acre, Rondonia, and Mato Grosso states of Brazil (see www.irrdb.com). Of these materials, 37.5% of the seeds went to Malaysia and 12.5% to Coˆte d’Ivoire and half of the collections was retained in Brazil. The clonal selections were brought to Malaysia and Coˆte d’Ivoire after

Table XI Estimated Wood Volume from Potential Clones, Accessions of Brazilian Amazonian and Allied Species Age (year)

Clear bole volume (m3/tree)

Canopy wood volume (m3/tree)

Total wood volume (m3/tree)

RRIM 910 RRIM 912 RRIM 931 PB 235 PB 355 RRIM 2008 RRIM 2014 Clones of Brazilian Amazonia RO/OP/4-20/125 AC/F/5-21/197 MT/C/5-12/137 AC/F/21-64/221 Allied species H. pauciflora H. guianensis H. nitida

PB 5/51 £ RRIM 623 PB 5/51 £ RRIM 623 PB 5/51 £ RRIM 713 PB 5/51 £ PB S/78 PB 235 £ PR 107 RRIM 623 £ PB 252 RRIM 717 £ PR 261

22 22 20 20 22 14 14

0.76 0.75 0.68 0.80 0.93 0.33 0.53

0.57 0.75 0.68 0.80 2.32 0.99 0.80

1.33 1.50 1.36 1.60 3.25 1.32 1.33

– – – –

13 13 13 13

1.259 1.403 1.054 1.137

1.159 1.052 1.318 1.364

2.518 2.455 2.372 2.501

– – –

24 24 24

1.13 1.45 1.04

0.41 2.18 1.04

1.14 3.64 2.08

BREEDING HEVEA BRASILIENSIS

Parentage

Clone

After Arshad et al. (1995).

383

384

P. M. PRIYADARSHAN

quarantine measures for SALB. Other member countries introduced material depending on their request. IRRDB supports germplasm centres based in Malaysia and Coˆte d’Ivoire to conserve these materials. Between 1945 and 1982, at least 10 collections from Brazil (mostly Rondonia) were undertaken (Goncalves et al., 1983). Crosses between Wickham and Amazonian accessions could introduce more variation. Breeding in Coˆte d’Ivoire (IRCA—Institut de recherches sur le caoutchouc en Afrique ) had been oriented towards utilisation of Amazonian accessions. Preliminary observations suggested they include great deal of diversity in vigour, foliage and disease reactions (Ong et al., 1983). Assuming that useful genetic combinations are randomly distributed in the Amazonian collections, Simmonds (1989) gave a response equation for exploitation of diversity: p XN ¼ X þ i h2 ·sG where XN is high future performance, which shall depend on high starting mean (X ), and high genetic variability (sG ). In this exercise, selection for yield is through test tapping, where clones equivalent to Malaysian primary clones are expected to occur. It is also presumed that due to inter population heterosis for vigour, a better yielder when crossed to Wickham clones shall give outstandingly vigorous families (Ho and Ong, 1981; Simmonds, 1989). The use of polycross is another option to induce, select and sustain useful diversity. This shall be otherwise a relaxed mass selection. Several dwarfs and semi-dwarfs have been identified in the principal population (Ong et al., 1983), perhaps dominant or semi-dominant, which may therefore, be useful to be crossed with high yielding genetic backgrounds to derive wind fast clones. Yet another strategy for utilising genetic diversity is towards exploitation of mtDNA variation. Since most of the oriental clones possess cytoplasm of either PB 56 or Tjir 1, introduction of diverse cytoplasm after DNA analysis must show good potentiality for higher yield.

D. MOLECULAR BREEDING 1. Molecular Diversity Several biological constraints impede the elucidation of the genetics in Hevea, viz., long growth cycle, poor seed set, vegetative propagation, amphidiploidy and severe inbreeding depression on selfing. Molecular breeding, especially the deciphering of molecular genetic maps can be employed to understand the genetic basis of yield potential and to identify genetic factors involved in partitioning the product of photosynthesis. This information can be used to choose parents with greatest breeding value, guide breeding decisions for multiple trait improvement and combine complementary genes with the hope of achieving new recombinants.

BREEDING HEVEA BRASILIENSIS

385

Efforts for breeding Hevea at molecular level commenced since Low and Bonner (1985) characterised nuclear genome containing 48% of most slowly annealing DNA (putative single copy) and 32% middle repetitive sequences with remaining highly repetitive or palindromic. Also, the whole genome size was calculated as 6 £ 108 base pairs. Further, Besse et al. (1994), using 92 clones of Amazonian prospection and 73 Wickham clones did an assessment of RFLP profiles. RFLP profiles were separated through ribosomal RNA probes and 25 low copy sequences of Hevea genome. Interestingly, the wild accessions could be categorised into genetic groups according to their geographic origin (Acre, Rondonia, Mato-Grosso). On the other hand, cultivated clones conserved relatively high level of polymorphism, despite narrow genetic base and continuous assortative mating and selection. As expected, polymorphism is very prevalent among allied species of Hevea. A comparison of isozyme analysis (Lebrun and Chevallier, 1990) with that of DNA markers showed much similarity (Besse et al., 1994). Identification of all Wickham clones could be done with 13 probes associated with restriction enzyme Eco RI (Besse et al., 1993a). However, the cultivated clones are genetically near to Mato-Grosso. Rondonia and MatoGrosso clones are more polymorphic as per RFLP data (Besse et al., 1994; Seguin et al., 1996b). A Rondonia clone (RO/C/8/9) shows eight specific restriction fragments and a unique malate dehydrogenase (MDH) allele, indicating that this clone is of interspecific origin. Such molecular markers are useful in Hevea breeding since no distinct morphological traits exist. RFLPs and DAFs were also used for identification of progeny with two common parents such as PR 255 and PR 261; RRIM 901 and RRIM 905; RRIM 937 and RRIM 938 (Low et al., 1996). Polymorphisms in microsatellites were detected in H. pauciflora, H. guianensis, H. camargoana, H. benthamiana and H. brasiliensis (Low et al., 1996). These polymorphisms must have played a role in delineating species during the course of evolution. A microsatellite-enriched library was constructed in H. brasiliensis involving four types of simple sequence repeats like (GACA)n (10%), (GATA)n (9%), (GA)n (34%) and (GC)n (9%) (Atan et al., 1996). Such exercises must contribute towards isolating clones that are diversified and can be used in recombination breeding and selection. Mitochondrial DNA (mtDNA) polymorphism was analysed in 345 Amazonian accessions, 50 Wickham clones and two allied species (H. benthamiana, H. pauciflora; Luo et al., 1995). While the variation in wild accessions was considerable, the cultivated clones formed only two clusters. Geographic specificity is shown both in nuclear and organelle RFLP profiles. It has also been shown that ribosomal DNA (rDNA) has relatively high level of variability than wild clones (Besse et al., 1993b). The aforesaid observations amply indicate that the selection was indirectly towards nuclear DNA polymorphism, while evolving modern clones. Luo et al. (1995) argue that the geographic specificity towards nuclear and mtDNA polymorphism is due to the greater level of genetic structuring among natural populations in the Amazon forests in relation to hydrographic network. In wild

386

P. M. PRIYADARSHAN

accessions, seed dispersal and selection are as per the environmental conditions, where fluctuations are less. Thus, much of the variations produced in natural habitat are being lost due to selection pressure of environmental factors. This is a matter of concern since the wild accessions have no contributions in evolving high yielding clones so far, after inhabiting to other parts of the globe. On the other hand, Wickham clones exhibited much nuclear DNA polymorphism, perhaps, due to breeding under differential geo-climatic zones with varied environmental factors. In fact, the nuclear genome has been forced to enhance variation to suite the diverse hydrothermal situations of newly introduced areas. mtDNA of Wickham clones has lesser variation for their female progenitors are all primary clones, naturally bred under the similar environmental conditions of Malaysia and Indonesia. Moreover, cytoplasmic donors for most of the improved clones are either PB 56 or Tjir 1. Obviously, this is the reason for the mtDNA profile of clones showing only two clusters. A possible explanation for greater polymorphism in mtDNA of wild accessions is that they must have been evolved through interspecific hybridisation. mtDNA polymorphism in wild accessions needs to be exploited fully. One way is to look for competent variations in their progeny and the seedlings of Wickham £ Brazilian Amazonian.

2.

Tissue Specific Gene Expression

The inquisitiveness to synthesise artificial rubber, of late, has increased the knowledge on rubber biosynthesis and on the genes involved. Genes responsible for the key enzyme for polymerisation of polyisoprenes—the rubber transferase—is one of the most abundantly expressed genes in the latex. Genes expressed in the latex can be broadly categorised into three based on their function: (a) defence genes, (b) genes for rubber synthesis, and (c) genes for allergenic proteins (Han et al., 2000). Hevein, a chitin-binding protein is one of the defence proteins that plays a crucial role in the protection of wound sites from fungal infestation. A cDNA clone (HEV 1 ) encoding Hevein was isolated by using polymerase chain reaction (PCR; Broekaert et al., 1990). HEV 1 is of 1018 base pairs and includes an open reading frame of 204 aminoacids with a signal sequence of 17 amino acid residues followed by 187 amino acid polypeptide. This polypeptide is found to contain striking features like an amino terminal region (43 amino acids) with a homology to other chitin-binding proteins and amino acid termini of wound inducible proteins in potato and poplar. It was also seen that their genes are well expressed in leaves, stems and latex (Broekaert et al., 1990). Nearly 12.6% of the proteins available in the latex are defence related (Han et al., 2000). Mainly three rubber synthesis related genes are expressed in the latex, viz., rubber elongation factor (REF; Dennis and Light, 1989; Goyvaerts et al., 1991), HMG CoA reductase (Chy et al., 1992) and small rubber particle protein

BREEDING HEVEA BRASILIENSIS

387

(SRPP; Oh et al., 1999). They constitute the 200 odd distinct polypeptides (Posch et al., 1997). The most abundantly expressed gene is that of REF (6.1%) followed by SRPP (3.7%) (Han et al., 2000). These expressed sequences (expressed sequence tags—ESTs) were compared with public databases of identified genes. About 16% of the database matched ESTs encoding rubber biosynthesis related proteins. Analysis of ESTs revealed that rubber biosynthesis-related genes are expressed maximum followed by defence-related genes and protein-related genes (Han et al., 2000). Unlike photosynthetic genes, transcripts involved in rubber biosynthesis are 20 –100 times greater in laticifers than in leaves (Kush et al., 1990). On the other hand, transcripts for chloroplastic and cytoplasmic forms of glutamine synthase are restricted to leaves and laticifers, respectively (Kush et al., 1990), indicating thereby that the cytoplasmic form of glutamine synthase plays a decisive role in amino acid metabolism of laticifers. Studies on laticifer specific gene expression have important implications on selection and breeding. It would be worthwhile to use transcript levels as molecular markers for early selection (Kush et al., 1990). The transcript levels of hydrolytic enzymes, viz., polygalacturonase and cellulase shall be taken as indicators for better laticifer development. It is felt that extensive studies on expression of genes are mandatory to unravel the intricacy of latex production. Detection and evaluation of more molecular markers must also help to breed Hevea at molecular level, to derive clones exclusively for marginal areas.

3.

Molecular Linkage Maps and QTLs

A comprehensive genetic linkage map of H. brasiliensis has been formulated recently with the help of RFLPs, AFLPs, microsatellites and isozyme markers (Lespinasse et al., 2000a). This was accomplished through a double pseudo-test cross as per the methodology of Grattapaglia and Sederoff (1994) and a map was constituted separately for each parent. Further, homologous markers segregating in both parents were ascertained and consensus map prepared. The parents used were PB 260 (PB5/51 £ PB 49) and RO 38 (F4542 £ AVROS 363). F4542 is a clone of H. benthamiana. The F1 synthetic map of 717 markers was distributed in 18 linkage groups. This comprised of 301 RFLP, 388 AFLP, 18 microsatellite and 10 isozyme markers (Fig. 8). Identification of loci was based on mobility of electrophoretic bands, necessitating verification of consistency of the location of alleles in both parental maps. The genetic length of 18 chromosomes was fairly homogeneous with an average map length per chromosome of 120 cM. Many AFLP markers were seen in clusters, which were attributed as reduced recombination frequency regions. Though the RFLP markers were well distributed all over the 18 linkage groups, these were insufficient to saturate the map. AFLPs and few microsatellites

388 P. M. PRIYADARSHAN Figure 8 F1 synthetic map of 717 markers distributed in 18 linkage groups. This map encompasses 301 RFLP, 388 AFLP, 18 microsatellite and 10 isozyme markers (after Lespinasse et al., 2000a). Hb CR RFLP probe, RGA R gene RFLP probe, EM AFLP, M microsatellite. Lx suffix duplicate loci, PB and RO suffix parents (PB 260 and RO 38) for markers present in both parents. Bridge markers are indicated in bold italic.

Continued.

389

Fig. 8

BREEDING HEVEA BRASILIENSIS

390

P. M. PRIYADARSHAN

together enriched saturating the map. However, these exercises are the initial steps for making a total genetic linkage map of Hevea in future. The isozymes were found to inherit following 1:1 ratio (Chevallier, 1988). On the other hand, a partially non-random arrangement of duplicate loci was observed by Lespinasse et al. (2000a) in their RFLP profiles with certain chromosome pairs indicating that they have homology descending from a common ancestor. There are reasons to believe that these duplications may have occurred during the course of evolution. This would also indicate there are regions of homoeology, whose origin is still unknown and H. brasiliensis continues to behave as a diploid. QTLs for resistance to SALB (M. ulei ) were mapped using 195 F1 progeny derived from a cross between PB 260 (susceptible) and RO 38 (resistant) clones (Lespinasse et al., 2000b), which was done in continuation to a genetic analysis done earlier (Seguin et al., 1996a). Eight QTLs were identified for resistance in RO 38 map through Kruskel –Wallis marker-by-marker test and interval mapping method (Lander and Botstein, 1989). The F1 consensus map confirmed the results obtained in parental maps. Lespinasse et al. (2000b) further rationalised that the resistance (alleles) of RO 38 have inherited from the wild grand parent (H. benthmiana ) and no favourable alleles came from AVROS 363, the Wickahm parent. Eight different QTLs for five strains of fungi were available in RO 38, with specificity of resistance to different strains. More durable resistance shall be available in other allied species and wild accessions of Hevea. However, the selection of clones with durable resistance with polygenic determinism is of much importance while undertaking such studies (Rivano, 1997). Darmono and Chee (1985) while studying the lesion size on leaf discs, identified SIAL 263, an illegitimate progeny of RRIM 501 as resistant to SALB. 4.

Direct Gene Transfer

The stable introduction of foreign genes into plant cells through direct gene transfer systems has opened up incredible avenues in the improvement of crops, especially perennial species, and rubber is no exception. While the in vitro plant regeneration system in rubber is getting standardised in few laboratories worldwide, efforts have been made to transform Hevea cells through Agrobacterium tumefaciens in order to complement plant breeding efforts to increase genetic variation (Arokiaraj et al., 1994). The anther-derived calli were transformed with A. tumefaciens having b-glucuronidase (gus ) gene and neomycin phosphotransferase (npt II) genes. Fluorometric assay and enzymelinked immunosorbent assay (ELISA) were performed to prove the expression of genes and npt II genes, respectively, in calli and embryoids (Arokiaraj et al., 1996). Further, the expression of foreign proteins in Hevea latex was also demonstrated in 1998 (Arokiaraj et al., 1998). This transformation appeared stable even after three vegetative generations with no chimeras, indicating

BREEDING HEVEA BRASILIENSIS

391

thereby, that a single transformed plant is sufficient to have a population achieved through budding. But this exercise would not take care of the stock – scion interaction and ensuing yield variation in a clonal population.

VI. CONCLUSIONS Rubber breeding has been successful in achieving substantial yield improvements. However, research needs to be reconstructed through a multifaceted approach, that concerted efforts must take rubber into new hard areas. The conclusions drawn from the review are as follows: (1) Selection for yield per se is the final criterion for breeding higher yield under any environment because, yield is an output from a complex holistic system (Wallace and Yan, 1998). In short, the increased knowledge about the components that govern yield will not shorten the time required to breed new clones. Creation of superior genetic segregates and evaluating them for environmental constraints give a holistic approach. (2) The spectrum of useful genetic variation need to be enlarged, especially through utilising variable cytoplasmic donors like RO/C/8/9, since most oriental clones received cytoplasm either from PB 56 (through PB 5/51) or Tjir 1. One of the options shall be to cross better yielders with new cytoplasmic donors ascertained after a molecular analysis of mtDNA variation. The exercise of backcrosses would become inevitable to retain the cytoplasm and the desirable nuclear genes. Large scale clone trials (LSCTs) can be directly laid for assessing the performance of newer genetic combinations. Yield system analysis through AMMI (Gauch, 1992) or pattern analysis (Yan and Hunt, 1998) is the superior way to select genetic diversity of parental germplasm for maximising the number of segregates. (3) There is a need to augment research on direct transfer of genes for apomixis to gain somatic seeds. Though sizeable work has been carried out at CIRAD, France, on micropropagation and acclimatisation of more than 13,000 plants under differential climatic conditions, exploitation of somaclonal variation is still primitive due to want of appropriate regeneration protocols. Any effort to achieve genetic diversity is substantially recognisable. The utility of apomixis, a natural phenomenon by which embryos are formed without meiosis or fertilisation needs to be explored since apomictically produced embryos are genetically identical to the female parent and analogous to somatic embryos. The case of guayule (P. argentatum ) is a fine example. While in guayule the expression of apomixis is evident and prominent, H. brasiliensis owns recession. Polyploid forms of guayule are obligate apomicts and diploids are sexually reproducing. Three pairs of genes are accounted to be involved in the determination of breeding behaviour. The gene a in homozygous condition leads to the formation of unreduced egg and gene b prevents fertilisation and gene c stimulates egg to develop without fertilisation.

392

P. M. PRIYADARSHAN

Plants with AAbbcc and aaBBcc can have unreduced eggs but cannot develop into embryos in the absence of fertilisation. Plants with AABBcc will have a normal sexual behaviour. Only plants with a genetic makeup of aabbcc will be apomictic (Bhojwani and Bhatnagar, 1992). Since apomictic and non-apomictic biotypes are morphologically and cytologically distinguishable, characterisation of genes at molecular level will also be possible. Our studies with immature embryos of Hevea demonstrated ovules-lodging abortive embryos have the tendency to induce adventive embryony from nucellus, exercising an extreme chance for reproduction and continuation of generations (Sowmyalatha et al., 1997). However, embryos are seen to be degenerating, which amply indicates the presence of genes meant for apomixis, but lack of proper activation/stimulus stands as a constrain in expressivity. Thus, research on apomixis needs further consideration at molecular level. In addition to achieving homogenous populations, apomictic seeds would ensure a tap root system and nullify expenditure towards raising of bud grafted poly bag plants. (4) Research on molecular markers that can be used in early selection of high yielding clones in order to shorten the breeding cycle needs to be augmented. The higher transcript levels of hydrolytic enzymes like polygalacturonase and cellulase can be the indicators for better laticifer development. (5) Allied species shall be incorporated in recombination breeding. H. camporum, H. guianensis, H. pauciflora, H. rigidifolia and H. spruceana exhibit attributes like partial defoliation that exempts infestation of powdery mildew. Similarly, such attributes must be the expression of abilities towards circumventing moisture and low temperature stresses. There is a potential for developing latex timber clones from allied species and a few Amazonian accessions. (6) International co-operation to have joint research programs need to be initiated especially in the expensive areas like biotechnology through scientists exchange programs. The aforesaid aspects, in addition to the ongoing need to be integrated into the research programs being pursued worldwide.

ACKNOWLEDGEMENTS The author is thankful to a number of scientists for providing information on the performance of rubber under various marginal environments, especially to Dr P.de S. Goncalves, of Instituo Agonomico Campinas, Brazil, Dr ClementDemange, Dr M. Seguin, Dr D. Lespinasse of CIRAD, France, Dr Ridha Arizal of Rubber Technology Centre, Bogor, Indonesia, Dr T.T.T. Hoa of The Rubber Research Institute of Vietnam and Dr Keith Chapman, FAO, Bangkok. The facilities and encouragements rendered by Dr N.M. Mathew, Director, Rubber Research Institute of India are gratefully acknowledged.

BREEDING HEVEA BRASILIENSIS

393

REFERENCES Anonymous. (1996). Rubber wood industry heading for a big boom. Rubber Asia 10(1), 93–96. Arokiaraj, P., Jones, H., Cheong, K. F., Coomber, S., and Charlwood, B. V. (1994). Gene insertion into Hevea brasiliensis. Plant Cell. Rep. 13, 425 –431. Arokiaraj, P., Jones, H., Jaafar, H., Coomber, S., and Charlwood, B. V. (1996). Agrobacterium mediated transformation of Hevea anther calli and their regeneration in to plants. J. Nat. Rubber Res. 11, 77–87. Arokiaraj, P., Yeang, H. Y., Cheong, K. F., Hamzah, S., Jones, H., Coomber, S., and Charlwood, B. V. (1998). Ca MV 35 S promoter directs b-glucuronidase expression in the laticiferous system of transgenic Hevea brasiliensis (rubber tree). Plant Cell Rep. 17, 621–625. Arshad, N. L., Othamn, R., and Yacob, A. R. W. (1995). Hevea wood availability in Peninsular Malaysia. RRIM Planters Bull. 224–225, 73–83. Atan, S., Low, F. C., and Saleh, N. M. (1996). Construction of a microsatellite enriched library from Hevea brasiliensis. J. Nat. Rubber Res. 11, 247– 255. Backhaus, R. A. (1985). Rubber formation in plant—a mini review. Israel J. Bot. 34, 283–293. Barlow, C. (1997). Growth, structural change and plantation tree crops: The case of rubber. World Develop. 25, 1589–1607. Barry, R. G., and Chorley, R. J. (1976). In “Atmosphere, weather and climate” (2nd ed.). Methuen, London. Baulkwill, W. J. (1989). The history of natural rubber production. In “Rubber” (C. C. Webster and W. J. Baulkwill, Eds.), pp. 1 –56. Longman, Essex. Besse, P., Lebrun, P., Seguin, M., and Lanaud, C. (1993a). DNA fingerprints in Hevea brasiliensis (rubber tree) using human minisatellite probes. Heredity 70, 237–244. Besse, P., Seguin, M., Lebrun, P., and Lanaud, C. (1993b). Ribosomal DNA variations in wild and cultivated rubber tree (Hevea brasiliensis ). Genome 36, 1049–1057. Besse, P., Seguin, M., Lebrun, P., Chevallier, M. H., Nicolas, D., and Lanaud, C. (1994). Genetic diversity among wild and cultivated populations of Hevea brasiliensis assessed by nuclear RFLP analysis. Theor. Appl. Genet. 88, 199 –207. Bhojwani, S. S., and Bhatnagar, S. P. (1992). In “The Embryology of Angiosperms”. Vikas Publishing House, New Delhi, India. Brazil (1971). Ministe´rio da Indu´stria e Come´rcio. Superintendeˆncia da Borracha. O geˆnero Hevea, descric¸a˜o das espe´cies e distribuic¸a˜o geogra´fica. Rio de Janeiro, Sudhevea, 1971 (Plano Nacional da Borracha, anexo 7). Broekaert, N., Lee, H., Kush, A., Chua, N. H., and Raikhel, N. (1990). Wound induced accumulation of mRNA containing a hevein sequence in laticifer of rubber tree (Hevea brasiliensis ). Proc. Natl. Acad. Sci. USA 87, 7633–7637. Cain, M. (2001). 39th IRSG Assembly in Antwerp. Indian Rubber J. 52, 76 –77. Chandrashekar, T. R. (1997). Stomatal responses of Hevea to atmospheric and soil moisture stress under dry sub humid climatic conditions. J. Plant. Crops 25, 146 –151. Chandrashekar, T. R., Jana, M. K., Thomas, J., Vijayakumar, K. R., and Sethuraj, M. R. (1990). Seasonal changes in physiological characteristics and yield in newly opened trees of Hevea brasiliensis in North Konkan. Indian J. Nat. Rubber Res. 3, 88 –97. Chandrashekar, T. R., Marattukalam, J. G., and Nazeer, M. A. (1996). Growth reaction of Hevea brasiliensis to heat and drought under dry sub humid climatic conditions. Indian J. Nat. Rubber Res. 9, 1–5. Chandrashekar, T. R., Nazeer, M. A., Marattukalam, J. G., Prakash, G. P., Annamalainathan, K., and Thomas, J. (1998). An analysis of growth and drought tolerance in rubber during the immature phase in a dry sub humid climate. Exp. Agric. 34, 1–14.

394

P. M. PRIYADARSHAN

Chapman, K. (2000). In “FAO Fact-finding Mission on Rubber in Vietnam—Final Mission Report”, p. 30. FAO, Bangkok. Chee, K. H. (1976). Assessing susceptibility of Hevea clones to Microcyclus ulei. Ann. Appl. Biol. 84, 135–145. Chevallier, M. H. (1988). Genetic variability of Hevea brasiliensis germplasm using isozyme markers. J. Nat. Rubber Res. 3, 42–53. Chye, M. L., Tan, C. T., and Chua, N. H. (1992). Three genes encode 3-hydroxy-3-methyl glutarylcoenzyme A reductase in Hevea brasiliensis. hmg1 and hmg3 are differentially expressed. Plant Mol. Biol. 19, 473– 484. Clement-Demange, A., Chapuset, T., Legnate, H., Costes, E., Doumbia, A., Obouayeba, S., and Nicolas, D (1998). Wind damage: the possibilities of an integrated research for improving the prevention risks and the resistance of clones in the rubber tree. In “IRRDB Symposium 1997,” General, Soils and Fertilization, Breeding and Selection, 14 –15 October 1997, Vol. I, pp. 182 –199. Ho Chi Minh City. Clement-Demange, A., Legnate, H., Seguin, M., Carron, M. P., Guen, V. Le., Chapuset, T., and Nicolas, D. (2000). Rubber Tree. Collection Reperes. In “Tropical Plant Breeding”. (A. Charrier, M. Hamon, S. Hamon, and D. Nicolas, Eds.), pp. 455 –480. CIRAD-ORSTOM, Montpellier, France. Cornish, K., and Siler, D. J. (1996). Hypoallergenic guayule latex provides the only immediate natural rubber solution to life threatening Hevea latex allergy, In “Proceedings of the Third International Conference on Industrial Crops and Products,” pp. 262–266. USDA-ARS, USA. Cornish, K., Siler, D. J., Grosjean, O., and Goodman, N. (1993). Fundamental similarities in rubber particle architecture and function in three evolutionarily divergent plant species. J. Nat. Rubber Res. 8, 275 –285. Costa, R. B., Resende, M. D. V., Araujo, A. J., Goncalves, P. de S., and Martins, A. L. M. (2000). Genotype–environment interaction and the number of test sites for the genetic improvement of rubber trees (Hevea ) in Sao Paulo State, Brazil. Genet. Mol. Biol. 23, 179–187. Darmono, T. W, and Chee, K. H. (1985). Reaction of Hevea clones to races of Microcyclus ulei in Brazil. J. Rubber Res. Inst. Malaysia 33, 1–8. Davies, W. (1997). The rubber industry’s biological nightmare. Fortune 136, 86. Dea, G. B., Keli, Z. J., Eschbach, J. M., Omont, H., and Canh, T (1997). Rubber tree (Hevea brasiliensis ) behaviour in marginal climatic zones of Coˆte d’Ivore: Assessment of ten years observations. In “IRRDB Symposium on Agronomical Aspects of the Cultivation of Natural Rubber (Hevea brasiliensis ),” pp. 44 –53. Beruwele, Sri Lanka. Dean, W. (1987). In “Brazil and the Struggle for Rubber”. Cambridge University Press, Cambridge. Dennis, M. S., and Light, D. R. (1989). Rubber elongation factor from Hevea brasiliensis. Identification, characterization and role in rubber biosynthesis. J. Biol. Chem. 264, 18608–18617. Devakumar, A. S., Sathik, M. B. M., Jacob, J., Annamalainathan, K., Prakash, P. G., and Vijayakumar, K. R. (1998). Effects of atmospheric and soil drought on growth and development of Hevea brasiliensis. J. Rubber Res. 1, 190–198. Dijkman, M. J. (1951). In “Hevea—Thirty Years of Research in the Far East”. University of Miami Press, Coral Gables, Florida. Edingon, J. A. S. (1991). In “Rubber in West Africa”, pp. 267. Rex Collings, Lagos. Eschbach, J. M., Clement-Demange, A., and Hoa, T. T. T (1998). The potential for rubber small holder development on the Vietnam highlands and a proposal for an adaptive research programme. In “IRRDB Symposium on Natural Rubber,” Vol. III, pp. 6 –13. Rubber Small Holdings, Natural Rubber Processing, Quality and Technology Sessions. Ho Chi Minh City, Vietnam. Estilai, A., and Ray, T. (1991). Genetics, cytogenetics and breeding of guayule. In “Guayule Natural Rubber: A Technical Publication with Emphasis on Recent Findings”. (J. W. Whitworth and E. E. Whiteread, Eds.), pp. 47 –91. USDA, University of Arizona, USA.

BREEDING HEVEA BRASILIENSIS

395

Gauch, H. G. (1992). In “Statistical Analysis of a Regional Yield Trial—AMMI Analysis of Factorial Designs”. Elsevier, New York. George, P. J., and Panikkar, A. O. N. (2000). Rubber yielding plants. In “Natural Rubber: Agro management and Crop Processing.” (P. J. George and C. K. Jacob, Eds.), pp. 20–28. Rubber Research Institute of India, Kottayam, India. Gilbert, N. E., Dodds, K. S., and Subramaniam, S. (1973). Progress of breeding investigations with Hevea brasiliensis. V. Analysis of data from earlier crosses. J. Rubber Res. Inst. Malaysia 23, 365 –380. Gohet, E., Prevot, J. C., Eschbach, J. M., Clement-Demange, A., and Jacob, J. L. (1995). Clone, growth and stimulation: Latex production factors. Plantations, Recherche´, Developpement January–February, 36–38. Goncalves, P. de S., Paiva, J. de R. (1983). Retrospectiva e Atualidade do Melhoramento Genetico da Seringueira (Hevea spp.) no Brasil e em paises Asiatics. EMBRAPA-CNPSD Documentos 2, Manaus, 1983. Goncalves, P. de S., Cardoso, M., and Ortolani, A. A. (1990). Origin, variability and domestication of Hevea. Pesquira Agropecuaria Brasileira Brasilia 25, 135–156. Goncalves, P. de S., Bataglia, O. C., Santoa, W. P. do S., Ortolani, A. A., Segnini, I., Jr., and Shikasho, E. H. (1998). Growth trends, genotype–environment interaction and genetic gains in six year old rubber tree clones (Hevea ) in Sao Paulo State, Brazil. Genet. Mol. Biol. 21, 15–122. Goncalves, P. de S., Bortoletto, N., Sambugaro, R., Furtado, E. L., Bataglia, O. C., Ortolani, A. A., and Godoy, G. G. (2001). Performance of Amazonian rubber tree clones in the plateau region of the state of Sa˜o Paulo, Brazil. Pesquisa Agropecuaria Brasileira, Brasilia 36, 1469– 1477. Goyvaeerts, E., Dennis, M., Light, D., and Chua, N. H. (1991). Cloning and sequencing of cDNA encoding the rubber elongation factor of Hevea brasiliensis. Plant Physiol. 97, 317–321. Grattapaglia, D., and Sederoff, R. (1994). Genetic linkage maps of Eucalyptus grandis and Eucalyptus urophylla using a pseudo-test cross mapping strategy and RAPD markers. Genetics 137, 1121–1137. Greek, B. F. (1991). Rubber demand is expected to grow after 1991. C&E News 69, 37 –54. Han, K., Shin, D. O., Yang, J., Kim, I. J., Oh, S. K., and Chow, K. S. (2000). Genes expressed in the latex of Hevea brasiliensis. Tree Physiol. 20, 503 –510. Haridas, G. (1985). Streamflow measurements in a small watershed to estimate evaporation from a stand of rubber, In “Proceedings of the Rubber Conference 1985,” pp. 670–681. Kuala Lumpur. Heywood, V. H. (1978). In “Flowering Plants of the World”. Oxford University Press, Oxford. Ho, C. Y. (1986). Breeding for Durable Resistance in Perennial Crops. In “Rubber, Hevea brasiliensis”. Vol. 70, pp. 85–114. FAO. Ho, C. Y., and Ong, S. H. (1981). Potentials of wide crosses in Hevea breeding. In “Third International Congress of SABRAO,” pp. 447–469. Kuala Lumpur. Ho, C. Y., Khoo, S. K., Meiganaratnam, K., and Yoon, P. K. (1979). Potential new clones from mother tree selection. In “Proceedings of the Rubber Research Institute, Malaysia Planters’ Conference”, pp. 201. Kuala Lumpur. Hoa, T. T. T., Tuy, L. M., Duong, P. H., Phuc, L. G. T., and Truong, V. V. (1998). Selection of Hevea clones for the 1998–2000 planting recommendation in Vietnam. In “Proceedings of the IRRDB Symposium on Natural Rubber,” General, Soils and Fertilization and Breeding and Selection Sessions, Vol. I, pp. 164 –177. Ho Chi Minh City, Vietnam. Hoa, T. T. T., Tuy, L. M., Duong, P. H., and Lam, L. V. (2002a). Performance of promising Hevea clones in sub-optimal zones of Vietnam. In “Proceedings of the Conference on Future of Perennial Crops,” November 2001, Cote d’Ivoire (CD-ROM of the conference session 5.3 is published by Bureau National d’ Etudes Techniques et de Developpement, Coˆte d’ Ivoire; Centre de Cooperation Internationale en Recherche Agronomique pour le Developpement, France; Economic and Social Institute, Netherlands).

396

P. M. PRIYADARSHAN

Hoa, T. T. T., Hue, N. T., and Enjalric, F. (2002b). Rubber development in Vietnam: current status and strategy for 2001–2010. In “Proceedings of the Conference on Future of Perennial Crops,” November 2001, Cote d’Ivoire (CD-ROM of the conference session 4.3 is published by Bureau National d’ Etudes Techniques et de Developpement, Coˆte d’ Ivoire; Centre de Cooperation Internationale en Recherche Agronomique pour le Developpement, France; Economic and Social Institute, Netherlands). Huasun, H., Qiuto, C., and Yutong, W. (1998). A statistical analysis of potentials and performance of some new Chinese Hevea clones. In “Proceedings of IRRDB Symposium on Natural Rubber,” General, Soils and Fertilization and Breeding and Selection Sessions, Vol. I, pp. 140–148. Ho Chi Minh City, Vietnam. Jacob, J. L., Prevot, J. C., Lacrotte, R., Clement-Demange, A., Serres, E., and Gohet, E. (1995). Clonal typology of laticifer functioning in Hevea brasiliensis. Plantations, Recherche´, Developpement September–October, 48–49. Jacob, J., Annamalainathan, K., Alam, B., Sathik, M. B. M., Thapliyal, A. P., and Devakumar, A. S. (1999). Physiological constraints for cultivation of Hevea brasiliensis in certain unfavourable agroclimatic regions of India. Indian J. Nat. Rubber Res. 12, 1–16. Jayasekera, N. E. M., and Karunasekera, K. B. (1984). Proceedings of IRRDB Meeting on Hevea Physiology, Exploitation and Breeding. In “Effect of environment on clonal performance with respect to early vigour and yield in Hevea brasiliensis”, pp. 250–255. IRRDB, Montpellier, France. Jiang, A. (1984). A geo-ecological study of rubber tree cultivation at high altitude in China. In “Proceedings of the International Rubber Conference,” pp. 117–130, Colombo, Sri Lanka. John, P. (1992). Biosynthesis of the Major Crop Products. In “Rubber,” pp. 114 –126. Wiley, England. Johnston, A. (1989). Diseases and pest. In “Rubber”. (C. C. Webster and W. J. Baulkwill, Eds.), pp. 415 –458. Longman, New York. Krisanasap, S. and Dalkit, P (1989). Rubber new plantings in the semi arid zone in Thailand. In “Proceedings of the Rubber Institute Malaya Rubber Growers’ Conference,” pp. 228– 237. Malacca, Malaysia. Kush, A., Goyvaerts, E., Chye, M., and Chua, N. H. (1990). Laticifer specific gene expression in Hevea brasiliensis (rubber tree). Proc. Natl. Acad. Sci. USA 87, 1787–1790. Lander, B. S., and Botstein, D. (1989). Mapping Mendelian factors underlying quantitative traits using RFLP linkage maps. Genetics 121, 185–199. Lebrun, P., and Chevallier, M. H. (1990). In “Starch and Polyacrylamide Gel Electrophoresis of Hevea brasiliensis—A Laboratory Manual”. IRCA/CIRAD, Montpellier, France. Lee, C. K., and Tan, H. (1979). Daily variations in yield and dry rubber content in four Hevea clones. J. Rubber Res. Inst. of Malaysia 27, 117–1126. Lespinasse, D., Rodier-Goud, M., Grivet, L., Leconte, A., Legnate, H., and Seguin, M. (2000a). A saturated genetic linkage map of rubber tree (Hevea spp.) based on RFLP, AFLP, microsatellite, and isozyme markers. Theor. Appl. Genet. 100, 127–138. Lespinasse, D., Grivet, L., Troispoux, V., Rodier-Goud, M., Pinard, F., and Seguin, M. (2000b). Identification of QTLs involved in the resistance to South American Leaf Blight (Microcyclus ulei ) in the rubber tree. Theor. Appl. Genet. 100, 975–984. Lewinsohn, T. M. (1991). The geographical distribution of plant latex. Chemoecology 2, 64– 68. Liyanage, A. de S., Gibb, A., and Weerasinghe, A. R. (1984). A crop weather calendar for rubber. In “Proceedings of International Rubber Conference Colombo,” pp. 354 –366, Sri Lanka. Low, F. C., and Bonner, J. (1985). Characterization of the nuclear genome of Hevea brasiliensis. In “International Rubber Conference 1985,” pp. 1 –9, Kuala Lumpur, Malaysia. Low, F. C., Atan, S., Jaafar, H., and Tan, H. (1996). Recent advances in the development of molecular markers for Hevea studies. J. Nat. Rubber Res. 11, 32–44.

BREEDING HEVEA BRASILIENSIS

397

Luo, H., Van Coppenolle, B., Seguin, M., and Boutry, M. (1995). Mitochondrial DNA polymorphism and phylogenetic relationships in Hevea brasiliensis. Mol. Breed. 1, 51–63. Majumder, S. K. (1964). Chromosome studies of some species of Hevea. J. Rubb. Res. Inst. Malaysia 18, 269–273. Meenattoor, J. R., Vinod, K. K., Krishnakumar, A. K., Sethuraj, M. R., Potty, S. N., and Sinha, R. R. (1991). Clone £ environment interaction during early growth phase of Hevea brasiliensis. I. Clonal stability on girth. Indian J. Nat. Rubber Res. 4, 51– 54. Meenattoor, R. J., Sasikumar, S., Soman, T. A., Gupta, C. K., Meti, S., Meenakumari, T., Nair, R. B., Licy, J., Saraswathyamma, C. K., and Brahmam, M. (2000). Genotype £ environment interactions in Hevea in diverse agroclimatic conditions in India—preliminary results. In “Proceedings of the International Planters’ Conference May 2000,” pp. 183 –195. Mohanakrishana, T., Bhaskar, C. V. S., Sanjeeva Rao, P., Chandrashekar, T. R., Sethuraj, M. R., and Vijayakumar, K. R. (1991). Effect of irrigation on physiological performance of immature plants of Hevea brasiliensis in North Konkan. Indian J. Nat. Rubber Res. 4, 36–45. Mondal, G. C., Das, K., Singh, R. P., Mandal, D., Gupta, C. K., Gohain, T., Deka, H. K., and Thapliyal, A. P. (1999). Performance of Hevea clones in Assam. Indian J. Nat. Rubber Res. 12, 55 –61. Monteny, B. A., Barbier, J. M., and Bernos, C. M. (1985). Determination of the energy exchanges of forest type culture: Hevea brasiliensis. In “Forest Atmospheric Interaction”. (B. A. Hutchinson and B. B. Hicks, Eds.), pp. 211–233. Reidel, Dordrecht. Moraes, V. H. F. (1977). Rubber. In “Ecophysiology of Tropical Crops”. (A. T. Alvin and H. Kozlowski, Eds.), pp. 315–331. Academic Press, New York. Mydin, K. K., Nazeer, M. A., George, P. J., and Panikkar, A. O. N. (1994). Long term performance of some hybrid clones of rubber with special reference to clonal composites. J. Plant. Crops 22, 19 –24. Nazeer, M. A., George, P. J., Premakumari, D., and Marattukalam, J. G. (1991). Evaluation of certain primary and secondary Hevea clones in large scale trial. J. Plant. Crops 18, 11–16. Nga, B. H., and Subramaniam, S. (1974). Variation in Hevea brasiliensis. 1. Yield and girth data of the 1937 hand pollinated seedlings. J. Rubb. Res. Inst. Malaysia 24, 69–74. Oh, S. K., Kang, H. S., Shin, D. H., Yang, J., Chow, K. S., Yeang, H. Y., Wagner, B., Breiteneder, H., and Han, K. H. (1999). Isolation characterization and functional analysis of a novel cDNA clone encoding a small rubber particle (SRPP) from Hevea brasileinsis. J. Biol. Chem. 274, 17132–17138. Oldeman, L. R., and Frere, M. (1982). A study of the agroclimatology of the humid tropics of South East Asia. FAO/UNESCO/WMO interagency project on agroclimatology. Technical Report, Rome, Tech Note No. 179 WMO No. 597, 230 p. Ong, S. H., Ghani, M. N. A., Tan, A. M., and Tan, H. (1983). New Hevea germplasm—its introduction and potential. In “Proceedings of the Rubber Research Institute Malaysia Rubber Planters’ Conference 1983”. pp. 1–14. Kuala Lumpur. Ong, S. H., Othman, R., and Benong, M. (1998). Breeding and selection of clonal genotypes for climatic stress conditions. In “Proceedings of the IRRDB Symposium on Natural Rubber (Hevea brasiliensis ),” General, Soils and Fertilization and Breeding and Selection, Vol. I, pp. 149–154. Ho Chi Minh City, Vietnam. Ortolani, A. A., Sentelhas, P. C., Camargo, M. B. P., Pezzopane, J. E. M., and Goncalves, P. de S. (1998). Agrometeorological model for seasonal rubber tree yield. Ind. J. Nat. Rubber Res. 11, 8 –14. Paardekooper, E. C., and Sookmark, S. (1969). Clonal variation in latex yield and dry rubber content in relation to saturation deficit of air. J. Rubber Res. Inst. Malaya 21, 341–347. Pakianathan, S. W., Haridas, G., and d’Auzac, J. (1989). Water relations and latex flow. In “Physiologyof Rubber Tree Latex.” (J. d’Auzac, J. L. Jacob and H. Chrestin, Eds.), pp. 233–256. CRC Press, Boca Raton.

398

P. M. PRIYADARSHAN

Pires, J. M. (1973). Revisa˜o do geˆnero Hevea: descric¸a˜o da espe´cies e distribuic¸a˜o geogra´fica. Relato´rio Anual, 1972. Bele´m, Instituto de Pesquisa Agropecua´ria do Norte, 1973, pp. 6 –66 (Projeto de Botaˆnica—Subprojeto revisa˜o do geˆnero Hevea. Sudhevea/Dnpea (Ipian)). Posch, A., Chen, Z., Wheeler, C., Dunn, M. J., and Baur, X. (1997). Characterisation and identification of latex allergens by two-dimensional electrophoresis and protein micro sequencing. J. Allerg. Chem. 99, 385–395. Priyadarshan, P. M., and Goncalves, P. de S. (2002). Hevea gene pool in rubber tree (Hevea brasiliensis Muell.- Arg.) breeding. The Planter, Kuala Lumpur. 78, 123–138. Priyadarshan, P. M., Vinod, K. K., Rajeswari, M. R., Pothen, J., Sowmyalatha, M. K. S., Sasikumar, S., Raj, S., and Sethuraj, M. R. (1998a). Breeding Hevea brasiliensis Muell. Arg. in Tripura (N.E. India). I. Performance of few stress tolerant clones in the early phase. In “Developments in Plantation Crops Research.” (N. M. Mathew and C. Kuruvilla Jacob, Eds.), pp. 63–68. Allied Publishers, New Delhi, India. Priyadarshan, P. M., Sowmyalatha, M. K. S., Sasikumar, S., Varghese, Y. A., and Dey, S. K. (1998b). Relative performance of six Hevea brasiliensis clones during two yielding regimes in Tripura. Indian J. Nat. Rubber Res. 11, 67–72. Priyadarshan, P. M., Sowmyalatha, M. K. S., Sasikumar, S., Varghese, Y. A., and Dey, S. K. (2000a). Evaluation of Hevea brasiliensis clones for yielding trends in Tripura. Indian J. Nat. Rubber Res. 13, 56– 63. Priyadarshan, P. M., Sasikumar, S., Nair, R. B., and Dey, S. K. (2000b). Long term stability in yielding potential in clones of Hevea brasiliensis in Tripura. In “International Conference on Plantation Crops, PLACROSYM XIV Hyderabad, December 12– 15 (abst.)”. Priyadarshan, P. M., Sasikumar, S., and Goncalves, P. de S. (2001). Phenological changes in Hevea brasiliensis under differential geo-climates. The Planter, Kuala Lumpur 77, 447 –459. Pushparajah, E. (1983). Problems and potentials for establishing Hevea under difficult environmental conditions. The Planter, Kuala Lumpur 59, 242–251. Pushparajah, E. (2001). Natural rubber. In “Tree Crop Ecosystems” (vol. 19 Ecosystems of the World Series) (F. T. Last, Ed.), pp. 379–407. Elsevier Science, Amsterdam. Rao, P. S., and Vijayakumar, K. R. (1992). Climatic requirements. In “Natural Rubber”. (M. R. Sethuraj and N. M. Mathew, Eds.), pp. 200– 219. Elsevier, Amsterdam. Rao, G. G., Rao, P. S., Rajagopal, R., Devakumar, A. S., Vijayakumar, K. R., and Sethuraj, M. R. (1990). Influence of soil, plant and meteorological factors on water relations and yield of Hevea brasiliensis. Int. J. Biometeorol. 34, 175 –180. Rao, P. S., Jayaratnam, K., and Sethuraj, M. R. (1993). An index to assess areas hydrothermally suitable for rubber cultivation. Indian J. Nat. Rubber Res. 6, 80–91. Rivano, F. (1997). South Amercian ‘blight of Hevea. I. Variability of Microcyclus ulei pathogenicity. Plantations, Recherche´, Developpement 4, 104– 114. Saengruksowong, C., Dansagoonpan, S., and Thammarat, C. (1983). Rubber planting in the North Eastern and Northern regions of Thailand. In “Proceedings of the IRRDB Symposium 1983,” Beijing, China, 12–14 May. Sasikumar, S., Priyadarshan, P. M., Dey, S. K., and Varghese, Y. A. (2001). Evaluation polyclonal seedling population of Hevea brasiliensis (Willd. Ex. Adr. De Juss.) Muell. Arg. in Tripura. Indian J. Nat. Rubber Res. 14, 125–130. Schultes, R. E. (1970). The history of taxonomic studies in Hevea. Bot. Rev. 36, 197–211. Schultes, R. E. (1977). Wild Hevea: An untapped source of germplasm. J. Rubber Res. Inst. Sri Lanka 54, 227 –257. Seguin, M., Lespinasse, D., Rodier-Goud, M., Legnate, H., Troispoux, V., Pinard, F., and Cle´mentDemange, A. (1996a). Genome mapping in connection with resistance to the South American Leaf Blight in rubber tree (Hevea brasiliensis ). Third ASAP Conference on Agricultural Biotechnology. National Centre for Genetic Engineering and Biotechnology, BIOTECH. Hua-Hin, Thailand, 10–15 November 1996, p. 3.

BREEDING HEVEA BRASILIENSIS

399

Seguin, M., Besse, P., Lespinasse, D., Lebrun, P., Rodier-Goud, M., and Nicolas, D. (1996). Hevea molecular genetics. Plantations, Recherche and Developpment March–April, 85–87. Serres, E., Lacrotte, R., Prevot, J. C., Clement, A., Commere, J., and Jacob, J. L. (1994). Metabolic aspects of latex regeneration in situ for three Hevea clones. Indian J. Nat. Rubber Res. 7, 79–88. Sethuraj, M. R., Potty, S. N., Vijayakumar, K. R., Krishnakumar, A. K., Rao, P. S., Thapliyal, A. P., Mohanakrishna, T., Rao, G. G., Chaudhuri, D., George, M. J., Soman, T. A., and Meenattoor, J. R. (1989). Growth performance of Hevea in the non-traditional regions of India. In “Proceedings of the Rubber Research Institute of Malaysia, Rubber Growers’ Conference, 1989,” pp. 212–227. Malacca, Malaysia. Shamshuddin, J. (1988). The suitability of T2 terrace soils of Peninsular Malaysia for growth of rubber, oil palm and cocoa. Malaysian Agric. J. 54, 97 –115. Shangpu, L. (1986). Judicious tapping and stimulation based on dynamic analysis of latex production. In “Proceedings of the IRRDB Rubber Physiology and Exploitation Meeting.” pp. 230–239. SCATC, Hainan, China. Shuochang, A., and Yagang, G. (1990). Exploration of the high yield physiological regulation of Hevea brasiliensis in Xishuangbanna. In “Proceedings of the IRRDB Symposium on Physiology and Exploitation of Hevea brasiliensis, 1990,” pp. 83 –92. Kuning, China. Simmonds, N. W. (1982). Some ideas on botanical research on rubber. Trop. Agric. 59, 2–8. Simmonds, N. W. (1986). Theoretical aspects of synthetic/polycross populations of rubber seedlings. J. Nat. Rubber Res. 1, 1 –15. Simmonds, N. W. (1989). Rubber breeding. In “Rubber”. (C. C. Webster and W. J. Baulkwill, Eds.), pp. 85–124. Longman, New York. Sowmyalatha, M. K. S., Priyadarshan, P. M., Dey, S. K., and Varghese, Y. A. (1997). Low fruit set in Hevea brasiliensis in Tripura: implications offloral attributes. Indian J. Nat. Rubber Res. 10, 15–26. Stewert, W. D., Watchel, W. L., Shipman, J. J., and Hanks, J. A. (1955). Synthesis of rubber by fungi. Science 122, 1271. Strahler, A. N. (1969). In “Physical geography” (3rd ed.). Wiley, New York. Tan, H. (1979). A biometrical approach to study crown-trunk relationships in Hevea. J. Rubber Res. Inst. Malaysia 27, 79– 91. Tan, H. (1981). Estimates of genetic parameters and their implications in Hevea breeding. Proceedings of the 4th International Congress of SABRAO, Kuala Lumpur. In “Crop Improvement Research”. (T. C. Yap, K. M. Graham and J. Sukanu, Eds.), pp. 439 –446. Tan, H. (1987). Strategies in rubber tree breeding. In “Improving Vegetatively Propagated Crops”. (A. J. Abbott and R. K. Atkin, Eds.), pp. 28–54. Academic Press, London. Tan, H., Khoo, S. K., and Ong, S. H. (1996). Selection of advanced polycross progenies in Hevea improvement. J. Nat. Rubber Res. 11, 215 –225. Thanh, D. K., Wang, N. N., Truong, D. X., and Nghia, N. A. (1998). Seasonal yield variations of rubber tree Hevea brasiliensis in climatic conditions of major rubber growing areas of Vietnam. In “IRRDB Symposium,” Vol. II. Physiology, Exploitation and Crop Production & Planting Materials, pp. 26–37. Ho Chi Minh City, 14–15 October, 1997. Tuy, L. M., Hoa, T. T. T., Lam, L. V., Duong, P. H., and Phuc, L. G. T. (1998). The adaptation of promising rubber clones in the central highlands of Vietnam. In “IRRDB Symposium on Natural Rubber,” Vol. I. General, Soils and Fertilization and Breeding and Selection Sessions. pp. 155–163. Ho Chi Minh City, 14 –15 October, 1997. Vinod, K. K., Meenattoor, J. R., Priyadarshan, P. M., Pothen, J., Chaudhuri, D., Krishankumar, A. K., Sethuraj, M. R., and Potty, S. N. (1996). Early performance of some clones of Hevea brasiliensis in Tripura. Indian J. Nat. Rubber Res. 9, 123– 129. Wallace, D. H., and Yan, W. (1998). “Plant breeding and whole system crop physiology”. CAB International, UK. Watson, G. A. (1989). Climate and soil. In “Rubber” (C. C. Webster and W. L. Baulkwill, Eds.), pp. 125–164. Longman, Essex.

400

P. M. PRIYADARSHAN

Whitby, G. S. (1919). Variation in Hevea brasiliensis. Ann. Bot. 33, 313–321. Wycherly, P. R. (1968). Introduction of Hevea to the Orient. The Planter, Kuala Lumpur 44, 1–11. Wycherly, P. R. (1969). Breeding of Hevea. J. Rubber Res. Inst. Malaya 21, 38–55. Wycherley, P. R. (1992). The genus Hevea—botanical aspects. In “Natural Rubber: Biology, Cultivation and Technology” (M. R. Sethuraj and N. M. Mathew, Eds.), pp. 50–66. Elsevier, Amsterdam. Yan, W., and Hunt, L. A. (1998). Genotype by environment interactions and crop yield. Plant Breeding Rev. 16, 135–178. Yee, H. C., Peng, N. A., and Subramaniam, S. (1969). Choice of clones. Planters Bull. 104, 226–247. Zongdao, H., and Xuequin, Z. (1983). Rubber cultivation in China. In “Proceedings of the Rubber Research Institute Malaysia Planters’ Conference 1983,” pp. 31–43. Kuala Lumpur, Malaysia. Zongdao, H., and Yanqing, P. (1992). Rubber cultivation under climatic stresses in China. Natural Rubber: Biology, Cultivation and Technology. (M. R. Sethuraj and N. M. Mathew, Eds.), pp. 220 –238. Elsevier, The Netherlands.

Index A 2,4-D, 262 abiotic stress see geo-climatic stress advocacy, soil quality management, 41 –7 Africa, rubber cultivation, 363, 366 aggregate stability arbuscular mycorrhizal fungi, 196–7, 202 –3 catch crops and green manures, 263 agricultural technologies, 48, 50 –1 AM see arbuscular mycorrhizal fungi Amazonian basin climate, 356 –7 ammonia, 253 –4 ammonia-oxidisers, 107–8 ammonification, 72, 73 analytical solutions, gross nitrogen flux calculation, 91–2 anthropogenic activities, selenium, 121, 132 –4 aphids, 337, 338, 339–40 apomixis, 391 –2 aquifers, 230–1, 289 –91 arbuscular mycorrhizal (AM) fungi, 185 –225 catch crops and green manures, 262 crop and soil management impact, 199– 206 different agricultural system effects, 206–8 future, 209–10 introduction, 186–8 sustainable agriculture management, 198–9, 209 –10 role, 189–98 atmosphere, selenium, 150–1 Australasia seleniferous soils, 129, 135, 136 subterranean clover, 303 –50 B bacteria ammonia-oxidisers, 107–8 arbuscular mycorrhizal fungi, 188, 197 bioremediation of selenium, 166–7 methylation, 169– 72 bacterium-like organisms (blos), 188 barometric process separation of 15N, 82–6 bioactivity indices, 21 bioavailability, selenium, 148 –50

biodynamic farming, arbuscular mycorrhizal fungi, 206 –8 bioremediation of seleniferous soils, 165 –74 methylation, 169 –72 water, 167–9 biotic stress, rubber cultivation, 366–8 blos see bacterium-like organisms boron arbuscular mycorrhizal fungi, 191 subterranean clover, 329 Brassica sp., 161 –2, 163–4, 172, 173 Brazil, rubber cultivation, 357–9, 366, 368, 371–2, 373, 374, 382, 384 break crops, 264–5 breeding Hevea brasiliensis, 351–400 burial, subterranean clover seeds, 305, 315, 316, 331 burrs, subterranean clover, 305, 306, 316 C C/N ratio see carbon/nitrogen ratio calcium, 191– 2 carbon cycling, 102–4 carbon/nitrogen (C/N) ratio, 246–7, 272, 277–8 catch crops, 227– 302 cropping sequences and whole crop rotations, 266–91 desired and undesired effects, 230 model simulations in crop rotation, 280 –8 nitrogen supply for subsequent crops, 241–59 nitrogen uptake and soil depletion, 232–41 nutrient effects other than nitrogen, 260– 1 policy scale placement, 289–91 reasons for use, 228 –32 soil biological activity, 262 species choice, 273 –4, 278–9, 291 temperate zone nitrogen management, 227–302 China rubber cultivation, 357–9, 362–3, 373, 375, 381 seleniferous soils, 129, 135, 136 selenosis, 154–5 climate catch crops, 247–8, 280–8

401

402

INDEX

rubber cultivation, 356– 66, 373–5 subterranean clover, 308–11, 314, 329 –30 CMN see common mycelial networks cold tolerance, catch crops, 275 collembola, 262 commercial utilization, selenium, 123 –5 common mycelial networks (CMN), 192– 3 compaction arbuscular mycorrhizal fungi, 203 soil quality, 35 contamination by selenium see seleniferous soils conventional vs sustainable agriculture, 198–9, 206–8 Coˆte d’Ivoire, rubber cultivation, 357– 9, 366, 370, 373, 382, 384 cover crops see catch crops; green manures covering, seleniferous soils, 157–8 cows see grazing livestock Cretaceous rocks, 127 –8 crop nutrition arbuscular mycorrhizal fungi, 189– 93, 203–5 catch crops and green manures, 231– 2, 260–1 history, 7 crop protection, 193 –4, 205 crop rotation arbuscular mycorrhizal fungi, 199– 201 catch crops, 266– 9, 280–8 crop water relations, 194–5 crop yield rubber cultivation, 352– 3, 371–6, 391 soil quality, 22, 27 crops see also catch crops catch crop effect on nitrogen supply, 241–59 market value per acre, 14 phytoaccumulation, 161 –2 phytovolatilization, 163–4 production, 9, 17, 48, 51 selenium absorption capacity, 160 selenium content, 145 –6 US subsidies, 18, 19–20 crucifer catch crops, 234, 235, 274 cultivars see varieties and cultivars cultivation, arbuscular mycorrhizal fungi, 202– 3 D DAISY model, 243, 280 –8 databases, soil quality, 24

decomposition gaseous nitrogen loss, 253– 4 nitrogen mineralisation, 247, 248– 50 decontamination by selenium see phytoremediation deficiency diseases, 120, 144 defoliation arbuscular mycorrhizal fungi, 206 subterranean clover, 316, 327–8 deselenification, 169– 72 digestibility, subterranean clover, 335 direct gene transfer, 390 –1 diseases arbuscular mycorrhizal fungal protection, 193, 197 catch crops and green manures, 264 –5 rubber cultivation, 366 –7 subterranean clover, 337– 9 disinfection by-products (DPBs), 31–2 diversity arbuscular mycorrhizal fungi, 198 Hevea, 382, 384 –6, 391 domestic animals, selenium toxicity, 122, 152 –3 dormancy see embryo dormancy; hardseededness DPBs see disinfection by-products drought, 194 –5 dry powder application of 15N, 82 E earthworms, 33–5 economic issues, arbuscular mycorrhizal fungi, 209 ecosystems health, 25– 7 selenium, 121 embryo dormancy, 311 –12 emissions, selenium, 131 enrichment, 15N pool, 86–7 environment catch crops and nitrogen, 230 –1 selenium poisoning, 122 –3 selenium sources, 130 –4 selenium threat, 137 –51 equilibrium, nitrogen pools, 76–7 error evaluation, gross nitrogen flux, 93–4 essential poison, selenium, 120, 122, 152–4 establishment of catch crops, 269–70 eutrophication, 230– 1, 289–91 evapotranspiration, 286, 287

INDEX experimental design, 15N pool dilution, 77–80 extra-radical hyphae, 189–90, 191, 196– 7 F fallow periods, 200–1 false strikes, 312 –13 fauna see soil fauna FE see fertilizer equivalent fertiliser equivalent (FE), 256–8 fertiliser replacement value, 254, 256–8 fertilisers arbuscular mycorrhizal fungi, 203–5, 208 catch crop field effects, 254–9 fertility see also soil fertility; soil quality subterranean clover oestrogenicity, 340–1 field effects, catch crops, 254–9 fixation see nitrogen fixation flowers arbuscular mycorrhizal fungi, 195 subterranean clover, 304 –5, 309– 10, 311, 327 –8 fodder crops, 160 foliar fungal diseases, 338–9 food quality, 8 supply, 48 French Guiana, rubber cultivation, 358 fungi see also arbuscular mycorrhizal fungi subterranean clover diseases, 338–9 G gas application of 15N, 81–2 genetics, Hevea species, 384–91 geo-climatic stress, rubber cultivation, 361 –6 geochemistry, selenium, 141 –3 germination, subterranean clover, 311 –16, 331 girth increment, rubber cultivation, 369–71, 375 Gloeosporium leaf disease, 366 glomalin, 196–7 Glomeromycete fungi, 186 see also arbuscular mycorrhizal fungi grass and grasses, 308 –9, 318, 325–6, 328–9, 335 –7 grazing livestock arbuscular mycorrhizal fungi, 205–6 subterranean clover, 317, 318–19, 322, 327 –8, 335, 336, 340– 1

403

green manures, 227–302 see also catch crops nitrogen mineralisation, 245– 6 nitrogen uptake, 240 –1 recovery of catch crop nitrogen, 255– 6 temperate zone nitrogen management, 227–302 Green Revolution, 50 gross immobilisation, 94– 6 gross mineralisation, 73, 74 gross nitrogen fluxes, 69–118 15 N pool dilution applications, 96–109 ammonia-oxidisers, 107–8 calculation, 90 –6 analytical solutions, 91–2 immobilisation, 94–6 numerical solutions, 92–4 carbon and nitrogen mineralisation, 102–4 MIT fluxes, 70, 72, 99–100 nitrogen saturation index, 108–9 organic nitrogen assimilation, 100–2 osmotic potential, 98 plant residues, 78–80 quantification, 99–102 soil disturbance, 105–7 soil organic matter, 77– 8 substrate availability, 104–5 temperature, 97 –8 water potential, 98 groundwater, 143–4 see also water growth catch crops and nitrogen uptake potential, 233–5 Hevea immature phase, 369–71 root, 235 –7 guayule, 353, 391– 2 H hardseededness, 312 –15, 316, 329–31 herbage, 332–5 Hevea brasiliensis (para rubber tree), 351– 400 biotic stresses, 366 –8 breeding programs, 376 –91 constraints, 361– 8 geo-climatic stresses, 361 –6 growing conditions, 356– 61 introduction, 352–6 marginal conditions, 357–61, 369–76

404

INDEX

Hevea species, 353, 354 –5, 360, 382, 384– 6, 392 history selenium poisoning, 121– 3 soil quality management, 7–10 subterranean clover, 306–7, 316–19 holism, soil quality, 23–7 horses, selenium poisoning, 122 human activities see anthropogenic activities humans, selenium toxicity, 154 –6 hygiene, soils, 37 I igneous rocks, selenium content, 126 immobilisation gross, 94 –6 nitrogen supply after catch crops, 241–2, 246, 248 selenium, 159 incorporation time (kill date), 270–3, 277 incubation period, 15N pool dilution, 87–8 India rubber cultivation, 357– 9, 361–2, 368–70, 371–2, 373, 374, 375 seleniferous soils, 135–7 selenosis, 155– 6 indices soil productivity, 20– 3 soil quality, 3, 4, 6, 11–13, 19, 20–3, 29–39, 51–2 Indonesia, rubber cultivation, 358, 359 industrial effluent, selenium, 132–3 injection systems, 15N delivery, 83–5, 88–90 inoculation, arbuscular mycorrhizal fungi, 206 inputs, soil quality management, 16 –20 insects arbuscular mycorrhizal fungal protection, 193–4 subterranean clover, 339–40 intercropping, 201 interplant nutrient transfer, 192 –3 Ireland, seleniferous soils, 128, 135, 136 irrigation, seleniferous soils, 128 isotopic discrimination, 76 Israel, seleniferous soils, 135 K kick-on property, subterranean clover, 321 kill date (incorporation time), 270–3, 277

L latex, rubber cultivation, 353, 373–5, 382, 386– 7 leaching, subterranean clover germination, 312 leaching loss, nitrogen, 230–1, 237–40, 259, 270– 1, 276, 280 –91 legumes catch crops and green manures, 233, 240 –1, 268, 272 subterranean clover, 303– 50 Liberia, rubber cultivation, 359 light, subterranean clover response, 308–9, 329 liming, 205 linkage maps see molecular linkage maps livestock see grazing livestock M macronutrients, 189–92 magnesium, 191– 2 Malaysia, rubber cultivation, 358, 359 manganese, 192 manure see also fertilisers; green manures arbuscular mycorrhizal fungi, 204 soil quality, 31, 36 –7 maturity grading (MG), subterranean clover, 320– 2 meta-organism analogy, 23 –7 methylation, selenium, 169 –72 MG see maturity grading microbial nitrogen pathways, 71– 3 microbiological pollution, 36 Microcyclus ulei see South American Leaf Blight micronutrients arbuscular mycorrhizal fungi, 192 catch crops and green manures, 261 microorganisms ammonia-oxidisers, 107 –8 arbuscular mycorrhizal fungi, 188, 197 bioremediation of selenium, 165–72 nitrogen pathways, 71–3 pathogens, 193, 197, 264– 5 soil quality, 36 mineral fertilizer equivalent, 256– 8 mineralisation carbon and nitrogen cycling, 102– 4 gross, 73, 74

INDEX incorporation time, 272 nitrogen supply after catch crops, 241 –3, 244, 245–50, 258–9, 277–8 mineralisation–immobilisation turnover (MIT), 70, 72–3, 99–100 minerals, selenium association, 127, 141–2 MIT see mineralisation–immobilisation turnover mites catch crops and green manures, 262 subterranean clover, 339 mixing, seleniferous soils, 158 moisture see also soil water; water subterranean clover, 313 molecular breeding, 384–91 molecular linkage maps, 387– 90 monocots, growth and nitrogen uptake potential, 234, 235 N 15

N pool dilution amount and enrichment of 15N, 86–7 applications, 71, 96–109 ammonium-oxidisers, 107 –8 carbon and nitrogen mineralisation, 102 –4 magnitude of nitrogen cycling, 99–102 nitrogen saturation index, 108 –9 regulating factors, 96 –9 soil disturbance, 105 –7 substrate availability, 104–5 assumptions, 75 –7 future research, 109–10 rapid consumption, 88 spatial heterogeneity, 90 techniques, 77 –90 experimental design, 77 –80 injection systems, 83 –5, 88–90 methodological considerations, 86–90 soil application methods, 71, 80–6 theory, 73–5 undisturbed soil, 88–90 15 N-enrichment, 74 natural sources, selenium, 131 Neff see nitrogen effect nematodes, 264 New Zealand see Australasia nif genes, 188 Nigeria, rubber cultivation, 359

405

nitrate leaching loss, 230 –1, 282, 287, 288, 289–91 water contamination, 28–9, 289–91 nitrogen see also 15N-enrichment; gross nitrogen fluxes arbuscular mycorrhizal fungi, 190–1 catch crops and green manures, 227–302 agricultural effects, 231–2 crop rotation placement, 267–9 depth distribution, 245, 250 –3, 272 environmental effects, 230–1 gaseous losses, 253 –4 nitrate leaching and aquifers, 230 –1, 289–91 nitrogen effect, 242– 4, 250, 252–3, 254, 256–7, 266, 270–2, 278–9, 291 recovery, 255– 6 soil depletion, 232 –41 supply for subsequent crops, 241– 59 uptake, 232 –41, 274– 5 nitrogen effect (Neff), 242 –4, 250, 252–3, 254, 256–7, 266, 270–2, 278–9, 291 nitrogen fixation, 240 –1, 275 nitrogen fluxes, 69–118 nitrogen leaching loss, 230 –1, 237 –40, 259, 270–1, 276, 280–91 nitrogen saturation index, 108–9 nitrous oxide, 253–4 no-till systems see tillage North America, seleniferous soils, 127–8, 135, 136 numerical solutions, gross nitrogen flux calculation, 92–4 nutrients arbuscular mycorrhizal fungi, 189–93, 203–5 catch crops, 231–2, 260–1 plants, 7 selenium, 120, 122, 144 O oestrogenicity, 322, 340– 1 Oidium heveae see powdery mildew organic farming arbuscular mycorrhizal fungi, 198, 206 –8 nitrogen effects of catch crops, 231–2 organic nitrogen assimilation, 100–2 osmotic potential, 98 –9 oxidation, methane, 107 –8

406

INDEX P

para rubber tree see Hevea brasiliensis parent material, seleniferous soils, 127–9 pathogens see also diseases arbuscular mycorrhizal fungi, 193, 197 catch crops and green manures, 264– 5 pea root rot, 265 pesticides catch crops and green manures, 262 soil quality, 30 pests catch crops and green manures, 264 subterranean clover, 337–40 pH see soil pH phenology, rubber cultivation, 367 –8 phosphorus arbuscular mycorrhizal fungi, 189– 90, 192, 195–6, 200, 203– 4 catch crops and green manures, 260– 1 phytoavailability, selenium, 148–50 phytoremediation of seleniferous soils, 160–5 phytoaccumulation, 161 –3 phytovolatilization, 163–4 rhizofiltration, 164 –5 plants populations, arbuscular mycorrhizal fungi, 198 reproduction, arbuscular mycorrhizal fungi, 195–6 residues, gross nitrogen fluxes, 78–80 selenium absorption, 160 accumulating species, 145–6 metabolism, 147 –8 occurrence, 144–50 phytoremediation, 160– 5 toxicity, 151–2 poisoning, selenium, 120 –2, 152– 4 policy issues, catch crops, 266, 267, 289–91 polyclonal Hevea seedlings, 376– 8 polymorphism, Hevea, 385– 6 potassium arbuscular mycorrhizal fungi, 191– 2 catch crops and green manures, 261 potatoes, 268 –9 powdery mildew, 361, 367 pre-emptive competition, 243, 244–5, 250, 263–4, 290, 291

precipitation catch crop incorporation time, 270–2 catch crops and green manures, 263 –4, 290 nitrogen leaching loss, 238, 239 rubber cultivation, 364, 370 winter and nitrogen availability for subsequent crops, 252–3 productivity see also soil productivity subterranean clover, 332– 7 protection see crop protection Q QTL see quantitative trait loci quality of product, 209 quality soil management, 1– 68 quantification, gross nitrogen fluxes, 99 –102 quantitative trait loci (QTL), SALB resistance, 390 R rainfall see precipitation rapid consumption of 15N, 88 recombination breeding, 378 –81 remediation technologies, 156 –72 remineralisation, 15N, 91–2 research and development, arbuscular mycorrhizal fungi, 209 –10 rocks, selenium content, 125–7 roots arbuscular mycorrhizal fungi, 186 –7, 189– 90, 193–4, 196, 197 catch crops, 235 –7, 239, 245, 265 diseases, 265, 339 rotation see crop rotation rubber see also Hevea brasiliensis biosynthesis, 386 –7 source, 353– 6 S SALB see South American Leaf Blight salinity, soils, 38–9 scientific analysis, soil quality management, 41– 7 seasons rubber cultivation, 367 –8 subterranean clover, 334

INDEX secondary leaf fall see powdery mildew sedimentary rocks, selenium content, 126 seedlings polyclonal Hevea, 376–8 subterranean clover, 312, 325–6 seeds arbuscular mycorrhizal fungi, 195–6 size and nitrogen uptake, 274 subterranean clover, 310 –16, 323 –5, 326 –31 seleniferous soils, 119 –84 formation, 127–9 geographical distribution, 134–7 inputs/outputs, 130, 133 management, 156–72 bioremediation, 165–72 competitive ions, 159 –60 mixing/covering, 157–8 phytoremediation, 160–5 plant absorption, 160 Se immobilization, 159 thermal treatment, 158–9 washing, 158 research needs, 173–4 selenium content, 138–43 seleniferous water, 167–9 selenium atmospheric occurrence, 150 –1 biological uses, 125 biomethylation, 169–72 chemical forms, 139–40 crustal distribution, 119–20, 121 deficiency diseases, 120, 144 environmental sources, 130–4 geochemistry and mobility, 141–3 historical aspects, 121 –3 immobilization, 159 metabolism in plants, 147–8 phytoavailability, 148–50 plant occurrences, 144–50 properties and uses, 123 –5 rock and mineral occurrences, 125–7 soil content, 138–43 toxicity effects, 120–2, 151–6 water occurrences, 143 –4 selenium toxicity, 120– 2 animals, 152–4 humans, 154 –6 plants, 151–2 selenosis see selenium toxicity semantics, soil quality, 1–68

407

sheep see grazing livestock silage maize, 268 small-scale clone trials (SSCTs), 378, 380 sodium, 191–2 soil earthworms, 33–5 15 N application, 80–6 seleniferous, 119– 84 subterranean clover, 307 soil depletion, 232–41 soil disturbance, 105–7 soil erosion, 34 soil fauna, 262 soil fertility, 7, 20, 45 see also soil quality soil formation, 127 –9 soil management, 156 –72 see also soil quality management soil microorganisms, 193, 197, 262, 264 –5 soil nutrition see nutrients soil orders, United States, 5 soil organic matter (SOM), 7–8, 29– 31, 32, 38, 70 gross nitrogen fluxes, 77–8 weeds 31 soil pathogens, 193, 197, 264–5 soil pH, arbuscular mycorrhizal fungi, 191, 205 soil productivity, 8–9, 17, 20– 3, 27 soil quality assessment, 22–3, 24 concept, 2–7, 13–15, 40–1, 43, 53 crop production, 9, 17 definitions, 10–12, 24–5, 27, 44 evaluation, 16, 44 functions, 11, 12, 21–3, 25, 52 health, 25–7, 44 –5 holism, 23–7 indices, 3, 4, 6, 11–13, 19, 20–3, 29–39, 51–2 management, 1–68 advocacy, 41–7 global perspective, 47– 9 historical basis, 7–10 input argument, 16–20 priorities, 39– 41 scientific analysis, 41–7 terminology, 10–16 paradigm failures, 27–9 parameters, 23–4 purity, 15, 37 sustainability, 38, 49–50

408

INDEX

soil respiration, 38 soil salinity, 38 –9 soil structure, 196–7, 202 –3, 263 soil temperature, 30–1, 97 –8, 247– 50 soil water catch crops and green manures, 263– 4 gross nitrogen fluxes, 98 subterranean clover, 310–11 transport, 35 solution application of 15N, 81 SOM see soil organic matter South America, rubber cultivation, 356 –7, 358, 359, 366 South American Leaf Blight (SALB), 356, 366–7, 390 spatial heterogeneity, 15N pool dilution, 90 Sri Lanka, rubber cultivation, 359 SSCTs see small-scale clone trials stewardship, soil, 49 strophiole, 313, 315 subsidies, US crops and conservation, 18, 19–20 substrate availability, nitrogen cycling, 104–5 subterranean clover (Trifolium subterraneum L.), 303–50 history of use, 306 –7, 316– 19 initial sowing rates, 324–5 management, 316 –31 oestrogenicity, 322, 340– 1 pests and diseases, 337–41 physiological response to the environment, 308–16 productivity, 332– 7 seed production, 326– 9 species characteristics, 304–7 sugar beet, 268 –9 sulphur arbuscular mycorrhizal fungi, 191 catch crops and green manures, 261 sustainability arbuscular mycorrhizal fungi, 185– 225 conventional agriculture comparison, 198– 9, 206–8 soil quality, 38, 49– 50 symbiosis, arbuscular mycorrhizal, 186

temperature catch crop nitrogen uptake, 275 gross nitrogen fluxes, 97–8 nitrogen mineralisation, 247–50 rubber cultivation, 362 –3, 364, 373–5 soil, 30–1, 97 –8, 247–50 subterranean clover, 308, 309–10, 314 –15, 329– 30 terminology, soil quality, 10– 16 Thailand, rubber cultivation, 357–9, 362, 373 thermal treatment, seleniferous soils, 158– 9 THM see trihalomethane tillage, 202– 3 timber, 382, 383 tissue specific gene expression, 386 –7 toxicity, selenium, 120–2, 151–6 Trifolium subterraneum L. see subterranean clover trihalomethane (THM), 31–2 U undersowing, catch crops, 269–70, 275, 279 undisturbed soil, 15N pool dilution, 83– 5, 88 –90 United States crop and conservation subsidies, 18, 19–20 crop market value per acre, 14 dominant soil orders, 5 soil quality index, 4 V VAM (vesicular–arbuscular mycorrhiza) see arbuscular–mycorrhizal fungi varieties and cultivars arbuscular mycorrhizal fungi, 201 –2 subterranean clover, 317– 18, 319–24 vernalisation, subterranean clover, 309–10, 311 vesicular–arbuscular mycorrhiza (VAM) see arbuscular–mycorrhizal fungi Vietnam, rubber cultivation, 357 –9, 362, 370, 371– 2, 373, 375 viruses, 337–8 W

T taxonomy, soil quality, 52 temperate zones, catch crops and green manures, 227–302

washing, seleniferous soils, 158 water see also soil water bioremediation, 167 –9

INDEX catch crops, 230–1, 263 –4, 282, 287, 288, 289 –91 nitrate contamination, 28–9, 289 –91 selenium occurrence, 143–4 water balance, 194 –5 weathering, 127 weeds catch crops and green manures, 265–6, 279 soil organic matter, 31 subterranean clover, 336 –7 West Africa, rubber cultivation, 363, 366 wind, rubber cultivation, 363, 365, 373, 381

409

winter precipitation leaching loss, 238, 239 nitrogen availability for subsequent crops, 252–3 wood, Hevea, 382, 383 Y yield rubber cultivation, 352–3, 371–6, 391 soil quality, 22, 27