Contents CONTRIBUTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PREFACE . . . ...
41 downloads
717 Views
2MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Contents CONTRIBUTORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PREFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ix xi
KEEPING IN TOUCH: MICROBIAL LIFE ON SOIL PARTICLE SURFACES Aaron L. Mills I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Nature of Soil Particles Related to Microbial Attachment . . . . . . . . . . . . . . A. Particle-size Distributions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Chemical Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. Benefits of Living on Particles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Particle Surfaces as a Physical Substrate for Collection of Nutrients. . . . B. Utilization of the Particle as a Chemical Substrate. . . . . . . . . . . . . . . . . IV. Importance of Attached Microbes in Soil . . . . . . . . . . . . . . . . . . . . . . . . . . A. Numbers of Attached Versus Free-living Microbes . . . . . . . . . . . . . . . . B. Phylogeny of Attached Versus Free-living Microbes . . . . . . . . . . . . . . . C. Quantitative Considerations of Activity of Attached Versus Non-attached . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Diversity of Modes of Attachment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Reversible Versus Non-reversible . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Electrostatics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Appendages and Cements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI. Effects of Saturated Versus Unsaturated Conditions. . . . . . . . . . . . . . . . . . . VII. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2 4 4 7 11 11 14 17 17 18 18 19 19 19 31 33 35 36
THE HISTORY AND SUCCESS OF THE PUBLIC- PRIVATE PROJECT ON GERMPLASM ENHANCEMENT OF MAIZE (GEM) Linda M. Pollak I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. The Need for Maize Enhancement . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. The Latin American Maize Project . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. GEM’s Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Public/private US Agricultural Research . . . . . . . . . . . . . . . . . . . . . . . . B. Public and Private Interaction to Organize GEM . . . . . . . . . . . . . . . . . . C. GEM’s Objective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. GEM’s Administration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Funding Mechanism. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
46 46 49 50 50 51 54 55 55 58
vi
CONTENTS
IV. Breeding Activities and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Value-added Trait Analyses and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . A. The Need for Improving Value-added Traits . . . . . . . . . . . . . . . . . . . . . B. The Value-added Trait Research Component of GEM . . . . . . . . . . . . . . C. Grain Composition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Starch Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . E. Oil Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI. Public Cooperator Research and Results . . . . . . . . . . . . . . . . . . . . . . . . . . . A. European Corn Borer Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Characterizing LAMP Accessions and Their Crosses for Wet-milling Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Other Significant Public Cooperator Findings . . . . . . . . . . . . . . . . . . . . VII. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Factors Responsible for GEM’s Successful Public/private Collaboration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Extending GEM’s Concept. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. GEM’s Future . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
58 67 67 68 68 70 72 74 75 77 79 80 80 81 83 83
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS FOR ASSESSING QUALITY OF ACID SOILS Zhenli He, X. E. Yang, V. C. Baligar and D. V. Calvert I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Acid Soil Distribution in the World . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. Quality Characteristics of Acidic Soil. . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Definition and Attributes of Soil Quality . . . . . . . . . . . . . . . . . . . . . . . . B. Quality Characteristics of Acidic Soils . . . . . . . . . . . . . . . . . . . . . . . . . IV. Measurement of Microbiological and Biochemical Parameters in Acidic Soils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Microbial Biomass Carbon, Nitrogen, and Phosphorus. . . . . . . . . . . . . . B. Microbial Turnover of Carbon, Nitrogen, and Phosphorus . . . . . . . . . . . C. Microbial Community Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Soil Enzyme Activity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Microbiological and Biochemical Indicators of Acid Soil Quality . . . . . . . . A. Microbial Biomass. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Microbial Biomass Turnover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Microbial Biomass-related Indicators . . . . . . . . . . . . . . . . . . . . . . . . . . D. Microbial Community Structure Indicators . . . . . . . . . . . . . . . . . . . . . . E. Enzyme Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI. Soil pH Versus Microbiological and Biochemical Indicators . . . . . . . . . . . . VII. Development of Acid Soil Quality Indexing Systems . . . . . . . . . . . . . . . . . VIII. Limitations and Prospective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
90 91 92 92 95 96 97 102 103 104 105 108 115 116 118 119 121 124 128 129 129
CONTENTS
vii
POLYPLOIDY AND THE EVOLUTIONARY HISTORY OF COTTON Jonathan F. Wendel and Richard C. Cronn I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Taxonomic, Cytogenetic, and Phylogenetic Framework . . . . . . . . . . . . . . . . A. Origin and Diversification of the Gossypieae, the Cotton Tribe . . . . . . . B. Emergence and Diversification of the Genus Gossypium . . . . . . . . . . . . C. Chromosomal Evolution and the Origin of the Polyploids . . . . . . . . . . . D. Phylogenetic Relationships and the Temporal Scale of Divergence. . . . . III. Speciation Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. A Fondness for Trans-oceanic Voyages . . . . . . . . . . . . . . . . . . . . . . . . B. A Propensity for Interspecific Gene Exchange. . . . . . . . . . . . . . . . . . . . IV. Origin of the Allopolyploids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Time of Formation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Parentage of the Allopolyploids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Polyploid Evolution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Repeated Cycles of Genome Duplication . . . . . . . . . . . . . . . . . . . . . . . B. Chromosomal Stabilization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Increased Recombination in Polyploid Gossypium . . . . . . . . . . . . . . . . . D. A Diverse Array of Genic and Genomic Interactions . . . . . . . . . . . . . . . E. Differential Evolution of Cohabiting Genomes . . . . . . . . . . . . . . . . . . . VI. Ecological Consequences of Polyploidization . . . . . . . . . . . . . . . . . . . . . . . VII. Polyploidy and Fiber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VIII. Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
140 142 142 144 148 150 155 155 155 158 158 161 165 165 168 168 169 173 175 176 178 179
DEVELOPMENT OF ACIDIC SUBSURFACE LAYERS SOIL UNDER VARIOUS MANAGEMENT SYSTEMS Keryn I. Paul, A. Scott Black and Mark K. Conyers
OF
I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Widespread Occurrence of Acidic Subsurface Layers . . . . . . . . . . . . . . . . . III. Detrimental Effects of Acidic Subsurface Layers on Agricultural Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Water and Nutrient Limitations due to Poor Root Growth . . . . . . . . . . . B. Suppression of N Mineralisation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Poor Root Nodulation of Legumes . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Poor Growth Response to Topdressing of P Fertiliser . . . . . . . . . . . . . . E. Poor Growth Response to Lime Application . . . . . . . . . . . . . . . . . . . . . IV. Rate of Development of Acidic Subsurface Layers . . . . . . . . . . . . . . . . . . . V. Cause of Development of Acidic Subsurface Layers . . . . . . . . . . . . . . . . . . A. Plant N Uptake . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Plant Residue Return . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Mn Reduction and Oxidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
188 189 189 189 192 192 193 193 193 195 195 197 200
viii
VI.
VII.
VIII. IX.
CONTENTS D. Urine Excretion from Grazing Stock. . . . . . . . . . . . . . . . . . . . . . . . . . . E. Soil pH Buffering Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Environmental Factors Affecting the Difference in pH Between Surface and Subsurface Layers . . . . . . . . . . . . . . . . . . . . . . A. Soil Fertility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Initial Soil pH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Rainfall and Fluctuations in Soil Moisture Content . . . . . . . . . . . . . . . . D. Earthworm Populations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Management Factors Affecting the Difference in pH Between Surface and Subsurface Layers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Agricultural Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Plant Species Grown . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Productivity and the Quantity of Plant Residues Added to the Soil Surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . D. Minimum Soil Disturbance and Tillage. . . . . . . . . . . . . . . . . . . . . . . . . E. Fertiliser Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . F. Lime Application. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Management Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
200 201 201 201 201 202 203 204 204 205 208 208 209 209 210 211 212
SOIL ACIDIFICATION AND LIMING INTERACTIONS WITH NUTRIENT AND HEAVY METAL TRANSFORMATION AND BIOAVAILABILITY Nanthi S. Bolan, Domy C. Adriano and Denis Curtin I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Processes of Acid Generation in Soils . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Natural Ecosystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Managed Ecosystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. Effect of Soil Acidity on Nutrient and Heavy Metal Transformation in Soils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Plant Nutrients. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Heavy Metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. Amelioration of Soil Acidity Through Liming . . . . . . . . . . . . . . . . . . . . . . A. Liming Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Effects of Liming. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Lime, Nutrient and Heavy Metal Interactions . . . . . . . . . . . . . . . . . . . . . . . A. Plant Nutrients. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Heavy Metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VI. Conclusions and Future Research Needs . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
216 218 218 223 230 231 234 237 237 239 242 242 256 258 260
INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
KEEPING IN TOUCH: MICROBIAL LIFE ON SOIL PARTICLE SURFACES Aaron L. Mills Laboratory of Microbial Ecology, Department of Environmental Sciences, University of Virginia, Charlottesville, Virginia 22904-4123, USA
I. Introduction II. Nature of Soil Particles Related to Microbial Attachment A. Particle-size Distributions B. Chemical Distribution III. Benefits of Living on Particles A. Particle Surfaces as a Physical Substrate for Collection of Nutrients B. Utilization of the Particle as a Chemical Substrate IV. Importance of Attached Microbes in Soil A. Numbers of Attached Versus Free-living Microbes B. Phylogeny of Attached Versus Free-living Microbes C. Quantitative Considerations of Activity of Attached Versus Non-attached V. Diversity of Modes of Attachment A. Reversible Versus Non-reversible B. Electrostatics C. Appendages and Cements VI. Effects of Saturated Versus Unsaturated Conditions VII. Summary References Microorganisms in unsaturated soil live in a world dominated by the presence of extensive surfaces, both solid and gas –liquid interfacial surfaces. Particle attachment in soils is similar to particle attachment in aquatic systems, which, because of the high abundance of suspended populations has been widely studied. Although there seems to be a general advantage to the microbes living at the interfaces in terms of enhanced nutrient concentrations and the potential to use many of the physical substrata themselves as energy or nutrient sources, the thickness of the water films in unsaturated conditions leaves the microbes little option except to adhere to the surfaces. Initial attachment to the surfaces appears to be dominated by electrostatic and hydrophobic effects that are described by Derjaguin-Landau-Verwey-Overbeek (DLVO) theory for negatively charged cells and particles. These effects result in reversible attachment and the cells are subject to rapid detachment with slight changes in solution chemistry and removal by hydraulic shear. Coatings play an important role in attachment, with metal oxide coatings 1 Advances in Agronomy, Volume 78 Copyright q 2003 by Academic Press. All rights of reproduction in any form reserved 0065-2113/02$35.00
2
A. L. MILLS conferring a positive charge to the particle surface resulting in a much tighter adhesion of the microbial cells to the surface. Attachment of the organisms to the particles by direct surface contact through appendages such as fimbriae or deposition of polysaccharidic slime results in irreversible attachment that can lead to buildup of colonies and biofilms. In this chapter, considerations of theory are presented as they pertain to soil organisms, and abundant use of examples from aquatic habitats exemplifies principles and ideas not easily studied in unsaturated soil. The importance of attachment to the gas – liquid interface is also highlighted. q 2003 Academic Press.
I. INTRODUCTION The soil habitat represents a unique but extensive environment in which microorganisms live and carry out biogeochemical reactions critical to the maintenance of ecosystems. The uniqueness of the soil is related to the vast amount of particle surface area contained there. Other habitats also contain particles, often with a large surface area, but soils are dominated by surfaces and the matrix is generally hydrologically unsaturated. The combination of particle surface area with thin water films makes soils different in many respects from their saturated counterparts in aquatic sediments and aquifers formed from unconsolidated materials. The purpose of this chapter is to examine the relationship of soil microorganisms to soil particles in terms of their tendency to attach themselves to the particles. Much of the chapter will examine mechanisms of attachment, but this information is best understood in the context of why the bacteria attach to the particles. To accomplish this goal, it will be necessary to consider particle attachment in general, including information from other habitats such as in lakes and marine environments. Thus, while the focus of this discussion is attachment in soils, principles will be derived from other environments as needed and justified. Fletcher recently published an excellent volume (Fletcher, 1996) that examined in general the attachment of microbes to surfaces from both an ecological and physiological viewpoint. Part of this chapter formed part of a contribution to this volume, and the book also served as an important starting point for a number of topics covered here. Readers are encouraged to use the chapters contained in this reference for a more detailed coverage of the general topic of microorganisms and surfaces than can be accomplished here. Attachment to particles could arise due to three possible reasons. Attachment might occur completely as a result of serendipity. Bacterial attachment might confer neither an advantage nor a disadvantage to the organisms, and there might be no active mechanism that pulls bacteria to the surface of particles. Experience teaches us that while events frequently arise spontaneously, their persistence in biological systems generally arises from some selective advantage conferred on the organisms involved. Furthermore, the degree to which bacteria attach to
MICROBES AND SOIL PARTICLE SURFACES
3
particles in soils and in aquatic ecosystems and the strength of the association once established, argue strongly against happenstance as the causative agent for bacterial attachment. A second possible reason for the high frequency of particle association by bacteria is that there is a selective advantage to the organisms to live in close association with a particle surface. In biological systems, behaviors that are not advantageous to the populations are generally lost over evolutionary time. It is possible that attachment represents a neutral behavior (a rationale for which is discussed in the following paragraph), but we can certainly be sure that particle attachment does not represent a behavior that is generally detrimental to the cells. If this were the case, the attached bacteria would quickly be eliminated by competition with the suspended bacteria for limiting resources. Indeed, it is likely that attachment represents an advantage to the organisms in some major way; they may obtain some essential element from the particle grains, they might obtain energy from organic or inorganic molecules tightly sorbed to the mineral grain surface, or they might benefit from living in a chemically richer aqueous environment due to enhanced concentrations of soluble nutrients in the proximity of the grain surface. In some cases, attachment to particles might provide partial or complete protection from grazing by bacterivorous organisms. All these possibilities will be discussed in detail later. A third possible reason for particle attachment in soils also exists. In unsaturated soils, microbes have little choice but to exist at or near the surface of the soil particles. Although filamentous fungi and actinomycetes have been observed to span unsaturated voids, single-celled organisms are limited to total immersion to be active (Metting, 1993); furthermore, nutritional uptake requires an aqueous phase for all phenotypes (Harris, 1981). Even under the so-called optimal soil moisture conditions in which the pore space of a silt loam soil is approximately 50% filled with water, the amount of water associated with each particle does not leave much room for the microbes to move a great distance from the surface of a particle. A simple calculation illustrates this point. Consider a soil composed entirely of uniformly spherical particles that are in the mid-silt-size range, i.e., 0.025 cm in diameter. Consider further that the soil has a bulk density of 1.2 g cm23. A final assumption is that all the water is perfectly uniformly distributed on the entire surface of all particles (note that the particles do not actually touch under this assumption). Under these oversimplified assumptions, each particle is coated with a film of water that is only about 6.4 mm thick. While no such soil exists in reality, and the actual thickness of water films under realistic soil conditions varies from a few molecules to millimeters, there is not much volume in the soil that permits the microbes to be very far from the surface of a mineral or organic soil particle. Indeed, Mills and Powelson (1996) estimated from literature data (Gardner, 1956; Green et al., 1964; Holmes et al., 1960; Kemper and Rollins, 1966) that at field capacity (soil moisture tension of 0.33 bar), one might expect a film thickness of only 0.2 –0.3 mm assuming
4
A. L. MILLS
uniform coverage of the grains by the water film. Metting (1993) calculated that at a soil matric potential of 2 0.01 MPa, capillaries less than 30 mm in diameter would be saturated, but at 2 0.03 MPa, saturation would be only in pores of less than 4 mm diameter. When the potential is less than 2 0.5 MPa there is only a film of water a few molecules thick. The principal point is that the water film thins quickly as the degree of saturation decreases, forcing the microbial cells even closer to the grain surface. The argument presented in the preceding paragraph might be quite compelling as a complete explanation if it were not for the fact that soils are not the only place where particle association is observed. In soils, attachment is so heavily dominant because the soil habitat is completely dominated by particle surface area surrounded by the thin films of water. In all systems the degree of particle association appears to be correlated to both the number of bacterial cells and the number of particles present. We cannot conclude that either selective advantage or necessity is the single reason for attachment as the way of life in soils. Obviously, the answer is a combination of the two factors. The fact that water films are generally thin and vegetative bacteria are forced to live, therefore, close to the particle surfaces is obvious. The ensuing discussion, therefore, will concentrate on factors that confer advantage to the organisms living on particle surfaces in soil.
II. NATURE OF SOIL PARTICLES RELATED TO MICROBIAL ATTACHMENT A. PARTICLE-SIZE DISTRIBUTIONS A number of factors influence the attachment and permanent association of bacteria with soil particles. In addition to particle composition (discussed later), particle size seems to play an important role in determining the distribution of microbial populations in soil aggregates. A study by Hattori (1973) showed the strong quantitative relationship between clay particles and bacterial cells (Fig. 1). A number of studies have shown that both the cell number and the bacterial biomass tend to be most concentrated in the smaller size silt and clay fractions (Jocteur Monrozier et al., 1991; Kandeler et al., 2000, 2001; vanGestel et al., 1996). Obviously, therefore, the bacteria are mainly present in micropores of 5– 30 mm (Amato and Ladd, 1992; Hassink et al., 1993; Kirchmann and Gerzabek, 1999). Analysis of the distribution of microbial enzyme activities suggest that the bacterial activities are dominant in the silt and clay fractions, whereas enzyme activities that indicate fungi are highest in the sand fraction (Gerzabek et al., 2002; Kandeler et al., 1999, 2000; Stemmer et al., 1998, 1999). There may be, however, even more qualitative selection for particle sizes than at a cell domain
MICROBES AND SOIL PARTICLE SURFACES
5
Figure 1 Adhesion of cells of E. coli to particles of sodium pyrophyllite as a function of particle concentration. Both cells and clay had effective mean diameters of 0.8 and 0.9 mm, respectively. Figure redrawn from Marshall (1980); original data from Hattori (1973). Reproduced with permission of John Wiley & Sons.
level, i.e., bacteria versus fungi. A recent article by Sessitsch et al. (2001) reported that not only were the numbers of attached bacterial cells greater in the finer textured fraction of Dutch soils, but also the community composition differed. Termmal Restriction Fragment Length Polymorphism (T-RFLP) analysis of the communities associated with the different size fraction indicated that different organisms were the dominant inhabitants of the coarser particles as compared with the fine materials. These authors also suggested that diversity of the amplifiable genotypes in the clay fractions was greater than that in the coarser
6
A. L. MILLS
fractions based on the number of fragments recovered in T-RFLP analysis of the particle associated DNA in each size fraction. Much of the difference noted was attributed to organic amendments in the several soils examined, and to possible competition with fungi in the coarser particle sizes. There is good reason why clay fractions would have the maximum interactions with bacteria. The particles’ small size yields an enormous surface area per unit weight of solid, and the crystal structure of clays tends to engender a strong net negative charge on the surface that can attract nutrients, organics, and under the right circumstances, the bacterial cells themselves. Quartz and feldspars, materials with relatively inert chemical behavior usually dominate sand-size grains. As particles weather to smaller silt and clay-size particles, their composition changes to layer silicates; the smaller soil particles present not only a larger total surface area, but also a more reactive one as well. The influence of the finer textured materials is, therefore, a combination of the surface area increase and the specific mineralogy of the particles. Table I shows how cation exchange capacity (CEC) increases with increasing fineness of texture. The range of values for exchange capacity for each textural category reflects different mineralogy and different amounts of organic matter present in individual soils. Most soil particles do not present surfaces with the reactivity reflecting only the base mineralogy of the particle. Many particles have some portion of their surface coated with reactive materials, such as iron, aluminum, and manganese oxides and hydroxides, and organic matter. These coatings can alter the reactive surfaces of the particles, in some cases changing the negative surface charge to neutral or positive, and they can otherwise add reactivity to only slightly reactive surfaces. In this way, even quartz sand can become highly reactive by adsorption of a coat of negative metal oxide or organic matter. The issue of coating and soil particles’ tendency to sorb bacteria will be discussed later.
Table I Change in CEC with Change in Soil Texture Exchange capacity (Cmol g21) Textural classification Sand Sandy loam Loam Silt loam Clay and clay loam
No. of soils
Average
Range
2 6 4 8 6
2.8 ^ 1.1 6.8 ^ 5.8 12.2 ^ 3.6 17.8 ^ 5.6 25.3 ^ 20.3
2.0– 3.5 2.3– 17.1 7.5– 15.9 9.4– 26.3 4.0– 57.5
Note. Data for averages are expressed as the mean and standard deviation for the soils, and the range represents the minimum and maximum values reported within the textural category. Source. Mills and Powelson (1996) based on data taken from Brady (1984). Reproduced with permission of Wiley–Liss.
MICROBES AND SOIL PARTICLE SURFACES
7
Understanding the particle-size distribution in soil generally does not confer a strong predictive ability in relation to growth and interactions of the microbes, as habitats available to the organisms are defined by the structural organization of the particles into aggregates (Marshall, 1980).
B. CHEMICAL DISTRIBUTION MINERAL The most common minerals found in soils are listed in Table II. In general, the aluminosilicates predominate with some inclusions of carbonates, sulfates, and iron and aluminum sesquioxides. Sulfates and carbonates are much more soluble than either silicates or sesquioxides; soils tend to lose the carbonates and sulfates Table II Common Soil Minerals, and their Chemical Formulae Name
Chemical formula
Quartz Feldspar Mica Amphibole Pyroxene Olivine Epidote Tourmaline Zircon Rutile Kaolinite Smectite Vermiculite Chlorite Allophane Imogolite Gibbsite Goethite Hematite Ferrihydrite Birnessite Calcite Gypsum
9 > = > ;
Importance
SiO2 (Na,K)A1O2[SiO2]3 CaA12O,[SiO2]2 K,A12O5[Si2O5]3Al4(OH)4 K2A12O5[Si2O5]3(Mg,Fe)6(OH)4 (Ca,Na,K)2.3(Mg,Fe,Al)5(OH)2 [Si,Al4O11]2 (Ca,Mg,Fe,Ti,Al)(Si,Al)O3 (Mg,Fe)2SiO4 9 > Ca2(Al,Fe)3(OH)Si3O12 = NaMg3Al6B3Si6O27(OH,F)4 > ZrSiO4 ; TiO2
Abundant in sand and silt Abundant in soil that is not leached extensively Source of K in most temperate-zone soils Easily weathered to clay minerals and oxides Easily weathered Easily weathered Highly resistant to chemical weathering; used as “index mineral” in pedologic studies
Si4Al4O10(OH)8 Mx(Si,Al)8(Al,Fe,Mg)4O20(OH)4, where M ¼ interlayer cation
Abundant in clay as products of weathering; source of exchangeable cations in soils
Si3A14O12·n H2O Si2Al4O10·5H2O Al(OH)3 FeO(OH) Fe2O3 Fe10O15·9H2O (Na,Ca)Mn7O14·2.8H2O CaCO3 CaSO4·2H2O
9 = ;
Abundant in soils derived from volcanic ash deposits Abundant in leached soils Most-abundant Fe oxide Abundant in warm regions Abundant in organic horizons Most-abundant Mn oxide Most-abundant carbonate Abundant in arid regions
Source: Sposito (1989). Reproduced with permission of Oxford University Press.
8
A. L. MILLS
first, while silicates are altered from primary minerals such as feldspars and micas to secondary minerals (especially the clay minerals like montmorillonite and kaolinite). Mature soils tend to be dominated by silicates, but as the soils age, the sesquioxides take on a greater importance. The ultisols of the southeastern United States are rich in iron oxide and clays (i.e., kaolinitic minerals) characteristic of old soils. The oxisols of the tropics (formerly called lateritic soils) are dominated by metal oxides and residual quartz. Charges associated with the primary minerals are generally low due to the mineralogy and the low surface area presented by the larger rock fragments from which the soil is weathered. The charge (in this case a negative charge usually expressed as the CEC) increases through a maximum for 2:1 minerals such as montmorillonite (typically 70– 150 Cmol g21) through kaolinite (typically 10 Cmol g21) to the iron and aluminum oxides which often express a positive charge (i.e., anion exchange capacity). Microorganisms interact differently with the minerals, in part because of the charge differences, but also because of mineral constituents that are in the crystal lattice or that are adsorbed to exchange sites on the mineral surface or in the interlayer spaces.
ORGANIC The decaying mass of plant material is broken down into particles of ever decreasing size. The chemical action of microbes along with physical disintegration facilitated by invertebrate feeding, results in the pulverization of particulate organic matter into units comparable to the finest mineral particles. Many changes occur as the material is decomposed and disintegrated. The initial material is fresh or senescent plant material, but the actions of the soil biota soon generates material that has little semblance to the original plant from which it came. As the readily metabolizable components are removed, the remnants take on the character of both the plant-produced refractory compounds that remain and of the microbial cells generated during the decay process. The remnants have both dissolved and solid-phase components; the dissolved constituents often end up as coatings on all of the particles in the soil matrix. The heterogeneous nature of starting materials for soil organic matter formation combined with the various organisms involved and conditioned by the local environmental properties operating over multiple time scales make the exact chemical composition of soil organic matter difficult to determine for any given site over long periods of time. With us, a general description of soil organic matter composition can only be discussed in the most general terms. Alexander (1977) reported that about 15% of the mass is identified as polysaccharides, polypeptides, and phenols. This value comprises about 20% carbohydrates, 20% amino acids and amino sugars, and 10 –20% aliphatic fatty acids (Paul and Clark,
MICROBES AND SOIL PARTICLE SURFACES
9
1989). The remainder of soil organic matter is humic materials, a dark amorphous substance derived from the degradation of organic residues. The process of conversion of plant material to soil organic matter is often referred to as humification and the organic matter itself is called humus. These references attest to the overall importance of humic acids in the mature organic material. As pointed out below, the hydrophobic and electrostatic properties of soil particulates dictate, to a large extent, whether or not microbes will sorb rapidly to their surfaces. Soil organic matter has properties that include both hydrophobic and electrostatic effects. Fulvic acids are moderately size, reactive molecules with average molecular weights of 800– 1500 Da. Humic acids are larger molecules with average mo lecular weights of 1500 –4000 Da (Beckett et al., 1987), although there are reports of weights up to 200,000 Da (Thurman, 1985). Both of these classes of compounds have many reactive sites. The sites are dominated by carboxyl- and phenolic hydroxyl groups that dissociate in water to yield a polyvalent anion (see Table III). Soil organic matter (exemplified by a model humic acid) also contains amines that can carry a positive charge at moderate to low pH (Fig. 2). The magnitude of the charge generated in soil organic matter can be quite large. Depending on the existing soil conditions, soil organic matter can have a high exchange capacity for either cations or anions. Negative charges are generated by the dissociation of a proton from a carboxyl or a phenolic hydroxyl group. Protonation of amine groups (R – NH2 þ Hþ!R–NHþ 3 ) results in a positively charged site. The involvement of protons and protonation – deprotonation reactions gives rise to a substantial pH-dependent charge. Mineral particles demonstrate a combination of permanent and pH-dependent charge, but the most highly charged particles are dominated by permanent charges generated by substitution within the crystal lattice. The CEC of organic particles may
Table III Elemental and Functional Group Analysis of Humic and Fulvic Acids Elemental analysis (%) Sample Fulvic acids Humic acids
C
H
N
S
O
Ash
49.5 56.4
4.5 5.5
0.8 4.1
0.3 1.1
44.9 32.9
2.4 0.9
Functional group analysis (meq g21)
Fulvic acids Humic acids
OCH3
COOH
Phenolic OH
Total acidity
0.5 1.0
9.1 4.5
3.3 2.1
12.4 6.6
Source: Smith (1993), from data in Tate (1987). Reproduced with permission of Marcel Dekker.
10
A. L. MILLS
Figure 2 Structure of a sample humic molecule. Note the high proportion of aromatic groups and the opportunities for both positive and negative charges on the molecule as a result of the carboxyl and amine groups. Redrawn from Stevenson (1982). Reproduced with permission of John Wiley & Sons.
double over the range of pH 4.0 –8.0 (Smith et al., 1993); the exchange capacity of organics is also much higher (2 – 30 times) than that of mineral colloids (Smith et al., 1993). Soil organic matter does not comprise only humic materials. Depending on the nature of the starting plant material, it may contain substantial amounts of readily available carbon compounds and nitrogen and other essential elements may also be abundant. The amount of such nutrients is highly variable as is the overall degradability of the material. The most degradable materials, sugars, free amino acids, etc., are soluble and leach quickly from the particulate fraction. Little of the material is lost, however, because microorganisms colonizing the detritus surface or in close proximity to the decaying particles rapidly utilize the dissolved material as soon as it is lost from a particle. Indeed, rapid microbial colonization of organic particles probably occurs because of the readily available carbon and nitrogen compounds leaching from them. Once the pool of readily available (i.e., soluble) compounds has been depleted, the microbial community turns its attention to the solid phase material that remains, mineralizing carbon and assimilating other nutrients during the humification process.
ORGANIC-COATED MINERAL As described later, sorption due to surface charge or hydrophobicity or some combination of the two would seem to be a likely phenomenon to be affected by the presence of a coating over the substrate. Ferric coatings, for example, were found, to greatly enhance bacterial sorption, by Mozes et al. (1987), Scholl et al. (1990), Mills et al. (1994), Knapp et al. (1998), and Bolster et al. (2001), among others. The importance of metal oxide coatings will be described in detail later. One of the most ubiquitous of all coatings is that of organic matter. Hunter (1981) found that, without exception, the suspended particles in river and gesturing
MICROBES AND SOIL PARTICLE SURFACES
11
waters were negatively and quite uniformly charged. Since the particles themselves varied widely as to composition, Hunter concluded that this was likely due to a coating of organic matter or metal oxide. Immersing positively charged hydrous iron oxide particles in natural water containing organic matter results in a negative charge being imparted to the particles, presumably due to the sorption of organic molecules (Loder and Liss, 1985; Tipping, 1981; Tipping and Cooke, 1982). The effect of organic coatings on microbial sorption is variable, as will be described later. In some cases, sorption is enhanced, but in many cases, the addition of organics to the surface of minerals decreases the sorptive capacity of the grains. Other effects such as hydrophobicity and active adhesion as to organic particles may have an overall stimulatory effect on microbial attachment to the coated grains.
III. BENEFITS OF LIVING ON PARTICLES A. PARTICLE SURFACES AS A PHYSICAL SUBSTRATE FOR COLLECTION OF NUTRIENTS A common observation is that, given access to surfaces and interfaces, microbes quickly colonize those habitats. Basic ecological principles dictate that if organisms inhabit a site, there must be some advantage accruing to the organisms. The high frequency of microbial association with surfaces, therefore, must be interpreted as advantageous to the surface-associated cells. Although the exact nature of that advantage is not completely understood, it is commonly accepted that the adhesion of microorganisms to surfaces allows the organisms to utilize higher concentrations of nutrients, especially energy sources, that are also found to be associated with the interface. Since the mid-1930s, microbiologists have recognized the overwhelming tendency of bacteria (especially heterotrophs) to associate with particles or surfaces. Waksman and Carey (1935) noted that bacteria grew rapidly when water samples were placed in bottles and stored for even short periods of time, and ZoBell and Anderson (1936) demonstrated that the increases in growth in the bottles was proportional to the surface-to-volume ratio of the storage containers. The so-called “bottle effect” was thought to be related to the tendency of nutrients, especially organics, to collect on the walls of the container. This speculation was confirmed by Stark et al. (1938), who reported that clean glass slides accumulated organic matter when immersed in sterile lake water. Later studies observed a stimulation of bacterial growth in the presence of increased surface area provided by the additional particular materials to liquid cultures with low nutrient concentrations when compared with similar cultures that lacked particles (Heukelekian and Heller, 1940; Jannasch, 1973; ZoBell, 1943). In the presence
12
A. L. MILLS
of high nutrient concentrations, the effect was not observed. The nutrient enhancement principle applied to the effect of surfaces has persisted, even though it may not completely explain the rapid growth of suspended organisms in sampling containers. Kaper et al. (1978) observed a doubling of the number of suspended cells in polyethylene sampling bags within 20 min of sample collection from the Chesapeake Bay. Wall-associated organisms were not enumerated, so neither attachment nor growth on the bag’s surfaces was determined. Given that some fraction of the suspended cells probably became attached during the brief incubation, the results represented an underestimate of the actual growth of bacteria in the bag. Although the container effect as described by the early workers does not account for prolific growth of suspended organisms, that growth does not contradict the potential effects of the surfaces on enclosed samples. Nutrient enhancement associated with particle surfaces is of great importance in soil. Metting (1985, 1993) has described the soil outside of the rhizosphere as “in essence a nutritional desert.” He further describes the microbial lifestyle as one in which activity is ephemeral and sensitive to fluctuations in substrate and nutrient availability, along with microscale variations in physical and chemical conditions. Microbial life exists in microhabitats that exist on or near particle surfaces on the exterior and interior of soil aggregates. In some cases, association of organic compounds with surfaces can actually cause a decrease in biodegradation. Otherwise metabolizable compounds are sometimes rendered non-degradable when associated with particle surfaces. Mills and Eaton (1984) noted nearly complete inhibition of degradation of bromobenzene when sand was added to the reaction mixture. Guerin and Boyd (1997) observed reduced degradation of naphthalene in some of the soils tested to determine the effect of particle sorption on bioavailability of the compound, but in no case was degradation enhanced when the soil particles were present. Other work examining different contaminant compounds showed reduced or completely inhibited biodegradation activity in the presence of sorptive particles (Gordon and Millero, 1984; Ogram et al., 1985). Difference in degradability is generally considered to be related to the availability of a given compound. For example, bovine serum albumin, a rapidly sorbing protein complex, was degraded by attached bacteria but not by unattached bacteria, whereas suspended bacteria metabolized a readily desorbable dipeptide that was less available to the attached bacteria (Griffith and Fletcher, 1991). Based on observations from the literature, Mills and Powelson (1996) speculated that for situations in which availability to organisms is decreased by sorption to particles, there appears a competition between the microbes in surface for the compound. If the microbes can extract the compound from the surface, there may be no observable change in, or even a possible enhancement of, degradation. But, if the surface attraction for the compound is stronger than the ability of the microbes to extract it from the surface, then
MICROBES AND SOIL PARTICLE SURFACES
13
the compound will be less available. Similarly, sorption to the interlayer spaces of expanding lattice clays may further affect the situation. For example, Weber and Coble (1968) observed that Diquat could be degraded when it was sorbed to the external exchange sites of kaolinite (a non-expanding clay), but it was not available when bound in the interlayer spaces of montmorillonite (an expanding lattice clay mineral). Evidence exists for the strong attachment of nitrifying bacteria, in particular the ammonium oxidizers, to soil particles. Ammonium (NHþ 4 ) is strongly sorbed to negatively charged soil particles (it is equivalent to Kþ in charge and radius), and bacterial residence on the particle surface could facilitate uptake of NHþ 4 directly from the surface of the colloids. Indeed, Aakra et al. (2000) observed that indigenous ammonia-oxidizing bacteria in a clay loam soil were extremely difficult to release from soil particles compared to most of the heterotrophic bacteria. Less than 1% of indigenous NHþ 4 -oxidizers were extractable by the dispersion-density-gradient centrifugation technique, at least 10-fold less than the extractability of heterotrophic bacteria. When urea was applied to the soil, the authors observed a 5-fold increase in the potential ammonia oxidation rate, with the concomitant result of in a much higher percentage (8%) extractability of NHþ 4 -oxidizers. The newly grown oxidizers in the urea-treated soil seemed less strongly attached to the soil particles, suggesting that the strong attachment of indigenous oxidizers is either a gradual process taking place due to a long residence time (infrequent/slow cell division) compared to heterotrophic organisms, or that there were differences in species composition of the original community compared with that growing in response to urea inputs. Although sorption of bacteria and chemicals to surfaces can represent an increased availability to the bacteria, at the same time, the soil minerals may also serve to immobilize some toxic materials from the soil. Clay minerals have been shown to provide protection to NHþ 4 -oxidizers from the effects of the organic nitrification inhibitor nitrapyrin (Powell and Prosser, 1991). Similarly, if toxic metal ions are sorbed so strongly to the minerals as to make them unavailable to the microbes, the presence of the surfaces can have a secondary beneficial effect for the microbes (Stotzky, 1979). Incorporation of the clay minerals, kaolinite or montmorillonite, into synthetic media (Babich and Stotzky, 1977) or soil (Babich and Stotzky, 1977) reduced the toxicity of cadmium to a variety of organisms, including Bacillus megaterium, Agrobacterium tumifaciens, Nocardia corallina, Fomes amnosus, and several other fungi and bacteria tested. The reduction in the toxicity of Cd was correlated with the CEC of the clays. Although particle association by the cells themselves is not an element in this phenomenon, the fact that they are also present on the surface suggests that attraction of toxic levels of (at least heavy metal) contaminants is not likely to inhibit microbial activity and may enhance it when the toxic materials might otherwise have a damaging effect.
14
A. L. MILLS
B. UTILIZATION OF THE PARTICLE AS A CHEMICAL SUBSTRATE For many materials, both mineral and organic, the soil particles serve as more than a physical habitat for the microorganisms. In some cases, the particles also serve as the source of nutritional components for cell growth, including essential nutrients and even carbon, electron, and energy sources. In these cases, the organisms must possess enzymes capable of accelerating the dissolution of the particle, in effect, extracting the desired materials from the solid phase, or they must possess the ability to rapidly incorporate dissolved products of weathering reactions as rapidly as they are formed.
MINERALS The most common soil minerals are listed in Table II. The substances there do not represent sources of macronutrients to microbes, although some elements such as base cations might be derived by acid leaching of the minerals or by contact exchange of sorbed ions. It is well documented that bacteria can generally 32 obtain adequate PO32 4 by dissolution of PO4 -containing minerals through acid excretion (Alexander, 1977). Some elements, however, can serve as energy sources that can be extracted by some bacteria directly from crystalline material. The oxidation of metal sulfides such as pyrite by chemoautolithotrophs is an excellent example of microbes attaching to inorganic particles for the purpose of extracting and exploiting the elements held therein as a source of energy. Some bacteria, such as Thiobacillus ferrooxidans, attach to pyrite or sulfur by means of extracellular lipopolysaccharides. The primary attachment to pyrite at pH 2 is mediated by exopolymer-complexed iron(III) ions in an electrochemical interaction with the negatively charged pyrite surface. The extracellular polymers from sulfur-grown cells possess increased hydrophobic properties compared with that of cells grown on pyrite and the polymers do not attach to pyrite, indicating that organisms can adapt their attachment ability to match the substratum (Gehrke et al., 1998). Further evidence for adaptation of attached organisms is given by Knickerbocker et al. (2000) who observed “blebbing” (sloughing of outer membrane vesicles) in Thiobacillus thiooxidans attached to sulfur grains, but not when grown with sulfite as an energy source. Because the sulfite is dissolved, the authors concluded that the cells formed the blebs to assist in the dissolution of the solid phase substrate. Edwards et al. (2000) examined the growth of Thiobacillus caldus on pyrite and marcasite and observed that more cells attach to marcasite than pyrite and the authors suggested that was because of the greater solubility of the former. Edwards et al. (2000) also concluded that preferential colonization of surfaces relative to solution and oriented cell attachment on the sulfide surfaces
MICROBES AND SOIL PARTICLE SURFACES
15
suggest that T. caldus may chemotactically select the optimal site for chemoautotrophic growth on sulfur (i.e., the mineral surface). Using an enzyme-linked immunofiltration assay, Dziurla et al. (1998) were able to estimate directly and specifically T. ferrooxidans attachment on sulfide minerals. The mean value of bacterial attachment was about 105 bacteria mg21 of pyrite at a particle size of 56– 65 mm. The geometric coverage ratio of pyrite by T. ferrooxidans ranged from 0.25 to 2.25%. From their results, Dziurla et al. (1998) inferred attachment of T. ferrooxidans on the pyrite surface to welldefined limited sites with specific electrochemical or surface properties. This conclusion was supported in a laboratory study (Sanhueza et al., 1999) of the attachment of a pure strain of T. ferrooxidans on films of synthetic pyrite. Pyrite films representing a wide range of structural and electronic properties were produced by sulfuration of pure iron films at different annealing temperatures, viz., 250, 300, 350, 400, 450, and 5008C. The patterns and degree of attachment of T. ferrooxidans to the synthetic pyrite depends strongly on the degree of crystallization of the sulfide films, which varies with the sulfuration temperature. In the low range of sulfuration temperatures (250, 300, 3508C), where there is a major presence of amorphous pyrite, elongated clusters of densely packed bacteria attach to the films. In the range of sulfuration temperatures (400, 4508C) where formation of highly crystallized pyrite predominates, bacteria attachment occurs as isolated bacteria or short bead-like chains. The percentage of pyrite surface coverage by T. ferrooxidans is lower at high sulfuration temperatures, where pyrite is fully crystallized. The microorganisms seem to attach preferentially to the less crystallized or amorphous zones of the pyrite films which provide a better availability of sulfide ions for bacterial oxidation. No examination of attachment to sulfur is presented in this work. From the above, and other evidence, it is clear that many metal sulfide oxidizing bacteria colonize the mineral grain surfaces, often selecting a specific site on the crystal that provides the maximum access to the elements required by the organism (Andrews, 1988), and some work suggests that treatment of the mineral surfaces with surfactant compounds (Jiang et al., 2000; Nyavor et al., 1996) can inhibit the formation of acid mine drainage, at least in part by reducing attachment and thereby the colonization of the particle surfaces.
ORGANICS The colonization of organic particles by microorganisms is a well-documented phenomenon in both aquatic and terrestrial systems. Both aquatic detritus and soil organic matter are made up of particles of various sizes that comprise decaying plant material along with the microorganisms that are the mediators of decay. As the particles decay their chemistry becomes less and less plant-like and more
16
A. L. MILLS
and more microbial in nature as the organic material in the particles is assimilated and converted to microbial biomass (both living and dead). It is usually observed that organic matter decomposition includes a shift in elemental ratios from those of the native plant material to the equilibrium value associated with well decomposed soil organic matter, i.e., about 10– 1 (roughly 5% by weight). This value is also a typical one observed for a large number of soil microbes (Alexander, 1977). Thus, the close association of microbial cells with soil organic matter particles is well accepted. Microscopic observations of organic particles from aquatic habitats (i.e., detritus) generally yield bacterial abundance values of about 108 cells g21, and those organisms are counted only after rigorous extractions to separate them from the detrital particles. For routine microscopic techniques such as epifluorescence counting, it is difficult to separate the organic from the inorganic particles in soil for the purpose of counting the attached microorganisms. It is reasonable to assume, however, that the number of bacteria associated with organic particles in soil are not dissimilar from those in detritus systems. Given the chemical nature of the organic compounds composing the particles, attachment of the microbes through electrostatic mechanisms is not direct. Electrostatic bonding occurs between charges created on the surface of the bacterial cell and the organic particle by functional groups associated with the molecules, make up the structures associated with each entity. Surfaces of bacteria are dominated by teichoic acid (gram positive strains) or polysaccharides (gram negative strains) (Brock and Madigan, 1991). The most common functional groups on the bacterial surface are carboxyl and some amine groups (although a variety of other, less common groups may also be involved), whereas charge-generating groups on humified particles tend to be dominated by carboxyl groups and phenolic hydroxyls combined with a lower percentage of amine groups than found on the bacterial cells (Stevenson, 1982). Given that acid carboxyl groups on humic materials are largely dissociated at about pH 5.0 and above, and that phenolic hydroxyls are undissociated at pH values below about 10, the overall charge of both the organic particles and the bacterial cells is negative (Plette et al., 1995; Posner, 1964; Stevenson, 1982). Amines can be protonated at a variety of pH values (depending on the specific amino acids with which the groups are associated), but they are rarely so heavily protonated at typical soil pH values that the net charge becomes positive (see Figs. 2 and 3). Given the net negative charge of both particles, it is clear that some intermediate bridge must be used to span and join the two negatives together. Polyvalent cations generally provide such a bridge, and complexation of ions such as Ca2þ, Mg2þ, etc., with the negative charges on either the cell or particle surface can leave exposed positive charges to complex with negative charges on the corresponding particle. While the presence of a large proportion of polyvalent ions tends to cause electrostatic attraction and flocculation of particles in suspensions, the organometallic complexes are much stronger than simple
+ M
-CO O + M+
+ M+ 2
-NH
-CO O
-CO O + M+
2
+ M
R-NH3+
-NH
-CO O + M+ -CO O -+ M
2
-NH
-CO O + M
+ M
O -CO
2
-CO O + M+
2
-NH
R-COO– + H+
17
+ M+
+ M+ + M+
+ M
+ M
R-COOH
-CO O + M -CO O
-NH
-CO O + M+
-CO O
+ M
+ M 2
-NH
-CO - + O M
+ M+
+ M+
MICROBES AND SOIL PARTICLE SURFACES
R-NH2 + H+
Figure 3 Generation of charges on the surface of bacterial cells and on organic particle surfaces. The degree of positive and negative charge on any cell will reflect the relative proportion of acids and bases exposed to the solution, the pKa for any proton donating–accepting group and the pH of the surrounding solution. Other groups may be involved with this type of reaction; carboxyl groups and amines are common groups and are shown for simplicity. The presence of polyvalent cations can generate a net positive charge that can lead to a strong ionic bonding of the particles together.
electrostatic attractions and attachment is irreversible unless some major change in ionic strength or composition occurs.
IV. IMPORTANCE OF ATTACHED MICROBES IN SOIL A. NUMBERS OF ATTACHED VERSUS FREE-LIVING MICROBES The total abundance of bacteria in soil is typically 108 g21 or greater, depending on conditions of soil moisture, organic concentration, pH, etc. The bacterial biomass is generally on the order of 500 kg ha21. It is difficult to determine what fraction of these organisms reside attached to the soil particle surfaces as opposed to being suspended in the soil solution. As pointed out above, the liquid volume in unsaturated soils is small; in a soil with a porosity of 0.4 at 50% saturation, the amount of water in a cm3 of the soil would only be 0.2 ml. If the solution contains about 106 cells ml21 (a value often associated with surface waters), the ratio of attached to unattached cells would be determined as: total cells 2 suspended cells 108 2 2 £ 105 ¼ ¼ 5000 total cells 2 £ 105
ð1Þ
Data to support this calculation are not readily available. Classical counting methods are grossly inaccurate for determining such a ratio; extraction of cells for direct microscopic enumeration fails to differentiate between truly suspended cells and those that are loosely associated with the particle surface via electrostatic or hydrophobic interactions and are easily removed. It is not clear
18
A. L. MILLS
if use of newer techniques can overcome preparation artifacts to help determine the fraction of cells in intimate contact with particle surfaces. However, calculations such as the one presented above suggest that the role of attached organisms must be dominant in soils. Such a conclusion would be consistent with the findings related to particle attachment in aquatic environments. While typical counts of suspended bacteria are on the order of 106 ml21, counts associated with organic particles are generally two orders of magnitude larger (i.e., in the order of 108 g21).
B. PHYLOGENY OF ATTACHED VERSUS FREE-LIVING MICROBES Given the lack of a real suspended phase for microorganisms as suggested above, one might anticipate that there would be no phylogenetic differences in the attached and suspended soil microbial community. However, whenever there is an opportunity for the two communities to develop there does seem to be a different set of microbes found attached to particles as opposed to remaining in suspension. In aquatic environments where detrital particles are dominant in the water column, molecular analysis has shown important differences in the attached and suspended communities. Examination of ribosomal RNA (both amplified and cloned) (DeLong et al., 1993) and low-molecular-weight (transfer and 5S ribosomal) RNA (Bidle and Fletcher, 1995) has shown that particleassociated communities differ from those in suspension. More recent work by the latter group, however, suggested that there was little difference in the communities during the summer, perhaps due to a rapid exchange of organisms between the attached and suspended phases (Noble et al., 1997). At this point, similar analyses have not been reported for soil; it may be that the lack of a clear suspended phase makes the question uninteresting. It may also be that concern over the removal of particle-associated microbes (which are most certainly the predominant in soil) during the extraction process may make the results of such an exercise suspect. One would logically expect, however, that differences in attached and suspended organisms in soil would be much less than found in aquatic environments.
C. QUANTITATIVE CONSIDERATIONS OF ACTIVITY OF ATTACHED VERSUS NON-ATTACHED There is little information for soils on the relative contribution to microbial activity of attached versus unattached cells. In most cases, it is simply assumed that all the organisms in a soil sample are associated with particle surfaces. Indeed, all other things being equal, the earlier calculation which suggested that 5000 times more microbial cells are associated with the mineral grain surfaces
MICROBES AND SOIL PARTICLE SURFACES
19
and in suspension would suggest that metabolic activity should be divided on a similar ratio. Although it is not examination of soil, a study by Hopkinson et al. (1989) in Georgia coastal waters, supports the concept that in the presence of large particulate surface areas, metabolic activity associated with particle surfaces is greater than comparative activity in the suspended phase. In the estuary, 80% of heterotrophic activity was associated with particles . 3.0 mm and 20% of the activity was associated with particles less than 3.0 mm. These results suggest that the organisms (and their activities) are primarily associated with particles. In the open ocean, however, the circumstances were reversed. Eighty percent of the metabolic activity passed a 3.0 mm filter, indicating nonparticle association. This seems perfectly reasonable, given that the particulate load in estuarine waters can be orders of magnitude higher than in the open ocean.
V.
DIVERSITY OF MODES OF ATTACHMENT A. REVERSIBLE VERSUS NON-REVERSIBLE
Reversible attachment describes the situation in which microbes are easily removed from particle surfaces by shear forces from tangential flow across the surface, or by small changes in the aqueous phase chemistry that causes a desorption of cells attached by electrochemical forces. Non-reversible attachment, or permanent adhesion, of microbes to surfaces occurs due to the formation of polymer bridges between the cells and the surfaces to which they attach (Marshall, 1980; Marshall et al., 1971). Non-reversible sorption also includes the formation of metal – organic complexes that do not dissociate except under conditions of significant chemical alteration, i.e., major change in ionic strength or composition such that the binding polyvalent metals would be displaced by monovalent forms through mass action, or loss of ionic strength through dilution. Hydraulic shear is not considered sufficient to break the complexes holding the bacteria to the surface.
B. ELECTROSTATICS ELECTROSTATICS AND HYDROPHOBIC EFFECTS: DLVO THEORY Reversible sorption to particles is often explained by a combination of electrostatics combined with hydrophobic effects to overcome the natural repulsion of bacteria and particles that arises from the similar charges expressed
20
A. L. MILLS
at their surfaces. The surface charges of both mineral particles and bacterial cells are slightly to strongly negative at common soil pH values (4 – 8). For an excellent review of surface charge interactions in soil, see Bolan et al. (1999). The pH at which positive and negative electrical charges on a particle balance is the isoelectric pH (pHiep), also referred to frequently as the point of zero charge or ZPC. Harden and Harris (1953) found that bacterial pHiep varied from 1.75 to 4.15 for 18 gram-positive species and from 2.07 to 3.65 for 13 gram-negative species. Typical soil solid pHiep values are 2.0 for quartz and 4.6 for kaolinite (Bolan et al., 1999; Stumm and Morgan, 1996). Both bacteria and common soil components have pHiep values lower than the typical pH of soil solution, pH 5– 8 (Table IV). Consequently, bacteria and solids generally have net negative charges and will repel each other electrostatically. It is important to keep in mind that some soil components may be positively charged in near-neutral pH conditions, e.g., amorphous Fe(OH)3, which has a pHiep of 8.5 (Stumm and Morgan, 1996). Scholl et al. (1990) found that attachment of negatively charged bacteria was much greater to positively charged surfaces of limestone, Fe(OH)3-coated quartz, and Fe(OH)3-coated muscovite than to uncoated quartz and muscovite. The negative charge on bacteria and solids is neutralized by a swarm of cations that becomes progressively less dense away from the surface, resulting in a diffuse double layer of charge (see Fig. 4). The approximate thickness of the diffuse layer is k 21. At 208C,
k<
I 0:5 0:28 nm
ð2Þ
where I is the ionic strength expressed in units of molality (m; note that many authors use molarity, M, to quantify ionic strength; although not strictly correct,
Table IV Surface Charges at Typical Soil pH Values
Soil mineral Kaolinite Illite Montmorillonite Vermiculite Muscovite Quartz Al hydroxide Gibbsite Fe hydroxide Humic acid
Surface charge and pH at which charge was measured (Cmol kg21) 213 (pH 7.0) 221 (pH 7.0) 290.4 to 2127.6 (pH 7.0) 2195.3 (pH 7.0) 222 (pH 7.0) 22 (pH 7.0) þ 24 (þ 16.0) to þ10.0 (pH 6.0) þ 7.2 (pH 6.0) to 20.88 (pH 9.0) þ 34 (pH 5.8) 2330 to 2340
Source: Bolan et al. (1999).
Reference Hendershot and Lavkulich (1983) Hendershot and Lavkulich (1983) Cowan et al. (1992) Bouabid et al. (1991) Hendershot and Lavkulich (1983) Hendershot and Lavkulich (1983) Hendershot and Lavkulich (1983) Hingston et al. (1974) Bolan et al. (1999) Posner (1964)
MICROBES AND SOIL PARTICLE SURFACES
B
UN
CONCENTRATION
CO TE
RI
O S
CO
N
UN
TE R IO
CO
IO N
S
S C O IO NS
I2
CONCENTRATION
A
N
21
I1 d2
d2 d1
d1 Wf
Figure 4 Distribution of charges (ions) near a charged surface. (A) Represents the distribution at two different ionic strengths (I ) representing two equilibrium concentrations of ions in the soil solution. Co-ions are those with the same charge as the surface and counterions are those with a charge opposite to that of the surface. d represents the thickness of the diffuse double layer near the particle surface (i.e., the distance from the surface at which the concentration of counter and co-ions equals that of the equilibrium solution). Note that as the ionic strength increases, the thickness of the double layer decreases. (B) Represents the situation in which part of the water layer has been removed due to drying. The initial thickness of the water film is greater than the thickness of the double layer. Note that the integral of the ion concentrations from 0 to d1 and d2 remains the same as no salts are removed upon evaporation of the water. Concept from Bolt and Bruggenwert (1976). Reproduced with permission of Elsevier Science Publishers.
molarity is a good approximation for dilute solutions) (Stumm and Morgan, 1996). For example, when I ¼ 0:001 m; k21 ¼ 8:9 nm: If a bacterial suspension with pH . pHiep is placed in an electric field, the cells will be drawn towards the positive pole, and the cations farthest away from the surface of a cell will be sheared off. The resulting potential at the outside of the diffuse layer, determined from the bacterial velocity, is called the zeta potential. This potential is dependent on the ionic strength and pH of the suspension as well as density of charge on the bacteria. Gannon et al. (1991) measured zeta potentials for 19 bacterial strains suspended in deionized water that ranged from 2 8 to 2 36 mV. Due to the diffuse layer, the electrostatic potential that repels ions of like charge (negatively charged particles, for the case considered here) increases as the particle approaches the solid. For the case of a negatively charged, 1 mm
22
A. L. MILLS
bacterium approaching a soil particle, the two surfaces may be approximated as spheres. The repulsive electrostatic potential (VR) is VR ¼
cr 2 e2kx ð2r þ xÞk2
ð3Þ
where c is a constant for fixed temperature and decreases the separation distance for a given level of repulsion. If electrostatic repulsion is not strong enough to keep bacteria away from a solid, close approach may result in strong adsorption due to van der Waals force. This force is sometimes referred to as electrodynamic or fluctuating dipole – induced dipole, since it results from the motion of electrons in their orbitals (Weber et al., 1991). This potential (VA) is always attractive; for spheres, VA ¼
Ar 12x
ð4Þ
where A is the Hamaker constant (about 100 g nm2 s22) (Stumm and Morgan, 1996). The total interaction potential is the sum of the attractive and the repulsive forces, VR þ VA. This is the basis of what is called the DLVO theory (for the researchers Derjaguin and Landau (1941), and Verwey and Overbeek (1948)). According to the theory, for most groundwater (I , 0.01 m), suspended bacteria are electrostatically repelled from solids with increasing force as they approach the solid, until a separation distance of about 1 nm is reached. Brownian motion, convection, or motility may push the bacteria over the repulsive barrier. If the separation distance becomes less than 1 nm, the potential rapidly decreases to a strongly adsorptive van der Waals minimum. At very close range, other mechanisms such as hydrogen bonding may also contribute to adsorption. Increasing ionic strength decreases the magnitude of the barrier (equation 2). Mills et al. (1994) found that increasing the ionic strength from 0.001 to 0.01 reduced recovery of bacteria from sand columns by as much as an order of magnitude. Similarly, Gross et al. (1995) found that increasing ionic strength from 0.01 to 1 M increased adsorption of bacteria to borosilicate beads by a factor of 9. For high ionic strengths (generally I . 0.1 M) the diffuse-layer potential is so weak that the total potential is always negative (attractive), and there will be no net energy barrier to contact. Murray and Parks (1980) discuss details of the DLVO equations for the interaction of a virus particle and an oxide surface. They emphasize the importance of differences in van der Waals potentials for different materials, with metals having strong potential and organics weak potential. In addition to the primary van der Waals minimum described above, a “secondary minimum” may be important for reversible adsorption of bacteria (see Fig. 5). The total interaction potential of a sphere and a plate (using representative values for the Hamaker constant and electrostatic potential and for ionic strengths around 0.01 M) is negative at about 5 –20 nm separation, as well
MICROBES AND SOIL PARTICLE SURFACES
ENERGY OF INTERACTION Attraction Repulsion
a
c
b GE
23
GE GE
Gtotal
GA
Gtotal H
Gtotal
H GA
Secondary Minimum
H
GA
Primary Minimum
Figure 5 Repulsive and attractive relationships near a charged surface. The three diagrams represent conditions of: (a) very low ionic strength; (b) intermediate ionic strength; (c) high ionic strength. Repulsive forces that keep same charged particles far from the surface, i.e., unadsorbed, dominate surfaces covered with very dilute solutions. Surfaces associated with solutions of intermediate ionic strength have a “secondary minimum” in the repulsive forces in which like-charged particles can rest in a “reversibly sorbed” state, subject to shear stresses. Under high ionic strength conditions, the diffuse double layer has collapsed to allow a deep primary minimum to form at the surface. Particles that reside in that region are considered irreversibly sorbed, although changes in pH, ionic strength, etc., can change the distribution of repulsive and attractive forces that can cause desorption of the particles. Concept from Van Loosdrecht et al. (1990). Reproduced with permission of The American Society of Microbiology.
as at , 1 nm (McDowell-Boyer et al., 1986; Van Loosdrecht et al., 1989). Support for this hypothesis is found in the attachment energy calculated from adsorption isotherms. In many cases, bacteria reversibly adsorb to solids such that there is a linear relationship between suspension concentration and adsorbed cells per gram of solid (the adsorption isotherm). The slope of this relationship (Kd) and the number of adsorption sites per gram (M ) may be used to calculate the energy of bacterial adhesion with the van’t Hoff equation: K DH 0 ln d ¼ 2 þC ð5Þ M RT where DH 0 is the standard state enthalpy, R the gas constant, T the absolute temperature, and C a constant (Hendricks et al., 1979). van Loosdrecht et al. (1989) found that the adhesion energy calculated from adsorption isotherms is about the same as the predicted potential at the secondary minimum using the DLVO equations and suggested that initial bacterial adhesion takes place at the secondary minimum.
24
A. L. MILLS
Electrostatic repulsion may result in enhanced transport of microbes relative to the average mass of water by keeping them exposed to greater velocities in the central “fast lane” of flowing soil pore water. Water adheres to pore walls and slows adjacent flow according to Newton’s law of shear. Consequently, water and entrained substances in the center of a pore move the microbes most rapidly. At the molecular level this entire process is called anion exclusion. Microbes have hydrophobic regions on their surfaces where there are high densities of hydrocarbon (C – H) groups. Water molecules “prefer” (have lower free energy) to associate with other water molecules or hydrophilic surfaces. Water molecules in contact with hydrophobic surfaces exhibit a surface tension that tends to minimize water contact with hydrophobic regions. Surface tension may squeeze hydrophobic particles out of the water environment and into other hydrophobic regions. Murray and Parks (1980) defined hydrophobic bonding as “the aggregation of non-polar surfaces resulting from the minimization of reoriented and thus higher (than bulk) free energy water structure adjacent to the non-polar surfaces.” For a bacterium attaching to a solid surface, this change in free energy is DGadh ¼ gBS 2 gBL 2 gSL
ð6Þ
where gBS is interfacial tension of the bacterium –solid surface, gBL is that of the bacterium –liquid, and gSL is that of the solid – liquid surface. Values of the interfacial tensions are calculated from Young’s equation by the relationship
gSV 2 gSL ¼ gLV cos u
ð7Þ
where gSV, gSL, and gLV are the solid –air, solid – water, and air –water interfacial tensions and u the contact angle of the water on the solid (Marshall, 1990; Neumann et al., 1980). Bacterial hydrophobicity is, therefore, most appropriately evaluated by contact angle measurement, but indexes of cell surface hydrophobicity have also been obtained by measuring bacterial adherence to hydrocarbons (BATH) and by hydrophobic interaction chromatography (HIC) (Rosenberg and Doyle, 1990). The bacterial contact angle is the angle formed when a drop of water contacts a surface such as a lawn of bacterial cells. More hydrophobic bacterial surfaces cause the water drop to “ball up,” resulting in larger contact angles. Wan et al. (1994) concentrated three strains of bacteria on filters and used a goniometer eyepiece to observe the angle formed by a drop of 1 mM NaNO3 on the bacteria. The contact angles ranged from 77.18 (relatively hydrophobic bacteria) to 24.78 (relatively hydrophilic). Huysman and Verstraete (1993) used the BATH method with octane to measure hydrophobicity of Escherichia coli, Streptococcus faecalis, and seven strains of Lactobacillus. Hydrophobicity, measured as percent removal by octane, ranged from 2% for E. coli to 94% for Lactobacillus strain Lc4. They then tested bacterial adsorption to sand and transport through sand columns and found
MICROBES AND SOIL PARTICLE SURFACES
25
significant correlations between hydrophobicity and adhesion to sand. The hydrophobic characteristics of poliovirus (which has a protein surface) were evaluated by Murray and Parks (1980) by testing its sorption to C2C13F3. There was no significant difference in virus concentration between a viral suspension mixed with this hydrophobic liquid and a control suspension similarly mixed without it. The authors concluded that the protein surface was very hydrophilic. They neglected to account for sorption of virus to the air – water interface, however, which could explain an equal loss in virus concentration for the control and C2C13F3 treatments. HIC measures the amount of bacteria retained by a hydrophobic gel. Gannon et al. (1991) used this method as well as the BATH method to estimate hydrophobicity of 19 bacterial strains. The HIC assay retained from 7 to 91% of the cells, and the BATH assay retained from 5 to 85% of the cells, although the correlation between the HIC and BATH assays was poor. Furthermore, there was no significant relationship between the HIC or BATH results and transport of cells through loam columns. Hydrophobic interactions are not simple relationships since humic material (generally the most important hydrophobic component of soil solids) occurs in solution as well as associated with solids. Dissolved organic molecules may compete with microbes for adsorption sites or modify interfacial tensions and thereby interfere with hydrophobic adsorption (Powelson et al., 1991). Although there is little controversy over the fact that both electrostatic effects and hydrophobic effects influence bacterial sorption to surfaces, there are different opinions about which is more important. Both Stenstro¨m (1989) and Van Loosdrecht et al. (1987a) found that increased hydrophobicity resulted in increased sorption and that hydrophobicity was more important than surface charge in determining the extent of sorption. Van Loosdrecht et al. (1987a,b) used negatively charged polystyrene with a contact angle of 708, making it relatively hydrophobic. Stenstro¨m (1989), however, found that sorption increased with cell hydrophobicity even with a hydrophilic substratum such as quartz. On the other hand, Fletcher and Loeb (1979) found sorption of a Pseudomonas sp. was lessened considerably on hydrophilic surfaces relative to hydrophobic surfaces. The sorption of a hydrophobic cell would be less with a hydrophilic surface would seem to be borne out by the findings of Absolom et al. (1983). One notable difference between studies which have found an electrostatic effect and those which have found a hydrophobic effect is the ionic strength at which they have been run. Studies which have found an electrostatic effect, whether it is due to ionic strength of the suspending liquid (Fontes et al., 1991; Marshall et al., 1971; Scholl et al., 1990) or the surface charge of the bacteria (Sharma et al., 1985), have used ionic strengths below 0.1 M. Studies which have observed a hydrophobic effect (Absolom et al., 1983; Fletcher and Loeb, 1979; Stenstro¨m, 1989; Van Loosdrecht et al., 1987a,b) have been carried out at ionic strengths greater than 0.1 M, except for that performed by Mozes et al. (1987),
26
A. L. MILLS
who used distilled water. Gordon and Millero (1984) found that the nature of electrostatic interactions changed at 0.1 M. Below this level, bacterial sorption increased with ionic strength, as predicted by DLVO theory. Above 0.1 M, DLVO theory was no longer applied; bacterial sorption was unaffected by ionic strength, but decreased with increasing cation concentration. This may help to explain why electrostatic interactions have been found to be, at best, of secondary importance at higher ionic strengths. It may also serve to explain why Scholl et al. (1990) and Mozes et al. (1987) observed changes in sorption with pH, while Stenstro¨m (1989) did not.
PRIMARY MINERAL DIFFERENCES Soil minerals comprise a variety of chemical classes (Sposito, 1989), but the dominant ones are silicates, carbonates, sulfates, and oxides of Fe and Al (Table II). In addition, PO32 4 and a few sulfides can also be found in some environments. There has not been a great deal of work comparing bacterial attachment rates on a variety of pure minerals, but some literature reports indicated differences among the minerals that have been tested. In a field study, Mills and Maubrey (1981) examined the effect of mineralogy on the initial colonization of chips of freshly exposed rock submerged in lakes or ponds. The chips were made by sawing parallel faces on pieces of rock or of museum-grade specimens of single mineral crystals, then carefully cleaning the surfaces to remove any organic contaminants left as a result of the processing. The chips were left in place for only 24 h because after that time, the film was invariably more than one cell thick in some places, precluding obtaining an accurate count of the attached bacteria. Mills and Maubrey (1981) observed that chips immersed in flowing water always accumulated more bacteria than chips of the same material placed in still water, a fact that probably has little relevance to the unsaturated soil situation in which shear and boundary layer considerations are not the important factors. Quartz chips were colonized more rapidly than calcite. When limestone and hematitic sandstone were compared, either no difference was observed, or the quartz-based rock was colonized more slowly than the carbonate. Environmental factors notwithstanding, this set of seemingly contradictory observations were deemed by the authors to be consistent with the findings of Marszalek et al. (1979) that surfaces which are biologically inert (such as glass) tend to be colonized more rapidly than those which are biologically active (i.e., that leach ions toxic to the colonizing community). Mills and Maubrey (1981) suggested that the presence of potentially leachable Fe in hematitic sandstone may increase the activity of the rock surface and thereby inhibit microbial colonization. In a similar study, the same research group exposed chips of quartz, muscovite, and limestone to a suspension of an organisms isolated from
MICROBES AND SOIL PARTICLE SURFACES
27
groundwater and measured the rate of attachment over a 16 h period (Scholl et al., 1990). In doing so they obtained results in direct opposition to the earlier work of Mills and Maubrey (1981). Attachment to the silicates was lower than to the limestone (Fig. 6), which compared favorably with quartz and muscovite after coating with iron hydroxide. Clearly the speculation of Mills and Maubrey (1981) about the leachable iron was incorrect. Scholl et al. (1990) observed that the attachment to the various materials corresponded with the surface charge of the mineral at the experimental pH. The latter finding is consistent with expectations based on DLVO theory and is supported by nearly all of the work on the importance of metal oxides that followed.
EFFECT OF MINERAL COATINGS (e.g., IRON OXIDES) Most temperate soils have silicate minerals making up the bulk of their composition (Table II). Other minerals occur in soil with high frequency, notably oxides of iron or aluminum. As phyllosilicates weather, the silicon-containing layers are stripped away leaving behind abundant aluminum mixed with residual oxides of iron. Most highly weathered soils are characterized by accumulation of 450 Muscovite Quartz Limestone Coated Quartz Coated Muscovite
400
Number of Bacteria (cells mm–2)
350 300 250 200 150 100 50 0 0
2
4
6
8
10
12
14
16
18
Hours of Exposure Figure 6 Attachment of a gram negative bacterium isolated from ground water to uncoated chips of quartz, muscovite and limestone and to quartz and muscovite chips coated with Fe-hydroxide. Values are mean ^ 1SEM; some fell within the size of the point as drawn. Figure redrawn from Scholl et al. (1990). Reproduced with permission of Elsevier Science Publishers.
28
A. L. MILLS
metal sesquioxides as part of the normal pedogenic process. In mature soils, often even in young soils, the oxides and hydroxides exist as coatings on the surfaces of silicate mineral grains. Podsolization of soils is the downward translocation of Fe, Al, and organic matter. The iron and aluminum are dissolved in the surface horizons (often with the assistance of complexation by soil organic matter) and are moved downward in the profile where they are deposited as coatings on sand-, silt-, and clay-sized particles. There is a substantial opinion that nearly all mineral grains have coatings that differ somewhat from the bulk mineralogy of the grains. Given that the surfaces of pure quartz and silicate minerals express a net negative charge, it is reasonable to assume that cations such as Al(OH)2þ or Fe(OH)þ 2 (products of the hydrolysis reactions of Al3þ and Fe3þ) would be readily attracted to the silicate surface. Indeed, coatings of amorphous iron and aluminum sesquioxides are common on silicate grains. At typical soil pH values, these layers of amorphous material, as well as the more crystalline coatings such as Al(OH)3 (gibbsite) or Fe(OH)3 (goethite), can block negative charges from the silicate and can often act as bases accepting protons and imparting a positive charge to the mineral surface (Fig. 7). Coatings do not completely cover the entire grain surface; a soil may retain its net negative charge even though there are abundant sites of less negative or even positive charge distributed over the surface. The presence of metal oxide coatings on mineral grains has a profound effect on the interaction of those surfaces with bacterial cells. Scholl et al. (1990) BULK SOLUTION
-
+
++
PLANE OF SHEAR ++
-
++
ADSORBED HUMIC LAYER
-
-
HYDROUS Fe or Al LAYER
+
-
+
-
-
+
-
-
+
+
-
+
-
-
-
-
MINERAL GRAIN
Figure 7 Changes in charge distribution upon coating of a silicate mineral with metal oxides and subsequent over-coating with organic materials. While soil particles normally have a net negative charge at most soil pH values, coating with metal sesquioxides or organics may generate regions of net positive charge at those same pH values. This figure illustrates the type of surface exposure expected given a discontinuous coat of sesquioxide or humic material over the surface of the mineral grain. Note the importance of (especially) polyvalent cations in conferring a positive charge over the organic layer. Figure reproduced from Mills and Powelson (1996) with permission of Wiley–Liss.
MICROBES AND SOIL PARTICLE SURFACES
29
demonstrated that the deposition of iron sesquioxide on quartz and muscovite greatly increased the sorption of bacteria from suspension. The retention of bacteria in columns and clean quartz sand coated with iron sesquioxide was much greater than for the uncoated analog (Scholl et al., 1990). Mills et al. (1994) demonstrated that bacterial sorption to clean sand followed a linear isotherm for the range of concentration of bacterial cells used but that sorption to iron-coated sand was nearly complete for all concentrations up to the sorption limit of 108 cells g21. Using that limit, Mills et al. (1994) constructed a simple model for sorption of bacteria to partially coated sand. Initial sorption was complete and irreversible up to the sorption limit imposed by the amount of sesquioxide coating available (i.e., 108 cells g21 of coating included) in the porous medium. Once the coating was saturated with cells, sorption was controlled by the linear isotherm developed for the clean quartz sand. The intercept for the linear isotherm was the maximum number of cells that could be sorbed by the coated sand in the mixture used for the experiment. The model differs significantly from standard saturation models (e.g., Langmuir isotherm) which is concentration independent up to the saturation level of the coated sand. The theory that underlies the simple model can describe any mixture of reversible and irreversible sorption processes.
EFFECT OF ORGANIC COATINGS ON ATTACHMENT The effect of organic matter on bacterial sorption is not well studied, although there is substantial evidence to indicate that the presence of a “conditioning film” of organic matter enhances and may even be a prerequisite for bacterial colonization (see Marshall (1996) for a thorough discussion of conditioning films). Pringle and Fletcher (1986) found that a variety of macromolecules inhibited attachment of bacteria to polystyrene when the macromolecules were present in the suspension during the attachment, and a number of the macromolecules also inhibited attachment when the surfaces were preconditioned with them. Scholl and Harvey (1992) observed that retention of bacteria in artificial groundwater was greatest in sand which had been leached of organic matter but still retained an iron oxyhydroxide coating (and was therefore positively charged), as opposed to sand which had both an organic and iron coating or sand which had neither. They also obtained less sorption to sand that was coated with iron oxyhydroxides and organic matter when organic matter was present in the suspension (but not with iron-coated sand with the organic matter coating removed, as might be expected for competition mechanism). Scholl and Harvey (1992) concluded that experiments with better defined components were necessary. Both Lance and Gerba (1984) and Powelson et al. (1991) found that organic matter in solution (derived from sewage sludge in both studies, and from natural humic material as well in the latter study) decreased sorption of virus,
30
A. L. MILLS
and they concluded that the effect was due to competition between the viruses and the organic matter for sorption sites. Richardson et al. (1998) used two strains of bacteria of differing hydrophobicity and surface charge in experiments designed specifically to look at the effect of humic acid on attachment of bacteria to sand. The presence of dissolved humate added to the bacterial suspension did not affect sorption of cells to clean quartz sand at the concentrations studied (up to 1.0 mg l21) in batch experiments using 50 ml of artificial ground water (AGW), 25 g sand, and an incubation period of 3 h. A strain effect was observed, with 50% of the added cells of the relatively more hydrophilic, less negatively charged strain (S138) retained across the treatments, but only 2.3% of the added cells of the more hydrophobic, more negatively charged strain (S139) retained (Table V). The authors concluded that the humic acid did not sorb strongly to the clean sand. However, when humate was sorbed to iron-coated quartz sand at two concentrations, but was not present in the AGW, bacterial sorption to ironcoated sand decreased, and the sorption decreased with increasing humate concentration in the range examined. An effect of strain was not seen. The presence of a humate “overcoat” over a metal oxide coating modifies the high sorptive properties of the oxide with respect to bacterial cells. The results suggested that the organic matter blocks otherwise available sorption sites, and sorption of the bacterial cells to the organic coating does not make up for the loss of the sites associated with the metal oxide. Figure 7 illustrates how organic coatings can block the electric positive effect of metal oxide coatings. Table V The Effect of Na-humate on Bacterial Sorption to Sand Percentage of Cells Sorbed to Sand. Humate was Either Added to the Aqueous Phase in the Case of Clean Quartz Sand, or “Overcoated” on Iron-oxyhydroxide-coated Quartz Sand Dissolved Na-humate (mg l21) Strain
0
S138 S139
53 ^ 2 2.0 ^ 5.9
0.1
1
Clean quartz sand 61 ^ 2 22.0 ^ 2.8
37 ^ 8 68 ^ 3.9
Concentration of Na-humate (mg g21 sand) 0
9.4 £ 1023
2.1 £ 1022
Iron-coated sand S138 S139
94 ^ 0 98 ^ 0
80 ^ 3 87 ^ 1
Note. All values are mean of three replicate treatments ^ 1 standard error of the mean. Source: Richardson et al. (2000).
64 ^ 4 60 ^ 8
MICROBES AND SOIL PARTICLE SURFACES
31
This concept is consistent with reports in the literature that coatings of mineral grains remove sorptive effects due to the properties of the bulk mineral and replace them with those of the coating material. Different colloids revert to a similar electronegative surface charge after exposure to dilute solutions of humic substances (Beckett et al., 1987).
C. APPENDAGES AND CEMENTS A variety of microfibrillar structures, usually called fimbriae or pili may be found on the surfaces of various bacteria. While they are composed of proteins similar to those of flagella, these hair-like structures tend to be straighter, thinner (0.004 – 0.008 mm diameter), and shorter than flagella (Joklik et al., 1992). These structures are similar to flagella in that they are composed of self-aggregating monomers that originate from the membrane. The fimbriae are known to be associated with cellular attachment to surfaces, although they have other purposes as well. Because different fimbriae act in different ways to confer survival advantage, they are often categorized by their functional role. The beststudied roles of fimbriae are in host– pathogen relationships, and fimbriae have been grouped in categories relative to their function. The classes include adhesins, lectins, evasins, agressins, and sex pili; the latter three categories have little to do with bacterial attachment in nature. Evasins are structures that assist pathogens (such as Neisseria gonorrhoeae ) to evade the host with immune system by covering large portions of the cell with non-antigenic materials. Fimbriae serve multiple purposes. In some strains of Streptococcus pyogenes, a surface virulence factor, the M protein, serves as an adhesin allowing the organism to colonize the pharynx, and this microfibril can also prevent phagocytosis (evasin) and is leukocidal (agressin). Many adhesins are directly associated with pathogenicity; the greatest volume of literature concerning adhesins is related to the role these molecules play in conferring virulence on (especially) opportunistic pathogens such as Pseudomonas aeruginosa (Prince, 1996), E. coli (Sokurenko et al., 2001), Streptococcus pneumoniae (Jado et al., 2001), Helicobacter pylori (Ikehara et al., 2001), etc. Microfibrils and their importance are not well studied in soils, but the fact that adhesins help cells bind strongly to hydrophobic surfaces suggests these structures may play a role in bacterial attachment to soil particles. Lectins, on the other hand, are known to be important in the soil environment. Lectins are adhesins which bind to specific sugars on cell surfaces. In soil, the lectins are best studied in rhizosphere situations, particularly those of the Rhizobium – legume association. In rhizosphere habitats, many of the organisms actively attach to the root surface. Many of the organisms adhere to specific plants at specific locations. Dazzo et al. (1976) found that Rhizobium leguminosarum biovar trifolii which nodulates the roots of clover, adhered specifically to those roots. However,
32
A. L. MILLS
binding to pea roots was observed in R. leguminosarum biovar vicae which nodulates pea, and by R. leguminosarum biovar trifolii and A. tumifaciens neither of which nodulates the plant (Smit et al., 1986, 1989). Smit et al. (1987) attributed the attachment to microfibrils. Matthysse (1996) asserts, however, that no fimbrial adhesions have been found in R. leguminosarum. She indicated that there are occasionally cellulose fibrils that are involved in adhesion that have been mistaken for the proteinaceous fimbriae (Smit et al., 1987). On the other hand, Bradyrhizobium japonicum does produce fimbriae at times that coincide with the cells’ ability to bind to soybean root hairs (Vesper and Bauer, 1986). In an attempt to exclude specific symbionts by adding large number of competing but non-nodulating cells, Wall and Favelukes (1991) found that only about 10% of the inoculated alfalfa nodulator Rhizobium meliloti sorbed to the root hairs, but addition of a 1000-fold excess of R. leguminosarum did not reduce the sorption of the proper symbiont. The role of lectins in defining the specificity of the diazotroph-plant symbiosis is well accepted (Matthysse, 1996), although detailed mechanisms are not yet worked out for all bacteria –plant pairs. In some cases, the lectin is produced by the bacterium (e.g., B. japonicum; Loh et al., 1993) to allow it to recognize the plant, and in other cases the lectin is produced by the plant, presumably to allow identification of the bacterium (Lodeiro and Favelukes, 1995; Wall and Favelukes, 1991). In either case, the protein serves as an attachment point for the bacterium to bind to the hairs of the plant root. The decay of organic matter, much of which is cellulose, in soil requires the use by the cells of extracellular enzymes to initiate the depolymerization of the large molecules that make up the fibers. Many cellulolytic organisms attach directly to cellulose by means of an organelle called the cellulosome (Bayer et al., 1996). This discrete multifunctional multicomponent surface protein complex is not only responsible for the adhesion of cells to cellulosic material, but it also provides the catalytic power for conversion of cellulosic substrates to cellobiose which can be readily taken up and assimilated by the bacterial cell (Bayer et al., 1983, 1996; Lamed and Bayer, 1988a,b, 1991; Lamed et al., 1987; Strobel et al., 1995). After initial adhesion, bacteria may become more permanently moored with polymers that extend beyond the bacterial and solid surfaces. Because of their small diameter, these fibers may be able to overcome electrostatic repulsion to link the bacterium with the solid by a process called bridging (Van Loosdrecht et al., 1989). Many bacterial cells secrete a polysaccharide layer that may effectively cement them to solids and to each other. Gannon et al. (1991), however, found that the presence of polysaccharide capsules or the presence of flagella was not correlated with soil transport at a flow rate of 2.5 cm h21. This is a somewhat surprising result in the light of the aquatic literature which suggests that high fluid flows promote the secretion of bacterial slime layers and the development of active biofilms on solid surfaces (Geesey et al., 1977; Ladd et al., 1979).
MICROBES AND SOIL PARTICLE SURFACES
33
VI. EFFECTS OF SATURATED VERSUS UNSATURATED CONDITIONS As pointed out earlier, the further away from saturation a soil is, the thinner is the water film. But at the same time, additional surface for sorption is produced. A number of investigators have provided ample evidence of sorption to the gas – water interface (GWI) in unsaturated soils. The observation is supported by evidence from surface layer films found in open water bodies. Guerin (1989) found that the total heterotrophic bacteria and humic materials were concentrated in an estuary microlayer, and particulate amino acids were found to adsorb to the sea surface microlayer (Hendricks and Williams, 1985). In lakes, organic N, P, and C in the microlayer are enriched by factors of 1.6 – 45 (concentration in the microlayer divided by concentration in the subsurface), and bacteria by factors of 6.4– 10.7, compared with subsurface water (Sodergren, 1993). In the ocean near sewage outfalls, Plusquellec et al. (1991) found bacterial surface enrichment of 32– 341, and many of the surface bacteria may be attached to particles, which in turn are concentrated in the surface microlayer. Harvey and Young (1980) found that the degrees to which bacteria were concentrated into the surface microlayer were linearly dependent on surface concentration of particulate material. There may be also a phylogenetic discrimination by the microlayer. Hardy and Apts (1984) found not only that the sea surface microlayer is enriched in total microalgae, but also that it contains distinct types of algae compared with the bulk water. While adhesion of microorganisms to the surface microlayer in open water bodies has a strong enriching factor over the bulk water, it may be that the attachment of microorganisms to the GWI and soils may have an even greater enrichment factor. In open water, the microorganisms are subjected to a number of stress factors such as ultraviolet light, airborne pollutants, and rapid temperature and salinity changes that are not found or are not as severe in soil (Lion and Leckie, 1981). Gas phases (i.e., bubbles) are often present in contact with particles in porous media and may act as additional immobile adsorptive surfaces. Bacterial adsorption to the GWI may be an important and often overlooked factor affecting adhesion in transport of bacteria in porous media (Mills and Powelson, 1996; Powelson and Mills, 1996, 1998; Wan and Wilson, 1994; Wan et al., 1994). Wan and Wilson (1994) employed etched glass micromodels of porous media to observe colloidal polystyrene beads, clay particles, and bacteria concentrating at the air – water interfaces under flow conditions. In these experiments, sorption appeared to increase with particle hydrophobicity, solution ionic strength, and positive electric charge of the particles. The authors suggested that initial adsorption was due to van der Waals and electrostatic interactions, followed by essentially irreversible adsorption due to capillary forces.
34
A. L. MILLS
Thus, sorption to the GWI is both extensive and very strong; Wan and Wilson (1994) predicted that “for relatively hydrophobic strain of bacteria, even a small amount of residual gas can dramatically reduce . . . transport” of particles through the porous media. In another study, bubbles moving through a microscope field were observed to sweep glass and polymer surfaces clean of adsorbed bacteria (Pitt et al., 1993). It is not know if GWIs and particle surface – water interfaces (SWIs) interact with bacteria in a similar manner. The two interfaces are likely to react differently in several ways. The SWI generally has an electrostatic component that may dominate adsorption, unlike the essentially non-charged GWI. Although not conclusively demonstrated, it is likely that that the bacterial association with the GWI is dominated by hydrophobic effects. If so, surfactants may have particularly strong effects on adsorption at the GWI. Polystyrene beads have been used to elucidate some of the basic mechanisms of particle interaction with the GWI. Butt (1994) directly measured the force between hydrophobic particles in water and air bubbles. The particles snapped into the air bubble by a process that is irreversible based on thermodynamic considerations. The figures presented by Butt (1994) indicate that the force necessary to pull a 20 mm hydrophobic particle (contact angle 110– 1208) away from the air – water interface into water was about 1000 nN, while suspension agitation (probably Brownian motion) appeared to provide less than 2 nN; once the particle is stuck in the GWI, it is there permanently. Williams and Berg (1992) observed polystyrene beads accumulating and aggregating at the surface of a water drop. They found that the beads arrived at the GWI at a constant rate over 1 h period and that the rate increased with increasing salt content. Beads aggregated at the GWI surface at salt concentrations that were only 1% of that required for aggregation of the beads in the bulk suspension. Organic matter, especially surfactant compounds, have important effects on microbial adsorption at the GWI. Marshall (1976) suggested that hydrophobic interactions likely dominate at the GWI and therefore hydrophobic bacteria should be more attracted to the gas phase and that surfactants should reduce adsorption. Powelson et al. (1991) found that dissolved organic matter increased the transport of MS2 virus in unsaturated soil columns by nearly an order of magnitude, and offered several possible explanations for the effect. First, organic matter may reduce surface tension, thereby reducing the strength of attachment of surface-adsorbed virus proteins. Almost all organic substances found in natural waters reduce the interfacial tension (Lion and Leckie, 1981). Hunter and Liss (1981) reported that adsorption of surface-active species on lake water microlayers can reduce the surface tension by 40%. MacRitchie and Alexander (1963) found that the adsorption rate of proteins at air interfaces declined logarithmically with surface tension. Second, the increase in virus transport (i.e., the decrease in sorption) with added dissolved organic matter might be due to the effect of surface-active organic molecules on the structure of water. In pure
MICROBES AND SOIL PARTICLE SURFACES
35
water, layers thicker than 1 mm from a surface can be considered to have properties of the bulk liquid (Clifford, 1975). The presence of surface-active organic films, however, can cause water molecules next to the interface to orient into ice-like, clathrate, lattice structures, resulting in additional stabilization of the boundary layer to a depth of 50 mm (Hardy et al., 1987). It is possible that the stabilization reduces access of the partly hydrophobic virus to the GWI. Third, the presence of organic matter may inhibit virus sorption because organic matter may compete with virus particles for GWI adsorption sites (Trouwborst et al., 1974). Powelson and Gerba (1995) reviewed several studies that compared microbial concentrations in porous media after exposure to water saturated and unsaturated conditions and found that, in every case, recovery of microbes was less in unsaturated conditions. Some authors (Boyd et al., 1969; Kibbey et al., 1978) reported a correlation of loss of microbes with the degree of unsaturation, and Powelson and Gerba deemed that observation consistent with the hypothesis of strong sorption to the GWI. Later work (Powelson and Gerba, 1995; Powelson et al., 1990, 1993; Wan and Wilson, 1994) directly attributed the loss of the microbes to adsorption to the GWI. From a practical point of view, bacterial adsorption to a GWI in soil may be detrimental by slowing the transport of organisms intended to degrade a pollutant or may be beneficial by slowing the transport of pathogens to drinking water aquifers. In the vadose zone, it may be possible to “chase” pollutants spilled on the ground surface with bacteria capable of degrading the chemicals by minimizing bacterial adsorption to GWIs.
VII. SUMMARY Soil microorganisms live in a particle-surface-dominated environment, and the microbes exist there associated either with the solid – liquid interface or the gas – liquid interface (GLI). They do so because it is advantageous for them; the particles attract and concentrate energy sources and nutrients, but they also associate with surfaces because the thin water films leave little space otherwise in which they can exist. To this end, there appears to be very little contribution to soil activities by unattached bacteria. Initial attraction to surfaces is largely a combination of electrostatic and hydrophobic effects, but irreversible attachment occurs at points where opposing charge centers appear (as when mineral grains are coated with metal oxides) or when the organisms attach via microfibrils (fimbriae) or form slime layers. Microbes likely move from location to location in soils under unsaturated conditions by association with GLIs that sweep the organisms along as the GLI moves under moisture tension gradients.
36
A. L. MILLS
REFERENCES Aakra, A., Hesselsoe, M., and Bakken, L. R. (2000). Surface attachment of ammonia-oxidizing bacteria in soil. Microb. Ecol. 39, 222 –235. Absolom, D. R., Lamberti, F. V., Policova, Z., Zingg, W., van Oss, C. J., and Neumann, A. W. (1983). Surface thermodynamics of bacterial adhesion. Appl. Environ. Microbiol. 46, 90–97. Alexander, M. (1977). In “Introduction to Soil Microbiology”. Wiley, New York. Amato, M., and Ladd, J. N. (1992). Decomposition of C-14-labeled glucose and legume material in soils—properties influencing the accumulation of organic residue-C and microbial biomass-C. Soil Biol. Biochem. 24, 455– 464. Andrews, G. F. (1988). The selective adsorption of thiobacilli to dislocation sites on pyrite surfaces. Biotechnol. Bioeng. 31, 378–381. Babich, H., and Stotzky, G. (1977). Reductions in the toxicity of cadmium to microorganisms in clay minerals. Appl. Environ. Microbiol. 33, 696–705. Bayer, E. A., Kenig, R., and Lamed, R. (1983). Studies on the adherence of Clostridium thermocellum to cellulose. J. Bacteriol. 156, 818–827. Bayer, E. A., Morag, E., Shoham, Y., Tormo, J., and Lamed, R. (1996). The cellulosome: A cell surface organelle for the adhesion to and degradation of cellulose. In “Molecular and Ecological Diversity of Bacterial Adhesion”. (M. Fletcher Ed.), pp. 25– 57. Wiley–Liss, New York. Beckett, R., Zhang, J., and Giddings, J. C. (1987). Determinations of molecular weight distributions of fulvic and humic acid using field-flow fractionation. Environ. Sci. Technol. 21, 289–295. Bidle, K., and Fletcher, M. (1995). Comparison of free-living and particle-associated bacterial communities in the Chesapeake Bay by stable low-molecular-weight RNA analysis. Appl. Environ. Microbiol. 61, 944–952. Bolan, N. S., Naidu, R., Syers, J. K., and Tillman, R. W. (1999). Surface charge and solute interactions in soils. Adv. Agron. 67, 87–140. Bolster, C. H., Mills, A. L., Hornberger, G. M., and Herman, J. S. (2001). Effect of surface coatings, grain size, and ionic strength on the maximum attainable coverage of bacteria on sand surfaces. J. Contam. Hydrol. 50, 287–305. Bolt, G. H., and Bruggenwert, M. G. M. (1976). In “Soil Chemistry: A. Basic Elements”. Elsevier Scientific Publishing Company, Amsterdam. Boyd, J. W., Yoshida, T., Vereen, L. E., Cada, R. L., and Morrison, S. M. (1969). Report No. 5. In “Bacterial Response to the Soil Environment”. Colorado State University, Fort Collins, CO. Brock, T. D., and Madigan, M. T. (1991). In “Biology of Microorganisms”. Prentice-Hall, Englewood Cliffs, NJ. Butt, H. -J. (1994). A technique for measuring the force between a colloidal particle in water and a bubble. J. Colloid Interf. Sci. 166, 109–117. Clifford, J. (1975). Properties of water in capillaries and thin films. In “Water: A Comprehensive Treatise” (F. Franks, Ed.). Plenum Press, New York. Dazzo, F. B., Napoli, C. A., and Hubbell, D. H. (1976). Adsorption of bacteria to roots as related to host specificity in the Rhizobium-legume symbiosis. Appl. Environ. Microbiol. 32, 166–171. DeLong, E. F., Franks, D. G., and Alldredge, A. L. (1993). Phylogenetic diversity of aggregateattached versus free-living marine bacterial assemblages. Limnol. Oceanogr. 38, 924–934. Derjaguin, B. V., and Landau, L. (1941). Theory of stability of strongly charged lyophobic sols and of the adhesion of strongly charged particles in solutions of electrolytes. Acta Physicochim. URSS 14, 633 –662. Dziurla, M. A., Achouak, W., Lam, B. T., Heulin, T., and Berthelin, J. (1998). Enzyme-linked immunofiltration assay to estimate attachment of thiobacilli to pyrite. Appl. Environ. Microbiol. 64, 2937–2942.
MICROBES AND SOIL PARTICLE SURFACES
37
Edwards, K. J., Bond, P. L., and Banfield, J. F. (2000). Characteristics of attachment and growth of Thiobacillus caldus on sulphide minerals: A chemotactic response to sulphur minerals? Environ. Microbiol. 2, 324– 332. M. Fletcher (Ed.) (1996). In “Bacterial Adhesion: Molecular and Ecological Diversity”, pp. 1–361. Wiley–Liss, New York. Fletcher, M., and Loeb, G. I. (1979). Influence of substratum characteristics on the attachment of a marine pseudomonad to solid surfaces. Appl. Environ. Microbiol. 37, 67–72. Fontes, D. E., Mills, A. L., Hornberger, G. M., and Herman, J. S. (1991). Physical and chemical factors influencing transport of microorganisms through porous media. Appl. Environ. Microbiol. 57, 2473–2481. Gannon, J. T., Manlial, V. B., and Alexander, M. (1991). Relationship between cell surface properties and transport of bacteria through soil. Appl. Environ. Microbiol. 57, 190–193. Gardner, W. R. (1956). Calculation of capillary conductivity from pressure plate outflow data. Soil Sci. Soc. Am. Proc. 20, 317–320. Geesey, G. G., Richardson, W. T., Yoemans, H. G., Irvin, R. T., and Costerton, J. W. (1977). Can. J. Microbiol. 23, 1733–1736. Gehrke, T., Telegdi, J., Thierry, D., and Sand, W. (1998). Importance of extracellular polymeric substances from Thiobacillus ferrooxidans for bioleaching. Appl. Environ. Microbiol. 64, 2743–2747. Gerzabek, M. H., Haberhauer, G., Kandeler, E., Sessitsch, A., and Kirchmann, H. (2002). Third Symposium ISMOM. In “Response of Organic Matter Pools and Enzyme Activities in Particlesize Fractions to Organic Amendments in Long-term Field Experiments”. Naples, Capri, Italy (in press). vanGestel, M., Merckx, R., and Vlassak, K. (1996). Spatial distribution of microbial biomass in microaggregates of a silty-loam soil and the relation with the resistance of microorganisms to soil drying. Soil Biol. Biochem. 28, 503 –510. Gordon, A. S., and Millero, F. J. (1984). Electrolyte effects on attachment of an estuarine bacterium. Appl. Environ. Microbiol. 47, 495– 499. Green, R. E., Hanks, R. J., and Larson, W. E. (1964). Estimates of field infiltration by numerical solution of the moisture flow equation. Soil Sci. Soc. Am. Proc. 28, 15–19. Griffith, P. C., and Fletcher, M. (1991). Hydrolysis of protein and model dipeptide substrates by attached and nonattached marine Pseudomonas spp. strained NCIBM 2021. Appl. Environ. Microbiol. 57, 2186–2191. Gross, M. J., Albinger, O., Jewett, D. G., Logan, B. E., Bales, R. C., and Arnold, R. G. (1995). Measurement of bacterial collision efficiencies in porous media. Water Res. 29, 1151–1158. Guerin, W. C. (1989). Phenanthrene degradation by estuarine surface microlayer and bulk water microbial populations. Microb. Ecol. 17, 89– 104. Guerin, W. F., and Boyd, S. A. (1997). Bioavailability of naphthalene associated with natural and synthetic sorbents. Water Res. 31, 1504–1512. Harden, V. P., and Harris, J. O. (1953). The isoelectric point of bacterial cells. J. Bacteriol. 65, 198 –202. Hardy, J. T., and Apts, C. W. (1984). The sea-surface microlayer: Phytoneuston productivity and effects of atmospheric particulate matter. Mar. Biol. 82, 293– 300. Hardy, J. T., Crecelius, E. A., Antrim, L. D., Broadhurst, V. L., Apts, C. W., Gurtisen, J. M., and Fortman, T. J. (1987). The sea-surface microlayer of Puget Sound. Part II. Concentrations of contaminants and relation to toxicity. Mar. Environ. Res. 23, 251–271. Harris, R. F. (1981). Effect of water potential on microbial growth and activity. In “Water Potential Relations in Soil Microbiology”. (J. F. Parr, W. R. Gardner and L. F. Elliot Eds.), pp. 23–95. Soil Science Society of America, Madison, WI. Harvey, R. W., and Young, L. Y. (1980). Enrichment and association of bacteria and particulate in salt marsh surface water. Appl. Environ. Microbiol. 39, 894 –899.
38
A. L. MILLS
Hassink, J., Bouwman, L. A., Zwart, K. B., Bloem, J., and Brussaard, L. (1993). Relationships between soil texture, physical protection of organic-matter, soil biota, and C-mineralization and N-mineralization in grassland soils. Geoderma 57, 105–128. Hattori, T. (1973). In “Microbial Life in the Soil: An Introduction”. Dekker, New York. Hendricks, S. M., and Williams, P. M. (1985). Dissolved in particulate amino acids and carbohydrates in the sea surface microlayer. Mar. Chem. 17, 141 –163. Hendricks, D. W., Post, F. J., and Khairnar, D. R. (1979). Adsorption of bacteria on soils: Experiments, thermodynamic rationale, and application. Water Air Soil Pollut. 12, 219–232. Heukelekian, H., and Heller, A. (1940). Relation between food concentration and surface for bacterial growth. J. Bacteriol. 40, 547– 558. Holmes, J. W., Greacen, E. L., and Gurr, G. C. (1960). In Transactions of the Seventh International Congress of Soil Science vol. 1, 188–194. Hopkinson, C. S., Sherr, B., and Wiebe, W. J. (1989). Size fractionated metabolism of coastal microbial plankton. Mar. Ecol.: Prog. Ser. 51, 155 –166. Hunter, R. J. (1981). In “The Zeta Potential in Colloid Science”. Academic Press, London, UK. Hunter, K. A., and Liss, P. S. (1981). Organic sea surface films. In “Marine Organic Chemistry”. (E. K. Duursma and R. Dawson, Eds.), pp. 259–298. Elsevier, Amsterdam. Huysman, F., and Verstraete, W. (1993). Water-facilitated transport of bacteria in unsaturated soil columns—Influence of inoculation and irrigation methods. Soil Biol. Biochem. 25, 91–97. Ikehara, Y., Nishihara, S., Yasutomi, H., Kitamura, T., Matsuo, K., Shimizu, N., Inada, K. I., Kodera, Y., Yamamura, Y., Narimatsu, H., Hamajima, N., and Tatematsu, M. (2001). Polymorphisms of two fucosyltransferase genes (Lewis and Secretor genes) involving type I Lewis antigens are associated with the presence of anti-Helicobacter pylori IgG antibody. Cancer Epidemiol. Biomarkers Prevention 10, 971 –977. Jado, I., Fenoll, A., Casal, J., and Perez, A. (2001). Identification of the psaA gene, coding for pneumococcal surface adhesin A, in viridans group streptococci other than Streptococcus pneumoniae. Clin. Diag. Lab. Immunol. 8, 895–898. Jannasch, H. W. (1973). Bacterial content of particulate matter in offshore surface waters. Limnol. Oceanogr. 18, 340 –342. Jiang, C. L., Wang, X. H., and Parekh, B. K. (2000). Effect of sodium oleate on inhibiting pyrite oxidation. Inter. J. Min. Process. 58, 305–318. Jocteur Monrozier, L., Ladd, J. N., Fitzpatrick, A. W., Foster, R. C., and Rauspach, M. (1991). Components and microbial biomass content of size fractions in soils of contrasting aggregation. Geoderma 49, 37–62. Joklik, W. K., Willett, H. P., Amos, D. B., and Wilfert, C. M. (1992). In “Zinsser Microbiology” (20th ed). Appleton & Lange, Norwalk, CT. Kandeler, E., Stemmer, M., and Klimanek, E. M. (1999). Response of soil microbial biomass, urease and xylanase within particle size fractions to long-term soil management. Soil Biol. Biochem. 31, 261–273. Kandeler, E., Tscherko, D., Bruce, K. D., Stemmer, M., Hobbs, P. J., Bardgett, R. D., and Amelung, W. (2000). Structure and function of the soil microbial community in microhabitats of a heavy metal polluted soil. Biol. Fert. Soils 32, 390 –400. Kandeler, E., Tscherko, D., Stemmer, M., Schwarz, S., and Gerzabek, M. H. (2001). Organic matter and soil microorganisms—Investigations from the micro- to the macro-scale. Bodenkultur 52, 117–131. Kaper, J. B., Mills, A. L., and Colwell, R. R. (1978). Evaluation of the accuracy and precision of enumerating aerobic heterotrophs in water samples by the spread plate method. Appl. Environ. Microbiol. 35, 756–761. Kemper, W. D., and Rollins, J. B. (1966). Osmotic efficiency coefficients across compacted clays. Soil Sci. Soc. Am. Proc. 30, 529 –534.
MICROBES AND SOIL PARTICLE SURFACES
39
Kibbey, H. J., Hagedorn, C., and McCoy, E. L. (1978). Use of fecal streptococci as indicators of pollution in soil. Appl. Environ. Microbiol. 35, 711 –717. Kirchmann, H., and Gerzabek, M. H. (1999). Relationship between soil organic matter in micropores in a long-term experiment at Ultuna, Sweden. J. Plant Nutr. Soil Sci. 162, 493– 498. Knapp, E. P., Herman, J. S., Hornberger, G. M., and Mills, A. L. (1998). The effect of distribution of iron-oxyhydroxide grain coatings on the transport of bacterial cells in porous media. Environ. Geol. 33, 243– 248. Knickerbocker, C., Nordstrom, D. K., and Southam, G. (2000). The role of “blebbing” in overcoming the hydrophobic barrier during biooxidation of elemental sulfur by Thiobacillus thiooxidans. Chem. Geol. 169, 425 –433. Ladd, T. I., Costerton, J. W., and Geesey, G. G. (1979). Determination of the heterotrophic activity of epilithic microbial populations. In “Native Aquatic Bacteria: Enumeration Activity and Ecology. ASTM STP 695”. (J. W. Costerton and R. R. Colwell Eds.), pp. 180– 195. American Society for Testing and Materials, Philadelphia. Lamed, R., and Bayer, E. A. (1988a). The cellulosome concept: Exocellular/extracellular enzyme reactor centers for efficient binding and cellulolysis. In “Biochemistry and Genetics of Cellulose Degradation”. (J. -P. Aubert, P. Be´guin, and J. Millet, Eds.), pp. 101–116. Academic Press, London, UK. Lamed, R., and Bayer, E. A. (1988b). The cellulosome of Clostridium thermocellum. Adv. Appl. Microbiol. 33, 1–46. Lamed, R., and Bayer, E. A. (1991). The cellulosome concept—A decade later!. In “Genetics, Biochemistry and Ecology of Lignocellulose Degradation.” (K. Shimada, S. Hoshino, K. Ohmiya, K. Sakka, Y. Kobayashi, and S. Karita, Eds.), vols 1 –12. Uni Publishers, Tokyo. Lamed, R., Naimark, J., Morgenstern, E., and Bayer, E. A. (1987). Specialized cell surface structures in cellulolytic bacteria. J. Bacteriol. 169, 3792–3800. Lance, J. C., and Gerba, C. P. (1984). Effect of ionic composition of suspending solution on virus sorption by a soil column. Appl. Environ. Microbiol. 47, 484–488. Lion, L. W., and Leckie, J. O. (1981a). The biogeochemistry of the air–sea interface. Ann. Rev. Earth Planet. Sci. 9, 449 –486. Lodeiro, A. R., and Favelukes, G. (1995). Tenth International Congress of Nitrogen Fixation. In “Enhanced Adsorption to Soybean Roots, Infectivity, and Competition for Nodulation of Bradyrhizobium japonicum Pretreated with Soybean Lectin”. Loder, T. C., and Liss, P. S. (1985). Control by organic coatings of the surface charge of estuarine suspended particles. Limnol. Oceanogr. 30, 484– 488. Loh, J. T., Ho, S. -C., de Feuter, A. W., Wang, J. L., and Schindler, M. (1993). Carbohydrate binding activities of Bradyrhizobium japonicum: Unipolar location of the lectin Bj38 on the bacterial cell surface. Proc. Natl Acad. Sci. USA 90, 3033–3037. MacRitchie, F., and Alexander, A. E. (1963). Kinetics of adsorption of proteins at interfaces. J. Colloid Sci. 18, 453 –469. Marshall, K. C. (1976). In “Interfaces in Microbial Ecology”. Harvard University Press, Cambridge, MA. Marshall, K. C. (1980). Adsorption of microorganisms to soils and sediments. In “Adsorption of Microorganisms to Surfaces”. (G. Bitton and K. C. Marshall Eds.) Wiley, New York. Marshall, K. C. (1990). Microbial processes occurring at surfaces. In “Surface and Colloid Chemistry on Matural Waters and Water Treatment”. (R. Beckett Ed.), pp. 21 –26. Plenum Press, New York. Marshall, K. C. (1996). Adhesion as a strategy for access to nutrients. In “Bacterial Adhesion: Molecular and Ecological Diversity”. (M. Fletcher Ed.), pp. 59– 87. Wiley–Liss, New York. Marshall, K. C., Stout, R., and Mitchell, R. (1971). Mechanism of the initial events in the sorption of marine bacteria to surfaces. J. Gen. Microbiol. 68, 337 –348. Marszalek, D. S., Gerchakov, S. M., and Udey, L. R. (1979). Influence of substrate composition on microfouling. Appl. Environ. Microbiol. 38, 987–995.
40
A. L. MILLS
Matthysse, A. G. (1996). Adhesion in the rhizosphere. In “Molecular and Ecological Diversity of Bacterial Adhesion”. (M. Fletcher Ed.), pp. 129 –153. Wiley–Liss, New York. McDowell-Boyer, L. M., Hunt, J. R., and Sitar, N. (1986). Particle transport through porous media. Water Resour. Res. 22, 1901–1921. Metting, F. B. (1985). Soil microbiology and biotechnology. In “Biotechnology: Applications and Research”. (P. A. Cherimisinoff and R. P. Ouellete Eds.), pp. 196 –214. Technomic Publishers, Lancaster, PA. Metting, F. B. (1993). Structure and physiological ecology of soil microbial communities. In “Soil Microbial Ecology: Applications in Agricultural and Environmental Management”. (F. B. Metting Ed.), pp. 3–25. Dekker, New York. Mills, A. L., and Eaton, W. D. (1984). Biodegradation of bromobenzene in simulated groundwater conditions. In “Biodegradation 6”. (G. C. Llewellyn and C. E. O’Rear Eds.), pp. 9–13. CAB International, Slough, UK. Mills, A. L., and Maubrey, R. (1981). The effect of mineral composition on bacterial attachment to submerged rock surfaces. Microb. Ecol. 7, 315–322. Mills, A. L., and Powelson, D. K. (1996). Bacterial interactions with surfaces in soil. In “Molecular and Ecological Diversity of Bacterial Adhesion”. (M. Fletcher Ed.), pp. 25 –57. Wiley–Liss, New York. Mills, A. L., Herman, J. S., Hornberger, G. M., and deJesus, T. H. (1994). Effect of solution ionic strength on mineral grains on the sorption of bacterial cells to quartz sand. Appl. Environ. Microbiol. 60, 3600–3606. Mozes, N., Marchal, F., Hermesse, M. P., Van Haecht, J. L., Reuliaux, L., and Leonard, A. J. (1987). Immobilization of microorganisms by adhesion: Interplay of electrostatic and nonelectrostatic interactions. Biotechnol. Bioeng. 30, 439–450. Murray, J. P., and Parks, G. A. (1980). Poliovirus adsorption on oxide surfaces. In Particulates in Water. (M. C. Kavenaugh and J. O. Leckie Eds.), vol. 89, pp. 97 –133. American Chemical Society, Washington, DC. Neumann, A. W., Hum, O. S., Francis, D. W., Zingg, W., and van Oss, C. J. (1980). Kinetic and thermodynamic aspects of platelet adhesion from suspensions to various substrates. J. Biomed. Mater. Res. 14, 499. Noble, P. A., Bidle, K. D., and Fletcher, M. (1997). Natural microbial community compositions compared by a back-propagating neural network and cluster analysis of 5S rRNA. Appl. Environ. Microbiol. 63, 1762–1770. Nyavor, K., Egiebor, N. O., and Fedorak, P. M. (1996). Suppression of microbial pyrite oxidation by fatty acid amine treatment. Sci. Total Environ. 182, 75– 83. Ogram, A. V., Jessup, R. E., Ou, L. T., and Rao, P. S. C. (1985). Effects of sorption on biological degradation rates of (2,4-dichlorophenoxy) acetic acid in soils. Appl. Environ. Microbiol. 49, 582–587. Paul, E. A., and Clark, F. E. (1989). In “Soil Microbiology and Biochemistry”. Academic Press, London, UK. Pitt, W. G., McBride, M. O., Barton, A. J., and Sagers, R. D. (1993). Air–water interface displaces adsorb bacteria. Biomaterials 14, 605 –608. Plette, A. C. C., Vanriemsdijk, W. H., Benedetti, M. F., and Vanderwal, A. (1995). Ph dependent charging behavior of isolated cell-walls of a gram-positive soil bacterium. J. Colloid Interf. Sci. 173, 354 –363. Plusquellec, A., Beucher, M., LeLay, C., LeGal, Y., and Cleret, J. J. (1991). Quantitative and qualitative bacteriology of the marine water surface microlayer in a sewage-polluted area. Mar. Environ. Res. 31, 227 –239. Posner, A. M. (1964). Titration curves of humic acid. In Eighth International Congress of Soil Science. vol. 3, pp. 161 –174. Publishing House of the Academy of the Socialist Republic of Romania, Bucharest.
MICROBES AND SOIL PARTICLE SURFACES
41
Powell, S. J., and Prosser, J. I. (1991). Protection of nitrosomonas-Europaea colonizing clay-minerals from inhibition by nitrapyrin. J. Gen. Microbiol. 137, 1923–1929. Powelson, D. K., and Gerba, C. P. (1995). Fate and transport of microorganisms in the vadose zone. In “Handbook of Vadose Zone Characterization and Monitoring”. (L. G. Wilson, L. G. Everett and S. J. Cullen Eds.), Lewis Publishers, Boca Raton, FL. Powelson, D. K., and Mills, A. L. (1996). Bacterial enrichment at the gas– water interface. Appl. Environ. Microbiol. 62, 2593– 2597. Powelson, D. K., and Mills, A. L. (1998). Bacterial transport through sand with variable water saturation and surfactant concentration. Soil Sci.. Powelson, D. K., Simpson, J. R., and Gerba, C. P. (1990). Virus transport and survival in saturated and unsaturated flow through soil columns. J. Environ. Qual. 19, 396 –401. Powelson, D. K., Simpson, J. R., and Gerba, C. P. (1991). Effects of organic matter on virus transport in unsaturated flow. Appl. Environ. Microbiol. 57, 2192–2196. Powelson, D. K., Gerba, C. P., and Yahya, M. T. (1993). Virus transport and removal in wastewater during aquifer recharge. Water Res. 27, 583–590. Prince, A. (1996). Pseudomonas aeruginosa: Versatile attachment mechanisms. In “Molecular and Ecological Diversity of Bacterial Adhesion”. (M. Fletcher Ed.), pp. 183–199. Wiley–Liss, New York. Pringle, J. H., and Fletcher, M. (1986). Influence of substratum hydration and adsorbed macromolecules on bacterial attachment to surfaces. Appl. Environ. Microbiol. 51, 1321–1325. Richardson, A. J., Stewart, C. S., and Gooday, G. W. (1998). Attachment to cellulose and maturation of attached thalli in the anaerobic rumen fungus Neocallimastix frontalis strain RE1. Mycol. Res. 102, 1119–1125. Rosenberg, M., and Doyle, R. J. (1990). Microbial cell surface hydrophobicity: History, measurement, significance. In “Microbial Cell Surface Hydrophobicity”. (R. J. Doyle and M. Rosenberg Eds.), pp. 1–37. American Society for Microbiology, Washington, DC. Sanhueza, A., Ferrer, I. J., Vargas, T., Amils, A., and Sanchez, C. (1999). Attachment of Thiobacillus ferrooxidans on synthetic pyrite of varying structural and electronic properties. Hydrometallurgy 51, 115–129. Scholl, M. A., and Harvey, R. W. (1992). Laboratory investigations on the role of sediment surface and groundwater chemistry in transport of bacteria through a contaminated sandy aquifer. Environ. Sci. Technol. 26, 1410– 1417. Scholl, M. A., Mills, A. L., Herman, J. S., and Hornberger, G. M. (1990). The influence of mineralogy and solution chemistry on attachment of bacteria to representative aquifer materials. J. Contam. Hydrol. 6, 331– 336. Sessitsch, A., Weilharter, A., Gerzabek, M. H., Kirchmann, H., and Kandeler, E. (2001). Microbial population structures in soil particle size fractions of a long-term fertilizer field experiment. Appl. Environ. Microbiol. 67, 4215– 4224. Sharma, M. M., Chang, Y. I., and Yen, T. F. (1985). Reversible and irreversible surface charge modification of bacteria for facilitating transport through porous media. Colloids Surf. 16, 193 –206. Smit, G., Kijne, J. W., and Lugtenberg, B. J. J. (1986). Correlation between extracellular fibrils and attachment of Rhizobium leguminosarum to pea root hair tips. J. Bacteriol. 168, 821 –827. Smit, G., Kijne, J. W., and Lugtenberg, B. J. J. (1987). Involvement of both cellulose fibrils and a Ca2þ-dependent adhesin in the attachment of Rhizobium leguminosarum to pea root hair tips. J. Bacteriol. 169, 4294–4301. Smit, G., Logman, T. J. J., Boerrigter, M. E. T., Kijne, J. W., and Lugtenberg, B. J. J. (1989). Purification and partial characterization of the Rhizobium leguminosarum biovar vicae Ca2þ dependent adhesin, which mediates the first step in attachment of the cells of the family Rhizobiacea to plant root hair tips. J. Bacteriol. 171.
42
A. L. MILLS
Smith, J. L., Papendick, R. I., Bezdicek, D. F., and Lynch, J. M. (1993). Soil organic matter dynamics and crop residue management. In “Soil Microbial Ecology: Applications in Agricultural and Environmental Management”. (F. B. Metting Ed.), pp. 65 –94. Dekker, New York. Sodergren, A. (1993). Role of aquatic surface microlayer in the dynamics of nutrients and organic compounds in lakes, with implications for their ecotones. Hydrobiologia 251, 217 –225. Sokurenko, E. V., Schembri, M. A., Trintchina, E., Kjaergaard, K., Hasty, D. L., and Klemm, P. (2001). Valency conversion in the type 1 fimbrial adhesin of Escherichia coli. Mol. Microbiol. 41, 675–686. Sposito, G. (1989). In “The Chemistry of Soils”. Oxford University Press, New York. Stark, W. H., Stadler, J., and McCoy, E. (1938). Some factors affecting the bacterial population of freshwater lakes. J. Bacteriol. 36, 653– 654. Stemmer, M., Gerzabek, M. H., and Kandeler, E. (1998). Organic matter and enzyme activity in particle-size fractions of soils obtained after low-energy sonication. Soil Biol. Biochem. 30, 9–17. Stemmer, M., Gerzabek, M. H., and Kandeler, E. (1999). Invertase and xylanase activity of bulk soil and particle-size fractions during maize straw decomposition. Soil Biol. Biochem. 31, 9–18. Stenstro¨m, T. A. (1989). Bacterial hydrophobicity, an overall parameter for the measurement of adhesion potential to soil particles. Appl. Environ. Microbiol. 55, 142 –147. Stevenson, F. J. (1982). In “Humus Chemistry”. Wiley, New York. Stotzky, G. (1979). In “Aquatic Microbial Ecology”. (R. R. Colwell and J. Foster Eds.), pp. 460. University of Maryland, Clearwater Beach, FL. Strobel, H. J., Caldwell, F. C., and Dawson, K. A. (1995). Carbohydrate transport by the anaerobic thermophile Clostridium thermocellum LQRI. Appl. Environ. Microbiol. 61, 4012–4015. Stumm, W., and Morgan, J. J. (1996). In “Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters” (3rd ed). Wiley, New York. Tate, R. L. (1987). In “Soil Organic Matter”. Wiley, New York. Thurman, E. M. (1985). In “Organic Geochemistry of Natural Waters”. Nijhoff –Junk, Dordrecht, The Netherlands. Tipping, E. (1981). The adsorption of aquatic humic substances by iron oxides. Geochim. Cosmochim. Acta 45, 191–199. Tipping, E., and Cooke, D. (1982). The effects of adsorbed humic substances on the surface charge of goethite (a-FeOOH) in freshwaters. Geochim. Cosmochim. Acta 46, 75–80. Trouwborst, T., Kuyper, S., DeJong, J. C., and Plantinga, A. D. (1974). Inactivation of some bacterial and animal viruses by exposure to liquid–air interfaces. J. Gen. Virol. 24, 155–165. Van Loosdrecht, M. C. M., Lyklema, J., Norde, W., Schraa, G., and Zehnder, A. J. B. (1987a). Electrophoretic mobility and hydrophobicity as a measure to predict the initial steps of bacterial adhesion. Appl. Environ. Microbiol. 53, 1898–1901. Van Loosdrecht, M. C. M., Lyklema, J., Norde, W., Schraa, G., and Zehnder, A. J. B. (1987b). The role of bacterial cell wall hydrophobicity in adhesion. Appl. Environ. Microbiol. 53, 1893–1897. Van Loosdrecht, M. C. M., Lyklema, J., Norde, W., and Zehnder, A. J. B. (1989). Bacterial adhesion: A physicochemical approach. Microb. Ecol. 17, 1 –15. Van Loosdrecht, M. C. M., Lyklema, J., Norde, W., and Zehnder, A. J. B. (1990). Influence of interfaces on microbial activity. Microbiol. Rev. 54, 75–87. Verwey, E. J. W., and Overbeek, J. T. G. (1948). In “Theory of the Solubility of Lyophobic Colloids”. Elsevier, Amsterdam. Vesper, S. J., and Bauer, W. D. (1986). Role of pili (fimbriae) in attachment of Bradyrhizobium japonicum to soybean roots. Appl. Environ. Microbiol. 52. Waksman, S., and Carey, C. L. (1935). Decomposition of organic matter in seawater by bacteria. 1. Bacterial multiplication in stored water. J. Bacteriol. 29, 531–545.
MICROBES AND SOIL PARTICLE SURFACES
43
Wall, L. G., and Favelukes, G. (1991). Early recognition in the Rhizobium meliloti—alfalfa symbiosis: Root exudate factor stimulates root adsorption of homologous rhizobia. J. Bacteriol. 173, 3492–3499. Wan, J., and Wilson, J. L. (1994). Visualization of the role of the gas– water interface on the fate and transport of colloids in porous media. Water Resour. Res. 30, 11–23. Wan, J., Wilson, J. L., and Kieft, T. L. (1994). Influence of the gas–water interface on transport of microorganisms through unsaturated porous media. Appl. Environ. Microbiol. 60, 509–516. Weber, J. B., and Coble, H. D. (1968). Microbial decomposition of diquat adsorbed on montmorillonite and kaolinite clays. J. Agric. Food Chem. 16, 475 –478. Weber, W. J., McGinley, P. M., and Katz, L. E. (1991). Sorption phenomena in subsurface systems: Concepts, models, and effects on contaminant fate and transport. Water Res. 25, 499 –528. Williams, D. F., and Berg, J. C. (1992). The aggregation of colloidal particles at the air–water interface. J. Colloid Interf. Sci. 152. ZoBell, C. E. (1943). The effect of solid surfaces upon bacterial activity. J. Bacteriol. 46, 39–56. ZoBell, C. E., and Anderson, D. Q. (1936). Observations on the multiplication of bacteria in different volumes of stored seawater and influence of oxygen tension and solid surfaces. Biol. Bull. Woods Hole 71, 324–342.
Preface Volume 78 contains six cutting-edge reviews in the crop and soil sciences. Chapter 1 discusses aspects of microbial-soil particle interfacial reactions including the benefits to microbes of living on surfaces, the importance of attached microbes in soil, the diversity of modes of attachment, and the effects of saturated vs. unsaturated conditions. Chapter 2 is a comprehensive review on the history and success of the public-private project on germplasm enhancement of maize (GEM). Aspects of the topic that are covered include: GEM’s development and administration, breeding activities and results, value added trait analyses and results, and public cooperator research and results. Chapter 3 deals with microbiological and biochemical indices for assessing quality of acid soils. Discussions on acid soil distribution worldwide, quality characteristics of acid soils, measurement of microbiological and biochemical parameters in acidic soils and of acidic soil quality, and development of acid soil quality indexing systems are included. In Chapter 4 the authors discuss polyploidy and the evolutionary history of cotton. Topics include: taxonomic, cytogenetic, and phylogenetic framework; speciation mechanisms; origin of the allopolyploids; polyploid evolution; and ecological consequences of polyploidization. Chapter 5 deals with the development of acidic subsurface layers of soil under different management systems. Discussions on the occurrence of acidic subsurface layers and their impacts on agricultural production; the rate and cause of development of the layers; and environmental and management factors that affect the pH differences in surface and subsurface layers are included. Chapter 6 provides a timely discussion on soil acidification and liming interactions with nutrient and heavy metal transformation and bioavailability. The processes of acid generation in soils; effect of soil acidity on nutrient and heavy metal transformations in soils; and lime, nutrient, and heavy metal interactions are discussed. I am grateful for the authors’ fine contributions. DONALD L. SPARKS
xi
Contributors Numbers in parentheses indicate the pages on which the authors’ contributions begin.
D. C. ADRIANO (215), University of Georgia, Savannah River Ecology Laboratory, Drawer E, Aiken, SC 29802, USA V. C. BALIGAR (89), USDA-ARS, Alternate Crops and System Research Laboratory, Beltsville Agricultural Research Center, Beltsville, MD 20704, USA A. S. BLACK (187), School of Agriculture, Charles Sturt University, Locked Bag 588, Wagga Wagga, NSW 2678, Australia NANTHI S. BOLAN (215), Fertilizer and Lime Research Centre, Massey University, Palmerston North, New Zealand D. V. CALVERT (89), University of Florida, Institute of Food and Agricultural Sciences, Indian River Research and Education Center, 2199 S. Rock Road, Fort Pierce, FL 34945, USA MARK K. CONYERS (187), Agricultural Institute, NSW Agriculture, PMB, Wagga Wagga, NSW 2650, Australia RICHARD C. CRONN (139), Pacific Northwest Research Station, USDA Forest Service, 3200 SW Jefferson Way, Corvallis, OR 97331, USA DENIS CURTIN (215), Crop and Food, Christchurch, New Zealand ZHENLI HE (89), University of Florida, Institute of Food and Agricultural Sciences, Indian River Research and Education Center, 2199 S. Rock Road, Fort Pierce, FL 34945, USA AARON L. MILLS (1), Laboratory of Microbial Ecology, Department of Environmental Sciences, University of Virginia, Charlottesville, VA 229044123, USA KERYN I. PAUL (187), CSIRO Forestry and Forest Products, PO Box E4008, Kingston ACT 2604, Australia LINDA M. POLLAK (45), USDA-ARS Corn Insect and Crop Genetics Research Unit, Department of Agronomy, Iowa State University, Ames, IA 50011, USA JONATHAN F. WENDEL (139), Department of Botany, Iowa State University, Ames, IA 50011, USA X. E. YANG (89), Department of Resource Science, College of Natural Resource and Environmental Sciences, Zhejiang University, Huajiachi Campus, 310029 Hangzhou, China
ix
THE HISTORY AND SUCCESS OF THE PUBLIC – PRIVATE PROJECT ON GERMPLASM ENHANCEMENT OF MAIZE (GEM) Linda M. Pollak USDA-ARS Corn Insect and Crop Genetics Research Unit, Department of Agronomy, Iowa State University, Ames, Iowa 50011, USA
I. Introduction A. The Need for Maize Enhancement B. The Latin American Maize Project II. GEM’S Development A. Public/private US Agricultural Research B. Public and Private Interaction to Organize GEM C. GEM’s Objective III. GEM’s Administration A. Organization B. Funding Mechanism IV. Breeding Activities and Results V. Value-added Trait Analyses and Results A. The Need for Improving Value-added Traits B. The Value-added Trait Research Component of GEM C. Grain Composition D. Starch Quality E. Oil Quality VI. Public Cooperator Research and Results A. European Corn Borer Resistance B. Characterizing LAMP Accessions and their Crosses for Wet-milling Efficiency C. Other Significant Public Cooperator Findings VII. Conclusions A. Factors Responsible for GEM’s Successful Public/Private Collaboration B. Extending GEM’s Concept C. GEM’s Future References
45 Advances in Agronomy, Volume 78 Copyright q 2003 by Academic Press. All rights of reproduction in any form reserved 0065-2113/02$35.00
46
L. M. POLLAK
I. INTRODUCTION The Latin American Maize Project (LAMP) was the first coordinated international project for evaluating a major world crop (Salhuana et al., 1991; Pollak, 1993; Eberhart et al., 1995; Salhuana and Sevilla, 1995; Salhuana et al., 1998). In LAMP, 12 countries cooperated to evaluate their native germplasm accessions. Before the LAMP accessions could be used as sources of breeding material in the USA, some process had to be established to enhance them such that they could enter commercial corn breeding channels. Therefore, a coordinated and cooperative effort, the germplasm enhancement of maize (GEM) project was organized among public and private sectors. The project provides the corn industry early breeding lines by using germplasm enhancement to improve and adapt useful exotic germplasm. The ultimate objective is to improve and broaden the germplasm base of corn hybrids grown by American farmers. Traits targeted for improvement are agronomic productivity, disease and insect resistance, and value-added characteristics. The project has grown to include international cooperators, both public and private.
A. THE NEED FOR MAIZE ENHANCEMENT Corn is the USA’s major crop (USDA –ERS, 1995) where over 30 million hectares are planted each year. USA is also the world leader in corn production, producing over 41% of the world’s production in 1989– 1990 (USDA, 1990). Corn is extremely important to the US economy due to the amount produced, its value to industry, and its export value. As a raw material, corn added over $17 billion to the economy in 1989. About 20% of the production is exported, providing a positive contribution to the nation’s trade balance (USDA, 1996). Approximately 17% is industrially refined. An additional $1.4 billion in refined products is exported. Through feeding livestock that is processed into meat and dairy products, corn affects nearly everyone in the American society. It has been estimated that 90% of domestic corn grain is used as food through the feeding of livestock (Hodge, 1982). Although corn is so valuable to the US economy, less than 1% of the US germplasm base consists of exotic germplasm (Goodman, 1985) leading to concerns about corn’s genetic vulnerability. Since the early 1960s, there have been frequent and urgent warnings about corn’s genetic vulnerability and the potential of exotic germplasm to decrease this vulnerability (National Academy of Science, 1972; Goodman, 1990; Walsh, 1981; Wilkes, 1989; Eberhart, 1971; Longquist, 1974; Brown, 1975; Crossa and Gardner, 1987). These concerns have been reinforced by studies documenting a reduction in genetic variability among lines and hybrids (Smith, 1988; Darrah and Zuber, 1985). History shows that problems can occur when the genetic base
GERMPLASM ENHANCEMENT OF MAIZE
47
of a crop becomes too narrow and changes in the environment, such as new pathogens, new insect pests, or unusual environmental stresses adversely affect the crop’s productivity. For example, the southern corn leaf blight (incited by Bipolaris maydis (Nisikado) Shoemaker, race T) epidemic in corn showed the devastating nature of such an event (National Academy of Science, 1972). In this case, the epidemic was due to the widespread genetic uniformity of the crop which sped development of the disease to epidemic proportions. Genetic variability is essential in plant breeding programs (Michelini and Hallauer, 1993), but improvements in crops by plant breeding are usually followed by decreased genetic diversity particularly in materials that reach commercial production. Thus, farmers’ hybrids are increasingly genetically vulnerable and farmers face increased economic risk. Corn Belt Dent is one of the most productive races of corn in the world, and has adequate genetic variability for resistance to most common pests. For this reason, US corn breeders have concentrated their effort on this race, which represents about 2% of the corn germplasm available in the world (Brown, 1975). In contrast to many other plant breeders, corn breeders have continued to focus on short-term breeding goals, largely because of the predominance of the private sector in corn breeding and its need for short-term results. This pattern resulted in the development of a very narrow genetic base of corn produced on the farm, with many companies selling closely related hybrids (Smith, 1988). This may lead to a yield plateau, greatly increase vulnerability to pests, and make it difficult meet new market demands. It is prudent to develop alternate breeding populations from exotic sources. Geadelmann (1984) suggested that incorporation of exotic strains into adapted germplasm would increase the available genetic variability and give rise to additional heterotic vigor, thereby lessening the chances for a yield plateau. In the US Corn Belt, exotic germplasm is usually considered to include unadapted domestic populations, and foreign temperate, tropical, and semi-tropical populations (Stuber, 1986). It is evident that conventional corn breeding using adapted materials has been extremely successful in the USA. Forty years ago average yields in the Corn Belt were 38 bu/acre (Johnson, 1991), but in 1987 the average yield was nearly 120 bu/acre with yields over 240 bu/acre reported (USDA, 1988). Russell (1986) summarized 15 studies indicating that nearly 60% of the recent yield gain was due to genetics. Although breeders are still making genetic gain for yield in adapted materials, there lately has been a decline in the rate of growth of cereal yield, possibly indicating a sign that breeding efforts are reaching the point of diminishing returns (Sehgal, 2000). The most recent survey of germplasm sources for the maize crop (Darrah and Zuber, 1985) found that 88% of 1984 US maize seed produced for 1985 planting included germplasm derived from one variety of the Corn Belt Dent race, the Reid Yellow Dent. Most current hybrids are derived from a few inbred lines of Iowa Stiff Stalk Synthetic origin crossed with a few lines of primarily Lancaster Sure Crop origin. Smith (1988) used
48
L. M. POLLAK
biochemical data to show that US corn breeding and production was dependent primarily on four adapted lines (B73, A632, Oh43, and Mo17), or closely related derivatives. It seems unreasonable to assume that most favorable alleles are concentrated in that sample. Another reason for the reduced genetic base of maize breeding programs was found by Jenkins (1978) to be a greater emphasis on developing recycled lines instead of lines developed from improved populations or synthetics. Private companies are, however, growing increasingly concerned about their narrow germplasm pools. Some companies have a germplasm-enhancement component to their breeding programs, although tough competition in the industry results in the tendency to focus on elite proprietary exotics from branch stations within a company. Throughout the world approximately 50,000 accessions of corn exist in germplasm banks (Goodman, 1983; Ayad et al., 1980) and, until recently, many had never been evaluated for useful traits. Obstacles limiting effective use of plant genetic resources include lack of evaluation data, lack of documentation and information, poor coordination of national policies, and poor linkages between gene banks and breeders. Evaluation is important in identifying potentially valuable traits in accessions, but most countries cite the lack of useful evaluation information as a major bottleneck to increasing germplasm utilization (Report on the State of the World’s Plant Genetic Resources, 1996). The assumption that well-evaluated and documented germplasm collections will lead to their increased use by breeders has been stated by many (Kannenberg, 1984; Wilkes, 1984; Plucknett et al., 1987; Smith and Duvick, 1989; Goodman, 1990; Anonymous, 1991; Salhuana et al., 1991). Germplasm bank accessions are at least 60 years behind currently used breeding populations for yield and standability. Genetic diversity exists in corn collections but little has been incorporated into elite breeding populations. LAMP evaluated over 12,000 accessions and the data are readily available to breeders. Resistance to using the elite LAMP accessions will decrease when they receive enough prebreeding to be attractive to commercial breeders. This prebreeding will be accomplished by GEM. Introgression of this germplasm into commercial materials will broaden the diversity of hybrids in farmers’ fields. Materials from LAMP have the potential to dramatically change the maize grown in the USA similar to the way materials from the Sorghum Conversion Program have changed grain and forage sorghum. Some of the important economic characteristics found in the germplasm released from the Conversion program and used to improve commercial sorghum hybrids are pest resistance, drought and saline tolerance, stalk strength, improved yield and yield stability, mold resistance, higher protein, a more desirable balance of amino acids, and superior food quality characteristics (Miller, 1979). Diversity of germplasm sources will aid the development of hybrids with more favorable amino acid compositions for food and feed uses, lower protein content
GERMPLASM ENHANCEMENT OF MAIZE
49
and higher starch yields for wet millers, and higher test weights, higher protein and harder endosperm for dry millers. Consumers will benefit by being assured of a stable food supply.
B. THE LATIN AMERICAN MAIZE PROJECT The GEM project in the United States would not exist if LAMP had not come first. LAMP provided the information necessary to efficiently select germplasm bank accessions for enhancement. In this regard, LAMP served as the first step to share promising maize materials from the germplasm banks with breeders. GEM will complete the process by returning to the germplasm bank enhanced materials developed from the accessions that can be directly used in applied breeding programs. LAMP involved the cooperative efforts of 12 countries (Argentina, Bolivia, Brazil, Colombia, Chile, Guatemala, Mexico, Paraguay, Peru, United States, Uruguay, and Venezuela) to evaluate their native maize germplasm accessions for yield and agronomic characteristics (Salhuana et al., 1991). The funding ($1.5 million) was donated in 1987 by the leading US producer of hybrid maize, Pioneer Hi-Bred International, under the incentive of its CEO, Dr William Brown. Dr Brown envisioned a collaborative effort among nations to characterize and regenerate maize accessions held in important germplasm banks of Latin America, leading to increased agricultural biodiversity. The funding was administered by the US Department of Agriculture (USDA), Agricultural Research Service (ARS) who also provided administrative support. Each participating country donated in-kind support consisting of a principal investigator and technical and support staff. LAMP was the first coordinated international project to deal with the evaluation of the genetic resources of a major world crop. During five stages, LAMP evaluated over 12,000 accessions (that belonged to 74% of total maize races) in locations divided into five homologous areas covering latitudes from 348S to 418N, longitudes from 448W to 1018W, and altitudes from 29 – 3300 m above sea level. In 1991, a catalog and CD-ROM of data of 12,113 accessions evaluated in LAMP’s first stage, and 2794 selected (primarily on yield) accessions evaluated in the second stage in 59 different locations of 32 regions of the 12 countries was published (LAMP, 1991). Based on this data the principal investigators in each country selected a total of 268 elite accessions that were crossed with the best testers of each region (Stage 3). Thirty-one testers were used for crossing with the elite accessions. Within a homologous area, principal investigators exchanged testcrosses among other principal investigators in the same homologous area, so that testcross evaluations were done in more than one country (Stage 4). Data from the testcross evaluations were published in a catalog and updated CD-ROM
50
L. M. POLLAK
in 1995 (Salhuana and Sevilla, 1995), and a final report published in 1997 (Salhuana et al., 1997). In the fifth stage of LAMP, each principal investigator was to enhance selected germplasm to meet his or her country’s breeding objectives, yet funding for only a year of small-scale enhancement was available. Enhancement of some of these selected accessions is an important activity for each country, because it is likely that the productive accessions identified by LAMP are very different from the breeding populations used in the country and that they can add new genes for productivity. This is true even though the selected LAMP accessions may be native to a country in question, because before LAMP few countries used their native germplasm in breeding programs (Salhuana et al., 1997). In the US, we used restriction fragment length polymorphisms to determine relationships among six Caribbean LAMP accessions and the adapted US inbred lines B73 and Mo17. It was very clear that the exotic accessions were very different from the Corn Belt lines, and that the Caribbean accessions were also very diverse from each other (Pollak and Salhuana, 1999).
II. GEM’S DEVELOPMENT A. PUBLIC/PRIVATE US AGRICULTURAL RESEARCH US agricultural research includes three sectors: federal programs, state programs in land-grant and other universities, and private industry. Both the state and federal programs are considered as the public sector. Federal programs include both intramural (e.g., USDA-ARS) and extramural research (through State Agricultural Experiment Stations). Objectives of intramural federal research (e.g., USDA-ARS) include conducting research that addresses national and regional problems where state incentives are low but social payoffs are high, and research that involves high-risk and long-term problems. Germplasm preservation, evaluation, and enhancement fit into these objectives. States typically direct their agricultural research toward problems and crops important in the state. The private sector focuses intensively on commercially-oriented problems, and has little incentive to do high-risk and long-term research on germplasm enhancement (Knudson, 2000). The amount of plant breeding effort in maize is in excess of 25% of total US plant breeding effort, calculated in scientific years (SYs). However, most of this effort is in the private sector. The private sector in 1994 employed 509.8 SYs, while the public sector employed 35.3 SYs (27.1 in the state universities, 8.2 in the USDA-ARS) (Frey, 1996). This represents a serious constraint in human resources and expertise preventing the public sector from carrying out a major maize enhancement effort without private involvement.
GERMPLASM ENHANCEMENT OF MAIZE
51
Legal changes in intellectual property rights and technology transfer during the past few decades allow greater cooperation between the public and private sectors. Intellectual property rights such as patents increase incentives for the private sector to invest in crop research. Technology transfer laws facilitate cooperation between federal and private laboratories. These legal changes have led to the emergence of a research consortium type of institution, whereby private and public collaborators jointly engage in basic and applied research, jointly own products from their research, and sometimes share the profits. GEM’s existence may be possible because of these changes in intellectual property and technology transfer laws (Knudson, 2000). The competitive nature of the seed industry made it unlikely that any one company would support an enhancement effort utilizing LAMP materials. Public breeders are poorly funded, there are very few of them compared to private breeders (Frey, 1996), and there are few if any grant sources for germplasm enhancement, so it was unlikely that public breeders could find the financial resources to support an enhancement effort. It was clear that a coordinated and cooperative effort among public and private sectors was needed before the LAMP materials would be used in the US breeding programs. It was also clear that coordination and primary funding of the effort would have to be by USDA-ARS because of the project’s high-risk long-term research and national scope.
B. PUBLIC AND PRIVATE INTERACTION TO ORGANIZE GEM As LAMP was nearing its conclusion in 1992, there was concern among maize germplasm scientists that unless an organized effort was put into place to enhance the best LAMP accessions they were unlikely to be used in breeding programs. Enhancement proposals were prepared by Linda Pollak of USDA-ARS, and by Major Goodman of North Carolina State University and Randy Holley of Northrup King (now Syngenta), and sent to the American Seed Trade Association’s (ASTA) Corn & Sorghum Basic Research Committee (CSBRC) for consideration at their December meeting. At the December 1992 meeting, the current concerns of the CSBRC were the diminishing role of the public sector in maize breeding, the federal and state budgetary constraints causing decreased financial support of public programs, and the relative lack of public support for maize research as compared to other crops (such as small grains) despite the greater economic importance of maize to the nation. The CSBRC recognized the need to develop a Congressional lobby for corn research and find approaches for funding an extension of LAMP to enhance the best accessions. Supporting an enhancement proposal that would be used to lobby Congress for research money was seen as the best action that could ease the concerns noted earlier. The CSBRC requested the public and private maize breeders who developed the initial enhancement proposals (Major Goodman, Randy Holley, Linda Pollak)
52
L. M. POLLAK
plus Douglas Tiffany and Wilfredo Salhuana (now retired) of Pioneer Hi-Bred International to develop a combined proposal for a public/private collaborative effort to enhance the accessions identified as useful by LAMP. During 1993 the proposal was used to generate interest and solicit in-kind support from ASTA’s member corn companies. To develop the framework of public/private cooperation, correspondence was initiated by Douglas Tiffany with scientists known to have maize germplasmenhancement interest, asking them to pass the information along to others. From this correspondence, the germplasm-enhancement (GE) network was compiled. This network has grown to include cooperators, administrators, politicians, and others, who receive notice of cooperator meetings, field days, updates to the GEM website, and other general information. At the GE network’s first meeting, held in association with the ASTA’s Corn & Sorghum December 1993 meeting, a technical committee to oversee enhancement efforts was elected. The committee included Wilfredo Salhuana as chair, Linda Pollak as coordinator, and Randy Holley, David Harper from Holdens Foundation Seeds as coordinator of a lobbying effort, Douglas Tiffany of Pioneer Hi-Bred, Kevin Montgomery of Golden Harvest, and Blaine Johnson of the University of Nebraska (Salhuana et al., 1994). Jim Parks of Wyffels Hybrids was added later to represent ASTA. Many have used mass selection to adapt tropical germplasm to temperate conditions (Hallauer and Sears, 1972; Longquist, 1974; Genter, 1976; Compton et al., 1979). But because the goal of this enhancement effort was for the products to be used in commercial programs as quickly as possible, we decided to adapt the germplasm by crossing to adapted inbred lines. During the winter of 1993– 1994, 46 LAMP accessions and tropical hybrids were crossed to the public inbred lines B73 and Mo17 of the Stiff Stalk (SS) and non-Stiff Stalk (nSS) heterotic patterns, respectively. Although these two inbred lines were no longer used commercially, they were considered by the technical committee to be the most representative of modern commercial inbred lines. The crosses were backcrossed to the same inbred line in 1994. These 50 and 25% breeding crosses were intended to be the initial breeding crosses from which enhanced S3 lines would be developed. During extensive discussions in developing a breeding protocol, however, the technical committee realized that these enhanced lines would be much less productive than those that could be developed by using modern commercial inbred lines to make breeding crosses. Each of the private members requested and eventually received permission from their companies to use proprietary inbred lines to make breeding crosses that would then belong to the cooperative project. This was an unprecedented use of proprietary inbred materials at this time, and demonstrated that the private cooperators were committed to making the project a success. In August 1994, project and breeding protocols (using a modified pedigree breeding procedure) were sent to all ASTA companies with maize breeding programs asking for their willingness to participate in GEM. Participation as
GERMPLASM ENHANCEMENT OF MAIZE
53
a private-sector collaborator in GEM involved signing an agreement and following the protocol. The protocol required them to: † contribute in-kind support for the breeding effort (winter and summer nursery rows, yield trial plots, and disease observation rows), † cross exotic accessions assigned by the coordinator to their proprietary inbred lines and return the crosses to the coordinator, and † not distribute proprietary £ exotic germplasm outside the company. The amount of donated in-kind support was determined by the company according to their investment in maize breeding. In-kind support by industry was considered important for lobbying the US Congress, for providing the necessary number of testing environments, for ensuring that the enhanced materials will have commercial relevance, and for providing public programs with routine cooperation and guidance. In return, a participating company received early access to GEM materials. Proprietary concerns were addressed by the following points: † A company’s S2 lines can be considered proprietary. † Pedigrees will be coded so that only the coordinator knows which company made a proprietary cross. † It is extremely difficult to extract an inbred line once it was hybridized, especially with an accession. † Publicly released material from company breeding efforts can be S2 synthetics, as opposed to S3 lines from public breeders. As the private-sector authors of the proposal, representing some of the largest US companies, had already signed the agreement before it was sent to the ASTA membership, company concerns that access to their proprietary inbreds would endanger their competitive position were reduced. The proposal was also sent to public-sector maize breeders giving them the opportunity to become collaborators, although they were not required to make breeding crosses (because the most elite US inbreds are proprietary) or provide in-kind support (because they lacked the financial resources). Signing an agreement allowed them access to GEM materials, potential financial or in-kind support to perform specific research or conduct breeding programs, and signaled to the US Congress and USDA-ARS their belief in the value of the project. The companies represented on the technical committee also considered that allowing public breeders access to the proprietary inbred line by exotic breeding crosses was a means by which modern commercial breeding material could be moved to the public sector. The technical committee knew that it was critical to obtain outside funding to help public cooperators participate. In addition, the proposal was sent to the countries that participated in LAMP, inviting their cooperation. This was in recognition of the role the countries had
54
L. M. POLLAK
played in identifying the germplasm used in GEM, and the largely Latin American source of the exotic germplasm. However, even though the countries were interested in participating, it was impossible without outside funding. As of this date, funding for their participation has not been found even though their participation has been encouraged by the CSBRC. Within a month and a half of the protocols and agreement being sent out, 17 seed companies and 14 public breeders had signed the agreement to become cooperators. The project has grown rapidly. In spite of rapid consolidation in the industry, in 2001 there were 39 cooperating companies. These companies included two starch processors, demonstrating the interest in value-added properties of the breeding materials. These companies included 11 popcorn companies that participated with several public cooperators in a popcorn sub-project of GEM. By 2001, 47 public scientists had signed the agreement, cooperating with varying levels of participation. There were also two international cooperators who had also participated in LAMP, Brazil and Argentina. A sub-committee from the CSBRC led by David Harper lobbied key legislators of the US Congress for permanent base funding to ARS to support the public effort at ARS and university locations, which they estimated at requiring $1,000,000 per year. Besides recognizing the value of the work, Congress was impressed by the $1.5 million contribution Pioneer Hi-Bred International made to LAMP, and the in-kind support from companies for GEM that was estimated by the companies to be worth approximately $450,000 per year in 1994. This amount underestimated the actual value of the contributions because it ignored the value of the proprietary germplasm, the overhead costs, and the advice and service of the private breeders. In 1995, $500,000 of permanent yearly funding was appropriated by Congress to support coordination of the enhancement effort at the Corn Belt ARS location in Ames, IA, research on value-added traits in Ames, data management of the project at Ames, a satellite location at a southern ARS location in Raleigh, NC, and support of public cooperators at other ARS and university locations.
C. GEM’S OBJECTIVE The objective of GEM is to provide to the maize industry materials developed using germplasm enhancement of useful exotic germplasm, with the ultimate aim of improving and broadening the germplasm base of maize hybrids grown by American farmers. GEM is an ongoing project, but to initiate enhancement 51 elite tropical and temperate LAMP accessions were chosen, plus seven commercial tropical hybrids were provided by DeKalb Genetics. The enhancement protocol is for one of the private cooperating companies to cross an exotic material by a proprietary inbred line to make a 50% exotic breeding cross, then for another private cooperator to cross the 50% cross with their proprietary line of the same heterotic pattern to make a 25% exotic breeding cross. All 50 and 25%
GERMPLASM ENHANCEMENT OF MAIZE
55
breeding crosses are evaluated for yield as testcrosses, and the best used to develop breeding lines by cooperators. Because proprietary germplasm is used to make breeding crosses, access to breeding materials is limited to GEM cooperators but the opportunity to become a cooperator is available to all. Data collected on GEM materials are freely available, and GEM enhanced lines and synthetics will be freely available through the US North Central Regional Plant Introduction Station (NCRPIS) after their public release. Traits targeted for improvement are agronomic productivity, disease and insect resistance, and value-added characteristics.
III. GEM’s ADMINISTRATION A. ORGANIZATION The organizational structure of GEM was based on LAMP. All GEM cooperators function similarly to the LAMP principal investigators, by being responsible for the project’s execution. A cooperators’ meeting is held once a year at the Corn & Sorghum ASTA meetings to discuss progress. Cooperators have the opportunity to view breeding materials at a field day held each year at Ames. Occasional field days are held by other cooperators. Public cooperators include both federal (USDA-ARS) and university faculty members, and include breeders, entomologists, plant pathologists, animal scientists, and food scientists. A technical steering group (TSG) meets three to four times a year to discuss policies, protocol, and results. The TSG is composed of seven members from industry representing large, regional, and foundation seed companies, and one public cooperator representing a university. USDA-ARS attendees were at first limited to two (GEM coordinator and North Carolina project leader), but have since increased to include others involved in the project. USDA-ARS attendees are ex-officio due to conflict of interest. Members from companies and universities serve a staggered three-year term. Wilfredo Salhuana and Linda Pollak served as chair and coordinator, respectively, and Marty Carson of USDA-ARS served as North Carolina project leader, since the beginning of the project. Starting in 2002, Mike Blanco is coordinating GEM at Ames while Linda Pollak is doing the GEM value-added research and breeding full time. In 2002, Joe Hudyncia started serving as the project leader in North Carolina, while Marty Carson is the Research Leader of the Ceral Rust Research Unit in St. Paul, Minnesota. The GEM coordinator in Ames is responsible for what is to be done, when, and at what cost. The coordinator manages seed curation, manages line evaluations and release, plans and organizes cooperative nurseries and yield tests using inkind support, manages data analysis and management, and coordinates public cooperator research and finances. A world-wide web home page has been developed (http://www.public.iastate.edu/~usda-gem/) that makes it easy to learn
56
Table I Public GEM Cooperator Research Projects Supported from Appropriated Funding 1995–2001 Name
Institution
Year(s)
Support
Research
Cornell University University of Illinois
1995–2001 1995–1998
$31,950 $20,700
John Ayers, Mel Johnson James Coors
Penn State University
1995–1996
$4400
Anthracnose stalk rot resistance Evaluation of tropical accessions as sources of genes to improve a Corn Belt Hybrid, leaf blight resistance Gray leaf spot resistance
University of Wisconsin
$35,720
Silage quality
Manjit Kang
University of Louisiana
1995–1997, 1999–2001 1995–1996
$4000
Richard Pratt
Ohio State
Resistance to Aspergillus flavus, maize weevil, southern rust Physical and compositional grain quality
Paul Williams, Frank Davis
USDA-ARS, Mississippi
1995–1996, 1998–2001 1995–1997
$28,174 $9000
Neil Widstrom
USDA-ARS, Georgia
1995–1998
$11,700
Billy Wiseman
USDA-ARS, Georgia
1995–1996
$8200
Larry Johnson
Iowa State University
1995–1996
$20,000
Dean Barry, Bruce Hibbard, Larry Darrah Jon Tollefson
USDA-ARS, Missouri
1995–1998
$24,845
Iowa State University
1995–1996
$9500
Resistance to fall armyworm, southwestern corn borer, and aflatoxin, southern yield tests Resistance to Aspergillus spp. infection and aflatoxin production, southern yield tests Resistance to corn earworm and fall armyworm Variation in proximate composition, grain physical properties, wet-milling properties, and starch functionality Resistance to first and second brood European corn borer and corn rootworm, evaluation of stalk and root quality Resistance to corn rootworm
L. M. POLLAK
Margaret Smith John Dudley, Donald White
Table I (continued) Name
Year(s)
Support
Research
Gary Munkvold Dennis West
Iowa State University University of Tennessee
1996–1999 1996–2001
$20,000 $28,000
Jim Hawk
University of Delaware
1996–2001
$50,000
Paul Scott Craig Abel, Richard Wilson Robert Lambert Jerry Sell
USDA-ARS, Ames USDA-ARS, Ames
1997–1998 1997
$5700 $5110
University of Illinois Iowa State University
1997–2001 1998
$25,000 $4041
Mark Campbell
Truman State University
1998, 2000–2001
$8143
Bruce Hibbard
USDA-ARS, Missouri
1999–2001
$13,500
Larry Darrah
USDA-ARS
1999–2001
$13,500
Javier Betran
Texas A&M
1999–2001
$12,500
Wenwei Xu
Texas A&M
1999–2001
$15,000
Ken Russell
University of Nebraska
2000–2001
$10,000
Fusarium ear rot resistance Southern yield tests, virus resistance, and breeding for food corn Southern yield tests, drought resistance, Stewart’s wilt Protein quality Non-DIMBOA European corn borer resistance Breeding for high-starch composition Evaluation of high protein lines as feed for broiler chickens NIR calibrations for amylose, modifying genes influencing ae and su2 alleles Resistance to western corn rootworm and European corn borer Development of lines for good rind penetrometer resistance and vertical root pulling resistance Development of food-grade corn germplasm with superior grain quality and adaptation Drought tolerance and corn earworm resistance Concentration of total phosphorus
GERMPLASM ENHANCEMENT OF MAIZE
Institution
Source: GEM, public cooperator summaries.
57
58
L. M. POLLAK
about GEM, contact GEM cooperators, obtain data, and order seed. In addition, the Ames location conducts value-added trait research and develops enhanced lines for improved yield and value-added traits. The Raleigh project leader manages enhancement of the 50% tropical breeding crosses for the southern USA and coordinates public cooperator projects in the southern USA.
B. FUNDING MECHANISM GEM received $500,000 from the federal government each year since the US Congress’s appropriation in the 1995 budget. After overhead, the Ames location received approximately $300,000 and Raleigh received approximately $150,000 each year. Of this amount, approximately $60,000 from Ames has gone each year to help support public cooperators (Table I). A comparable amount has been used from the Raleigh budget to help support southern public cooperators conduct yield tests and to support the breeding effort of Major Goodman. In-kind support from private cooperators each year provides approximately 6500 summer nursery rows, 2000 winter nursery rows, and 2600 disease observation rows. Over 7000 rows of yield test plots throughout the eastern, southern, and Corn Belt sections of the USA provide a wide diversity of environments. Additional GEM-related nursery rows and yield test plots are grown in the Ames and Raleigh locations.
IV. BREEDING ACTIVITIES AND RESULTS GEM’s standard breeding protocol developed by GEM’s TSG is presented in Table II.1 The numbers of S1s and S2s in Table II that need to be selected and evaluated are goals, and due to the unadapted nature of many of the breeding 1
The accessions that are used for breeding crosses will be assigned to the cooperators at random by the coordinator. The coordinator will recommend the selected material to be used for breeding to each group. If the number of nursery rows or yield trial plots for a given company is too few to handle an entire project, the coordinator will split the project into parts to accommodate those numbers. Requests for seed of the accessions are made directly through NCRPIS and for the breeding and testcrosses made through the coordinator. The distribution of the seed will be according to the utilization of the material. (a) Breeding crosses. We will send 400 kernels of the accession, necessary to pollinate a minimum of 200 ears with pollen of the inbred line. The cooperator will send seed of balanced bulks of the ears as directed by the coordinator, with the remaining seed returned to the coordinator. (b) Breeding procedure. We will send 2000 kernels necessary to self 1000 plants. Public sector cooperators will send seed of their selected S2 lines to the coordinator, and private sector cooperators will send the coordinator their synthetic based on recombined selected S2 lines. (c) Yield testing. Each experiment will require six locations, one replication, two rows per plot, and 40 kernels per row, which requires 480 kernels. Each cooperator will receive enough for the number of locations they are growing. Each cooperator will send to the coordinator a hard copy and diskette (or email attachment) of the data as soon as possible after harvesting.
GERMPLASM ENHANCEMENT OF MAIZE
59
Table II GEM Breeding Protocol for Developing S3 Lines or S2 Synthetics Season Winter 1
Summer 2
Winter 2 Summer 3
Winter 3
Summer 4
Winter 4
Summer 5
Winter 5
Description Private cooperator crosses the assigned accessions with a proprietary line of the same heterotic group to form the breeding cross A different private cooperator crosses the accession £ proprietary line breeding crosses with their proprietary line of the same heterotic group to form the three-way cross Cross all breeding and three-way crosses with two inbred testers Conduct yield trials of proprietary breeding and three-way topcrosses. Yield trials are conducted in a minimum of six locations, using five common and two local hybrid checks. Results permit selection of the best breeding or three-way crosses for further work Self-pollinate about 1000 plants of the breeding or three-way crosses selected from topcross trials. Shell S1 seed from each ear separately; do not treat seed. Send all seed to coordinator unless otherwise specified (coordinator may split project into a minimum of 250 ear row sections to match in-kind support) Plant the seed from the available ears ear-to-row and self-pollinated some plants in each row. Handle this material in the nursery as you would any of your own breeding projects (such as inoculating for diseases, selection procedures, etc.). Select among and within S1 families to obtain about 200 S2s. (For example, choose 200 rows and one ear from each row or 50 rows and an average of about four ears per row, or whatever combination fits your breeding philosophy and style. The target is about 200 ears. If a cooperator is doing 1/4 of a population, the target is about 50 ears; if 1/2 of a population, select about 100 ears.) Public cooperators send the coordinator 100 kernels of the selected S2s (about 200 ears). For the private cooperators, the S2s can be considered proprietary material Make 200 S2 testcrosses with an appropriate inbred tester of the opposite heterotic groupp. If the private cooperator does not have in-kind winter nursery rows available, contact the coordinator to get testcrosses made Yield test all 200 S2 testcrosses (minimum of six locations, one replication per location, need a minimum of four good locations for data summary). If the cooperator does not have enough in-kind yield trial plots to fully test their testcrosses, contact the coordinator to organize these yield trials The best S2 lines (the target is about 10 lines) will be recombined by private companies and the seed sent to the coordinator. (If a private company is working part of a population, such as 1/4 or 1/2, send the selected S2 lines to the coordinator to recombine with the other lines selected from that population). For the public cooperator projects, the best S2 lines will be selfed to the S3 by the coordinator and bulked (continued on next page)
60
L. M. POLLAK
Table II (continued) Season Summer 6 or Winter 6
Summer 7
Winter 7
Description The coordinator will use seed from the private-sector’s synthetics of recombined S2s and the public-sector’s S3s for seed increase, for evaluations of value-added traits, and for disease and insect resistance, using the in-kind support of cooperators who expressed interest in these evaluations. Cooperator will cross the synthetics of recombined S2s and a bulk pollen sample of the S3s to two testers not used previously Yield test testcrosses made in summer or winter 6 (minimum of six locations, one replication per location, need a minimum of four good locations for data summary). Send to the coordinator a hard copy and diskette (or email attachment) of the data as soon as possible after harvesting The coordinator will write a release article to be published in Crop Science. When the article is published, the coordinator will release the privatesector’s synthetics of recombined S2s, the public-sector’s S3s and all relevant data to NCRPIS
crosses are sometimes difficult to obtain. For example, a few of the breeding crosses have produced lines with a high percentage of male-sterile tassels, or lines with poor synchronization between male and female flowering. The coordinator has a great deal of latitude to help cooperators modify the protocol to fit their special requirements. The LAMP US Principal Investigator, Linda Pollak, and Technical Advisor, Wilfredo Salhuana, selected 51 highly productive LAMP accessions from temperate and lowland tropical areas as starting materials for GEM. In addition, DeKalb Genetics donated seven tropical hybrids. In 1994, the coordinator assigned a minimum of four accessions or tropical hybrids to each private cooperator to cross to a proprietary inbred line in the 1994– 1995 winter nursery. The coordinator specified the heterotic pattern (SS or nSS) to be used based on previous information from LAMP testcrosses (Salhuana and Sevilla, 1995; Salhuana et al., 1998, unpublished data). If the heterotic pattern of the exotic material was unknown, the coordinator had crosses with both heterotic patterns made. Private cooperators returned exotic £ private inbred breeding crosses to the coordinator from winter nurseries in the form of balanced bulks for future breeding and for making three-way crosses, and leftover bulks for other evaluations. Balanced samples of the breeding crosses were sent to two different companies for making three-way SS or nSS breeding crosses in 1995 summer nurseries or day-neutral locations. Each private cooperator was assigned a minimum of eight crosses. New private cooperators were assigned four accessions to make breeding crosses. In the 1995 –1996 winter nurseries, two companies (Golden Harvest and Pioneer) made nSS testcrosses with all available SS breeding crosses, and two
GERMPLASM ENHANCEMENT OF MAIZE
61
companies (Cargill and Holdens) made SS testcrosses with nSS breeding crosses. Cooperative yield testing during 1996 involved 16 experiments, each grown in more than six locations, evaluating 564 testcrossed 50 and 25% exotic breeding crosses. From these results, breeding crosses were selected for advancement to line development. Results for the best topcrosses of SS and nSS breeding crosses are shown in Tables III and IV, respectively, expressed in percentage over the five check hybrids (LH195/LH212 and LH195/LH59 from Holdens Foundation Seeds, and Pioneer Brand Hybrids 3489, 3525, and 3163). Some results are similar to the mean of the checks or better, thus we expect the testcrosses of the best lines developed from these breeding crosses to be superior to the check hybrids. Any line selected using this procedure will probably need agronomic improvement; hence, they are best used as breeding lines in a commercial breeding program. In Ames and North Carolina, before the above data were available, line development was started in a few breeding crosses based on LAMP data (Salhuana and Sevilla, 1995). In 1997, the lines were grown as testcrosses in yield tests in the Corn Belt and southeast. Approximately 130 S2 or S3 lines from GEM breeding crosses were better than the average of commercial check hybrids in trials managed by Raleigh, and approximately 65 S2 lines had yields similar to or greater than commercial check hybrid means in trials managed by Ames (GEM, 2000). Results for some of these lines are presented in Table V. After the initial stages of developing and evaluating breeding crosses, GEM’s seasonal breeding program developed into a balance of the elements discussed (developing and evaluating new breeding crosses), ongoing line development at various stages with public and private cooperators, and maintenance activities such as regenerating breeding crosses and increasing lines. To date, we have developed approximately 500 breeding crosses, and instituted pedigree breeding for line development in many of these. Yield data are available in CD-ROM for 1997– 1999 experiments (GEM, 2000), and after that on our website (http://www. public.iastate.edu/~usda-gem/). Results indicate that these lines have true yield potential. For example, in experiments grown by Pioneer Hi-Bred International in 1998 to test S2 lines from CUBA164 breeding crosses testcrossed with a Pioneer nSS inbred line, a CUBA164 topcross was the second highest yielding entry behind Pioneer Brand Hybrid 3525, in six replications. Seven other lines yielded more than Pioneer 3163, ranking five to eleven after Pioneer checks 3489 and 34G81. In another experiment with seven replications, eight testcrosses from a CUBA164 breeding cross yielded more than Pioneer Brand Hybrid 3163, ranking two to nine after Pioneer Brand Hybrid 33A14. Another experiment of four replications grown by DeKalb Genetics evaluated lines from a 50% exotic breeding cross with an accession from Argentina, AR01150, testcrossed to a DeKalb SS inbred line. Six AR0150 topcrosses were the highest yielding entries, beating the highest yielding check, Pioneer 3163. Four more AR01150 topcrosses beat the next highest yielding check, DK621 (Pollak and Salhuana, 2001).
62
Table III Yields and Moisture Values of Some GEM Breeding Crosses Testcrossed to an Elite nSS Line and Evaluated in 1996 GEM
Experiment
Percentage over five check hybrids (%)
Mean
Grain
Yield (kg/ha)
Moisture (g/kg)
Breeding cross
Yield
Grain moisture
50% tropical breeding crosses Golden Harvest tester
9
CUBA164:S15 CHIS775:S19 BR52051:S17 DKB844:S16 LSD(0.05)
97 96 92 90
100 102 103 100
9841 9697 9358 9145 22.1
214 218 221 213 1.8
50% tropical breeding crosses Pioneer tester
5
DKB844:S16 CUBA117:S15 CUBA164:S15 LSD(0.05)
98 98 90
116 108 115
10,807 10,788 9935 23.7
261 242 259 1.6
25% tropical breeding crosses Golden Harvest tester
8
CHIS740:S1411a CHIS740:S1415 LSD(0.05)
101 96
91 98
10,920 10,456 17.7
185 198 1.3
25% tropical breeding crosses Pioneer tester
4
CUBA164:S52008a LSD(0.05)
103
93
11,873 20.2
211 1.5
50% temperate breeding crosses Golden Harvest tester
8
AR16035:S19 FS8B(S):S03 AR16026:S17 LSD(0.05)
100 96 91
104 105 99
10,851 10,494 9923 15.2
201 203 192 1.4
L. M. POLLAK
Locations
Table III (continued) GEM
Mean
Grain
Yield (kg/ha)
Moisture (g/kg)
Locations
Breeding cross
Yield
Grain moisture
50% temperate breeding crosses Pioneer tester
5
FS8A(S):S09 UR10001:S18 UR13010:S13 AR16026:S17 LSD(0.05)
93 92 91 90
103 105 99 106
9829 9716 9653 9496 20.5
260 265 251 268 1.5
25% temperate breeding crosses Golden Harvest tester
7
AR16021:S0908b FS8A(S):S0915 UR13010:S1316 AR16026:S1719 CH04030:S0916 UR05017:S0415 UR10001:S1801 AR17056:S1216 UR10001:S1813 AR16026:S1716 LSD(0.05)
106 99 99 98 97 96 96 96 96 96
92 91 87 91 87 97 86 91 94 90
10,964 10,287 10,224 10,161 10,036 9992 9986 9986 9948 9929 17.8
187 186 177 185 178 197 176 186 191 183 1.4
GERMPLASM ENHANCEMENT OF MAIZE
Experiment
Percentage over five check hybrids (%)
Source: GEM 1996 yield data.
63
64
Table IV Yields and Moisture Values of Some GEM Breeding Crosses Testcrossed to an Elite SS Line and Evaluated in 1996 GEM
Experiment
Locations
Breeding cross
Percentage over five check hybrids (%) Yield
Grain moisture
Mean
Grain
Yield (kg/ha)
Moisture (g/kg)
6
ANTIG03:N12 CHIS775:N19 LSD(0.05)
96 92
110 105
9691 9377 25.5
240 229 2.2
50% tropical breeding crosses Holden’s tester
6
DKB844:N11b DREP150:N20 DKXL212:N11a DKXL370:N11a LSD(0.05)
96 94 93 90
107 117 108 114
10,174 9942 9797 9521 19.2
244 267 247 261 1.7
25% tropical breeding crosses Cargill tester
8
DKXL370:N11a20 CHIS775:N1920 BR52051:N0412 GUAD05:N0620 LSD(0.05)
101 101 99 98
109 105 108 102
11,058 10,995 10,801 10,738 10.5
230 222 228 215 4.8
25% tropical breeding crosses Holden’s tester
9
BR1501:N11a08d BR51501:N11a12 CHIS775:N1912 DREP150:N2011d ANTIG03:N1216 DREP150:N2012 LSD(0.05)
101 99 98 97 96 96
103 108 107 106 102 109
11,121 10,920 10,870 10,707 10,619 10,563 13.9
237 251 247 245 235 251 1.3
L. M. POLLAK
50% tropical breeding crosses Cargill tester
Table IV (continued) GEM
Experiment
Source: GEM 1996 yield data.
Mean
Grain
Locations
Breeding cross
Yield
Grain moisture
Yield (kg/ha)
Moisture (g/kg)
5
FS8B(T):N11a08a UR13085:N0215 CASH:N1410 AR16026:N1209 UR13010:N0602 LSD(0.05)
103 102 100 98 98
98 105 99 107 100
10,782 10,694 10,468 10,261 10,249 25.6
258 275 260 282 263 3.1
GERMPLASM ENHANCEMENT OF MAIZE
25% tropical breeding crosses Holden’s tester
Percentage over five check hybrids (%)
65
66
L. M. POLLAK
Table V Results from Experiments Evaluating Topcrosses of Lines Developed from Two Breeding Crosses using the Chilean Temperate Accession CH05015, Grown in 1997 in Corn Belt Locations Lodged (%) Pedigree
Tester
Yield (kg/ha)
Dropped
% H2 O
Test weight (lbs/bu)
Stalk
Root
Ears (%)
CH05015:N15-98-1 CH05015:N15-76-1 LH195/LH212 LH195/LH59 Pioneer 3163 Pioneer 3489 Pioneer 3525 Checks mean LSD(0.05)
Holden’s Holden’s
9041 8493 10,262 8644 10,175 8981 9210 9454 20.3
20.7 18.5 20.0 18.7 22.4 17.8 16.4 19.0
61.5 59.4 61.0 60.0 60.6 60.3 60.0 60.4
3.7 5.4 4.3 2.7 9.6 5.5 14.2 7.2
4.7 4.7 0.9 2.5 6.4 0.5 0.0 2.1
2.8 0.0 0.9 1.5 0.0 0.0 0.0 0.5
CH05015:N12-183 CH05015:N12-186 CH05015:N12-145 CH05015:N12-140 LH195/LH212 LH195/LH59 Pioneer 3163 Pioneer 3489 Pioneer 3525 Checks mean LSD(0.05)
Holden’s Holden’s Holden’s Holden’s
10,670 10,638 10,428 10,253 10,518 8607 10,774 10,309 10,684 10,178 32
20.4 19.8 20.8 21.1 19.3 18.5 20.4 17.6 16.4 18.5
49.8 54.5 54.4 56.6 54.6 54.0 49.1 57.3 53.4 53.7
4.8 6.8 7.1 5.7 6.5 6.2 17.9 15.2 8.6 10.9
0.0 0.0 2.6 0.0 0.0 0.0 0.0 0.0 1.2 0.2
0.8 1.6 0.0 0.6 0.6 1.2 0.0 0.0 0.6 0.5
Source: GEM 1997 yield data.
Similar results have been obtained for experiments testing lines from other breeding crosses in later years. Each year’s yield tests coordinated from Ames are a mixture of S2 testcross evaluations from public cooperators (primarily from the coordinator’s breeding program, Linda Pollak), from private cooperators, retests of previously selected lines, and research experiments. The yield tests coordinated from Raleigh are primarily evaluations of lines from the project leader’s (Marty Carson) and primary southern public cooperator’s (Major Goodman) breeding programs. From each year’s public cooperator experiments in Ames approximately 10 S2 lines per experiment are selected for further testing and possible future public release, approximately 50– 60 total each year. As soon as the lines were increased to S3 bulks, they are listed on our website and available to GEM cooperators for crossing and evaluating with their own inbred testers, for disease or insect
GERMPLASM ENHANCEMENT OF MAIZE
67
evaluations, for breeding and for research. Similarly, public lines are available to cooperators from the Raleigh breeding program. From GEM’s inception it has been understood that participating companies can freely use in their breeding programs genetic materials obtained either through seed orders of breeding crosses or S3 lines for possible future release, or through protocol breeding assignments as long as mutually agreed upon materials are returned to the coordinator. Companies are encouraged but not required to share data from evaluating materials through seed orders, or data taken on protocol breeding materials that were not part of the protocol.
V. VALUE-ADDED TRAIT ANALYSES AND RESULTS A. THE NEED FOR IMPROVING VALUE-ADDED TRAITS Although LAMP was the first coordinated international project for evaluating a major world crop (Salhuana et al., 1991; Pollak, 1993; Eberhart et al., 1995), only grain yield and agronomic data were collected (Salhuana and Sevilla, 1995; Salhuana et al., 1997). Studies indicate that significant variability for quality (Tello et al., 1965; Jellum, 1970; Zuber et al., 1975; White et al., 1990; Hameed et al., 1994; Campbell et al., 1995; Dunlap et al., 1995) and biomass for paper pulp production (Hammes and Pendleton, 1984) is present in maize genetic resources. Because much of the exotic germplasm has undergone selection for many indigenous uses (feed, foods, beverages, etc.) by various cultures, it seems likely that new grain quality characteristics will be found in exotic rather than the narrow-based germplasm now used. Limited variability for feed quality is found in present day hybrids and thus in elite breeding materials, according to composition trait values collected in the Iowa Corn Yield Tests since 1988, and 7399 samples collected from 27 locations in North America from 1987 to 1993 (Ertl and Orman, 1994). Breeding for high-yielding hybrids using limited genotypes with reduced variability results in a high degree of uniformity in grain type and nutritional content. Uniformity is valued by farmers but may not serve the needs of users and processors. Corn is the major feed for livestock and will continue to be important. The livestock industry is becoming increasingly concentrated as production moves from small independent operators to large integrated corporations, and thus are beginning to reach the point where minor changes in raw materials can provide significant changes in overall costs (Wheat, 1992). As more foreign customers demand meat products to improve the diets of their people, an increase in grain output becomes vital. USA exports as much as 20% of its corn crop with over onehalf of this export grain fed to livestock (Nutrient Content and Feeding Value of
68
L. M. POLLAK
Iowa Corn, 1991). Hill (1981) emphasized the need for improved corn quality traits if the United States is to maintain its present world market export share.
B. THE VALUE-ADDED TRAIT RESEARCH COMPONENT OF GEM During LAMP, the USA Principal Investigator had a laboratory with analytical capability to analyze oil quality of corn kernels, and had research colleagues with analytical capability to analyze seed composition, wet-milling characteristics, and starch quality. Along with the agronomic LAMP evaluations, many LAMP and other accessions and resulting breeding materials developed from the accessions were analyzed for these value-added traits (Campbell et al., 1995; Dunlap et al., 1995; Hameed et al., 1994; Ng et al., 1997; Pollak and White, 1997; White et al., 1990). These studies indicated that there was a great potential for improving adapted corn for these traits by introgressing exotic materials. Therefore, including a value-added research component to GEM was a primary objective during GEM’s organization (Pollak, 1997). Support for valueadded research of GEM breeding materials, additional analytical equipment, and a laboratory manager was included in the lobbying effort from the beginning. At present the value-added trait laboratory in Ames has the capability to measure starch, oil, and protein composition, starch quality (thermal and viscosity) characteristics, amino acid composition, fatty acid composition, and other traits such as tocopherol, carotenoid, and vitamin levels. All GEM breeding crosses and the selected S3 lines are evaluated for grain composition, with those close to targeted values undergoing further starch, oil, or protein quality evaluations as appropriate. Those lines with unusual traits close to targeted values are then included in the value-added research and breeding component of GEM as appropriate. All released S3 lines will be released with all data that have been collected on them in the laboratory, thus the lines can be targeted to their intended use in commercial breeding programs.
C. GRAIN COMPOSITION Corn kernels are composed of approximately 73% starch, 10% protein, and 5% oil, with the remainder made up of fiber, vitamins, and minerals (Eckhoff and Paulsen, 1996). Corn’s contribution as a feed ingredient is primarily as energy provided by starch. Among feed grains, corn is one of the most concentrated sources of energy, containing more metabolized energy—or total digestible nutrients—because of its high starch – low fiber content (Watson, 1987). The major drawback of corn, however, is its low protein content. In addition, the protein is of low biological value, as it does not supply the essential amino acids either in adequate quantities or adequate proportions (Perry, 1988). Because of its
GERMPLASM ENHANCEMENT OF MAIZE
69
chemical composition, corn protein itself is not sufficient for optimum animal growth. The total protein of corn is deficient in lysine and tryptophan for monogastric animal species and requires supplementation for adequate performance (Church and Nipper, 1984; Watson, 1977). One advantage of corn is that it has high levels of the S-containing essential amino acids, cystine and methionine (Watson, 1988). To increase the energy content other foodstuff, such as food oil, is added to the feed. As nutrient requirements become more accurately defined through research, it is possible to formulate diets more precisely (Nutrient Requirements of Swine, 1988), thereby achieving more efficient animal production. Since lipid provides more energy than starch (9 versus 4 cal/g), it is beneficial to increase the lipid content to provide more energy. The average oil content of corn is 4.4% of whole corn kernels (dry basis) (Watson, 1987). Corn lines with oil content of 18% have been reported (Weber, 1987). This information is convincing evidence that it is possible to increase the oil content in corn along with other composition changes to improve the nutritional status of corn. In the feed industry, supplementation is made by adding foodstuff rich in lysine and tryptophan amino acids, such as oilseed, fish and component meat meals. When oilseeds such as soybean meal are added to a feed mixture, the high content of lysine and tryptophan in soybean meal compensates for their deficiency in corn protein, while the high cystine and methionine content of corn protein compensates for their low value in soybean meal (Watson, 1988). Supplementation usually takes up to 28% of the entire feed mixture (Watson, 1977), among which soybean meal is about 15% (Perry, 1988). If protein content in corn could be increased from 10 to 15% with more balanced amino acid contents, then, the use of soybean meal would be eliminated when 15% of protein is required in a feed mixture. This small increase of protein in corn could have a substantial economic influence for farmers because soybean meal is more expensive than corn ($180 versus $120 per ton; Church and Nipper, 1984). Taking into account 120 million tons of corn and 16 million tons of oilseed meals used for feed each year (Watson, 1988), the economic benefit to farmers is obvious. All livestock producers, large or small, must use supplementation to achieve optimum growth of animals. Supplementation adds extra costs of ingredients, transportation, storage, and feeding complexity. One way to minimize this problem is to improve the nutritional values of corn by breeding. The effectiveness of incorporating exotic germplasm into Corn Belt material to alter valueadded traits is illustrated in Table VI. In the Corn Belt, the typical values of protein, oil, and starch are 10, 5, and 73%, respectively (Ertl and Orman, 1994). Each step in our enhancement protocol, from the Argentine accession AR16035, to the breeding cross, to the lines developed from the cross, shows additional improvement. Without selection for composition, lines from this breeding cross already nearly meet, equal, or sometimes even exceed our target composition
70
L. M. POLLAK
Table VI Grain Composition Values for a GEM Breeding Cross and Lines Developed from an Argentine Accession, AR16035 Genotype AR16035 AR16035:S02 AR16035:S02-36 AR16035:S02-519 AR16035:S02-535 Corn Belt means
Protein (%dm)
Oil (%dm)
Starch (%dm)
13.7 14.2 11.4 16.9 6.5 10.0
5.4 5.6 6.5 6.0 4.6 5.0
67.2 66.7 67.7 64.2 75.0 73.0
Source: GEM value-added trait laboratory.
values of 16, 7, and 75% for protein, oil, and starch, respectively. This example clearly shows that the Corn Belt material was effectively enhanced by exotic germplasm. We use near infrared reflectance spectroscopy (NIRS) to measure corn composition. NIRS is a rapid, non-destructive method for measuring the nutritional properties of grain. The instrument has a standard calibration based on scans from a wide variety of maize germplasm. The values are substantiated with wet chemistry analysis of constituents and used to evaluate unknown samples. Our samples are analyzed as whole grain for protein content, oil content, starch content, and moisture. Grain from GEM breeding crosses and selected lines for possible release are estimated as predicted by the calibration data. These early breeding materials with best compositional profiles are used in the value-added trait breeding and research program at Ames. The targeted values for grain quality are proteins of 13% or higher; oil of 6% or higher; and starch of 75% or higher. We usually try to select and advance both high and low values for these traits for research use. In screening 139 S1 lines from a special project designed to develop a few lines (approximately 10) from many breeding crosses for valueadded trait research and breeding, we obtained ranges of 2.0 –6.8 for oil (%dm), 8.7 – 15.8 for protein (%dm), and 65– 73.5 for starch (%dm) (unpublished data). In screening selected S3 bulk lines from 1997 to 1998 yield tests coordinated from Ames we obtained values of 1.9 – 5.3 for oil (%dm), 9.4 – 15.1 for protein (%dm), and 64.7– 73.1 for starch (%dm) (unpublished data). NIRS is a good method for screening for breeding material and as a selection tool in a breeding program, as 200 samples can be non-destructively analyzed each day.
D. STARCH QUALITY Corn starch is widely used in food and non-food industries. Examples of corn starch use range from providing the functionality or properties of anticaking,
GERMPLASM ENHANCEMENT OF MAIZE
71
dusting, molding, viscosity, texture, film-forming, colloid protection, sweeteners, or encapsulation in the food industry for supplying alcohol, ethanol, binders in gypsum board, additives in paper making, or serving as an ingredient in the development of thermoplastics and polyurethanes (Orthoefer, 1994; White, 1994). Corn starches are mixtures of linear and branched polysaccharides, which give very different physical properties. The starches composed of only one component have special properties that can be utilized in broader applications, and lead to specialized uses for which regular starches are not appropriate (Whistler, 1984; White, 1994). One approach to improve starch functionality is to chemically modify the physical and chemical characteristics of native starch to provide unique functionality. Still another approach is to develop corn varieties that naturally produce starches requiring less or even no chemical modification for unique use. This approach is especially valuable to the food industry as an opportunity to develop “all natural” convenience foods. Naturally modified starch in normal corn hybrids could be beneficial to farmers if the large yield reduction that occurs with mutant corn is avoided (Pollak and White, 1997). Oil and protein have commercial value as by-products from the production of corn starch in the food industry. Starch represents nearly 70% of the dry weight of the mature corn kernel and is the most economically important component. Therefore, it is essential to determine the endosperm variation and starch quality of selected GEM materials. The process has two steps. The first step is the extraction of the starch from the endosperm of single kernels using a modified mini wet-milling procedure developed in our laboratory (Krieger et al., 1997). The second step evaluates the starch qualities and structures by measuring the starch gelling properties with the differential scanning calorimeter (DSC) to predict functionality in a food or industrial use (Stevens and Elton, 1971; White et al., 1990; Pollak and White, 1997). The DSC allows direct measurement of the energy required to gelatinize starch. The gelatinized samples are stored for a week in a refrigerator and rescanned to determine the amount of retrogradation or recrystallization, and hence the stability of the starch gel. Because we use single kernel analysis, we do not use this analysis on breeding lines unless they are fairly inbred, S3 minimum. Lines with unusual DSC scans are used to develop specialty corn lines with different starch properties than normal corn starch, and give breeders indications as to which breeding crosses might produce lines with different starch properties. To fully understand the uniqueness of these lines, research involving additional laboratory analyses can be done. These analyses include rapid visco analysis (RVA) to determine the rheological and pasting properties (important for food applications), scanning electron microscopy to observe the starch granules in the kernel, image analysis to ascertain granule size and shape, and gel strength. Additional research can also be done to determine the starch structural properties that cause the starch to have unusual functional properties. Research that examines how thermal properties, as measured by DSC, relate to the structural and functional
72
L. M. POLLAK
characteristics of the starch is summarized by White (1994). Using the published research along with industry consultation, the thermal properties and their implications was compiled by Dr Pamela White, a public GEM cooperator from the Department of Food Science and Human Nutrition at Iowa State University, and is given in Table VII. Typical DSC values for checks used in our laboratory and some GEM lines with outstanding values (in bold) are shown in Table VIII.
E. OIL QUALITY Corn oil is commercially produced from corn germ as a by-product of wet or dry milling. The value of corn oil results mainly from the recognition of the importance of its high content of unsaturated fatty acids and flavor stability (Leibovitz and Ruckenstein, 1983; Watson, 1988). Corn oil is used primarily as a premium cooking and frying oil (Sonntag, 1979) without hydrogenation. This superiority of corn oil to other vegetable oils became important when Mensink and Katan (1990) reported that trans fatty acids in a diet increased the total and low-density lipoprotein – cholesterol and lowered the high-density lipoprotein – cholesterol levels compared to cis fatty acids in a diet. The effect of trans fatty acids may be similar to that of saturated fatty acids which are believed to be the most effective dietary factors in increasing plasma cholesterol (Dupont et al., 1991). The majority of trans fatty acids (80%) in the human diet comes from consumption of hydrogenated vegetable oils, with the average per capita intake ranging from 6.5 to 12.0 g/day of t FA (Borenstein, 1991). Oil with increased saturated fatty acid content may increase the corn oil oxidative stability further, resulting in specialty oil suitable for deep-fat frying in the food industry or for margarine production without hydrogenation. Naturally saturated corn oils should also have fewer processing costs and should result in more profit for the farmers and/or less cost for the consumers. Oil quality is determined by fatty acid composition measured on individual kernels. Because this analysis is done on individual kernels, it is usually reserved for fairly inbred breeding materials, usually S3 lines. Corn oil from individual kernels is extracted following a procedure modified from soybean (Hammond, 1991). Individual kernels are placed in an aluminum crushing plate in a hydraulic press and crushed. Hexane is used to extract the oil (Hammond and Fehr, 1984). Sodium methoxide in methanol is used to transesterify the fatty acids to fatty acid methyl esters (FAMES) which are analyzed on a gas chromatograph. Standards of the FAMES are injected to determine their retention times. There are certain parameters of importance indicating oil quality, such as high saturated fatty acid content (palmitic plus stearic acids), low saturated fatty acid content, and oleic acid. Corn usually has fatty acid values of palmitic 11.0%, stearic 1.7%, oleic 25.8%, linoleic 59.8%, linolenic acid 1.1% and total saturated fatty acids of 12.7% (Strecker et al., 1996). The targeted values for lines for our breeding
GERMPLASM ENHANCEMENT OF MAIZE
73
Table VII Desired Characteristics of Differential Scanning Calorimetry (DSC) Parameters DSC parameter TpG (peak onset of gelatinization)
DHG (enthalpy of gelatinization)
RnG (range of gelatinization)
PHI (peak height index) (enthalpy/1/2 range)
DHR (enthalpy of retrogradation)
%R (percentage of retrogradation)
Shape of the thermogram
Source: White, Iowa State University.
Desired characteristic Low value means less energy required in starch cooking process Target ,608C Low value means less energy required in starch cooking process Target ,2.5 cal/g High value means extensive thickening power Target .4.0 g/cal Low value means starch granules are likely from a homogeneous population. Starch cooking can occur quickly within a brief range Target ,58C High value means starch granules are likely from a heterogeneous population. Starch cooking occurs over a wide temperature range Target .158C High value means the thermogram has a tall narrow peak, which suggests high thickening power within a narrow temperature range Target .1.2 Low value means the starch is not subject to aligning and recrystallizing. Starch is likely stable in frozen and refrigerated products Target ,1 cal/g Low value gives similar meaning as previous parameter. More information about the relation to the original enthalpy is obtained Target ,20% High value may indicate presence of resistant starch Target .80% Double or triple peaks suggest two or more populations of starch granules. Starch cooking occurs over a wide temperature and may have multiple functions Target: detection
74
L. M. POLLAK Table VIII Lines with Unusual DSC Values (Starch Thermal Traits) TpG (8C)
DHG (cal/g)
RnG (8C)
PHI
DHR (cal/g)
%R
CUBA164:S2008a-9-1-2 DK212T:S0610-25-1-1 CUBA164:S2008a-9-1-2 CHIS775:S1911b-16-1-1 DK212T:S0610-8-1-3 CHIS775:S1911b-37-1-1 DK212T:S0610-25-1-1 DK212T:S0610-10-1-3 CUBA164:S1511b-38-1-3 CUBA164:S1511b-34-1-3 CUBA164:S2008a-31-1-3
62.9 69.0 62.9 65.4 66.9 66.1 69.0 68.5 65.9 65.6 65.0
2.9 3.5 2.9 2.3 4.1 3.0 3.5 3.3 2.9 3.2 3.1
14.4 4.2 14.4 10.5 7.5 9.3 4.2 4.3 9.3 10.1 10.0
0.4 1.7 0.4 0.4 1.10 0.64 1.7 1.6 0.6 0.6 0.6
1.9 2.0 1.9 1.2 1.9 1.1 2.0 1.3 2.4 2.7 2.6
66.2 55.8 66.2 53.8 46.8 38.7 55.8 40.2 83.8 86.7 83.9
Sigma starch Mo17 Pioneer 3394 Pioneer 3489
65.1 70.4 64.1 64.5
9.1 5.7 11.1 10.9
0.7 1.1 0.5 0.5
1.1 1.3 1.2 1.5
38.8 40.0 48.2 51.8
Pedigree
Corn Belt Checks 2.9 2.9 2.7 3.0
Source: GEM value-added trait laboratory.
program are lines with , 6% for low total saturated oil, . 17% for high total saturated oil, and . 65% for oleic acid for high monounsaturated oil, with our eventual goals for developing specialty oils that may be used in an application listed in Table IX.
VI. PUBLIC COOPERATOR RESEARCH AND RESULTS Research conducted by public cooperators usually is related to their area of expertise and covers a wide range (Table I). The following research areas are priorities set by the TSG. † first and second generation European corn borer (ECB1 and ECB2, respectively) resistance, † corn rootworm resistance, † gray leaf spot resistance, † Stewart’s wilt resistance, † anthracnose stalk rot resistance, † fusarium ear rot resistance, † virus resistance, † silage quality, and † grain quality.
GERMPLASM ENHANCEMENT OF MAIZE
75
Table IX Targeted Values of Oil Quality for Food and Industrial Applications Fatty acid
Goal (%)
High saturated fatty acids Low saturated fatty acids
30 6
Oleic
80
Potential application
Current best value
Trans-fatty acid free margarine 21.0%, FS8A(T):N1801-47 Low saturated fat labeling on 9.7%, CH05015:N1206-35-1 food products Heart healthy oil similar to 60.6%, FS8A(T):N1801-1 olive oil
Source: GEM value-added trait laboratory.
A. EUROPEAN CORN BORER RESISTANCE The GEM-related research of Drs Craig Abel and Richard Wilson, entomologists at NCRPIS in Ames (now USDA-ARS at Stoneville, MS and retired, respectively) focused on the European corn borer, which causes large losses in maize. It is important to find exotic sources of resistance and study ways to introgress it into elite material. Peruvian LAMP accessions previously found resistant to leaf feeding by ECB1 (Abel et al., 1995) were evaluated for resistance to ECB2, corn ear worm, southwestern sugarcane borer, and western corn root worm (Wilson et al., 1995). The LAMP accessions Lambayeque 42, Piura 208 and Libertad 3 were resistant to ECB2 and corn ear worm, whereas Lambayeque 29 was resistant to ECB2 and western corn root worm (Wilson et al., 1995). The multiple pest resistance showed that these accessions would be useful in pest management. The Corn Belt germplasm sources of ECB resistance are based primarily on the chemical DIMBOA but the resistance of these Peruvian accessions is not based on DIMBOA (Abel and Wilson, 2000). Larval feeding tests indicated that water-soluble factors from the Peruvian accessions inhibited the growth, development time, and survival of the ECB larvae (Binder et al., 1999). The accessions showing resistance were crossed to two public inbred lines (B94 and B97). Public lines were used instead of private lines because the initial crosses were made before 1994, when GEM received permission from private companies to use proprietary germplasm in crosses. Plants of the F1 generation were infested with neonate ECB larvae and only the resistant plants were backcrossed to lines B94 or B97. The same procedure was utilized to make the second backcross using only resistant plants. Seed of the second backcross was planted and plants selfed. Descriptive statistics for the evaluation of ECB1 and ECB2 in the best plants for first and second backcrosses (BC1 and BC2, respectively) (Table X) show that ECB resistance was maintained through introgression. ECB1 scores are qualitative and range from 1 to 9. Ratings of 1 – 3 are considered resistant, 4– 6 intermediate, and 7– 9 susceptible. ECB2 values
76 Table X European Corn Borer Scores (ECB1) and Tunneling (ECB2) of Selected Plants in the First Backcross Generation (BC1), and Lines Developed from the Second Backcross Generation (BC2)† BC-1 Line
ECB1 scores
ECB2 tunnel length (in.)
ECB1 scores
ECB2 tunnel length (in.)
Lambayaque 29 Lambayaque 29 Lambayaque 29 Lambayaque 29
B94 B94 B94 B94
Mean Standard deviation Minimum Maximum
1.9 0.3 1 2
1.1 1.4 0 3
1.3 0.4 1 2
2.2 1 0.5 4.5
Lambayaque 42 Lambayaque 42 Lambayaque 42 Lambayaque 42
B94 B94 B94 B94
Mean Standard deviation Minimum Maximum
2 0 2 2
1.9 1.1 1 3.3
1.4 0.5 1 2
4.6 3.7 1.5 11.5
Libertad 3 Libertad 3 Libertad 3 Libertad 3
B97 B97 B97 B97
Mean Standard deviation Minimum Maximum
2.3 0.3 2 3
4 4.8 0 23
1.9 0.6 1 3
3.4 1.7 0.5 8.5
Source: GEM 1997 executive summary.
1 ¼ resistant 9 ¼ susceptible
1 ¼ resistant 9 ¼ susceptible
L. M. POLLAK
Accession
BC-2
GERMPLASM ENHANCEMENT OF MAIZE
77
represent inches of stalk tunneling; resistant 0 –6 in. of feeding, intermediate 6 – 12 in., and susceptible more than 12 in. The mean, minimum and maximum values for the plants in BC1 and BC2 show very good resistance for ECB1 and ECB2. The lines developed from the second backcross with B94 were topcrossed with an nSS inbred and the lines with a B97 background were topcrossed to an SS inbred. Yield trials of these topcrosses were grown in 1997 GEM in-kind yield tests to evaluate yield and other agronomic characteristics. We expected that the best yield of the topcrosses would not be superior to commercial hybrids because the resistant Peruvian accessions were not among those LAMP accessions selected for high yield, and public lines were used to make the breeding crosses. However, three topcrossed lines had mean yields over six Corn Belt locations equal to or exceeding the mean yields of the five check hybrids (Holden’s Foundation Seeds LH195/LH212 and LH195/LH59, Pioneer Brand Hybrids 3489, 3525, and 3163), and still exhibited ECB resistance. The lines resistant to ECB were used in a breeding program to develop germplasm lines for insect resistance. Resulting lines were resistant to leaf and collar feeding by ECB but had low levels of DIMBOA (Abel et al., 2000a). Evaluation of 15 experimental lines from this program were evaluated for corn earworm, fall armyworm, southwestern corn borer, and sugarcane borer (Abel et al., 2000b). Lines that were resistant to leaf feeding by fall armyworm and leaf and stalk feeding by southwestern corn borer were found. A line resistant to corn earworm showed maysin levels lower than those commonly found in earwormresistant lines, indicating a new source of resistance. One of the ECB-resistant lines became the first publicly released line from GEM (Abel et al., 2001).
B. CHARACTERIZING LAMP ACCESSIONS AND THEIR CROSSES FOR WET-MILLING EFFICIENCY The GEM-related research of Drs Suvrat Singh and Lawrence Johnson, food scientists at Iowa State University (Dr Singh now at Ruiz Foods) focused on wet-milling efficiency and other value-added traits. Forty-nine LAMP accessions, most of which were among the first group of accessions used in GEM, two commercial hybrids (Pioneer Brand Hybrids 3394 and 3489), and two inbreds (B73 and Mo17) were evaluated for their compositional, physical, and wetmilling properties (Singh et al., 2001a). The recovered starch was characterized to identify any unusual thermal, pasting, gelling, or retrogradation properties (Singh et al., 2001c). Heterosis of these traits was examined in 10 selected GEM accessions crossed with each of Mo17 and B73 inbreds (Singh et al., 2001b, unpublished data). The accessions contained 3– 6% less starch, 4 –6% more protein, and up to 2 – 4% more fat than the commercial hybrids (Singh et al., 2001a). Higher protein
78
L. M. POLLAK
and fat contents make the energy of GEM lines dense and good for animal feed. On an average, absolute densities were greater (1.32 versus 1.29 g/cc) and 1000 kernel weights were less for GEM than for the commercial hybrids; test weights were similar. Absolute density is highly correlated with protein content and kernel hardness. US dent corn in general is quite soft, while the accessions are much harder, making them less susceptible to breakage and more suitable for dry milling. Thus, the exotic genes may contribute these traits to the GEM lines. The wet-milling characteristics of the GEM accessions were not nearly as good as for the commercial hybrids (Singh et al., 2001a). Starch yields averaged only 54.3% for the GEM accessions versus 64.8% for the commercial hybrids. Proteins content of starches recovered from GEM were much greater than for commercial hybrids. Gluten yields were much greater while gluten protein contents were much lower for the GEM accessions than for the commercial hybrids due to difficulty in separating starch from gluten. Occasionally, high fiber yields were also obtained for the accessions, indicating that the starch did not separate well from the fiber. Thermal properties of starches recovered from GEM accessions had much wider variation than starch recovered from normal dent corn (Singh et al., 2001c). These differences were statistically significant but not of practical significance to the starch industry. However, these differences can be useful to corn breeders to expand genetic differences in starch which would then be of value to the starch industry. Starches isolated from GEM accessions had on average higher gelatinization temperatures, lower heats of gelatinization (enthalpy), and similar percentages of retrogradation. GEM accessions had on an average greater temperatures at peak viscosity, greater peak viscosities, and greater viscosity breakdowns. The gel strengths were typically greater for the GEM starches than for starches from commercial hybrids. Ten accessions were selected from the 49 original GEM lines and crossed with two dent corn inbreds, B73 and Mo17 (Singh et al., 2001b). When crossing with B73, protein contents and absolute densities were found to be greater for the cross than either parent; all other compositional, physical, and wet-milling properties were similar to the mean of the parents. When crossed with Mo17, starch contents, absolute densities, starch yields, and starch recoveries were greater, and gluten yields were lower than either parent; all other compositional, physical, and wet-milling properties were similar to the mean of the parents. Mo17 expressed poor wet-milling properties per se but produced superior crosses than did crosses with B73, which had better wet-milling properties as an inbred per se. As Mo17 belongs to the nSS heterotic pattern, breeders utilizing GEM lines to improve wet-milling characteristics may want to focus on using the lines from nSS breeding crosses. Similar results for starch properties were observed using both inbreds (unpublished data). Gelatinization peak temperature of starches from the crosses were similar to those of starch from the accessions. Enthalpies, peak height
GERMPLASM ENHANCEMENT OF MAIZE
79
indices, peak viscosities, viscosity breakdowns, and percentages of retrogradation were greater for the crosses than for either parent. Gel strengths were similar to the mean of both parents.
C. OTHER SIGNIFICANT PUBLIC COOPERATOR FINDINGS Some other significant findings by public cooperators include the following: † Mary Carson, the Raleigh Project Leader, found many 50% tropical GEM breeding crosses as well as advanced S2 and S3 lines that had gray leaf spot resistance equal to the most-resistant commercial check hybrid. † In a trial to estimate silage yield and nutritive value grown in three replications at two Wisconsin locations, a testcross of DXL212:N11a-3182-1 had the highest forage yield and milk per acre in the trial (9.35 tons per acre and 23,541 lbs milk per acre). Milk per acre was estimated based on MILK2000 equations (www.wisc.edu/dysci) developed by the University of Wisconsin Agronomy and Dairy Science Departments. A testcross of CUBA117:S1520156 also had excellent nutritional characteristics mostly due to low neutral detergent fiber and high in vitro neutral detergent fiber digestibility. This work was performed by the public cooperator James Coors at the University of Wisconsin. In previous trials, he found two testcross with both excellent yield and quality: a testcross with CUBA164:S15-64-10 and a testcross with CUBA164:S15-184-1. The former had the highest silage yield in that trial, which included check hybrid N4687, one of the highest yielding silage hybrids currently available in the north central region of the US. Furthermore, both GEM topcrosses had above average quality for all traits examined (low NDF, ADF, and high IVTD, IVNDFD, and protein). In particular, the CUBA164:S15-64-10 testcross had excellent digestibility on both whole-plant and fiber basis. † Mark Campbell, a public cooperator at Truman State University, is developing lines from GEM breeding crosses that have been converted with the recessive amylose-extender (ae) allele. His goal is to develop lines that have starch – amylose values at least 65% amylose. Numerous selections from GUAT209:S13 £ (Oh43 £ H99ae) and CUBA110:N1711 £ (Oh43 £ H99ae) have starch –amylose levels at or exceeding 70%. † Wenwei Xu, a public cooperator at Texas A&M in Lubbock, evaluated breeding crosses for insect resistance. His results showed that BVIR103:S04, DKXL380:S08a, DKB830:S19, GUAT209:N19, CUBA117:S15, and CUBA164:S20 may be new sources of corn earworm resistance. † Bruce Hibbard, a public cooperator with USDA-ARS at Columbia, MO, found the breeding cross AR16026:N1210 less damaged than the insecticide control in rootworm evaluations. In European corn borer stalk tunneling evaluations,
80
L. M. POLLAK
51 breeding crosses were less damaged than the resistant check, Mycogen 7250, and in leaf feeding evaluations, 12 breeding crosses were less damaged than the resistant checks Mycogen 7250 and Pioneer Brand Hybrid 3184. Although no significant differences between GEM breeding crosses were found in a Western Corn Rootworm evaluation, all but one cross were nominally less damaged than the susceptible control, B37 £ H84, and 22 crosses were nominally less damaged than the resistant control, NGSDCRW1(C4)S2. † Jerry Sell, a public cooperator at Iowa State University, evaluated experimental high-protein GEM lines in chicken feeding trials. Overall, the data show that the greater protein content of the experimental corn could prove advantageous economically for use in feeds of broiler chickens because of a decrease in the amount of the major protein source (soybean meal) needed in diets containing these corns. Additional research needs to be done with larger supplies of the experimental corns to obtain more definitive information about their feeding value.
VII. CONCLUSIONS A. FACTORS RESPONSIBLE FOR GEM’S SUCCESSFUL PUBLIC/PRIVATE COLLABORATION GEM provides the social returns (agricultural diversity) to justify its public support, and the potential for private returns to justify private participation (Knudson, 2000). One factor in its success seems to be its federal leadership which provides the funding certainty that would be difficult to achieve with university leadership and grant funding. This is critical because the project is too large to manage without funding. Funding certainty by the USDA is assured by the continued private lobbying. The USDA-ARS has incentives to support GEM because its size and national focus are some of the same reasons that the USDA involves itself in intramural research. Federal leadership also ensures secrecy of proprietary information, which private leadership would be unable to guarantee. The GEM developers were a small group of people who all knew each other, trusted each other, and communicated well with each other, so companies trust the coordinator and GEM staff to keep the identity of the proprietary lines in the breeding crosses a secret. GEM protocol is written to give private cooperators confidence that their intellectual and proprietary rights will be safeguarded. Another factor in GEM’s success is its collaboration between the public and private corn breeding sectors. Each sector of maize breeding research can contribute according to their strengths. The federal sector contributes the national
GERMPLASM ENHANCEMENT OF MAIZE
81
focus and funding, the university sector the specialized research projects that benefit regions or states, and the private sector the elite proprietary germplasm, in-kind support, and commercial relevance. The private sector can expect proprietary returns from their investment but yet also treat their enhancement contributions as a social good deed. Increased intellectual property rights may have had a dampening effect on formal and informal information exchange among plant breeders, but information exchange among scientists is critical to creative problem solving. GEM’s organization structure allows for formal and informal information exchange and leads to creative problem solving, in particular, through the TSG and germplasm network functions. In the past, maize germplasm evaluation and enhancement was done on an inefficient piecemeal basis, making it difficult to obtain positive results. Largely because of the lack of positive results, germplasm work was relegated to a low priority. There was little interaction among the few scientists working in the area, and almost no sources of outside funding. Now there is an organized network of scientists interested in maize germplasm, and many chances to exchange ideas. In-kind support has been critical to GEM’s success. Donation of in-kind support reinforces the value the company places on the project, which is useful for lobbying. It also enhances participation of the companies involved, because they have no incentive to having their resources used inefficiently. As the companies are so intimately involved in the project, they may have been more willing to contribute their proprietary germplasm. In-kind support greatly increases the amount of breeding effort devoted to the project, increases the rate of developing potentially useful lines, and increases the number of testing environments needed to identify good lines. Finally, GEM is careful to make sure that only the best germplasm is used. Exotic germplasm has been identified through LAMP, through companies (tropical hybrids), or through breeding programs outside the USA (e.g., CIMMYT). Other exotics used in GEM have undergone screening to identify traits important to maize breeders such as ECB resistance and value-added traits. Corn Belt germplasm included in breeding crosses is the best available: companies’ proprietary inbred lines.
B. EXTENDING GEM’S CONCEPT When the GEM proposal was sent to potential US collaborators in 1994, it was also sent to the LAMP principal investigators asking them to collaborate. The LAMP countries had a large stake in the enhanced materials, because it was due to their efforts that the initial germplasm used by GEM was identified. Although the LAMP countries were interested in participating, it was impossible because of the lack of funding for an international project.
82
L. M. POLLAK
Although LAMP has been completed, the principal investigators meet every two years to exchange ideas and maintain their international research network. At the LAMP meeting in 1996, Dr Wilfredo Salhuana, Technical Director of LAMP, Chair of GEM’s TSG, and Research Fellow at Pioneer Hi-Bred International (retired), presented a proposal for a new international collaborative project named the International Maize Germplasm Enhancement Project (IMGEP) that would build upon LAMP and GEM (Salhuana et al., 1997). GEM breeding crosses would bring participating countries of IMGEP genes for improved yield, agronomic characteristics, and value-added traits. Crosses made with participating international countries’ improved germplasm would benefit GEM. The exchange of germplasm and the joint effort in the selection would amplify the benefit to other sectors or regions of similar environmental conditions and permit selection of traits, besides yield, important to the country. Participation in IMGEP could benefit maize-growing countries of the AsiaPacific region. In a similar manner, incorporating the elements that made the public/private collaboration of GEM successful, the concept could be extended to other crops. GEM’s concept has already been extended to the public and private popcorn breeders in the USA. The popcorn group operates an enhancement effort to utilize dent GEM breeding materials and exotic popcorn populations in cooperation with the dent corn GEM. The popcorn breeders and companies are GEM cooperators and meet with the other cooperators at field days and cooperator meetings. The Popcorn Project Leader, Ken Ziegler, serves as an ex oficio member of the GEM TSG. GEM cooperators in Canada and the northern US have been discussing the initiation of a more focused effort in the north with earlier breeding crosses, having found that most of the current GEM breeding crosses are too late for their locations. Formation of new breeding crosses targeted to the northern effort has begun. Some of the starch quality cooperators have also discussed the concept of a more focused effort. One example of the concept of LAMP extended to other crops is the Soybean Asian Variety Evaluation Project, Project SAVE (Manjarrez-Sandoval et al., 1996). This project evaluates modern Asian soybean varieties for their potential as sources of new yield genes for US soybean breeding. Collaborators include USDA-ARS and land-grant universities, with support provided by the United Soybean Board, a commodity group. No single organization, nation, or region has the capacity to improve its agriculture to its optimum. There is a long history of international collaborations, germplasm exchanges, and mutual interdependence in agriculture. Our future depends not only on breeders but also on the general public and policymakers appreciating our mutual interdependence. As scientists, we must take the lead in developing collaborations that will safeguard our future.
GERMPLASM ENHANCEMENT OF MAIZE
83
C. GEM’S FUTURE GEM is a project designed to keep generating variability. The future of GEM is to continue contributing new germplasm with yield and other characteristics necessary for the demands of new markets. Since needs of farmers and industry change with time, it is necessary to start looking into new genetic resources that may have the desired characteristics for their needs. Increasing productivity and quality, insect and disease resistance, tolerance to stress conditions, and additional traits that add value to the grain (starch, protein, oil, etc.) are characteristics that need to be improved in the future. We anticipate that new technology will facilitate transfer of useful genes from unadapted germplasm to elite material, and will help in searching the world’s genetic resources for valuable traits.
REFERENCES Abel, C. A., and Wilson, R. L. (2000). Evaluation of 11 maize populations from Peru for mechanisms of resistance to leaf feeding by European corn borer. J. Kansas Entomol. Soc. 72, 149–159. Abel, C. A., Wilson, R. L., and Robbins, J. C. (1995). Evaluation of Peruvian maize for resistance to European corn borer (Lepidoptera: Pyralidae) leaf feeding and ovipositional preference. J. Econ. Entomol. 88, 1044– 1048. Abel, C. A., Berhow, M. A., Wilson, R. L., Binder, B. F., and Hibbard, B. E. (2000a). Evaluation of conventional resistance to European corn borer (Lepidopera: Crambidae) and western corn rootworm (Coleoptera: Chrysomelidae) in experimental maize lines developed from a backcross breeding program. J. Econ. Entomol. 93, 1814–1821. Abel, C. A., Wilson, R. L., Wiseman, B. R., White, W. H., and Davis, F. M. (2000b). Conventional resistance of experimental maize lines to corn earworm (Lepidoptera: Noctuidae), fall armyworm (Lepidoptera: Noctuidae), southwestern corn borer (Lepidoptera: Crambidae), and sugarcane borer (Lepidoptera: Crambidae). J. Econ. Entomol. 93, 982 –988. Abel, C. A., Pollak, L. M., Salhuana, W., Widrlechner, M. P., and Wilson, R. L. (2001). Registration of GEMS-0001 maize germplasm resistant to leaf blade, leaf sheath, and collar feeding by European corn borer. Crop Sci. 41, 1651–1652. Anonymous, (1991). In “Managing Global Genetic Resources. The US National Plant Germplasm System”. Committee on Managing Global Genetic Resources: Agricultural imperatives. National Academy Press, Washington, DC. Ayad, G., Toll, J., and Esquinas-Alcazar, J. T. (1980). Maize. In “Directory of Germplasm Collections. III. Cereals”. International Board for Plant Genetic Resources, Rome. Binder, B. F., Robbins, J. C., Wilson, R. L., Abel, C. A., and Hinz, P. N. (1999). Effects of Peruvian maize extracts on growth, development, and fecundity of the European corn borer. J. Chem. Ecol. 25, 1281–1294. Borenstein, B. (1991). Trans fatty acids in the US diet: Consumption and safety. In “Foods Oils and Fats” (C. I. Beck Ed.), pp. 1– 33. Stratecon, Winston-Salem, NC. Brown, W. L. (1975). A broader germplasm base in corn and sorghum. Ann. Corn Sorghum Res. Conf. Proc. 30, 81–89. Campbell, M. R., White, P. J., and Pollak, L. M. (1995). Properties of sugary-2 maize starch: Influence of exotic background. Cereal Chem. 72, 389–392.
84
L. M. POLLAK
Church, D. C., and Nipper, W. A. (1984). High-energy feedstuffs. In “Livestock Feeds and Feeding” (D. C. Church Ed.), pp. 107 –132. O & B Books, Inc., Corvallis, OR. Compton, W. A., Mumm, R. F., and Mathema, B. (1979). Progress from adaptive mass selection in incompletely adapted maize populations. Crop Sci. 19, 531 –533. Crossa, J., and Gardner, C. O. (1987). Introgression of an exotic germplasm for improving an adapted maize population. Crop Sci. 27, 187–190. Darrah, L. L., and Zuber, M. S. (1985). United States farm maize germplasm base and commercial breeding strategies. Crop Sci. 26, 1109–1113. Dunlap, F., White, P. J., and Pollak, L. M. (1995). Evaluation of Latin American Maize Project accessions for fatty acid content. J. Am. Oil Chem. Soc. 72, 989 –993. Dupont, J., White, P. J., and Feldman, E. B. (1991). Saturated and hydrogenated fats in food in relation to health. J. Am. Coll. Nutr. 10, 577–592. Eberhart, S. A. (1971). Regional maize diallels with US and semi-exotic varieties. Crop Sci. 11, 911–914. Eberhart, S. A., Salhuana, W., Sevilla, R., and Taba, S. (1995). Principles for tropical maize breeding. Maydica 40, 339–355. Eckhoff, S. R., and Paulsen, M. R. (1996). Maize. In “Cereal Grain Quality” (R. J. Henry and P. S. Kettlewell, Eds.), pp. 77–112. Chapman & Hall, London. Ertl, D. S., and Orman, B. A. (1994). Protein, oil and starch composition of corn hybrids sampled from 1987 to 1993. Agronom. Abst. 198. Frey, K. (1996). “National Plant Breeding Study—1.” Special Report 98. Iowa Agric. Home Econ. Exp. Sta. Geadelmann, J. L. (1984). Using exotic germplasm to improve Northern corn. Ann. Corn Sorghum Res. Conf. Proc. 39, 98–110. GEM. (2000). “Germplasm enhancement of maize 1997, 1998, and 1999 yield trial data.” CD-ROM. Genter, C. F. (1976). Mass selection in a composite of intercrosses of Mexican races of maize. Crop Sci. 16, 556–558. Goodman, M. M. (1983). In “An evaluation and critique of current germplasm programs”, pp. 195– 249. Pioneer Hi-Bred International, Johnston, IA. Goodman, M. M. (1985). Exotic maize germplasm: Status, prospects, and remedies. Iowa St. J. Res. 59, 497 –527. Goodman, M. M. (1990). Genetic and germplasm stocks worth conserving. J. Hered. 81, 11–16. Hallauer, A. R., and Sears, J. H. (1972). Integrating exotic germplasm into Corn Belt maize breeding programs. Crop Sci. 12, 203–206. Hameed, A., Pollak, L. M., and Hinz, P. N. (1994). Evaluation of Cateto maize accessions for grain yield and physical grain quality traits. Crop Sci. 34, 265–269. Hammes, J. L., and Pendleton, J. W. (1984). Photoperiod-sensitive tropical corn as a potential source of biomass or paper. Agron. J. 76, 159–160. Hammond, E. G. (1991). Organization of rapid analysis of lipids in many individual plants. In “Modern Methods of Plant Analysis. New Series, Vol. 12. Essential Oils and Waxes” (H. F. Jackson and J. F. Jackson, Eds.), Vol. 12, pp. 321. Springer-Verlag, Heidelberg, Germany. Hammond, E. G., and Fehr, W. R. (1984). Progress in breeding for low-linolenic acid soybean oil. In “Biotechnology for the Oils and Fats Industry” (C. Ratledge, P. Dawson, and J. Rattray, Eds.), pp. 89. American Oil Chemists’ Society, Champaign, IL. Hill, L. D. (1981). Quality problems in exporting US corn. Ann. Corn Sorghum Res. Conf. Proc. 36, 191–198. Hodge, J. E. (1982). Food and feed uses of corn. CRC Handbook of Processing and Utilization in Agriculture. Part 1 Plant Products (I. A. Wolff, Ed.), Vol. II, pp. 79 –87. CRC Press, Boca Raton, FL. Jellum, M. D. (1970). Plant introductions of maize as a source of oil with unusual fatty acid composition. J. Agric. Food Chem. 18, 365– 370.
GERMPLASM ENHANCEMENT OF MAIZE
85
Jenkins, M. T. (1978). Maize breeding during the development and early years of hybrid maize. In “Maize Breeding and Genetics” (D. B. Walden, Ed.), pp. 13–28. John Wiley and Sons, New York. Johnson, L. A. (1991). Corn, production, processing, and utilization. In “Handbook of Cereal Science and Technology”, pp. 55–131. Dekker, New York. Kannenberg, L. W. (1984). Utilization of genetic diversity in crop breeding. In “Plant Genetic Resources: A Conservation Imperative” (C. W. Yeatman, K. Kafton, and G. Wilkes, Eds.), pp. 93–109. Westview Press, Boulder, Co. Knudson, M. K. (2000). The research consortia model for agricultural research. In “Public–Private Collaboration in Agricultural Research: New Institutional Arrangements and Economic Implications” (K. Fuglie and D. E. Schimmelpfennig, Eds.), . Iowa St. University Press, Ames, IA. Krieger, K. M., Duvick, S. A., Pollak, L. M., and White, P. J. (1997). Thermal properties of corn starch extracted with different blending methods: Microblender and homogenizer. Cereal Chem. 74, 553 –555. LAMP, (1991). “Catalogo del germoplasma de maı´z, Tomo 1 & 2”. USDA-ARS, Washington, DC. Leibovitz, Z., and Ruckenstein, C. (1983). Our Experiences in Processing Maize (Corn) Germ Oil. J. Am. Oil Chem. Soc. 60, 347A–351A. Longquist, J. H. (1974). Consideration and experience with recombination of exotic and corn belt maize germplasm. Ann. Corn Sorghum Res. Conf. Proc. 29, 102–117. Manjarrez-Sandoval, P., Carter, T. E., Nelson, R. L., and Freestone, R. E. (1996). “Project SAVE, Agronomic Performance of Modern Asian Cultivars in the US”. USDA-ARS, Raleigh, NC. Mensink, R. P., and Katan, M. B. (1990). Effect of dietary trans fatty acids on high-density and lowdensity lipoprotein cholesterol levels in healthy subjects. N. Engl. J. Med. 323, 439 –445. Michelini, L. A., and Hallauer, A. R. (1993). Evaluation of exotic and adapted maize (Zea mays L.) germplasm crosses. Maydica 38, 275–282. Miller, F. R (1979). Utilization of introduced germplasm in crop improvement programs. Paper presented at American Society of Agronomy annual meetings, Fort Collins, CO, August 8, 1979. National Academy of Science (1972). Genetic vulnerability of major crops. “Report of Committee on Genetic Vulnerability of Major Crops.” Agric. Board, Div. Biol. Agric., NAS-NRC, Washington, DC. Ng, K. -Y., Pollak, L. M., Duvick, S. A., and White, P. J. (1997). Thermal properties of sixty-two exotic maize (Zea mays L.) lines grown in two locations. Cereal Chem. 74, 837–841. “Nutrient Content and Feeding Value of Iowa Corn.” (1991). Iowa Corn Growers Association, Iowa Corn Promotion Board, ISU Coopp. Extension Service, ISU Agric. and Home Econ. Exp. Stn. Extension Publication AS 598, 2nd ed. Nutrient Requirements of Swine, 9th ed (1988). National Academy Press, Washington, DC. Orthoefer, F. T. (1994). Corn starch modification and uses. In “Corn: Chemistry and Technology” (S. A. Watson, and P. E. Ramstad, Eds.), pp. 479 –499. American Association of Cereal Chemists, St. Paul, MN. Perry, T. W. (1988). Corn as a livestock feed. In “Corn and Corn Improvement” (G. F. Sprague and J. W. Dudley, Eds.), pp. 941–963. ASA-CSSA-SSA, Madison, WI. Plucknett, D. L., Smith, N. J. H., Williams, J. T., and Anishetty, N. M. (1987). “Gene banks and the world’s food.” Princeton University Press, Princeton, NJ. Pollak, L. M. (1993). Evaluation of Caribbean maize accession in Puerto Rico. Trop. Agric (Trinidad) 70, 8–12. Pollak, L. M. (1997). The US germplasm enhancement of maize project. In “LAMP Final Report” (W. Salhuana, Ricardo Sevilla, and Steva A. Eberhart, Eds.), pp. 111–123, Pioneer Hi-Bred Int., Inc. Johnston, JA Special Publication. Pollak, L. M., and Salhuana, W. (1999). Lines for improved yield and value-added traits—Results from GEM. Ann. Corn Sorghum Ind. Res. Conf. 53, 143–158.
86
L. M. POLLAK
Pollak, L. M., and Salhuana, W. (2001). US Germplasm enhancement for maize project (US-GEM). In “Broadening the Genetic Bases of Crop Production” (H. D. Cooper and C. Spillane, Eds.), pp. 319 –329. IPGRI/FAO, Rome. Pollak, L. M., and White, P. J. (1997). Thermal starch properties in corn belt and exotic corn inbred lines and their crosses. J. Cereal Chem. 74, 412–416. Report on the State of the World’s Plant Genetic Resources (1996). International Technical Conference on Plant Genetic Resources, Leipzig, Germany. Russell, W. A. (1986). Contribution of breeding to maize improvement in the United States, 1920s–1980s. Iowa St. J. Res. 61, 5 –34. W. Salhuana and R. Sevilla (eds) (1995). In “Latin American Maize Project (LAMP), Stage 4 Results from Homologous Areas 1 and 5 (Catalog and CD-ROM)”. National Seed Storage Laboratory, Fort Collins, CO. Salhuana, W., Jones, Q., and Sevilla, R. (1991). The Latin American Maize Project: Model for rescue and use of irreplaceable germplasm. Diversity 7, 40–42. Salhuana, W., Pollak, L. M., and Tiffany, D. (1994). Public/private collaboration proposed to strengthen quality and production of US corn through maize germplasm enhancement. Diversity 9–10, 1–2. Salhuana, W., Sevilla, R., Eberhart, S (1997). Latin American Maize Project (LAMP) Final Report. Pioneer Hi-Bred International Special Publication. Salhuana, W., Pollak, L. M., Ferrer, M., Paratori, O., and Vivo, G. (1998). Agronomic evaluation of maize accessions from Argentina, Chile, The United States, and Uruguay. Crop Sci. 38, 866–872. Sehgal, S. (2000). The seed industry and agricultural biotechnology: Some implications for food production and security. Diversity 16, 32–35. Singh, S. K., Johnson, L. A., Pollak, L. M., and Hurburgh, C. R. (2001a). Compositional, physical, and wet-milling properties of accessions used in the Germplasm Enhancement of Maize Project. Cereal Chem. 78, 330 –335. Singh, S. K., Johnson, L. A., Pollak, L. M., and Hurburgh, C. R. (2001b). Heterosis in compositional, physical, and wet-milling properties of adapted £ exotic corn crosses. Cereal Chem. 78, 36– 341. Singh, S. K., Johnson, L. A., White, P. J., Jane, J. -L., and Pollak, L. M. (2001c). Thermal properties and paste and gel behaviors of starches recovered from accessions used in the Germplasm Enhancement of Maize Project. Cereal Chem. 78, 315– 321. Smith, J. S. C. (1988). Diversity of United States hybrid maize germplasm: Isozymic and chromatographic evidence. Crop Sci. 28, 63 –69. Smith, J. S. C., and Duvick, D. N. (1989). Germplasm collections and the private plant breeder. In “The Use of Plant Genetic Resources” (A. H. D. Brown, O. H. Frankel, D. R. Marshall and J. T. Williams, Eds.), pp. 17–31. Cambridge University Press, Cambridge. Sonntag, N. O. V. (1979). Composition and characteristics of individual fats and oils. In “Bailey’s Industrial Oil and Fat Products” (M. W. Swern, Ed.), pp. 289–478. Wiley, New York. Stevens, D. J., and Elton, G. A. H. (1971). Thermal properties of the starch/water system. Part I. Measurement of heat of gelatinization by differential scanning calorimetry. Starch/Staerke 23, 8– 11. Strecker, L. R., Bieber, M. A., Maza, A., Grossberger, T., and Doskoczynski, W. J. (1996). Corn oil. In Bailey’s Industrial Oil and Fat Products (Y. H. Hui, Ed.), vol. 2, pp. 125– 128. Wiley, New York. Stuber, C. W. (1986). Use of exotic sources of germplasm for maize improvement. In “Breeding of Silage Maize” (O. Dolstra and P. Miedema, Eds.), pp. 19–31. Pudoc, Wageningen, The Netherlands. Tello, F., Alvarez-Tostado, M. A., and Alvarado, G. (1965). A study on the improvement of the essential amino acid balance of corn protein. I. Correlation between racial and variety characteristics and lysine levels of corn. Cereal Chem. 42, 368 –384.
GERMPLASM ENHANCEMENT OF MAIZE
87
USDA (1988). “Crop Production 1987 Summary”. National Agricultural Statistics Service. US Department of Agriculture, Washington, DC. USDA, (eds), (1990). In “Agricultural Statistics”. US Government Printing Office, Washington, DC. USDA, (eds), (1996). In “Agriculture Fact Book”. USDA Office Community, Washington, DC. USDA-ERS, (eds), (1995). In “Feed Situation and Outlook Yearbook, November”. USDA-ERS, Washington, DC. Walsh, J. (1981). Genetic vulnerability down on the farm. Science 214, 161 –164. Watson, S. A. (1977). Industrial utilization of corn. In “Corn and Corn Improvement” (G. F. Sprague, Ed.), pp. 721–763. ASA-CSSA-SSA, Madison, WI. Watson, S. A. (1987). Structure and composition. In “Corn: Chemistry and Technology” (S. A. Watson and P. R. Ramstad, Eds.), pp. 881 –940. American Association of Cereal Chemists, St. Paul, MN. Watson, S. A. (1988). Corn marketing, processing, and utilization. In “Corn and Corn Improvement” (G. F. Sprague and J. W. Dudley, Eds.), pp. 881 –940. ASA-CSSA-SSA, Madison, WI. Weber, E. J. (1987). Lipids of the kernel. In “Corn: Chemistry and Technology” (S. A. Watson and P. E. Ramstad, Eds.), pp. 311–350. American Association of Cereal Chemists, Inc., St. Paul, MN. Wheat, D. (1992). Corn last commodity to see market differentiation. Feedstuffs 34–43. February 17. Whistler, R. L. (1984). History and future expectation of starch use. In “Starch: Chemistry and Technology” (R. L. Whistler, J. N. Bemiller and E. F. Paschall, Eds.), pp. 1–9. Academic Press, Inc., Orlando, FL. White, P. J. (1994). Properties of corn starch. In “Specialty Corns” (A. R. Hallauer, Ed.), pp. 29–54. CRC Press, Inc., Boca Raton, FL. White, P. J., Abbas, I., Pollak, L. M., and Johnson, L. A. (1990). Intra- and interpopulation variability of thermal properties of maize starch. Cereal Chem. 67, 70–73. Wilkes, G. (1984). Germplasm conservation toward the year 2000: Potential for new crops and enhancement of present crops. In “Plant Genetic Resources, A Conservation Imperative” (C. W. Kafton, D. Kafton and G. Wilkes, Eds.), pp. 131–164. Westview Press, Inc., Boulder, CO. Wilkes, G. (1989). Germplasm preservation: Objectives and needs. In “Biotic Diversity and Germplasm Preservation, Global Imperatives” (L. Knutson and A. K. Stoner, Eds.), pp. 13–41. Kluwer Academic Publishers, Dordrecht, The Netherlands. Wilson, R. L., Abel, C. A., Wiseman, B. R., Davis, F. M., Williams, W. P., Barry, B. D., and White, W. H. (1995). Evaluation for multiple pest resistance in European corn borer (Ostrinia nubilalis ) resistant maize accessions from Peru.. J. Kansas Entomol. Soc. 68, 326– 331. Zuber, M. S., Skrdla, W. H., and Choe, B. -J. (1975). Survey of maize selections for endosperm lysine content. Crop Sci. 15, 93–94.
POLYPLOIDY AND THE EVOLUTIONARY HISTORY OF COTTON Jonathan F. Wendel1 and Richard C. Cronn2 1 Department of Botany, Iowa State University, Ames, Iowa 50011, USA Pacific Northwest Research Station, USDA Forest Service, 3200 SW Jefferson Way, Corvallis, Oregon 97331, USA
2
I. Introduction II. Taxonomic, Cytogenetic, and Phylogenetic Framework A. Origin and Diversification of the Gossypieae, the Cotton Tribe B. Emergence and Diversification of the Genus Gossypium C. Chromosomal Evolution and the Origin of the Polyploids D. Phylogenetic Relationships and the Temporal Scale of Divergence III. Speciation Mechanisms A. A Fondness for Trans-oceanic Voyages B. A Propensity for Interspecific Gene Exchange IV. Origin of the Allopolyploids A. Time of Formation B. Parentage of the Allopolyploids V. Polyploid Evolution A. Repeated Cycles of Genome Duplication B. Chromosomal Stabilization C. Increased Recombination in Polyploid Gossypium D. A Diverse Array of Genic and Genomic Interactions E. Differential Evolution of Cohabiting Genomes VI. Ecological Consequences of Polyploidization VII. Polyploidy and Fiber VIII. Concluding Remarks References The cotton genus (Gossypium ) includes approximately 50 species distributed in arid to semi-arid regions of the tropic and subtropics. Included are four species that have independently been domesticated for their fiber, two each in Africa – Asia and the Americas. Gossypium species exhibit extraordinary morphological variation, ranging from herbaceous perennials to small trees with a diverse array of reproductive and vegetative characteristics. A parallel level of cytogenetic and genomic diversity has arisen during the global radiation of the genus, leading to the evolution of eight groups of diploid (n ¼ 13) species (genome groups A– G, and K). The evolutionary history of the genus included multiple episodes of trans139 Advances in Agronomy, Volume 78 2003 Published by Academic Press. All rights of reproduction in any form reserved 0065-2113/02$35.00
140
J. F. WENDEL AND R. C. CRONN oceanic dispersal, invasion of new ecological niches, and a surprisingly high frequency of natural interspecific hybridization among lineages that are presently both geographically isolated and intersterile. Recent investigations have clarified many aspects of this history, including relationships within and among the eight genome groups, the domestication history of each of the four cultivated species, and the origin of the allopolyploid cottons. Data implicate an origin for Gossypium 5– 15 million years ago (mya) and a rapid early diversification of the major genome groups. Allopolyploid cottons appear to have arisen within the last million years, as a consequence of trans-oceanic dispersal of an A-genome taxon to the New World followed by hybridization with an indigenous D-genome diploid. Subsequent to formation, allopolyploids radiated into three modern lineages, including those containing the commercially important species G. hirsutum and G. barbadense. Genome doubling has led to an array of molecular genetic interactions, including inter-locus concerted evolution, differential rates of genomic evolution, inter-genomic genetic transfer, and probable alterations in gene expression. The myriad underlying mechanisms are also suggested to have contributed to both ecological success and agronomic potential. q 2003 Academic Press.
I. INTRODUCTION One of the most remarkable stories in the annals of crop domestication is the origin of cultivated cotton. Perhaps the most striking aspect of this history is that it is global in scope, involving ancient human cultures in both the Old and New Worlds and a convergent or parallel plant domestication process from divergent and geographically isolated wild ancestors. Indeed, cotton is unique among crop plants in that four separate species were independently domesticated (Brubaker et al., 1999a; Brubaker and Wendel, 1994; Percy and Wendel, 1990; Wendel, 1989; Wendel et al., 1992: Wendel et al., 1999) for the specialized single-celled trichomes, or fibers, that occur on the epidermis of the seeds. This parallel domestication process involved four species, two from Americas, Gossypium hirsutum and G. barbadense, and two from Africa – Asia, namely G. arboreum and G. herbaceum. In each of these four cases, aboriginal people discovered several thousand years ago that the unique properties of cotton fibers made them useful for ropes, textiles and other applications. As a consequence, cotton cultivation became increasingly widespread, such that over the millennia cotton became firmly established as the world’s most important fiber crop and an important source of seed oil and protein meal. Each of the four domesticated Gossypium species has its own unique history of domestication, diversification, and utilization. Many aspects of this history have been detailed elsewhere (Brubaker et al., 1999a; Hutchinson, 1951, 1954, 1959; Hutchinson et al., 1947; Wendel, 1995; Wendel et al., 1999), including the various stages in the domestication process, the origin of present patterns of
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
141
genetic diversity, the shape and severity of genetic bottlenecks that accompanied the development of landraces and cultivars, and the influence of recent human history on geographic patterns of cultivation. Although all four cotton species spread far beyond their ancestral homes during the last several millennia, one species, G. hirsutum, recently has come to dominate world cotton commerce, having supplanted the vast majority of cultivation of the other three species. Gossypium hirsutum presently is responsible for over 90% of the annual cotton crop internationally, having spread from its original home in Mesoamerica to over 50 countries in both hemispheres. As the world’s leading textile fiber plant, cotton forms a vital part of global agriculture and is a mainstay of the economy of the United States, as underscored by the fact that the $100 billion/year cotton agriculture and textile industry employs 300,000 Americans and ranks among the largest contributors to the US gross national product. Cotton is grown on about 12 million acres in the United States, more than all crops except maize, wheat, or soybean (Anonymous, 1997). Because of its economic importance, G. hirsutum has attracted considerable scientific interest, not only among plant breeders and agricultural scientists, but also among taxonomists, geneticists, and evolutionary biologists. Concomitant with these efforts has been the recent explosion in our understanding of the organization and structure of eukaryotic genomes. This work, representing the combined efforts of innumerable investigators in diverse disciplines during the last century, has increasingly clarified the evolutionary history of cotton and its genome. Fundamental to this understanding is the realization that both G. hirsutum (Upland cotton) and G. barbadense (Pima cotton, Egyptian cotton) have polyploid genomes, resulting from a truly remarkable chance biological reunion among ancestral diploid genomes that presently are geographically restricted to different hemispheres. One generalization that has emerged from the recent massive effort in genome sequencing and mapping in a diversity of organisms is that genome doubling through polyploidy is a prominent process in plant evolution and has played a major role in the evolution of eukaryotic nuclear genomes (Hughes et al., 2000; Makalowski, 2001; Sidow, 1996; Smith et al., 1999; Spring, 1997; Taylor and Brinkmann, 2001; Wolfe, 2001; Wolfe and Shields, 1997). Polyploidization has been especially active and ongoing in higher plants, with up to 70% of all angiosperms having experienced a relatively recent episode of genome doubling (Grant, 1981; Leitch and Bennett, 1997; Masterson, 1994; Soltis and Soltis, 1993, 1999; Soltis et al., 1992; Stebbins, 1950, 1971). Although there are various types of polyploidy (Grant, 1981; Stebbins, 1971), the most common is allopolyploidy, whereby two differentiated genomes, usually from different species, become reunited in a common nucleus as a consequence of a hybridization event. In the simplest case, allopolyploids have one complete diploid set of chromosomes derived from each parental species, and thus contain a doubled complement of genes (homoeologues). Examples of such polyploids abound and include many of
142
J. F. WENDEL AND R. C. CRONN
the world’s most important agricultural commodities (Hilu, 1993) including cotton. Gossypium hirsutum and G. barbadense are thus classic allopolyploids, resulting from the merger of two formerly isolated diploid genomes. This history may have promoted morphological, ecological, and physiological adaptation, mediated by natural selection on a greatly enhanced level of variation resulting from an instantaneously doubled complement of genes (Fryxell, 1965, 1979; Grant, 1981; Harland, 1936; Ohno, 1970; Otto and Whitton, 2000; Soltis and Soltis, 2000; Stebbins, 1950, 1971; Stephens, 1951a,b). For the same reasons, genome doubling may have offered novel opportunities for agronomic improvement through human selection (Hutchinson et al., 1947; Jiang et al., 1998, 2000b; Wright et al., 1998). The evolutionary history of polyploid cottons is reviewed here, with a focus on the recent insights gleaned from the happy marriage of phylogenetic analysis with genomic investigations. Our intent is to provide a convenient entry point into a burgeoning and dispersed literature, by summarizing evidence bearing on the origin and diversification of the cotton genus and the trans-oceanic voyage that led to the origin and subsequent evolution of the allopolyploids. In addition to conveying an understanding of organismal context and history, we also discuss briefly the evolution of the modern cotton genome. One surprising and recently revealed aspect of allopolyploid plants is that their genomes need not be strictly additive with respect to the genomes of progenitor diploids. Instead, in some cases the merger of two different genomes in a common nucleus is accompanied by considerable genomic reorganization and non-Mendelian genetic behavior (reviewed by Wendel, 2000; Wendel and Liu, 2002). Although the extent and significance of the various phenomena involved are not yet clear, the myriad underlying molecular mechanisms are clearly relevant to cotton. Accordingly, we draw attention to some of the mysteries associated with allopolyploidy, with a focus on novel gene and genome interactions, as well as implications for crop improvement.
II.
TAXONOMIC, CYTOGENETIC, AND PHYLOGENETIC FRAMEWORK
A. ORIGIN AND DIVERSIFICATION OF THE GOSSYPIEAE, THE COTTON TRIBE The origin of polyploid cotton can be fully appreciated within the context of the evolutionary history of the genus and its tribe. For more than a century traditional taxonomic methods have been used to explore the natural affinities of Gossypium, and more recently this knowledge base has been supplemented by modern approaches involving comparative analysis of DNA sequences. A synthesis of these data has led to a reasonably coherent taxonomic concept
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
143
of a group of genera that are aligned into a single small tribe, the Gossypieae. This tribe, which includes only eight genera (Fryxell, 1968, 1979), has traditionally been distinguished from other members of the Malvaceae on the basis of morphological features of the embryo, wood and seed-coat anatomy, and by the presence of the punctae or lysigenous cavities (“gossypol glands”) that are widely distributed throughout the plant body. More recently, the monophyly of the tribe has been confirmed using comparative analyses of chloroplast DNA restriction site variation (LaDuke and Doebley, 1995) and DNA sequence data (Seelanan et al., 1997; Wendel et al., 2002). Four of the eight genera in the Gossypieae are small, with restricted geographic distributions (Fryxell, 1968, 1979). Lebronnecia is a monotypic endemic from the Marquesas Islands. Cephalohibiscus, from New Guinea and Solomon Islands, also contains only a single species and is not found in cultivation. Two species from East Africa and Madagascar are described in Gossypioides. The Hawaiian endemic Kokia includes four species, of which one is extinct. In addition to these four small genera, the tribe includes four moderately sized genera with broader geographic ranges: Hampea comprises of 21 neotropical species, Cienfuegosia includes 25 species with an aggregate range that includes the neotropics and parts of Africa, and 17 species are recognized in the pantropically distributed Thespesia. The largest and most widely distributed genus in the tribe is Gossypium, which contains more than 50 species (Fryxell, 1992), including the four domesticated species described before. These cultivated species embody considerable genetic diversity, but this diversity is dwarfed by that included in the genus as a whole (detailed later), whose species have an aggregate geographic range that encompasses most tropical and subtropical regions of the world. Gossypium is distinguished from related genera by a combination of characters, including: an undivided style, coriacious capsule containing several seeds per locule, a somatic chromosome number of 26, and the presence of three foliaceous (usually) involucellar bracts subtending each flower. As each of these traits is found in related genera, no unique morphological characters define Gossypium. Recent molecular phylogenetic analyses have clarified several aspects of the evolutionary history of the tribe that are particularly germane to an exploration of polyploid cotton (Cronn et al., 2002b; Seelanan et al., 1997; Wendel et al., 2002). Most important has been the formal demonstration that the diverse group of species recognized as belonging to Gossypium do in fact constitute a single natural lineage (a monophyletic group), despite their worldwide distribution and extraordinary morphological and cytogenetic diversity (the latter discussed later). A second significant discovery has been the identity of the closest relatives of Gossypium. All phylogenetic data sets concur in revealing that two genera collectively constitute the phylogenetic sister-group to Gossypium, i.e., the African –Madagascan genus Gossypioides and the Hawaiian endemic genus
144
J. F. WENDEL AND R. C. CRONN
Kokia. These latter genera may thus be used as phylogenetic outgroups for purposes of studying gene and genome evolution within Gossypium. An additional insight that emerges from phylogenetic analysis concerns the temporal component to the genealogy, which is evident in sequence divergence data that serve as a proxy for time. Using sequence divergence data from the chloroplast gene ndhF and published sequence divergence rates to calibrate a molecular clock, Seelanan et al. (1997) suggested that Gossypium branched off from Kokia and Gossypioides approximately 12.5 mya, with the latter two genera becoming separated more recently, perhaps 3 mya. The early estimates are in broad agreement with those recently obtained from a larger data set based on synonymous site divergences calculated from nearly 8000 aligned sites representing 10 different nuclear genes (Cronn et al., 2002b). These latter data show that the mean divergence at synonymous sites between members of Gossypium and its closest relatives (Gossypioides þ Kokia ) is approximately 7.0%, with the latter two genera being about 2.8% divergent at synonymous sites. Although rates of mutation may vary widely among genes and across lineages (Gaut, 1998; Sanderson, 1998), the magnitude of divergence is consistent with a Miocene separation between Gossypium and its closest relatives and a rather recent, perhaps Pliocene divergence between Gossypioides and Kokia. We note that the relatively recent split of genera now geographically isolated from one another by thousands of kilometers of open ocean (Kokia from Hawaii and Gossypioides from Madagascar and East Africa) implies that trans-oceanic dispersal was involved in the evolution of one or both genera. In this respect the Gossypioides –Kokia example represents only the latest in a series of examples of long-distance, oceanic dispersal as a factor in the evolution of the cotton tribe and the genus Gossypium (DeJoode and Wendel, 1992; Fryxell, 1979; Stephens, 1958, 1966; Wendel, 1989; Wendel and Albert, 1992; Wendel and Percival, 1990; Wendel and Percy, 1990).
B. EMERGENCE AND DIVERSIFICATION OF THE GENUS GOSSYPIUM A GLOBAL RADIATION As noted earlier, a remarkable diversification accompanied the global radiation of Gossypium. These morphologies evolved in response to the demands of particular ecological settings and selective environments. Plant habit, for example, ranges from fire-adapted, herbaceous perennials in NW Australia to small trees in SW Mexico that escape the dry season by dropping their leaves. Corolla colors span a rainbow of blue to purple (G. triphyllum ), mauves and pinks (“Sturt’s Desert Rose,” G. sturtianum, is the official floral emblem of the
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
145
Northern Territory, Australia), whites and pale yellows (Mexico, Africa –Arabia) and even a deep sulphur-yellow (G. tomentosum from Hawaii). Seed coverings range from nearly glabrous to the naked eye (e.g., G. klotzschianum and G. davidsonii ), to short stiff, dense, brown hairs that aid in wind-dispersal (G. australe, G. nelsonii ), to long, fine white fibers that characterize highly improved forms of the four cultivated species. There are even seeds that produce fat bodies to facilitate ant-dispersal (section Grandicalyx cottons from NW Australia, Seelanan et al., 1999). Much of this morphological diversity is described in detail by Fryxell (1979). The foregoing discussion suggests that the cotton genus has a history that extends back millions of years, so perhaps it is not surprising that the genus achieved worldwide distribution, with several primary centers of diversity in the arid or seasonally arid tropics and subtropics (Table I). Particularly species-rich regions include Australia, especially the Kimberley region in NW Australia, the Horn of Africa and southern Arabian Peninsula, and the western part of central and southern Mexico. Recognition of these groups of related species and their individual constituents reflects accumulated scientific understanding that has emerged from a long history of basic plant exploration and taxonomic and evolutionary study. The taxonomy of the genus has been summarized in several useful volumes (Fryxell, 1979, 1992; Hutchinson et al., 1947; Saunders, 1961; Watt, 1907). The most recent and widely followed taxonomic treatments are those of Fryxell (1979, 1992), in which species are grouped into four subgenera and eight sections (Table I). This classification system is primarily based on morphological and geographical evidence, although most infrageneric alignments are congruent with cytogenetic and molecular data sets as well, as will be discussed later. At present, Gossypium includes approximately 50 species (Fryxell, 1992), but remarkably, new species continue to be discovered (Fryxell et al., 1992, Stewart,
Table I Diversity and Geographic Distribution of the Major Lineages of Gossypium Genome group A B C D E F G K AD
Number of species 2 3 2 13 7þ 1 3 12 5
Geographic distribution Africa, possibly Asia Africa, Cape Verde Islands Australia Primarily Mexico; also Peru, Galapagos Islands, Arizona Arabian Peninsula, Northeast Africa, Southwest Asia East Africa Australia NW Australia New World tropics and subtropics including Hawaii
146
J. F. WENDEL AND R. C. CRONN
Craven, Brubaker and Wendel, unpublished). It is a truism that the morphological and ecological breadth encompassed by the wild species of Gossypium must have parallels in physiological and chemical diversity. The wild species of cotton, consequently, represent an ample genetic repository for potential exploitation by the cotton breeders. Although these wild species remain a largely untapped genetic resource, examples abound of their productive inclusion in breeding programs (reviewed by McCarty and Percy, 2001). A brief introduction to the major groups of diploid cotton species follows.
AUSTRALIAN SPECIES Australian cottons (subgenus Sturtia ) comprise 16 named species as well as a new species that is yet to be named (Stewart, Craven, Brubaker, and Wendel, unpublished). Collectively, these taxa comprise the C-, G-, and K-genome groups, with two, three, and 12 species, respectively. These three groups of species are implicated by DNA sequence data (Liu et al., 2001b; Seelanan et al., 1997, 1999) to be natural lineages, consistent with their formal alignments into the taxonomic sections Sturtia (C-genome), Hibiscoidea (G-genome), and Grandicalyx (K-genome). Relationships among the three groups, however, remain unclear. Some data place G. robinsonii as basal within the entire assemblage of Australian species (Wendel and Albert, 1992), suggesting that radiation of Gossypium in Australia proceeded eastward from the westernmost portion of the continent. Whether this basal position will withstand the scrutiny of other data sets is an open question, as the most recent analyses (Liu et al., 2001b; Seelanan et al., 1997, 1999) are equivocal in this regard. With respect to the taxonomy within each of the three Australian genome groups, there is little uncertainty for the C- and G-genome groups, as these are well represented in collections and have been thoroughly studied (Fryxell, 1979, 1992; Liu et al., 2001b; Seelanan et al., 1997, 1999; Wendel and Albert, 1992; Wendel et al., 1991, and references therein). Much less certain is the taxonomy of the K-genome species, which are all placed in section Grandicalyx. Recent expeditions to the Kimberley area have enhanced our understanding of diversity within the group and have resulted in the discovery of at least seven new species, six of which have been formally described (Fryxell et al., 1992). These unusual species have a distinctive geography, morphology and ecology, and exhibit a syndrome of features that are characteristic of fire-adaptation. In particular, they are herbaceous perennials with a bi-seasonal growth pattern whereby vegetative growth dies back during the dry season, or as a result of fire, to underground rootstocks that initiate a new cycle of growth with the onset of the next wet season. Species in section Grandicalyx have pedicels that recurve following pollination so that the capsules are pendent and open inverted at maturity,
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
147
releasing sparsely vestitured, ant-dispersed seeds that bear elaiosomes to aid in attracting ants. Many of these species are poorly represented in collections and are incompletely understood from a taxonomic standpoint. Molecular phylogenetic analyses have yielded conflicting results regarding interspecific relationships in this group (Liu et al., 2001b; Seelanan et al., 1999).
AFRICAN – ASIAN SPECIES Fourteen species from Africa and Arabia are recognized in the most recent taxonomic treatment of the genus (Fryxell, 1992), collectively comprising the subgenus Gossypium. The taxonomic section Gossypium contains four subsections whereas section Serrata contains only G. trifurcatum, from deserts in eastern Somalia; this species is poorly understood taxonomically and cytogenetically. The unusual feature of dentate leaves raises the possibility that it may not belong to Gossypium, and may instead be better referred to Cienfuegosia (Fryxell, 1992), a possibility that requires future evaluation. This latter example underscores the provisional nature of much of the taxonomy of the African –Arabian species of Gossypium, which are sorely in need of basic plant exploration and systematic study. Within section Pseudopambak, species recognition and definition are in some cases based on limited material (e.g., G. benadirense, G. bricchettii, G. vollesenii ) and no analyses have been conducted on cytogenetic characteristics or molecular phylogenetic affinities. From a cytogenetic standpoint, the African – Arabian species exhibit considerable diversity, collectively accounting for four of the eight genome groups (A-, B-, E-, and F-). The two cultivated cottons of subsection Gossypium, G. arboreum and G. herbaceum, have been extensively studied (reviewed by Wendel et al., 1989) and comprise the A-genome. The three African species in subsection Anomala comprise the B-genome, as discussed before. The sole F-genome species, G. longicalyx, is cytogenetically distinct (Phillips and Strickland, 1966), morphologically isolated (Fryxell, 1971, 1992; Vollesen, 1987), and is perhaps adapted to more mesic conditions than any other diploid Gossypium species. The remaining species, those of subsection Pseudopambak, are considered to possess E-genomes, although few of these taxa have been sufficiently studied to verify this supposition.
AMERICAN DIPLOID SPECIES Subgenus Houzingenia contains two sections and six subsections, whose species collectively represent the New World D-genome diploids. Given their proximity to American taxonomists, these species have been thoroughly
148
J. F. WENDEL AND R. C. CRONN
collected and studied than most, and consequently their taxonomy is reasonably well understood. The subgenus has also received considerable phylogenetic attention (Cronn et al., 1996; Seelanan et al., 1997; Small and Wendel, 2000b; Wendel and Albert, 1992; Wendel et al., 1995b), which provides strong support for the naturalness of most of the recognized subsections. Evolutionary relationships among the apparently natural subsections are less evident, however, although increasing evidence (Cronn et al., 1996; Liu et al., 2001b; Small and Wendel, 2000b) suggests that G. gossypioides is basal-most within the subgenus (also Cronn and Wendel, unpublished data). As the center of diversity for the 13 species of D-genome diploids is western Mexico, it is likely that the lineage became established and initially diversified in this region. Later range extensions are inferred to have arisen from relatively recent (probably Pleistocene) long-distance dispersals, leading to the evolution of endemics in Peru (G. raimondii ) and the Galapagos Islands (G. klotzschianum ).
C. CHROMOSOMAL EVOLUTION AND THE ORIGIN OF THE POLYPLOIDS GENOME SIZE VARIATION AND THE CONCEPT OF GENOME GROUP As the genus diversified and spread, it underwent extensive chromosomal evolution, which has been studied by many researchers (reviewed by Endrizzi et al., 1985). Chromosome morphology is similar among closely related species, and this is reflected in the ability of related species to form hybrids that display normal meiotic pairing and high F1 fertility. In contrast, crosses among more distant relatives are often difficult or impossible to effect, and those that are successful are characterized by meiotic abnormalities. The collective observations of pairing behavior, chromosome sizes, and relative fertility in interspecific hybrids led to the designation of single-letter genome symbols (Beasley, 1941) for related clusters of species. At present, eight diploid genome groups (A – G and K) are recognized (Endrizzi et al., 1985; Stewart, 1995; Wendel et al., 1999). This cytogenetic partition of the genus is largely congruent with taxonomic and phylogenetic divisions, as discussed later. Although all diploid Gossypium species share the same chromosome number (n ¼ 13), there is more than a threefold variation in DNA content per genome (Bennett et al., 1997; Bennett et al., 1982; Edwards et al., 1974; Edwards and Mirza, 1979; Kadir, 1976; Michaelson et al., 1991), with 2C contents ranging from approximately 2 pg per 2C nucleus in the New World, D-genome diploids to approximately 7 pg per cell in Australian K-genome species (J. McD. Stewart, personal communication). A-genome species have intermediate values of approximately 3.8 pg per 2C nucleus. The range in genome sizes is even greater when other diploid members of the tribe are considered, nearly sevenfold
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
149
variation in DNA content is reported (Wendel et al., 2002) between the largest (T. populnea; 2C ¼ 8.2 pg) and smallest (Gossypioides kirkii and K. drynarioides, each with 2C ¼ 1.2 pg) genomes measured till date in the Gossypieae. This extraordinary variation in DNA content is widely believed to be caused by modification of the repetitive DNA fraction, with relatively little change in the absolute amounts of single-copy DNA (Geever et al., 1989). Increasing evidence indicates that both genome size expansion and contraction are common in evolution, not only in Gossypium and Gossypieae but also in angiosperms as a whole (Wendel et al., 2002).
GENOMIC COMPOSITION OF THE POLYPLOIDS Cytogenetic investigations, starting as early as the 1920s (Denham, 1924), revealed that in addition to species having a haploid complement of 13 chromosomes, Gossypium included taxa with a haploid number of 26. Longley (1933) noted that this doubled chromosome number suggested “a duplication of the chromosomes of an ancestral type.” Webber (1935) commented that the formation of 13 bivalents in a hybrid between wild and cultivated American species “support the hypothesis that the species having 26 pairs are allotetraploids,” and further suggested that the ancestral diploid donors involved “wild American species,. . . and Asiatic species.” This conclusion was also attained by Skovsted (1934, 1937) based on the analyses of chromosome sizes and pairing behavior in interspecific hybrids. Historically important confirmation of the allopolyploid nature of the American tetraploid cotton species emerged from the work of Beasley (1940) and Harland (1940), who synthesized experimental allotetraploids from A-genome (Asiatic) and D-genome (American) diploids and showed that these could form fertile hybrids with natural American tetraploids. These classic cytogenetic studies demonstrated that the American tetraploid species are true allopolyploids and that they contain two resident genomes, an Agenome from Africa or Asia, and a D-genome similar to those found in the American diploids. Additional and conclusive support (reviewed by Endrizzi et al., 1985) for the hypothesis of an allopolyploid origin of the American tetraploids emerged in subsequent decades from diverse sources of evidence, including genetic studies of duplicate factors controlling morphology (Stephens, 1944b, 1951a), meiotic pairing behavior in synthetic polyploids (Gerstel and Phillips, 1958; Phillips, 1963, 1964), phytochemical analysis (Parks et al., 1975), isozyme markers (Saha and Zipf, 1998), comparative genetic mapping (Brubaker et al., 1999b; Reinisch et al., 1994) and comparative analysis of DNA sequences (Cronn et al., 1999; Small et al., 1998; Small and Wendel, 2000a). These latter studies provide particularly compelling proof that the allotetraploid (AD-
150
J. F. WENDEL AND R. C. CRONN
genome) species formed from hybridization between A- and D-genome ancestors: most nuclear genes are duplicated in the AD-genome cottons, and when both copies are isolated and sequenced, they correspond phylogenetically and phenetically to those of the antecedent A- and D-genome diploids.
TAXONOMIC DIVERSITY OF THE POLYPLOIDS Following their initial origin, allopolyploid Gossypium spread and diversified, and at present, five distinct species are recognized. It is ecologically noteworthy that two of these are island endemics that must have originated following longdistance dispersal events. Gossypium darwinii is native to the Galapagos Islands, where it may form large and continuous populations in some areas (Wendel and Percy, 1990). Gossypium tomentosum, from the Hawaiian Islands, has a more diffuse population structure, occurring mostly as scattered individuals and small populations on several islands (DeJoode and Wendel, 1992). A third allopolyploid, G. mustelinum, has an island-like distribution in the sense that it is an uncommon species restricted to a relatively small region of northeast Brazil (Wendel et al., 1994). In addition to these three true wild species, there are two cultivated species (G. barbadense and G. hirsutum ), each of which has a large indigenous range, collectively encompassing a wealth of morphological forms that span the wild-to-domesticated continuum (Brubaker et al., 1999a; Brubaker and Wendel, 1994; Fryxell, 1979; Wendel et al., 1992). Gossypium hirsutum is widely distributed in Central and South America, the Caribbean, and even reaches distant islands in the Pacific (Solomon Islands, Marquesas). Gossypium barbadense has a more southerly indigenous range, centered in the northern third of South America but with a large region of range overlap with G. hirsutum in the Caribbean. Some have recognized a sixth allopolyploid species (Fryxell, 1979), G. lanceolatum ( ¼ G. hirsutum “race palmeri”), which is known only as a cultigen. Brubaker and Wendel (1993) reviewed the evidence that bears on the specific status of this taxon and concluded that G. lanceolatum is more appropriately considered a variant of G. hirsutum.
D. PHYLOGENETIC RELATIONSHIPS AND THE TEMPORAL SCALE OF DIVERGENCE A GENEALOGICAL FRAMEWORK FOR THE GENUS Relationships within and among the various genome groups have been addressed in a number of recent molecular phylogenetic investigations, using variation in cpDNA restriction sites (Wendel and Albert, 1992), DNA sequences for the 5S ribosomal genes and spacers (Cronn et al., 1996), DNA sequence data
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
151
for the cpDNA gene ndhF and the nuclear 5.8S gene and its flanking internal transcribed spacers (Seelanan et al., 1997), sequence analysis of Adh genes (Seelanan et al., 1999; Small and Wendel, 2000b) nuclear introns (Liu et al., 2001b) and most recently, by extensive analyses of DNA sequence data for 16 different loci from the nuclear and chloroplast genomes, collectively representing over 17,000 nucleotides per taxon (Cronn et al., 2002b). Each of these studies shows that the genealogical lineages of species are largely congruent with genome designations and geographical distributions. Accordingly, each genome group corresponds to a single natural lineage, and in most cases, these lineages are also geographically cohesive. This information has been embodied in a synthesis of relationships and genome evolution, as summarized in Fig. 1, which is largely based on the thorough molecular phylogenetic analysis of Cronn et al. (2002b). Several aspects of the phylogenetic history of Gossypium bear highlighting. First, there exist four major lineages of diploid species corresponding to three continents: Australia (C-, G-, K-genomes), the Americas (D-genome), and Africa/Arabia (two lineages: one comprising the A-, B-, and F-genomes, and a second containing the E-genome species). Second, the earliest divergence event in the genus separated the New World D-genome lineage from the ancestor of all Old World taxa. Thus, New World and Old World diploids are phylogenetic sister groups. Following this basal-most split in the genus, cottons comprising the Old World lineage was divided into three groups, namely, the Australian cottons (C-, G-, and K-genome species), the African – Arabian E-genome species, and the African A-, B-, and F-genome cottons. Third, the African F-genome clade, which consists of the sole species G. longicalyx, is definitively diagnosed as sister to the A-genome species. This identifies the wild forms most closely related to those first domesticated in the A-genome species G. arboreum and G. herbaceum. Because this relationship is revealed, prospects are raised for ultimately understanding the genetic basis of the origin of modern lint. Fourth, the serendipitous merger of the A- and D-genome at the time of allopolyploid formation represents a chance biological reunion of two genomes descended from the earliest split in the genus. Thus, the two constituent genomes of allopolyploid cotton evolved first in different hemispheres and diverged for millions of years in isolation from one another. Fifth, and finally, the accumulated data indicate that the major lineages of Gossypium were established in relatively rapid succession shortly after the genus originated and diverged from the Kokia – Gossypioides clade. The evolutionary picture thus envisioned is that there was a rapid and global radiation early in the history of the genus, with temporally closely spaced divergence events. There remains some uncertainty regarding several of the earliest branch points. These exist because the phylogenies inferred from different molecular data sets differ with respect to the resolution of the B- and E-genome species groups. As discussed by Cronn et al. (2002b), the failure to robustly resolve the E-genome probably reflects its early divergence from its nearest relatives; rapid
152
J. F. WENDEL AND R. C. CRONN
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
153
radiation translates into phylogenetically short interior branches, which are by definition more difficult to resolve with statistical confidence than longer branches. Perhaps more problematic is the absence of a consistent phylogenetic signal for the B-genome species. Chloroplast DNA data robustly place the B-genome lineage sister to the combined Australian (C þ G)-genome, whereas 12 nuclear loci place the B-genome lineage solidly into an African clade that includes A- and F-genome cottons. On the basis of evolutionary independence of the estimates provided by 12 different nuclear loci, the resolution of the B-genome inferred from analysis of the nuclear data probably more accurately reflects its true history. Taken together, the phylogenetic results suggest that all African –Arabian cottons comprise a single clade, with early and rapid radiations of the E-, B-, and A þ F-genome lineages (Fig. 1).
DATING THE DIVERGENCE EVENTS A temporal component to the evolutionary history shown in Fig. 1 may be developed from a different use of the DNA sequence data, namely sequence divergence amounts. Previous authors have recognized the extensive morphological variation and global distribution of the genus and have suggested that: (1) Gossypium was relatively ancient, with genome groups originating in the Cretaceous, 60– 100 mya (Endrizzi et al., 1985, 1989; Geever et al., 1989) and (2) the present distribution of the New World and Old World lineages reflects divergence arising from the breakup of the Gondwanan supercontinent (Endrizzi et al., 1985, 1989; Saunders, 1961). This hypothesis has been frequently cited even though it lacked support from alternative evidence such as macrofossils or palynological surveys. Indeed, the latter contradicts a Cretaceous diversification of the genus, as the oldest pollen referable to the Malvaceae is Eocene (38 –46 mya) in age (Muller, 1981, 1984). At present, Gossypium fossils are limited to leaf prints
Figure 1 Evolutionary history of Gossypium, as inferred from multiple molecular phylogenetic data sets. The closest relative of Gossypium is a lineage comprising the African–Madagascan genus Gossypioides and the Hawaiian endemic genus Kokia. Following its likely Miocene origin, Gossypium split into three major lineages: the New World diploids (D-genome); the African– Asian clade (A-, B-, E- and F-genomes); and the Australian group (C-, G-, and K-genomes). This global radiation was mediated by several trans-oceanic dispersal events and was accompanied by considerable morphological, ecological, and cytogenetic differentiation (2C genome sizes shown in circles). Interspecific hybridization is implicated in the evolution of approximately one-fourth of the genus. Allopolyploid cottons formed following trans-oceanic dispersal of an A-genome diploid to the Americas, where the new immigrant underwent hybridization, as female, with a native D-genome diploid. Polyploid cotton probably originated during the Pleistocene, with the five modern species representing the descendants of an early and rapid colonization of the New World tropics and subtropics.
154
J. F. WENDEL AND R. C. CRONN
in Hawaiian volcanic sediments dating to ca. 0.4 mya (Woodcock and Manchester, 1998), and fossil pollen has yet to be ascribed to the genus. Molecular data, initially derived from nuclear DNA thermal reannealing experiments (Endrizzi et al., 1989; Geever et al., 1989), and more recently from DNA sequence variation (Cronn et al., 2002b; Seelanan et al., 1997; Wendel and Albert, 1992), have been used to evaluate divergence times among diploid cottons and to estimate the time of formation of the allopolyploid members of the genus. Using thermal stability measures of nucleotide divergence, Geever et al. (1989) provided the first estimate of divergence for members of Gossypium, arguing for a Cretaceous origin of the genus. This suggestion has been refuted using later data and estimates derived from DNA sequence data, where sequence divergence amounts were used in conjunction with divergence rates (estimated from other taxa) to calibrate a Gossypium molecular clock. Direct sequencing of the chloroplast gene ndhF (Seelanan et al., 1997) resulted in a mean difference between Cgenome cottons and other diploid genome groups of 0.86%. Using a molecular rate calibration based on rbcL sequence divergence for other plants, for which the fossil record provides indications of divergence times, the earliest split in Gossypium was calculated to have taken place approximately 12 mya. A more extensive analysis was conducted by Cronn et al. (2002b), who first estimated rates of divergence at approximately 8000 synonymous sites from across multiple nuclear genes and then pegged these divergence amounts to divergence times using divergence rate calibrations from angiosperm Adh genes (see discussion in Cronn et al., 2002b). These calculations yielded results that were in broad agreement with those based on the earlier cpDNA sequence divergence data, suggesting that the genus originated in the last 5– 15 million years, probably in the Miocene. The molecular clock analyses further indicate that the initial split in the genus, namely, separation of the D-genome lineage from the Old World lineage, occurred within the last seven million years. This estimate provides an approximation of the amount of time that the two progenitor genomes (A and D) of allopolyploid cotton evolved in isolation prior to their merger during hybridization and polyploidization. To quantify the concept of “rapid divergence” among the major extant lineages of diploid cotton, Cronn et al. (2002b) used a maximum likelihood approach to estimate that the modern diploid lineages of cottons diverged within a time span of at most two million years. Although there are many sources of error in the calculations such as these, they offer useful approximations of absolute and relative dates of diversification events. Irrespective of the inherent imprecision in these estimates, two aspects of evolutionary history appear incontrovertible based on the phylogenetic results and the molecular dating analyses: first, that the extant Gossypium lineages diversified sufficiently recently that they achieved their global distribution via an evolutionary history involving at least several long-distance, trans-oceanic dispersals (as noted earlier); and second, that the major divergence events occurred on a temporally compressed scale relative to the age of the genus.
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
155
III. SPECIATION MECHANISMS A. A FONDNESS FOR TRANS-OCEANIC VOYAGES The aforementioned dispersal history of the genus is so remarkable that it bears further comment. As noted in Section II-A earlier, the apparent propensity for long-distance dispersal appears to be a characteristic of the entire cotton tribe, and as noted in the previous paragraphs, it certainly has played an important role in diversification of Gossypium itself. Consideration of the phylogeny in a temporal context and in light of plate tectonic history leads to several diagnoses of intercontinental, presumably trans-oceanic voyages. These would include at least one dispersal between Australia and Africa, another to the Americas (probably Mexico) leading to the evolution of the D-genome diploids, and a second, much later colonization of the New World by the A-genome ancestor of the AD-genome allopolyploids. Long-distance dispersal clearly has played an important role not only in diversification of major evolutionary lines but also in speciation within Gossypium genome groups. Examples include dispersals from southern Mexico to Peru (G. raimondii ), from northern Mexico to the Galapagos Islands (G. klotzschianum ), from western South America to the Galapagos Islands (G. darwinii ), from Africa to the Cape Verde Islands (G. capitis-viridis ), and from the neotropics to the Hawaiian Islands (G. tomentosum ) (DeJoode and Wendel, 1992; Wendel and Percival, 1990; Wendel and Percy, 1990). These latter examples, as well as those noted earlier for the origin of Kokia and Gossypioides from a common ancestor in the Pliocene, suggest a common dispersal mechanism of oceanic drift. In this respect, it is satisfying that seeds of many species of Gossypium are tolerant to prolonged periods of immersion in salt water (Stephens, 1958, 1966). It is astonishing that seeds of the Hawaiian endemic cotton, G. tomentosum, are capable of germination after three years immersion in artificial salt water (Fryxell, personal communication). Apparently seeds of some species may retain buoyancy for at least a couple of months, which may be insufficient for trans-oceanic dispersal, perhaps in some cases longdistance dispersal was mediated through natural rafting on floating debris (Stephens, 1966).
B. A PROPENSITY FOR INTERSPECIFIC GENE EXCHANGE In addition to the conventional mechanisms of geographic speciation (Fryxell, 1965, 1979) and speciation promoted by long-distance dispersal, hybridization has played an important role in diversification of Gossypium (Table II). This was first discovered during routine phylogenetic studies of Australian species, where
156
J. F. WENDEL AND R. C. CRONN Table II Species and Lineages of Gossypium that may have Hybrid or Introgressant Ancestries
Species or species group G. aridum G. gossypioides G. bickii G. cunninghamii B-genome: G. anomalum, G. capitis-viridis, G. triphyllum AD-genome allopolyploids: G. barbadense, G. darwinii, G. hirsutum, G. mustelinum, G. tomentosum
References DeJoode (1992) and Wendel and Albert (1992) Wendel et al. (1995b) and Zhao et al. (1998) Wendel et al. (1991) and Liu et al. (2001b) Seelanan et al. (1999) and Wendel and Albert (1992) Cronn et al. (2002b) Reviewed here
molecular markers from the plastid and nuclear genomes were used to document an unusual evolutionary history for G. bickii (Wendel et al., 1991). This species is one of three morphologically similar G-genome cottons (along with G. australe and G. nelsonii ) in section Hibiscoidea. In contrast to expectations based on this taxonomy, the maternally inherited chloroplast genome of G. bickii was shown to be nearly identical to the plastid genome of G. sturtianum, a morphologically distant C-genome species from a different taxonomic section (Sturtia ). In contrast, nuclear markers show the expected relationship, i.e., G. bickii shares a more recent common ancestor with its close morphological allies, G. australe and G. nelsonii, than it does with G. sturtianum. This discrepancy was explained by invoking a bi-phyletic ancestry for G. bickii, whereby G. sturtianum, or a similar species, served as the maternal parent in an ancient hybridization with a paternal donor from the lineage leading to G. australe and G. nelsonii. Interestingly, no G. sturtianum nuclear genes were detected in G. bickii, suggesting that the nuclear genomic contribution of the maternal parent was eliminated from the hybrid or its descendent maternal lineage (see also Liu et al., 2001b). This phenomenon of “cytoplasmic capture” has subsequently been implicated elsewhere in the genus (Table II). It may be that the entire B-genome speciesgroup has an introgressant ancestry, as suggested by Cronn et al. (2002b) in noting the conflicting phylogenetic signal for this clade offered by sequence data from the nuclear and cytoplasmic genomes. Another likely example concerns the K-genome species G. cunninghamii, which perhaps not coincidentally has an unusual morphology and is geographically widely disjunct from its close relatives. This species is restricted to the Cobourg Peninsula, approximately 500 km distant from the Kimberley region where all other K-genome taxa are found. Analogous to G. bickii, the chloroplast genome of G. cunninghamii appears to have been donated by the G. sturtianum lineage, although in this case the hybridization event appears to have been more ancient (Seelanan et al., 1999; Wendel and Albert, 1992).
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
157
A final example of cytoplasmic introgression resulting from hybridization involves G. aridum, one of four species of Mexican cottons that comprise subsection Erioxylon. These four species form a morphologically coherent and distinctive group of small trees whose shared common ancestry is supported, with one exception, by all molecular data sets (Cronn et al., 1996; DeJoode, 1992; Small and Wendel, 2000b; Wendel and Albert, 1992; Wendel et al., 1995b). The exception is that populations of G. aridum from the Mexican state of Colima have a chloroplast genome that is strikingly divergent from that found in the remainder of the species. This alien cytoplasm is inferred to have originated through an ancient hybridization with a member of the Integrifolia subsection, whose two extant species (G. davidsonii and G. klotzschianum ) have geographic ranges (Baja California and the Galapagos Islands, respectively) that are distant from the range of G. aridum. As was the case with G. bickii, the nuclear genome of G. aridum, including the Colima populations, exhibits no evidence of this introgression event. The most recently discovered and mysterious example of interspecific sexual contact in Gossypium is one involving not only cytoplasmic introgression between species but also recombination between diverged nuclear genomes. The species in question is G. gossypioides, the exclusive member of subsection Selera and a taxon with an apparently relictual range, occurring only in several small isolated populations in a single river drainage in Oaxaca, Mexico. Until recently, there was no indication that G. gossypioides had a noteworthy evolutionary history, in that the inferred relationships between G. gossypioides and the other D-genome species, based on comparative gross morphology, cytogenetic data, interfertility relationships, and allozyme analysis were congruent (reviewed by Wendel et al., 1995b). Wendel et al. (1995b), however, showed that the nuclear ribosomal DNA sequences from G. gossypioides are unlike those of any other D-genome taxon in fact, the sequence data implicated extensive recombination with rDNA sequences from A-genome cottons. Subsequent to this finding, G. gossypioides was discovered to contain a variety of repetitive DNAs that are shared with A-genome species but are otherwise unknown among D-genome species (Zhao et al., 1998). Complicating the story even further, recent phylogenetic analysis of DNA sequences reveals that G. gossypioides occupies a basal position within the D-genome clade (Cronn et al., 1996; Liu et al., 2001b; Small and Wendel, 2000b), yet it possesses a chloroplast genome very much like that found in G. raimondii (Wendel and Albert, 1992), which occupies a more terminal phylogenetic position with the D-genome assemblage. Wendel et al. (1995b) attributed these data to an ancient hybridization event, whereby G. gossypioides experienced contact with an A-genome, either at the diploid level, or at the triploid level as a consequence of hybridization with a New World allopolyploid, followed by repeated backcrossing of the hybrid into the G. gossypioides lineage, thereby restoring the single-copy component of the D-nuclear genome. It may be
158
J. F. WENDEL AND R. C. CRONN
that the G. gossypioides lineage was spawned by this process, and then later underwent secondary hybridization with a member of the G. raimondii lineage to acquire its chloroplast genome. Regardless of the details of this mysterious ancestry, this example highlights an utterly unexpected and as yet unexplained history of interspecific recombination. The foregoing examples underscore a remarkable biogeographic feature of the genus, namely, relatively frequent hybridization between geographically widely separated lineages that apparently have no opportunity for contact. It may well be that one or more of the foregoing species (or their ancestors) actually originated through an evolutionary process that was “seeded” by a hybridization event. Although this remains an open question with the diploid species just discussed, there remains no doubt about the allopolyploid species, which exemplify a different outcome of the hybridization process while reflecting this biogeographically wondrous capacity for interspecific travel and genomic reunion.
IV. ORIGIN OF THE ALLOPOLYPLOIDS A. TIME OF FORMATION As discussed in Section IIC, ample evidence establishes that the New World tetraploid cottons are allopolyploids containing one genome similar to those found in the Old World, A-genome diploids and a second genome like those of the New World, D-genome diploids (reviewed by Endrizzi et al., 1985; Wendel, 1989; Wendel et al., 1995b). This implies that the two genomes must have established physical proximity, at least ephemerally, at some time in the past. Because the two parental genomic groups exist in diploid species that presently are half a world apart, a classic botanical mystery emerged: how and when did allopolyploid cotton form? For over 50 years now, this riddle has stimulated interest and speculation. We previously noted that a number of authors have suggested that the Gossypium had an ancient, perhaps Cretaceous or early Tertiary origin (Endrizzi et al., 1985, 1989; Geever et al., 1989; Harland, 1939; Saunders, 1961; Skovsted, 1934; Stebbins, 1947). This idea appears to have been generated largely by the biogeographic realization that the genus has attained a global pattern of distribution during which it acquired impressive morphological variation (Saunders, 1961; Stebbins, 1947). Given these observations, and the present isolation of progenitor diploid genomes in different hemispheres, it seemed rational to suggest that allopolyploid formation was also ancient. According to this hypothesis, hybridization and polyploidization took place prior to the separation of the parental A- and D-genome lineages, which subsequently drifted apart as a consequence of plate tectonic movements. Under this scenario, then,
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
159
allopolyploids originated prior to the rifting of the South American and African continents, in the Cretaceous or perhaps the early Tertiary. At the other end of the spectrum, several authors have been strongly influenced by the occurrence of cultivated species in both the Old and New World, and by the supposition that agronomically advanced fiber probably was developed only once from a truly wild ancestor (Hutchinson, 1959; Hutchinson et al., 1947; Johnson, 1975). These observations were used to invoke a very recent origin of allopolyploids, forwarding a scenario that involved human transfer of an African or Asiatic A-genome cultigen to the New World, followed by deliberate or accidental hybridization with a wild D-genome species. Hutchinson et al. (1947), for example, concluded that allopolyploid cotton originated following human-mediated intercontinental transfer of a cultivated A-genome diploid, which was “. . .carried across the Pacific by man among the seeds of his crop plants and with the tools of his civilization.” According to this hypothesis, allopolyploid cotton first formed in agricultural times, perhaps within the last six millennia. In between these two extremes of a Cretaceous (perhaps 60 – 100 million) and recent (perhaps 6000 years) origin, which, remarkably, vary by four orders of magnitude, are other proposals. Endrizzi et al. (1989) argued for a probable Miocene origin (5 – 18 mya), based on thermal stability measurements in interspecific DNA-DNA hybridization experiments. This calculation was based, however, on the assumption of an early Cretaceous (100 mya) divergence between the parental diploid groups. Phillips (1963), in a review of cytogenetic evidence, including patterns of intra- and intergenomic chromosome differentiation, presents a persuasive case that polyploid cotton originated “in geologically recent times, probably since the start of the Pleistocene.” Fryxell (1965) reached a similar conclusion based on taxonomic and ecological evidence considered in light of our understanding of historical climate change, and stated: “. . .the amphidiploids, which combine at the tetraploid level the germinal lines of two anciently divergent parts of the genus, are relative newcomers in the evolution of the genus” (he further defines this as mid-Pleistocene). More recent experiments have uniformly supported this latter view, namely, that allopolyploid Gossypium originated prior to the evolution of modern humans but relatively recently in geological terms. Specifically, all molecular data sets support a Pleistocene origin, probably within the last two million years (Cronn et al., 1999; Seelanan et al., 1997; Small et al., 1998; Wendel et al., 1989; Zhao et al., 1995). The first estimate explicitly based on DNA sequence data was that of Wendel (1989), who assayed variation in 560 restriction sites in the chloroplast genomes of diploid and allopolyploid species and translated these data into sequence divergence percentages. These estimates permitted divergence times to be calculated by calibrating with rates from other plants; the conclusion reached was that allopolyploids formed in the mid-Pleistocene. A similar conclusion emerged from a later study (Seelanan et al., 1997) based on direct sequencing
160
J. F. WENDEL AND R. C. CRONN
of the cpDNA gene ndhF. Additional support was provided by Zhao et al. (1995), who reported minimal variation in repetitive DNAs among the various polyploid species. DNA sequence data for a variety of nuclear genes has recently become available, and notwithstanding the gene-to-gene variation in rates, a consistent picture has emerged regarding divergence amounts between the diploid genome groups and between the diploids and their orthologous counterparts in the allopolyploids. Cronn et al. (1999), in a study of 16 low-copy nuclear sequences, reported that mean sequence divergence between the A- and D-genome diploids averaged 2.2%, whereas that between the diploids and their counterparts in the allopolyploid averaged 0.68 and 1.05%, respectively, for the A- and D-genome comparisons. The minimum of these latter two values (0.68%) is the relevant one for the present purposes, in that it provides an estimate of the maximum age of Gossypium allopolyploids. Using the formula T ¼ K/2r, where K is the divergence amount and r is the rate of synonymous site divergence for nuclear genes from plants (perhaps 2.6 £ 1029 – 1.5 £ 1028 substitutions/synonymous site/yr—(Gaut, 1998; Koch et al., 2000; Morton et al., 1996)), one may estimate that allopolyploids formed 0.3– 1.3 mya. As the clock calibration utilizes synonymous sites, and because the sequence data include some, presumably more slowly diverging non-synonymous sites, this time estimate is moderately biased downward, perhaps by approximately one-third (Wendel and Cronn, unpublished). Also, given that the generation time is positively correlated with molecular evolutionary rates (Gaut, 1998) and that wild Gossypium species are long-lived perennials, it is likely that the more appropriate end of the rate spectrum to use is the slower rates. Hence, it seems probable that Gossypium allopolyploids formed in the Mid-Pleistocene, ca. 1 –2 mya. Given a Pleistocene origin for allopolyploid cotton species, one may infer that their morphological diversification and spread must have been relatively rapid following polyploidization. Recent phylogenetic analyses have demonstrated that since formation, allopolyploid cottons have radiated into three lineages collectively comprising five species (Cronn et al., 1996; DeJoode and Wendel, 1992; Small et al., 1999; Wendel, 1989; Wendel and Percy, 1990; Wendel et al., 1994, 1995a). The only living descendant of one branch of the first cladogenetic event in the allopolyploids is Gossypium mustelinum (Small et al., 1998; Wendel et al., 1994). The other branch is represented by two species-pairs, each containing one of the two cultivated species (G. barbadense and G. hirsutum ) and an island endemic that originated from long-distance dispersals; thus, G. barbadense is the sister-species of G. darwinii, from the Galapagos Islands (Wendel and Percy, 1990), while G. hirsutum is sister to G. tomentosum, from the Hawaiian Islands (DeJoode and Wendel, 1992). Collectively, these allopolyploid species exhibit great morphological diversity and have a geographic range that encompasses much of Central America and the Caribbean, the Hawaiian Islands, northern South America, and many distant Pacific Islands (Fryxell, 1979). Yet
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
161
molecular data sets indicate low levels of interspecific divergence. For example, in a survey of ca. 7000 nucleotides from introns and spacer sequences from the chloroplast genomes of each allopolyploid, relatively few differences were detected between species (Small et al., 1998 #168; Wendel, 1989). Similarly, low levels of interspecific sequence divergence are reported for nuclear genes (Cronn et al., 1996; DeJoode and Wendel, 1992; Small et al., 1998; Wendel et al., 1994; Wendel et al., 1995a). Collectively, the evidence indicates that there must have been a relatively rapid colonization of the New World tropics by the allopolyploids following their formation.
B. PARENTAGE OF THE ALLOPOLYPLOIDS Since the discovery that allopolyploid Gossypium species contain two genomes whose progenitors presently occur in different hemispheres, investigators have attempted to provide pieces to the puzzle of polyploid origin. One particular focus has been the question of parentage; that is, which of the modern species of A- and D-genome diploids best serve as models for the progenitor genome donors? Over the decades a diverse array of tools have been used in an effort to solve this question, from early studies of comparative morphology and segregation analysis, through cytogenetic, comparative phytochemistry, and protein electrophoretic studies, to modern phylogenetic investigations using DNA sequencing of homologous genes. The history of these efforts thus closely parallels the conceptual and methodological development of the fields of biosystematics and taxonomy. Much of this history was reviewed by Endrizzi et al. (1985), to which the reader is referred for a lucid discussion of the evidence generated up until that time; only a brief synopsis of this older work is needed here. Early efforts to identify the A- and D-genome donors focused on genetic and morphological evidence. Stephens (1944a,b), for example, compared allometric patterns of leaf development in intergenomic hybrids to those of the naturally occurring allopolyploids G. hirsutum and G. darwinii (as G. barbadense var. darwinii ). From these comparisons he offered perhaps the first explicit hypothesis of parentage, and stated (Stephens, 1944b) that “. . .either (G. klotzschianum, its close relative G. davidsonii, or G. raimondii ) in combination with G. arboreum would produce a hybrid showing considerable similarity to present-day New World cottons.” Additional support for the hypothesis of G. raimondii as the D-genome donor emerged from comparative analyses of plant habit and shape, floral features and extrafloral bract morphology in synthetic A £ D amphiploids, and from observations of lint characteristics and vigor of intergenomic hybrids (Hutchinson et al., 1945). A similar conclusion was later reached by Fryxell (1965), based on observations of the lint characteristics of diploid and wild polyploid species.
162
J. F. WENDEL AND R. C. CRONN
Cytogenetic studies also bolstered the view that G. raimondii was a reasonable model of the D-genome donor. Hutchinson et al. (1947) reported on multivalent frequencies in synthetic allopolyploid £ D-genome hexaploids, using five different D-genome species as parents in crosses with either G. hirsutum or G. barbadense. They argued that the exceptionally low fertility in the hexaploids involving G. raimondii arose from a higher frequency of multivalent formation than observed in the other synthetic hexaploids, implicating G. raimondii as closer to the D-genome than the other species tested. This approach, involving comparative analysis of diverse synthetic allohexaploids, became an important and widely used methodology to address the question of polyploid parentage in the subsequent 15 years (reviewed by Endrizzi et al., 1985). Gerstel (1958), for example, studied multivalent frequencies in hexaploids involving both of the two extant A-genome species (G. arboreum and G. herbaceum ) to argue that G. herbaceum was more closely related of the two to the A-genome of the natural allopolyploids. Particularly, clever extensions of the use of synthetic hexaploids involved the study of genetic segregation in testcrosses, under the assumption that the degree of fit to predicted autotetraploid ratios serves as an appropriate proxy for level of relatedness. Phillips (1963, 1964) summarized segregation data for between three and 10 loci in seven synthetic AD £ D hexaploids involving seven different D-genome species. Segregation in crosses involving G. raimondii exhibited segregation ratios closest to the theoretical expectations of autotetraploids, implicating this species as the closest living relative of the original D-genome donor. An additional perspective that became evident from the accumulating cytogenetic and segregation data is that the A-genome of allopolyploid cotton is more similar to that of the A-genome diploids than the D-genome of the allopolyploid is to that of the D-genome diploids. Thus, in synthetic allohexaploids multivalent frequencies are higher and genetic segregation more closely approximates autotetraploid ratios for A-genome chromosomes than D-genome chromosomes (Gerstel and Phillips, 1958; Phillips, 1964). Subsequent data from many sources has confirmed this observation. For example, in a recent survey of amplified fragment length polymorphisms (AFLPs) in diploid and polyploid cottons, 368 polymorphic fragments were observed in a collection of G. barbadense and G. hirsutum cultivars (Abdalla et al., 2001). Of these 368 bands, 143 were shared between at least some of the tetraploids and the A-genome diploids, whereas only 84 were similarly shared with G. raimondii. Cronn et al. (1999) quantified these relationships using 14,705 base pairs of sequence information for 16 nuclear loci isolated from the D-genome diploid G. raimondii, the A-genome diploid G. arboreum (or G. herbaceum ), and the AD-genome tetraploid G. hirsutum. Sequence divergence between the diploids and their corresponding genomes in the allopolyploid were 0.68 and 1.05% for the A- and D-genomes, respectively. Thus, G. arboreum and G. herbaceum may
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
163
be thought of as approximately 50% better model of the progenitor A-genome diploid than G. raimondii of the D-genome diploid. The classical inferences of the parentage of allopolyploid cotton have withstood the scrutiny of time and of other techniques. Thus, seed protein electrophoretic (Cherry et al., 1970, 1972), flavonoid (Parks et al., 1975), and isozyme (Saha and Zipf, 1998) surveys were consistent with or directly corroborated the earlier inferences based on genetic and cytogenetic studies. In the last several years, these studies of phenetics, or similarity, have been supplemented by formal phylogenetic analyses, using explicit methods of genealogical reconstruction. Liu et al. (2001b), for example, showed that G. raimondii is the sister group to a clade composed of all five allopolyploid species, with respect to the D-genome homoeologues of a fatty acid desaturase gene. Four other D-genome species are sequentially basal to this clade, thereby eliminating them as candidates for D-genome donor status. Cladistic analysis of 5S ribosomal DNAs (Cronn et al., 1996) and the internal transcribed spacer region of the 18S – 26S arrays (Seelanan et al., 1997) yield similar results, even when sampling all extant D-genome species (see, however Wendel et al., 1995b). Ongoing studies of other nuclear genes, including two different alcohol dehydrogenase genes (Small et al., 1998; Small and Wendel, 2000a,b), provide unambiguous evidence that G. raimondii is the closest living relative of the ancestral D-genome donor (Wendel, Cronn and Perkey, unpublished). With respect to the A-genome parent, most have considered G. herbaceum to be closer than G. arboreum to the allopolyploid A-genome (Endrizzi et al., 1985). It is important to recognize, however, that G. herbaceum is not the actual progenitor of the polyploid A-genome. This is evidenced by its chromosomal and molecular differentiation from the A-genome of the allopolyploids (Brubaker et al., 1999b; Cronn et al., 1996; Gerstel, 1953; Liu et al., 2001b; Small et al., 1998; Wendel, 1989; Wendel and Albert, 1992). Cytogenetic and comparative mapping studies have revealed that these genomes differ by at least two large translocations. Moreover, and more critically with respect to the issue of parentage, all cladistic analyses of molecular sequence data show that the two extant A-genome species are phylogenetically sister to each other and hence are genealogically equidistant from the A-genome of the allopolyploids (Cronn et al., 1996; Liu et al., 2001b; Wendel, 1989; Wendel and Albert, 1992). This leads to an important and perhaps self-evident truth namely, that the actual parents of the allopolyploids are extinct, and hence that reference to their parentage is more appropriately framed in terms of closest living descendants of the donor species. As noted above, the degree of divergence between the ancestral and modern genomes has been quantified by sequence data. The notion of polyploid parentage is entwined with the biogeography of their formation and is informed by cognizance of their Pleistocene origin. Cytogenetic data, combined with the observation that the only known wild A-genome cotton is African (G. herbaceum subsp. africanum ), has been used to support the suggestion
164
J. F. WENDEL AND R. C. CRONN
that polyploidization occurred following a trans-Atlantic introduction to the New World of a species similar to G. herbaceum. While trans-Atlantic transfer of an A-genome propagule is plausible, Wendel and Albert (1992) raised the possibility of a pre-Pleistocene A-genome radiation into Asia followed by a trans-Pacific, rather than trans-Atlantic dispersal to the Americas. This possibility is supported by the biogeography of the D-genome species, whose center of diversity is western Mexico, and by the occurrence of wild tetraploids in the Pacific region, including Hawaii and the Galapagos Islands. The relatively recent arrival of G. raimondii in Peru also suggests that the initial hybridization event may have taken place in Mesoamerica rather than South America, although other scenarios have been proposed (see discussion in Liu et al., 2001b). We note that present species ranges may be rather different from those that existed at the time that an A-genome seed or seeds managed to find their way to the New World, so by necessity these speculations must be considered tentative. One aspect of the history of the New World polyploid cottons that has become clear is that they all contain Old World cytoplasms. This was first suggested based on the observation that interspecific hybrids between species with these two genome types are more readily effected with the A-genome parent as female (Phillips, 1963). Subsequent analyses of chloroplast (Galau and Wilkins, 1989; Wendel, 1989) and mitochondrial (Small and Wendel, 1999) DNAs confirmed this early suggestion, thereby demonstrating that the seed parent in the initial hybridization event was an African or Asian A-genome taxon. Several authors have proposed that allopolyploids formed more than once, that is, they are polyphyletic (Johnson, 1975; Parks et al., 1975). These suggestions were made not on the basis of definitive evidence but were instead speculations for which the alternative of monophyly cannot be excluded (see discussion in Endrizzi et al., 1984, 1985). Indeed biogeographic considerations, with the progenitor genomes existing in different hemispheres, are such that polyphyly is extraordinarily unlikely, especially in light of the foregoing evidence for a Pleistocene origin and the implied requirement for trans-oceanic dispersal. The question of monophyly versus polyphyly has only recently been addressed using explicit cladistic methods and molecular sequence data. These data demonstrate that all five allopolyploid species possess a cytoplasm descended from a single source (Small and Wendel, 1999; Wendel, 1989), indicating that there was only a single seed parent in the initial hybridization event. Ongoing studies using nuclear (bi-parentally inherited) genes lead to the same conclusion (Wendel, Cronn and Perkey, unpublished). Hence, evidence indicates that allopolyploid cottons formed only once. A final and puzzling aspect of the parentage of the polyploids concerns the relationship and biogeography of G. raimondii and G. gossypioides. As a member of the D-genome diploids, G. raimondii, from Peru, belongs to an evolutionary lineage that is otherwise Mexican, and the species occupies a cladistically derived position within the subgenus. Hence, G. raimondii represents
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
165
the descendant taxon of a relatively recent dispersal to South America. Yet is appear to share a cytoplasm with G. gossypioides (Wendel and Albert, 1992), which is narrowly distributed in the SW Mexican state of Oaxaca. As detailed in Section III-B, the genome of G. gossypioides mysteriously contains a number of repetitive DNAs that are shared with A-genome species but are otherwise unknown among D-genome species (Wendel et al., 1995b; Zhao et al., 1998). As G. gossypioides is the only D-genome diploid that exhibits evidence of genetic “contact” with an A-genome plant, it must have acquired these introgressant genomic components after phylogenetic separation from the lineage leading to G. raimondii. Hence, the long-distance dispersal event that led to an ephemeral presence of an A-genome entity in the New World may have occurred after G. gossypioides diverged from G. raimondii, consistent with other indications of a Pleistocene allopolyploid origin. This evolutionary history raises the possibility that the G. gossypioides lineage was involved in the origin of allopolyploid cotton, as suggested earlier based on morphological considerations (Valicek, 1983). Indeed, A-genome introgression into G. gossypioides and initial allopolyploid formation may have been spatially and temporally associated events, as recently proposed (Wendel et al., 1995b). This scenario, however, is challenged by recent phylogenetic analyses of nuclear genes, which routinely place G. gossypioides as basal within the subgenus, distant from a lineage comprising G. raimondii and the D-genome of the allopolyploids. Thus, there are conflicting signals from the chloroplast and nuclear genomes with respect to the relationship between the latter two species, a full understanding of the parentage of the polyploids requires reconciliation of this incongruence, as well as a more complete understanding of the events that led to intergenomic contamination of G. gossypioides.
V. POLYPLOID EVOLUTION A. REPEATED CYCLES OF GENOME DUPLICATION In Section I, it was noted that polyploidy is a prominent process in plant speciation, having played a role in generating a relatively high percentage of existing angiosperm species diversity. Because genome doubling has been continuing since flowering plants first definitively appeared in the Cretaceous, many if not most angiosperm genomes have experienced several cycles of polyploidization at various times in the past. The most ancient of these historical genome doubling events may be difficult to discern, due to evolutionary restoration of diploid-like chromosomal behavior and/or other genomic changes following polyploidization. Nonetheless, most angiosperms appear to have “paleopolyploid” genomes, which are revealed as such through comparative
166
J. F. WENDEL AND R. C. CRONN
mapping or other approaches (Brubaker et al., 1999b; Devos and Gale, 2000; Gaut and Doebley, 1997; Go´mez et al., 1998; Grant et al., 2000; Lagercrantz, 1998; Muravenko et al., 1998; Paterson et al., 2000; Reinisch et al., 1994; Sossey-Alaouni et al., 1998; Wilson et al., 1999). Throughout the angiosperms, more recent polyploidization events have been superimposed on these more ancient genome doubling events, followed often by additional rounds of diploidization and evolutionary divergence among previously doubled genomic sequences. This cyclic process of duplication and divergence leads to a concept of the modern angiosperm genome as one characterized by a series of nested duplications of varying antiquity, only some of which descend to the present relatively unscathed by subsequent evolutionary disruption. Only the most recent genome duplications, such as that of modern allopolyploid cotton, are likely to be classically recognized as constituting polyploid speciation events. This episodic process of genome doubling is apparent in the genomes of many of our most important crop plants, including Brassica (Lagercrantz, 1998; Lagercrantz and Lydiate, 1996), soybean (Grant et al., 2000; Shoemaker et al., 1996), and many important cereals (Bennetzen and Freeling, 1997; Devos and Gale, 2000; Gaut and Doebley, 1997; Go´mez et al., 1998; Kellogg, 1998; Moore et al., 1995; Wilson et al., 1999). Comparative genetic mapping studies reveal that the genomes of diploid cotton species harbor multiple instances of nested, duplicated chromosome segments, indicating that allotetraploid cotton experienced rounds of polyploidization that are more ancient than the most recent one in the Pleistocene (Brubaker et al., 1999b; Paterson et al., 2000; Reinisch et al., 1994). Hence, present AD-genome allotetraploids appear evolutionarily to be at least paleooctaploid. The suggestion that diploid cotton itself is a paleopolyploid was first made over 70 years ago on the basis of secondary associations that are visible during meiotic metaphase in diploid cotton (Davie, 1933; Lawrence, 1931; Skovsted, 1933, 1937). Lawrence (1931) may have been the first to draw attention to the possibility that this reflects ancient polyploidy, citing observations of Denham (1924). Davie (1933) discussed at length the likelihood of ancient polyploidy, and additionally highlighted the presence of a single pair of chromosomes that was larger than the rest, as would be expected if the ancestral condition was n ¼ 7 and with modern diploid cotton having a haploid complement of 13 chromosomes. Later, Abraham (1940) noted that only “about seven pairs of chromosomes seem to be homologous” in interspecific hybrids between the A-genome species G. arboreum and the E-genome taxon G. stocksii, suggesting that the remaining unpaired chromosomes have a different origin, and that therefore diploid cottons “represent a secondary condition derived from a lower ancestral number.” Saunders (1961) cited genetic evidence, noting that diploid cotton contained duplicate genetic factors (for chlorophyll deficiency, among others), suggesting that this may reflect ancient polyploidy. More recently, BrdU-Hoechst – Giemsa chromosome banding techniques have been applied to diploid Gossypium,
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
167
leading to the recognition of pairs of chromosomes whose banding patterns are sufficiently similar that they were hypothesized to represent homoeologues from an ancient polyploidization event (Muravenko et al., 1998). Muravenko et al. (1998) suggested that “the ancestral cotton genome contained seven homologous chromosome pairs” (presently, n ¼ 13 in diploid cotton). In this respect we note that the entire tribe to which Gossypium belongs (the Gossypieae ) is based on a chromosome number of n ¼ 13, implying that the paleopolyploidization event evidenced in the chromosome banding and genetic mapping data occurred prior to the origin of the tribe, which may be 20 – 40 million years old (Seelanan et al., 1997). Paleopolyploidization may conceivably be even more ancient, perhaps even antedating the origin of the Malvaceae. These studies reveal a history of both ancient and recent polyploidization in Gossypium, events that must have profoundly impacted its morphological, ecological and physiological diversification. Given this central significance, it is of interest to attempt to evaluate the molecular evolutionary consequences of allopolyploidy, and to distinguish the phenomena and processes that might characterize the earliest stages of polyploid formation from those that are responsible for longer-term genomic changes. The most experimentally tractable examples of duplication trace to the most recent allopolyploidization event; this was the Pleistocene merger of two differentiated genomes (A and D) into a single nucleus in only one of the two parental cytoplasms (that of the A-genome parent, Small and Wendel, 1999; Wendel, 1989). Though we know relatively little about many if not most details of polyploidization, the initial stages of biological reunion evidently are molded by an array of molecular genetic mechanisms and processes that collectively lead to polyploid stabilization (Table III). This suite of mechanisms and processes may be rather different, at least in part, from those responsible for longer-term gene and genome evolution in polyploids, such as functional diversification of duplicated genes. In the following we review recent efforts to understand the allopolyploid genome of Gossypium, with a focus on the evolutionary consequences of gene and genome doubling. Table III Molecular Evolutionary Phenomena that Characterize Allopolyploid Evolution in Gossypium Molecular evolutionary phenomenon
References
Chromosome stabilization Enhanced genetic recombination Interlocus concerted evolution Independent evolution of duplicated genes Intergenomic epistasis Unequal expression of homoeologues Intergenomic “horizontal” transfer
Endrizzi et al. (1985) Brubaker et al. (1999b) Wendel et al. (1995a) Cronn et al. (1996); Liu et al. (2001a); Small and Wendel (2000a) and Small et al. (1998) Jiang et al. (2000b) Wendel lab (unpublished) Hanson et al. (1998) and Zhao et al. (1998)
168
J. F. WENDEL AND R. C. CRONN
B. CHROMOSOMAL STABILIZATION Classical cytogenetic evidence indicates that the A- and D-genomes of allopolyploid Gossypium are more distinct from one another than are the descendents of their diploid progenitors (reviewed by Endrizzi et al., 1985). For example, allopolyploid-derived haploids form an average of less than one bivalent per cell at meiotic metaphase, whereas chromosomes in hybrids between extant A- and D-genome diploids average 5.8 and 7.8 bivalents. These and similar observations indicate that natural selection has favored the evolution of mechanisms that promote exclusive bivalent formation in the allopolyploid. Neither the pace at which such mechanisms operate nor their nature are understood, but it seems rational to postulate that selection would be most intense in the first generations following allopolyploid formation, where the fitness cost of unbalanced gametes would be the greatest. Perhaps one component of genome stabilization after polyploidization is reorganization of the two resident genomes so that they are no longer capable of homoeologous pairing. To evaluate this possibility Brubaker et al. (1999b) employed a common set of RFLP probes and created genetic maps, using interspecific F2 progenies for both diploid and allopolyploid cottons (see also Reinisch et al., 1994). Direct comparisons of gene order and synteny among the A- and D-genome maps, as well as those for both genomes of the allopolyploid, permitted direct assessment of the types and magnitudes of structural changes that preceded and followed allopolyploid formation. As expected from the chromosome numbers, 13 sets of homoeologous linkage groups were identified. Map comparisons showed that the two reciprocal translocations that had previously been identified as distinguishing the A-genome diploids from their counterpart chromosomes in the allopolyploid (Endrizzi et al., 1985) arose in the diploid lineage after allopolyploid formation. More importantly, only 19 locus order differences (inversions) were detected among the two diploid and two allotetraploid genomes, and conservation of colinear linkage groups was the rule rather than the exception. Thus, allopolyploidy in Gossypium was not accompanied by extensive chromosomal rearrangement. One implication of this and similar studies (Paterson et al., 2000) is that in many cases gross structural rearrangement may not be a particularly significant aspect of the process of polyploid genome stabilization.
C. INCREASED RECOMBINATION IN POLYPLOID GOSSYPIUM Comparative genetic mapping experiments are useful not only for detecting evolutionary changes in synteny and gene order, but also provide valuable information on recombination. As noted earlier, A-genome diploids have twice the DNA content per cell as D-genome diploids, with a corresponding difference
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
169
in chromosome size. To a large extent, these differences are maintained in allopolyploid Gossypium (Endrizzi et al., 1985; Skovsted, 1934), although DNA content is not quite additive and the A-genome of allopolyploid cotton has chromosomes that are slightly smaller than those of their diploid antecedent (Davie, 1933; Endrizzi et al., 1985). This nearly twofold difference in genome size provides a natural experiment on the relationship between genome size and recombination. These two genomic attributes often are uncoupled (e.g., Ahn et al., 1993; Whitkus et al., 1992), and they appear to be in Gossypium as well. Brubaker et al. (1999b) showed that recombination rates are essentially conserved between the A- and D-diploid genomes; for a common set of markers, the genetic length of these two genomes by only about 6%. Similarly, at the tetraploid level, recombination in the two resident genomes differed by only 5%. This latter result verifies previous reports of a lack of correlation between genome size and total recombination in a particularly satisfying way, in that the two allopolyploid genomes are in the same nucleus, thereby controlling for the myriad life history, population genetic, and ecological covariables that might be suspected of affecting recombination rates. Although there is no significant difference in recombination between genomes that vary in size by a factor of two, at either the diploid or allopolyploid level, Brubaker et al. (1999b) reported increased recombination in allopolyploid cotton. Specifically, the A- and D-genomes of allopolyploid cotton had, for a common set of markers, map lengths that were 52 and 59% higher, respectively, than those of their diploid counterparts. This suggests that polyploidy itself has promoted higher rates of recombination in Gossypium. At present, neither the generality of this conclusion nor the responsible mechanism are understood, but the hypothesis that polyploidy is recombinogenic bears further investigation both in Gossypium and in other systems.
D. A DIVERSE ARRAY OF GENIC AND GENOMIC INTERACTIONS The most immediate and important genomic consequence of allopolyploid formation in Gossypium was simultaneous duplication of all nuclear genes. Many have addressed the potential evolutionary significance of gene duplication (Levin, 1983; Lewis, 1980; Ohno, 1970; Stebbins, 1950; Stephens, 1951a,b). As polyploidy is being so significant in cotton, insight into mechanisms of genomic change in polyploids may ultimately lead to applied benefits (Section VI). Theory suggests that one possible outcome of gene duplication will be relaxation of selection, allowing divergence between the duplicated genes (homoeologues) and the acquisition of new function (Ferris and Whitt, 1979; Hughes, 1994; Hughes et al., 2000; Li, 1985; Lynch and Conery, 2000; Lynch and Force, 2000; Ohno, 1970). Indeed, polyploidization is widely perceived to provide the raw material for the origin of physiological, ecological (Section V)
170
J. F. WENDEL AND R. C. CRONN
and morphological novelty (Barrier et al., 1999; Grant, 1981; Levin, 1983; Lewis, 1980; Schranz and Osborn, 2000; Stebbins, 1950). In this regard Stephens’ comment half a century ago about the possible significance of divergence following the duplication in cotton is noteworthy: “One might expect (still on a priori grounds) that a mechanism in which new functions could be added and the old ones retained would have considerable selective advantage” (Stephens, 1951b). An alternative outcome of gene doubling is that one member of the duplicated gene pair will become silenced and ultimately degenerate as a pseudogene (Lynch and Conery, 2000; Wendel, 2000). Duplicated genes may also maintain their original function, or aggregate function may be partitioned between the two duplicates (Force et al., 1999; Lynch and Force, 2000). From a phylogenetic perspective, these various fates of gene duplication may partially be modeled as shown in Fig. 2. The null hypothesis for sequence evolution in allopolyploids derives from the organismal history, if both duplicated genes evolve independently and at equal rates following allopolyploid formation, then each homoeologue should be phylogenetically sister to its orthologue from a diploid cotton, rather than to the other homoeologue. Similarly, if rates of sequence evolution are similar at the diploid and allopolyploid level, branch lengths for the two A-genome sequences (one from the diploid, the other from the allopolyploid) should be similar, as they should for the two D-genome sequences (Fig. 2, center). The utility of the null hypothesis lies in its falsification; if homoeologous sequences interact, for example (Fig. 2, top), a different tree may be recovered, or if there is strong directional selection or pseudogenization (Fig. 2, bottom), rate inequalities may become evident. Additional possibilities include silencing or loss of one of the duplicated copies (Fig. 2, top right) and “horizontal transfer” of sequences from one genome to the other (Fig. 2, bottom right). Recently there have been several tests of the null expectations of independence and rate equality in allopolyploid Gossypium as well as in other plants (reviewed by Wendel, 2000). Wendel et al. (1995a) demonstrated interaction among the 18S– 26S ribosomal genes that exist at multiple loci in the A- and D-genomes (Ji et al., 1999). Specifically, instead of evolving independently, as expected if they were sequestered in separate genomes of diploid plants, repeats at the different loci in allopolyploid cotton become homogenized to the same sequence (either “A-like” or “D-like”) by one or more processes of concerted evolution (reviewed by Elder and Turner, 1995). In four of the five allopolyploid species, interlocus homogenization has created exclusively D-genome like rDNAs, whereas in G. mustelinum nearly all rDNA repeats have been homogenized to an A-like form. This example showed that since polyploid formation in the Pleistocene, some 3800 repeats, each approximately 10 kb in length, were “overwritten” with the alternative form originating from the other parental genome, probably through unequal crossing over or gene conversion. Moreover, interlocus concerted evolution was bi-directional, operating in different directions in different allopolyploid lineages.
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON A At = Dt
171
D
Gene Conversion
AD-genome allopolyploids 1-2 mya
A
A A-genome diploids
At
Dt
D
Dt
D
Gene loss
D-genome diploids
Null hypothesis
A
At+Dt Dt
D
Common ancestor 5-10 mya Dt A
At
D
Intergenomic transfer
Unequal rates
Figure 2 Phylogenetic history of diploid and allopolyploid Gossypium species (left) and the various possibilities for the evolution of duplicated genes. Allopolyploids are expected to have duplicated copies (At and Dt) of most single-copy and low-copy genes, and duplicated suites of similar repetitive DNAs. In the absence of mutation or selection, homoeologous copies are expected to evolve at equivalent rates and independently of one another, such that they are phylogenetically sister to their counterparts from the progenitor diploids rather than to each other (center). This expectation provides a convenient null hypothesis for diagnosing molecular evolutionary phenomena that accompany genome doubling, such as gene conversion (top), accelerated evolutionary rates (bottom), transfer of sequences between genomes (bottom right), and gene silencing or gene loss (top right). See text for details.
This demonstration that some repeated sequences could interact across genomes in the allopolyploid nucleus led to additional investigations of the scope of the phenomenon. In an analogous study, Cronn et al. (1996) showed that the duplicated arrays of tandemly repeated 5S rDNA genes are not homogenized by concerted evolutionary forces in the allopolyploid, in contrast to the 18S –26S arrays. Similarly, low-copy nuclear genes duplicated by allopolyploidy appear to largely evolve independently of one another in the polyploid nucleus (Cronn et al., 1999; Small and Wendel, 2000a). In fact, to date there has been no
172
J. F. WENDEL AND R. C. CRONN
convincing demonstration of interlocus gene conversion for single-copy or lowcopy nuclear genes in cotton, although PCR-mediated recombinants are often recovered (Cronn et al., 2002a). More generally, the physiological significance of these observations is not clear. Perhaps interlocus interactions of the kind detectable by changes in duplicated gene sequences have not played a particularly prominent role in generating novel physiology, ecology or morphology in the allopolyploids. Perhaps instead other sorts of genomic interactions have been more important. Clues into the nature of these interactions may possibly be evidenced in the results of other recent investigations. In a study of experimental backcross populations between G. hirsutum and G. barbadense, Jiang et al. (2000b) noted large deficiencies of donor parent (G. barbadense ) transmission for some chromosomal regions. They attributed this to epistatic interactions affecting chromatin transmission, a high proportion of which were caused by interactions between alleles contributed by the two genomes. Other data show that dispersed repetitive elements have become mobilized as a consequence of polyploidization in cotton, possibly leading to novel regulatory changes or gene functions. The studies of Zhao et al. (1998) and Hanson et al. (1998) are noteworthy in this respect, using florescent in situ hybridization, they showed that dispersed repetitive sequences that are A-genome-specific at the diploid level have colonized the D-genome at the polyploid level. Similarly, Hanson et al. (1999) showed that a family of copia-like retrotransposable elements “horizontally” transferred across genomes following allopolyploid formation. These and other studies highlight the evolutionary possibility of transposable element spread across genomes following polyploid formation, and raise the possibility that this process has played a role in diversification and adaptation. In Gossypium these intergenomic interactions appear to arise on an evolutionary timescale as opposed to being an immediate consequence of hybridization and polyploidization (Feldman et al., 1997; Liu et al., 1998a,b; Ozkan et al., 2001; Shaked et al., 2001; Song et al., 1995). Liu et al. (2001a) used AFLP analysis to evaluate the extent of fragment additivity in nine sets of newly synthesized allotetraploid and allohexaploid Gossypium. Approximately 22,000 genomic loci were examined, yet fragment additivity was observed in nearly all cases, even when methylation sensitive and insensitive isoschizomers were used. These indications of genomic additivity and epigenetic stasis during allopolyploid formation provide a contrast to recent evidence from several model plant allopolyploids, most notably wheat (Feldman et al., 1997; Liu et al., 1998a,b; Ozkan et al., 2001; Shaked et al., 2001) and Brassica (Song et al., 1995), where rapid and unexplained genomic changes have been reported. In addition, the data contrast with the foregoing account of repetitive DNAs in Gossypium, some of which are subject to non-Mendelian molecular evolutionary phenomena such as interlocus concerted evolution and intergenomic colonization. Collectively, these and other recent studies have drawn attention to the “dynamic” (Soltis and Soltis,
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
173
1995) nature of polyploids, and underscored the relatively poorly understood and sometimes non-Mendelian mechanisms that may characterize gene and genome evolution in polyploids. In addition to evolutionary changes in gene and genome structure, a key component of polyploid evolution concerns the consequences of genome doubling on gene expression. This is a largely unexplored area in Gossypium, although ongoing studies indicate a range of responses from full co-expression to gene silencing (Wendel Lab, unpublished). In older polyploids, it is well documented that there is a slow decay of duplicate gene expression due to the deletional/substitutional processes of pseudogenization (Gottlieb, 1982; Haufler, 1987; Soltis and Soltis, 1989). In addition, hybridization and/or polyploidization may stimulate more rapid epigenetic changes that lead to bursts of transposable element activity (Liu and Wendel, 2000; O’Neill et al., 1998), although as discussed above there is little evidence for this in nascent Gossypium allopolyploids. Nonetheless, genomic incompatibilities of various kinds and epigenetic responses may accompany polyploid formation or stabilization, as has been shown in Arabidopsis (Comai, 2000; Comai et al., 2000; Lee and Chen, 2001) and Brassica (Chen and Pikaard, 1997). It is likely that different mechanisms affect gene expression evolution and gene silencing in nascent versus stabilized allopolyploids, and equally probable that these processes have been evolutionarily significant (Sections V and VI).
E. DIFFERENTIAL EVOLUTION OF COHABITING GENOMES Following the union of two genomes into a single nucleus as a consequence of allopolyploidization it is expected that over time some genes will become mutagenized into pseudogenes whereas others may diverge and acquire new function, as discussed above. On an average, however, one would expect that these and other phenomena that impact the molecular evolution of genes would be equally distributed in the two allopolyploid genomes. This leads to a useful null hypothesis, i.e., evolutionary rates will be equivalent for duplicated homoeologues. A corollary expectation is that both gene copies will accumulate infraspecific diversity at an equivalent rate. In any single case this need not be true, of course, as when there is strong directional selection on one gene copy or pseudogene formation. Nonetheless, the model may be helpful in informing a search for the underlying explanation for differential evolutionary rates or different levels of diversity when these are observed. For example, if one homoeologue becomes pseudogenized while the other remains under purifying selection, then nucleotide diversity is expected to increase in the former locus at a faster rate than in the latter. The fact that duplicated genes reside in the same nucleus greatly simplifies the challenge of isolating potentially important genomic forces from population-level factors that might effect patterns of
174
J. F. WENDEL AND R. C. CRONN
diversity, such as breeding system or effective population size. As populationlevel factors are neutral with respect to the two homoeologues, observed differences in diversity are more easily attributed to genetic or genomic processes. Gossypium allopolyploids offer a powerful model for these explorations, particularly in as much as the two genomes are known to be largely co-linear yet differ in genome size by a factor of two. An early suggestion of unequal evolutionary rate for the A- and D-genomes was stimulated by the observation that synthetic A-genome £ D-genome hybrids can be synthesized exclusively with the A-genome parent as female. Phillips (1963), on noting this, speculated: “If, in the ancestral amphidiploid the D-genome was contained in A-cytoplasm; during “shakedown” of the raw amphidiploid the D-genome might have been genetically and chromosomally more unstable than the A-genome leading to a more rapid genetical and cytological diploidization of the D-genome of the allopolyploid.” This remarkable comment, made nearly four decades ago, has surprisingly found recent support from divergent quarters (Table IV). For example, in a survey of RFLP polymorphism levels detected in allopolyploid cotton, 10% more probes revealed polymorphisms in the D-genome than the A-genome (Reinisch et al., 1994). Similarly, in two independent phylogenetic analyses (Liu et al., 2001b; Small et al., 1998), D-genome sequences in the allopolyploids were found to have longer branches (i.e., faster evolutionary rates) than their homoeologous A-genome sequences. Moreover, inferences of the location of loci controlling quantitative characters repeatedly suggest a higher evolutionary rate in the D-genome than the A-genome. This was shown for fiber-related traits, where 10 of 14 QTLs were located in the D-genome (Jiang et al., 1998), disease resistance, where five of six resistance genes were localized to the D-genome (Wright et al., 1998), and leaf morphology, where 14 of 21 QTL mapped to the D-genome (Jiang et al., 2000a). A more direct test of the null hypothesis of rate equivalence for homoeologous genes is provided by the measures of nucleotide diversity levels. If evolutionary forces are equivalent for duplicated genes, mutations should accumulate Table IV Evidence for Differential Rates of Evolution for the A- and D-Genomes in Allopolyploid Gossypium Type of data or evidence Artificial hybridization data RFLP polymorphism levels Branch lengths in phylogenetic analyses QTL analysis Surveys of nucleotide polymorphism levels
References Phillips (1963) Reinisch et al. (1994) Liu et al. (2001b) and Small et al. (1998) Jiang et al. (1998, 2000) and Wright et al. (1998) Small et al. (1999) and Small and Wendel (2002)
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
175
randomly with respect to homoeologue, and hence in a survey of allelic polymorphism in a sample of individuals, the number of alleles detected should be approximately equal for the two gene copies. This was the approach used by Small et al. (1999) in a study of approximately 1 kb of AdhA sequence for 22 accessions (44 alleles per genome) of G. hirsutum and for five accessions (10 alleles per genome) of G. barbadense. In both allopolyploid species, estimates of nucleotide diversity were higher for AdhA from the D-genome than from the Agenome, by a factor of two or more. In a follow-up study wherein a 1.3 kb section was sequenced of a second alcohol dehydrogenase gene (AdhC ) with a faster overall evolutionary rate, the same conclusion was even more emphatically reached (Small and Wendel, 2002). In a survey of 44 alleles from each genome of G. hirsutum, 24 different alleles were detected for the D-genome homoeologue whereas only seven allelic variants of the A-genome sequence were observed. To evaluate whether this was a species-specific effect, 12 alleles were sequenced from each genome of a second allopolyploid species, G. barbadense. Although diversity levels were lower, the same phenomenon of differential diversity was observed, with three and one alleles detected for the D- and A-genome homoeologues, respectively. These observations collectively suggest that there has been an overall acceleration in evolutionary rate in the D-genome relative to the A-genome of allopolyploid Gossypium. Although this rate enhancement is not always observed (Cronn et al., 1999), the emerging picture is that evolutionary forces operating on the two genomes may be fundamentally different. At present, the responsible forces and underlying molecular mechanisms are obscure, but a logical suggestion is that they are causally connected to the nearly twofold difference in genome size.
VI. ECOLOGICAL CONSEQUENCES OF POLYPLOIDIZATION The foregoing discussion of the genomic and genetic attributes of allopolyploid cotton demonstrates that polyploid formation has led to a diverse array of genetic and genomic responses, including non-Mendelian transmission. The question naturally arises as to the selective consequences of genome duplication: has allopolyploidy stimulated novel adaptation or physiological capacity? A voluminous literature in plants documents the frequency of polyploids in various habitats, their morphological and physiological attributes, and their ecological success relative to diploids (reviewed by Grant, 1981; Soltis and Soltis, 2000; Stebbins, 1950, 1971). One generalization that has emerged from this accumulated body of work is that polyploidy often is associated with broader ecological amplitude and novel evolutionary opportunity, perhaps
176
J. F. WENDEL AND R. C. CRONN
mediated by the increased “buffering” capacity afforded by duplicated genes and the enhanced vigor resulting from the “fixed heterozygosity” of their duplicated genomes. Perhaps, some of the more recently discovered genetic phenomena discussed above also play a role in the success of polyploids, but this direct connection remains to be demonstrated. With respect to Gossypium, allopolyploidy led to the establishment of a new and successful clade as well as the apparent invasion of a new ecological niche. In considering the Pleistocene origin of allopolyploid cotton, Fryxell (1965, 1979) noted that in contrast to the majority of diploid species, allopolyploid species typically occur in coastal habitats, at least those forms that arguably are truly wild (see also Brubaker and Wendel, 1994). Thus, among the five allopolyploid species, two are completely restricted to near coastlines, in that they are island endemics (G. darwinii and G. tomentosum ), and for two others (G. barbadense and G. hirsutum ), wild forms occur disparately in littoral habitats ringing the Gulf of Mexico, northwest South America, and even on distant Pacific Islands. We have previously mentioned this capacity for oceanic dispersal (Fryxell, 1965, 1979; Stephens, 1958, 1966), but here draw attention to the observation that in the case of allopolyploid Gossypium, this dispersal capacity was associated with specialization for establishment in coastal communities. Fryxell (1965, 1979) forwarded the tantalizing suggestion that following initial formation, adaptation of the newly evolved allopolyploid to littoral habitats enabled it to exploit the fluctuating sea levels that characterized the Pleistocene. This ecological innovation is envisioned to have not only permitted the initial establishment of the nascent polyploid lineage, but is also suggested to have provided a means for the rapid dispersal of the salt –water tolerant seeds. By this means, perhaps, the mobile shorelines of the Pleistocene facilitated the exploitation of a new ecological niche, and hence colonization of the New World tropics.
VII. POLYPLOIDY AND FIBER A final consequence of polyploidy is one of paramount agronomic importance, as it concerns fiber. As noted in Section I, four separate species of Gossypium were independently domesticated for their seed hairs. The characteristic that attracted the attention of the earliest domesticators, the seed lint itself, however, evolved only once in the progenitor of all four cotton species. This becomes evident from the account of organismal taxonomy and phylogeny provided in Section II, which highlights the fact that the ancestral condition of Gossypium species is to have seeds with epidermal seed trichomes that typically are short and tightly adherent to the seed. While mature seeds from wild species exhibit great diversity in fiber length, color, and other properties; it has recently been shown that the earliest developmental stages are similar among all species (Applequist
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
177
et al., 2001). Indeed, there is little difference in fiber density or developmental profile for the first several days following anthesis. To identify developmental differences that might account for variation in fiber length and to place these differences in a phylogenetic context, Applequist et al. (2001) conducted scanning electron microscopy of ovules at and near the time of flowering, and generated growth curves for cultivated and wild diploid and allopolyploid species. Trichome initiation was found to be similar in all species, with few notable differences in fiber density or early growth. Developmental profiles of the fibers of most wild species are similar, with fiber elongation terminating at about two weeks post-anthesis. In contrast, growth is extended to 3 weeks in the A- and F-genome diploids. When this observation is considered in light of the phylogeny of the genus (Fig. 1), it becomes clear that this prolonged elongation period represents a key evolutionary event in the origin of long fiber, and that it happened in the common ancestor of these two groups of diploid cottons prior to domestication in Africa. This observation has a fascinating implication; namely, that the domestication of the New World cottons that presently dominate cotton agriculture worldwide was first precipitated by a developmental switch that occurred millions of years ago in a different hemisphere. Analysis of fiber growth curves reveals that domestication itself has been associated with further prolongation of elongation at both the diploid and allopolyploid levels. This provokes the speculation that the effects of parallel artificial selection for long fiber in the four cultivated species resulted in a genetically convergent or parallel transformation in the developmental program that is responsible for this aspect of fiber development. Applequist et al. (2001) further showed that a second evolutionary innovation in fiber morphology was that absolute growth rate is higher in species with long fibers. Thus, wild forms of A-genome diploids, for example, have fiber elongation rates that exceed that of their closest relatives with shorter fiber. A final intriguing observation from Applequist et al. (2001) emerges from noting that fiber growth curves for wild AD-genome allopolyploids are similar to those of the wild A-genome species, but that the fiber of the cultivated allopolyploids is superior to that of the cultivated Old Word diploids. Also, domestication at the allopolyploid level was shown not only to have prolonged the elongation period beyond three weeks, but was also demonstrated to have increased the growth rate in the early stages of trichome expansion. Perhaps, these shifts in developmental profiles were mediated by recruitment of novel expression patterns for D-genome genes, or perhaps novel expression of Agenome genes. It may be well that the genome-wide gene duplication caused by allopolyploidization provided the raw material necessary for the evolution of novel gene expression patterns, which subsequently were exploited by the aboriginal domesticators (and perhaps modern plant breeders) of G. hirsutum and G. barbadense.
178
J. F. WENDEL AND R. C. CRONN
In light of the foregoing, several recent studies (Jiang et al., 1998, 2000b; Wright et al., 1998) have suggested that allopolyploidization provided novel opportunities for agronomic improvement. Jiang et al. (1998), for example, used quantitative genetic approaches (QTL analysis) to study fiber-related traits in an interspecific F2 progeny derived from a cross between G. hirsutum and G. barbadense. They reported that a majority of loci affecting fiber yield and quality are found in the D-genome rather than the A-genome, possibly explaining the superiority of the lint of the allopolyploids. These studies may constitute actual genetic evidence for a speculation forwarded 65 years ago by Harland (1936), who stated “If as a consequence of polyploidy a large number of genes become duplicated, and the characters governed by such genes are of importance to the species, one of the members may mutate, leaving the character unimpaired, with the further possibility that the mutation may be of benefit to the species.” An exciting prospect is that in the near future we will obtain information on the nature of these genes and the mutational or regulatory changes that underlie altered morphology and agronomic improvement.
VIII. CONCLUDING REMARKS In this review we have attempted to provide a synthesis of the evolutionary history of cotton and its genome. Quite remarkably, this history has included multiple episodes of trans-oceanic dispersal and a surprisingly high frequency of natural interspecific hybridization among lineages that presently are both geographically isolated and intersterile. Moreover, the resulting genomic reunions have led to an array of genetic mechanisms and adaptive responses that are not yet fully understood. We note with wonder the many implausibilities and improbabilities involved in this account, which were revealed over the decades by innumerable investigators in diverse disciplines. The exploration of Gossypium and its genome has truly been an interdisciplinary enterprise, enriched by investigations at all levels of biological organization, from molecular to ecological. Indeed, insights into the evolutionary history of Gossypium and its genome have not only been deepened by but also have been dependent upon what we termed in Section I “the happy marriage of phylogenetic analysis with genomic investigations.” It seems likely that additional insights will continue to emerge from this interplay between molecular biology and evolutionary systematics. Our understanding of allopolyploid formation is still in its infancy, as relatively little is known about the myriad consequences of genomic merger and the attendant short-term and long-term effects on patterns of gene expression and development. This by necessity limits our understanding of the effects of gene and genome doubling on morphological or physiological attributes and agronomic potential.
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
179
Powerful new technologies are beginning to be brought to bear on the problem, such as global expression screens using comparative microarray or cDNA –AFLP analysis. As documented in the present account of the history of our understanding of Gossypium, the most profound insights may emerge from a phylogenetically informed implementation of these and related technologies.
REFERENCES Abdalla, A. M., Reddy, O. U. K., El-Zik, K. M., and Pepper, A. E. (2001). Genetic diversity and relationships of diploid and tetraploid cottons revealed using AFLP. Theor. Appl. Genet. 102, 222 –229. Abraham, P. (1940). Cytological studies in Gossypium. I. Chromosome behavior in the interspecific hybrid G. arboreum £ G. stocksii. Ind. J. Agric. Sci. 10, 285–298. Ahn, S., Anderson, J. A., Sorrells, M. E., and Tanksley, S. D. (1993). Homoeologous relationships of rice, wheat and maize chromosomes. Mol. Gen. Genet. 241, 483–490. Anonymous, (1997). In “Agricultural Statistics Board, Annual Crop Summary”. NASS, USDA. Applequist, W. L., Cronn, R. C., and Wendel, J. F. (2001). Comparative development of fiber in wild and cultivated cotton. Evol. Dev. 3, 3–17. Barrier, M., Baldwin, B. G., Robichaux, R. H., and Purugganan, M. D. (1999). Interspecific hybrid ancestry of a plant adaptive radiation: Allopolyploidy of the Hawaiian silversword alliance (Asteraceae) inferred from floral homeotic gene duplications. Mol. Biol. Evol. 16, 1105–1113. Beasley, J. O. (1940). The origin of American tetraploid Gossypium species. Am. Nat. 74, 285–286. Beasley, J. O. (1941). Hybridization, cytology, and polyploidy of Gossypium. Chronica Botanica 6, 394 –395. Bennett, M. D., Smith, J. B., and Heslop-Harrison, J. S. (1982). Nuclear DNA amounts in angiosperms. Proc. R. Soc. Lond. B 216, 179–199. Bennett, M. D., Cox, A. V., Leitch, I. J (1997). Angiosperm DNA c-Values Database, http://www. rbgkew.org.uk/cval/database1.html. Bennetzen, J. L., and Freeling, M. (1997). The unified grass genome: synergy in synteny. Genome Res. 7, 301–306. Brubaker, C. L., and Wendel, J. F. (1993). On the specific status of Gossypium lanceolatum Todaro. Genet. Res. Crop Evol. 40, 165–170. Brubaker, C. L., and Wendel, J. F. (1994). Reevaluating the origin of domesticated cotton (Gossypium hirsutum; Malvaceae) using nuclear restriction fragment length polymorphisms (RFLPs). Am. J. Bot. 81, 1309–1326. Brubaker, C. L., Bourland, F. M., and Wendel, J. F. (1999a). The origin and domestication of cotton. In “Cotton: Origin, History, Technology and Production” (C. W. Smith and J. T. Cothren, Eds.), pp. 3–31. Wiley, New York. Brubaker, C. L., Paterson, A. H., and Wendel, J. F. (1999b). Comparative genetic mapping of allotetraploid cotton and its diploid progenitors. Genome 42, 184 –203. Chen, Z. J., and Pikaard, C. S. (1997). Transcriptional analysis of nucleolar dominance in polyploid plants: Biased expression/silencing of progenitor rRNA genes is developmentally regulated in Brassica. Proc. Natl. Acad. Sci. USA 94, 3442–3447. Cherry, J. P., Katterman, F. R. H., and Endrizzi, J. E. (1970). Comparative studies of seed proteins of species of Gossypium by gel electrophoresis. Evolution 24, 431 –447. Cherry, J. P., Katterman, F. R. H., and Endrizzi, J. E. (1972). Seed esterases, leucine aminopeptidases and catalases of species of the genus Gossypium. Theor. Appl. Genet. 42, 218–226. Comai, L. (2000). Genetic and epigenetic interactions in allopolyploid plants. Plant Mol. Biol. 43, 387 –399.
180
J. F. WENDEL AND R. C. CRONN
Comai, L., Tyagi, A. P., Winter, K., Holmes-Davis, R., Reynolds, S. H., Stevens, Y., and Byers, B. (2000). Phenotypic instability and rapid gene silencing in newly formed Arabidopsis allotetraploids. Plant Cell 12, 1551–1567. Cronn, R. C., Zhao, X., Paterson, A. H., and Wendel, J. F. (1996). Polymorphism and concerted evolution in a tandemly repeated gene family: 5S ribosomal DNA in diploid and allopolyploid cottons. J. Mol. Evol. 42, 685–705. Cronn, R., Small, R. L., and Wendel, J. F. (1999). Duplicated genes evolve independently following polyploid formation in cotton. Proc. Natl. Acad. Sci. USA 96, 14406–14411. Cronn, R. C., Cedroni, M., Haselkorn, T., Grover, C., and Wendel, J. F. (2002a). PCR-mediated recombination in amplification products derived from polyploid cotton. Theor. Appl. Genet. (in press). Cronn, R. C., Small, R. L., Haselkorn, T., and Wendel, J. F. (2002b). Rapid diversification of the cotton genus (Gossypium: Malvaceae) revealed by analysis of sixteen nuclear and chloroplast genes. Am. J. Bot. 89, 707 –725. Davie, J. H. (1933). Cytological studies in the Malvaceae and certain related families. J. Genet. 28, 33–67. DeJoode, D. R (1992). Molecular insights into speciation in the genus Gossypium L (Malvaceae). MS Thesis, Iowa State University, Ames. DeJoode, D. R., and Wendel, J. F. (1992). Genetic diversity and origin of the Hawaiian Islands cotton, Gossypium tomentosum. Am. J. Bot. 79, 1311–1319. Denham, H. J. (1924). The cytology of the cotton plant. II. Chromosome numbers of old and new world cottons. Ann. Bot. 38, 433– 438. Devos, K. M., and Gale, M. D. (2000). Genome relationships: the grass model in current research. The Plant Cell 12, 637 –646. Edwards, G. A., and Mirza, M. A. (1979). Genomes of the Australian wild species of cotton. II. The designation of a new G genome for Gossypium bickii. Can. J. Genet. Cytol. 21, 367–372. Edwards, G. A., Endrizzi, J. E., and Stein, R. (1974). Genomic DNA content and chromosome organization in Gossypium. Chromosoma 47, 309–326. Elder, J. F., and Turner, B. J. (1995). Concerted evolution of repetitive DNA sequences in eukaryotes. Q. Rev. Biol. 70, 297 –320. Endrizzi, J. E., Turcotte, E. L., and Kohel, R. J. (1984). Cotton, Agronomy Monograph No. 24. In “Qualitative genetics, cytology, and cytogenetics”, pp. 81 –129. ASA–CSSA–SSSA, Madison. Endrizzi, J. E., Turcotte, E. L., and Kohel, R. J. (1985). Genetics, cytogenetics, and evolution of Gossypium. Adv. Genet. 23, 271–375. Endrizzi, J. E., Katterman, F. R. H., and Geever, R. F. (1989). DNA hybridization and the time of origin of three species of Gossypium. Evol. Trends Plants 3, 115 –119. Feldman, M., Liu, B., Segal, G., Abbo, S., Levy, A. A., and Vega, J. M. (1997). Rapid elimination of low-copy DNA sequences in polyploid wheat: A possible mechanism for differentiation of homoeologous chromosomes. Genetics 147, 1381–1387. Ferris, S. D., and Whitt, G. S. (1979). Evolution of the differential regulation of duplicate genes after polyploidization. J. Mol. Evol. 12, 267–317. Force, A., Lynch, M., Pickett, F. B., Amores, A., Yan, Y. -L., and Postlethwait, J. (1999). Preservation of duplicate genes by complementary, degenerative mutations. Genetics 151, 1531–1545. Fryxell, P. A. (1965). Stages in the evolution of Gossypium. Adv. Front. Plant Sci. 10, 31 –56. Fryxell, P. A. (1968). A redefinition of tribe Gossypieae. Bot. Gazette 129, 296– 308. Fryxell, P. A. (1971). Phenetic analysis and the phylogeny of the diploid species of Gossypium L (Malvaceae). Evolution 25, 554– 562. Fryxell, P. A. (1979). In “The Natural History of The Cotton Tribe”. Texas A&M University Press, College Station, TX. Fryxell, P. A. (1992). A revised taxonomic interpretation of Gossypium L (Malvaceae). Rheedea 2, 108–165.
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
181
Fryxell, P. A., Craven, L. A., and Stewart, J. M. (1992). A revision of Gossypium sect. Grandicalyx (Malvaceae), including the description of six new species. Syst. Bot. 17, 91 –114. Galau, G. A., and Wilkins, T. A. (1989). Alloplasmic male sterility in AD allotetraploid Gossypium hirsutum upon replacement of its resident A cytoplasm with that of D species G. harknessii. Theor. Appl. Genet. 78, 23 –30. Gaut, B. S. (1998). Molecular clocks and nucleotide substitution rates in higher plants. Evolutionary Biology (M. K. Hecht Ed.), vol. 30, pp. 93–120. Plenum Press, New York. Gaut, B. S., and Doebley, J. F. (1997). DNA sequence evidence for the segmental allotetraploid origin of maize. Proc. Natl. Acad. Sci. USA 94, 6808–6814. Geever, R. F., Katterman, F. R. H., and Endrizzi, J. E. (1989). DNA hybridization analyses of a Gossypium allotetraploid and two closely related diploid species. Theor. Appl. Genet. 77, 553 –559. Gerstel, D. U. (1953). Chromosomal translocations in interspecific hybrids of the genus Gossypium. Evolution 7, 234 –244. Gerstel, D. U. (1958). Chromosomal translocations in interspecific hybrids of the genus Gossypium. Evolution 7, 234 –244. Gerstel, D. U., and Phillips, L. L. (1958). Segregation of synthetic amphiploids in Gossypium and Nicotiana. Cold Spring Harbor Sympp. Quant. Biol. 23, 225 –237. Go´mez, M. I., Islam-Faridi, M. N., Zwick, M. S., Czeschin, D. G., Hart, G. E., Wing, R. A., Stelly, D. M., and Price, H. J. (1998). Tetraploid nature of Sorghum bicolor (L.) Moench. J. Hered. 89, 188 –190. Gottlieb, L. D. (1982). Conservation and duplication of isozymes in plants. Science 216, 373 –380. Grant, V. (1981). In “Plant Speciation”. Columbia University Press, New York. Grant, D., Cregan, P., and Shoemaker, R. C. (2000). Genome organization in dicots: Genome duplication in Arabidopsis and synteny between soybean and Arabidopsis. Proc. Natl. Acad. Sci. USA 97, 4168– 4173. Hanson, R. E., Zhao, X. -P., Islam-Faridi, M. N., Paterson, A. H., Zwick, M. S., Crane, C. F., McKnight, T. D., Stelly, D. M., and Price, H. J. (1998). Evolution of interspersed repetitive elements in Gossypium (Malvaceae). Am. J. Bot. 85, 1364–1368. Hanson, R. E., Islam-Faridi, M. N., Crane, C. F., Zwick, M. S., Czeschin, D. G., Wendel, J. F., Mcknight, T. D., Price, H. J., and Stelly, D. M. (1999). Ty1-copia-retrotransposon behavior in a polyploid cotton. Chromosome Res. 8, 73 –76. Harland, S. C. (1936). The genetical conception of the species. Camb. Philos. Soc. Biol. Rev. 11, 83 –112. Harland, S. C. (1939). In “The Genetics of Cotton”. Jonathan Cape, London. Harland, S. C. (1940). New allopolyploids in cotton by the use of colchicine. Tropp. Agric. 17, 53–55. Haufler, C. H. (1987). Electrophoresis is modifying our concepts of evolution in homosporous pteridophytes. Am. J. Bot. 74, 953– 966. Hilu, K. W. (1993). Polyploidy and the evolution of domesticated plants. Am. J. Bot. 80, 1494–1499. Hughes, A. L. (1994). The evolution of functionally novel proteins after gene duplication. Proc. R. Soc. Lond. B 256, 119– 124. Hughes, A. L., A, G. J., Garbayo, J. M., and Roberts, R. M. (2000). Adaptive diversification within a large family of recently duplicated, placentally expressed genes. Proc. Natl. Acad. Sci. USA 97, 3319–3323. Hutchinson, J. B. (1951). Intra-specific differentiation in Gossypium hirsutum. Heredity 5, 161 –193. Hutchinson, J. B. (1954). New evidence on the origin of the Old World cottons. Heredity 8, 225–241. Hutchinson, J. B. (1959). In “The Application of Genetics to Cotton Improvement”. Cambridge University Press, Cambridge. Hutchinson, J. B., Stephens, S. G., and Dodds, K. S. (1945). The seed hairs of Gossypium. Ann. Bot. 9, 361 –368.
182
J. F. WENDEL AND R. C. CRONN
Hutchinson, J. B., Silow, R. A., and Stephens, S. G. (1947). In “The Evolution of Gossypium and the Differentiation of the Cultivated Cottons”. Oxford University Press, London. Ji, Y., De Donato, M., Crane, C. F., Raska, W. A., Nurul Islam-Faridi, M. N., McKnight, T. D., Price, H. J., and Stelly, D. M. (1999). New ribosomal RNA gene locations in Gossypium hirsutum mapped by meiotic FISH. Chromosoma 108, 200–207. Jiang, C., Wright, R., El-Zik, K., and Paterson, A. H. (1998). Polyploid formation created unique avenues for response to selection in Gossypium (cotton). Proc. Natl. Acad. Sci. USA 95, 4419–4424. Jiang, C., Wright, R. J., Woo, S. S., DelMonte, T. A., and Paterson, A. H. (2000). QTL analysis of leaf morphology in tetraploid Gossypium (cotton). Theor. Appl. Genet. 100, 409 –418. Jiang, C. -X., Chee, P. W., Draye, X., Morrell, P. L., Smith, C. W., and Paterson, A. H. (2000). Multilocus interactions restrict gene introgression in interspecific populations of polyploid Gossypium (cotton). Evolution 54, 798–814. Johnson, B. L. (1975). Gossypium palmeri and a polyphyletic origin of the New World cottons. Bull. Torrey Bot. Club 102, 340–349. Kadir, Z. B. A. (1976). DNA evolution in the genus Gossypium. Chromosoma 56, 85– 94. Kellogg, E. A. (1998). Relationships of cereal crops and other grasses. Proc. Natl. Acad. Sci. USA 95, 2005–2010. Koch, M. A., Haubold, B., and Mitchell-Olds, T. (2000). Comparative evolutionary analysis of chalcone synthase and alcohol dehydrogenase in Arabidopsis, Arabis, and related genera (Brassicaceae). Mol. Biol. Evol. 17, 1483–1498. LaDuke, J. C., and Doebley, J. F. (1995). A chloroplast DNA based phylogeny of the Malvaceae. Syst. Bot. 20, 259– 271. Lagercrantz, U. (1998). Comparative mapping between Arabidopsis thaliana and Brassica nigra indicates that Brassica genomes have evolved through extensive genome replication accompanied by chromosome fusions and frequent recombinations. Genetics 150, 1217–1228. Lagercrantz, U., and Lydiate, D. J. (1996). Comparative genome mapping in Brassica. Genetics 144, 1903–1910. Lawrence, W. J. C. (1931). The secondary association of chromosomes. Cytologia 2, 352–384. Lee, H. -S., and Chen, Z. J. (2001). Protein-coding genes are epigenetically regulated in Arabidopsis polyploids. Proc. Natl. Acad. Sci. USA 98, 6753–6758. Leitch, I. J., and Bennett, M. D. (1997). Polyploidy in angiosperms. Trends Plant Sci. 2, 470–476. Levin, D. A. (1983). Polyploidy and novelty in flowering plants. Am. Nat. 122, 1–25. W. H. Lewis (ed.) (1980). In “Polyploidy: Biological Relevance”. Plenum Press, New York. Li, W. -H. (1985). Accelerated evolution following gene duplication and its implications for the neutralist–selectionist controversy. In “Population Genetics and Molecular Evolution” (T. Ohta and K. Aoki, Eds.), pp. 333 –353. Springer, Berlin. Liu, B., and Wendel, J. F. (2000). Retroelement activation followed by rapid repression in interspecific hybrid plants. Genome 43, 874 –880. Liu, B., Vega, J. M., and Feldman, M. (1998). Rapid genomic changes in newly synthesized amphiploids of Triticum and Aegilops. II. Changes in low-copy coding DNA sequences. Genome 41, 535 –542. Liu, B., Vega, J. M., Segal, G., Abbo, S., Rodova, M., and Feldman, M. (1998b). Rapid genomic changes in newly synthesized amphiploids of Triticum and Aegilops. I. Changes in low-copy noncoding DNA sequences. Genome 41, 272–277. Liu, B., Brubaker, C. L., Mergeai, G., Cronn, R. C., and Wendel, J. F. (2001a). Polyploid formation in cotton is not accompanied by rapid genomic changes. Genome 43, 874– 880. Liu, Q., Brubaker, C. L., Green, A. G., Marshall, D. R., Sharp, P. J., and Singh, S. P. (2001b). Evolution of the FAD2-1 fatty acid desaturase 50 UTR intron and the molecular systematics of Gossypium (Malvaceae). Am. J. Bot. 88, 92–102. Longley, A. E. (1933). Chromosomes in Gossypium and related genera. J. Agric. Res. 46, 217 –227.
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
183
Lynch, M., and Conery, J. S. (2000). The evolutionary fate and consequences of duplicate genes. Science 290, 1151–1155. Lynch, M., and Force, A. (2000). The probability of duplicate gene preservation by subfunctionalization. Genetics 154, 459 –473. Makalowski, W. (2001). Are we polyploids? A brief history of one hypothesis. Genome Res. 11, 667 –670. Masterson, J. (1994). Stomatal size in fossil plants: Evidence for polyploidy in majority of angiosperms. Science 264, 421–424. McCarty, J. C., and Percy, R. G. (2001). Genes from exotic germplasm and their use in cultivar improvement in Gossypium hirsutum L. and G. barbadense L. In “Genetic Improvement of Cotton: Emerging Technologies” (J. N. Jenkins and S. Saha, Eds.), pp. 65–80. Science Publishers, Enfield, NH. Michaelson, M. J., Price, H. J., Ellison, J. R., and Johnston, J. S. (1991). Comparison of plant DNA contents determined by Feulgen microspectrophotometry and laser flow cytometry. Am. J. Bot. 78, 183–188. Moore, G., Devos, K. M., and Wang, Z. (1995). Grasses, line up and form a circle. Curr. Biol. 5, 737 –739. Morton, B. R., Gaut, B. S., and Clegg, M. T. (1996). Evolution of alcohol dehydrogenase genes in the palm and grass families. Proc. Natl. Acad. Sci. USA 93, 11735–11739. Muller, J. (1981). Fossil pollen records of extant angiosperms. Bot. Rev. 47, 1–142. Muller, J. (1984). Significance of fossil pollen for angiosperm history. Ann. Missouri Bot. Gard. 71, 419 –443. Muravenko, O., Fedotov, A. R., Punina, E. O., Federova, L. I., Grif, V. G., and Zelenin, A. V. (1998). Comparison of chromosome BrdU–Hoechst– Giemsa banding patterns of the A1 and (AD)2 genomes of cotton. Genome 41, 616–625. Ohno, S. (1970). In “Evolution by Gene Duplication”. Springer, New York. O’Neill, R. J. W., O’Neill, M. J., and Graves, J. A. M. (1998). Undermethylation associated with retroelement activation and chromosome remodelling in an interspecific mammalian hybrid. Nature 393, 68–72. Otto, S. P., and Whitton, J. (2000). Polyploid incidence and evolution. Annu. Rev. Genet. 34, 401–437. Ozkan, H., Levy, A. A., and Feldman, M. (2001). Allopolyploidy-induced rapid genome evolution in the wheat (Aegilops–Triticum ) group. Plant Cell 13, 1735–1747. Parks, C. R., Ezell, W. L., Williams, D. E., and Dreyer, D. L. (1975). The application of flavonoid distribution to taxonomic problems in the genus Gossypium. Bull. Torrey Bot. Club 102, 350 –361. Paterson, A. H., Bowers, J. E., Burow, M. D., Draye, X., Elsik, C. G., Jiang, C. -X., Katsar, C. S., Lan, Y. -R., Ming, R., and Wright, R. J. (2000). Comparative genomics of plant chromosomes. Plant Cell 12, 1523– 1539. Percy, R. G., and Wendel, J. F. (1990). Allozyme evidence for the origin and diversification of Gossypium barbadense L. Theor. Appl. Genet. 79, 529 –542. Phillips, L. L. (1963). The cytogenetics of Gossypium and the origin of New World cottons. Evolution 17, 460–469. Phillips, L. L. (1964). Segregation in new allopolyploids of Gossypium. V. Multivalent formation in New World £ Asiatic and New World £ wild American hexaploids. Am. J. Bot. 51, 324 –329. Phillips, L. L., and Strickland, M. A. (1966). The cytology of a hybrid between Gossypium hirsutum and G. longicalyx. Can. J. Genet. Cytol. 8, 91 –95. Reinisch, A. J., Dong, J., Brubaker, C. L., Stelly, D. M., Wendel, J. F., and Paterson, A. H. (1994). A detailed RFLP map of cotton, Gossypium hirsutum £ G. barbadense: Chromosome organization and evolution in a disomic polyploid genome. Genetics 138, 829 –847.
184
J. F. WENDEL AND R. C. CRONN
Saha, S., and Zipf, A. (1998). Genetic diversity and phylogenetic relationships in cotton based on isozyme markers. J. Crop Prod. 1, 79 –93. Sanderson, M. J. (1998). Estimating rate and time in molecular phylogenies: Beyond the molecular clock? In “Molecular Systematics of Plants. II. DNA Sequencing” (P. S. Soltis, D. E. Soltis and J. J. Doyle, Eds.), pp. 242 –264. Kluwer Academic, Dordrecht. Saunders, J. H. (1961). In “The wild species of Gossypium and their evolutionary history”. Oxford University Press, London. Schranz, M. E., and Osborn, T. C. (2000). Novel flowering time variation in the resynthesized polyploid Brassica napus. J. Hered. 91, 242 –246. Seelanan, T., Schnabel, A., and Wendel, J. F. (1997). Congruence and consensus in the cotton tribe. Syst. Bot. 22, 259–290. Seelanan, T., Brubaker, C. L., Stewart, J. M., Craven, L. A., and Wendel, J. F. (1999). Molecular systematics of Australian Gossypium section Grandicalyx (Malvaceae). Syst. Bot. 24, 183– 208. Shaked, H., Kashkush, K., Ozkan, H., Feldman, M., and Levy, A. A. (2001). Sequence elimination and cytosine methylation are rapid and reproducible responses of the genome to wide hybridization and allopolyploidy in wheat. Plant Cell 13, 1749–1759. Shoemaker, R. C., Polzin, K., Labate, J., Specht, J., Brummer, E. C., Olson, T., Young, N., Concibido, V., Wilcox, J., Tamulonis, J. P., Kochert, G., and Boerma, H. R. (1996). Genome duplication in soybean (Glycine subgenus soja ). Genetics 144, 329–338. Sidow, A. (1996). Gen(ome) duplications in the evolution of early vertebrates. Curr. Opin. Genet. Dev. 6, 715–722. Skovsted, A. (1933). Cytological studies in cotton. I. The mitosis and meiosis in diploid and triploid Asiatic cotton. Ann. Bot. 47, 227–251. Skovsted, A. (1934). Cytological studies in cotton. II. Two interspecific hybrids between Asiatic and New World cottons. J. Genet. 28, 407 –424. Skovsted, A. (1937). Cytological studies in cotton. IV. Chromosome conjugation in interspecific hybrids. J. Genet. 34, 5–42. Small, R. L., and Wendel, J. F. (1999). The mitochondrial genome of allotetraploid cotton (Gossypium L.). J. Hered. 90, 251– 253. Small, R. L., and Wendel, J. F. (2000). Copy number lability and evolutionary dynamics of the Adh gene family in diploid and tetraploid cotton (Gossypium ). Genetics 155, 1913–1926. Small, R. L., and Wendel, J. F. (2000). Phylogeny, duplication, and intraspecific variation of Adh sequences in New World diploid cottons (Gossypium L. Malvaceae). Mol. Phyl. Evol. 16, 73 –84. Small, R. L., and Wendel, J. F. (2002). Differential evolutionary dynamics of duplicated paralogous Adh loci in allotetraploid cotton (Gossypium ). Mol. Biol. Evol. 19, 597–607. Small, R. L., Ryburn, J. A., Cronn, R. C., Seelanan, T., and Wendel, J. F. (1998). The tortoise and the hare: Choosing between noncoding plastome and nuclear Adh sequences for phylogeny reconstruction in a recently diverged plant group. Am. J. Bot. 85, 1301–1315. Small, R. L., Ryburn, J. A., and Wendel, J. F. (1999). Low levels of nucleotide diversity at homoeologous Adh loci in allotetraploid cotton (Gossypium L.). Mol. Biol. Evol. 16, 491–501. Smith, N. G. C., Knight, R., and Hurst, L. D. (1999). Vertebrate genome evolution: A slow shuffle or a big bang? BioEssays 21, 697–703. Soltis, D. E., and Soltis, P. S. (1989). Polyploidy, breeding systems and genetic differentiation in homosporous pteridophytes. In “Isozymes in Plant Biology” (D. E. Soltis and P. S. Soltis, Eds.), pp. 241 –258. Dioscorides, Portland, OR. Soltis, D. E., and Soltis, P. S. (1993). Molecular data and the dynamic nature of polyploidy. Crit. Rev. Plant Sci. 12, 243– 273. Soltis, D. E., and Soltis, P. S. (1995). The dynamic nature of polyploid genomes. Proc. Natl. Acad. Sci. USA 92, 8089–8091. Soltis, D. E., and Soltis, P. S. (1999). Polyploidy: Origins of species and genome evolution. Trends Ecol. Evol. 9, 348–352.
POLYPLOIDY AND EVOLUTIONARY HISTORY OF COTTON
185
Soltis, P. S., and Soltis, D. E. (2000). The role of genetic and genomic attributes in the success of polyploids. Proc. Natl Acad. Sci. USA 97, 7051–7057. Soltis, P. S., Doyle, J. J., and Soltis, D. E. (1992). Molecular data and polyploid evolution in plants. In “Molecular Systematics of Plants” (P. S. Soltis, D. E. Soltis and J. J. Doyle, Eds.), pp. 177–201. Chapman & Hall, New York. Song, K., Lu, P., Tang, K., and Osborn, T. C. (1995). Rapid genome change in synthetic polyploids of Brassica and its implications for polyploid evolution. Proc. Natl Acad. Sci. USA 92, 7719–7723. Sossey-Alaouni, K., Serieys, H., Tersac, M., Lambert, P., Schilling, E., Griveau, Y., Kaan, F., and Berville´, A. (1998). Evidence for several genomes in Helianthus. Theor. Appl. Genet. 97, 422 –430. Spring, J. (1997). Vertebrate evolution by interspecific hybridization—are we polyploid? FEBS Lett. 400, 2–8. Stebbins, G. L. (1947). Evidence on rates of evolution from the distribution of existing and fossil plant species. Ecol. Monogr. 17, 149– 158. Stebbins, G. L. (1950). In “Variation and Evolution in Plants”. Columbia University Press, New York. Stebbins, G. L. (1971). In “Chromosomal Evolution in Higher Plants”. Edward Arnold, London. Stephens, S. G. (1944). The genetic organization of leaf-shape development in the genus Gossypium. J. Genet. 46, 28–51. Stephens, S. G. (1944). Phenogenetic evidence for the amphidiploid origin of New World cottons. Nature 153, 53–54. Stephens, S. G. (1951). Evolution of the gene—“Homologous” genetic loci in Gossypium. Cold Spring Harbor Sympp. Quant. Biol. 16, 131– 140. Stephens, S. G. (1951). Possible significance of duplication in evolution. Adv. Genet. 4, 247 –265. Stephens, S. G. (1958). Salt water tolerance of seeds of Gossypium species as a possible factor in seed dispersal. Am. Nat. 92, 83–92. Stephens, S. G. (1966). The potential for long range oceanic dispersal of cotton seeds. Am. Nat. 100, 199 –210. Stewart, J. M. (1995). Potential for crop improvement with exotic germplasm and genetic engineering. In “Challenging the Future: Proceedings of the World Cotton Research” (G. A. Constable and N. W. Forrester, Eds.), pp. 313– 327. CSIRO, Melbourne. Taylor, J. S., and Brinkmann, J. S. (2001). 2R or not 2R? Trends Genet. 17, 488–489. Valicek, P. (1983). A contribution to the botanical characteristics of the species Gossypium gossypioides (Ulbr.) Standl. Agric. Tropp. Subtropp. Univ. Agric. Praga 16, 185 –196. Vollesen, K. (1987). The native species of Gossypium (Malvaceae) in Africa, Arabia and Pakistan. Kew Bull. 42, 337–349. Watt, G. (1907). In “The wild and cultivated cotton plants of the world”. Longmans, Green and Co, London. Webber, J. M. (1935). Interspecific hybridization in Gossypium and the meiotic behavior of F1 plants. J. Agric. Res. 51, 1047–1070. Wendel, J. F. (1989). New World tetraploid cottons contain Old World cytoplasm. Proc. Natl Acad. Sci. USA 86, 4132–4136. Wendel, J. F. (1995). Cotton. In “Evolution of Crop Plants” (N. Simmonds and J. Smartt, Eds.), pp. 358–366. Longman, London. Wendel, J. F. (2000). Genome evolution in polyploids. Plant Mol. Biol. 42, 225 –249. Wendel, J. F., and Albert, V. A. (1992). Phylogenetics of the cotton genus (Gossypium L.): Characterstate weighted parsimony analysis of chloroplast DNA restriction site data and its systematic and biogeographic implications. Syst. Bot. 17, 115–143. Wendel, J. F., and Liu, B. (2002). Non-Mendelian phenomena in allopolyploid genome evolution. Curr. Genom. (in press).
186
J. F. WENDEL AND R. C. CRONN
Wendel, J. F., and Percival, A. E. (1990). Molecular divergence in the Galapagos Island–Baja California species pair Gossypium klotzschianum and G. davidsonii (Malvaceae). Plant Syst. Evol. 171, 99–115. Wendel, J. F., and Percy, R. G. (1990). Allozyme diversity and introgression in the Galapagos endemic Gossypium darwinii and its relationship to continental G. barbadense. Biochem. Syst. Ecol. 18, 517–528. Wendel, J. F., Olson, P. D., and Stewart, J. M. (1989). Genetic diversity, introgression and independent domestication of Old World cultivated cottons. Am. J. Bot. 76, 1795–1806. Wendel, J. F., McD, S. J., and Rettig, J. H. (1991). Molecular evidence for homoploid reticulate evolution in Australian species of Gossypium. Evolution 45, 694 –711. Wendel, J. F., Brubaker, C. L., and Percival, A. E. (1992). Genetic diversity in Gossypium hirsutum and the origin of Upland cotton. Am. J. Bot. 79, 1291–1310. Wendel, J. F., Rowley, R., and Stewart, J. M. (1994). Genetic diversity in and phylogenetic relationships of the Brazilian endemic cotton, Gossypium mustelinum (Malvaceae). Plant Syst. Evol. 192, 49–59. Wendel, J. F., Schnabel, A., and Seelanan, T. (1995a). Bidirectional interlocus concerted evolution following allopolyploid speciation in cotton (Gossypium ). Proc. Natl Acad. Sci. USA 92, 280–284. Wendel, J. F., Schnabel, A., and Seelanan, T. (1995b). An unusual ribosomal DNA sequence from Gossypium gossypioides reveals ancient, cryptic, intergenomic introgression. Mol. Phyl. Evol. 4, 298–313. Wendel, J. F., Small, R. L., Cronn, R. C., and Brubaker, C. L. (1999). Genes, jeans, and genomes: Reconstructing the history of cotton. Proceedings of the VII International Symposium of the International Organization of Plant Biosystematists. In “Plant Evolution in Man-made Habitats” (L. W. D. van Raamsdonk and J. C. M. den Nijs, Eds.), pp. 133–161. Hugo de Vries Laboratory, Amsterdam. Wendel, J. F., Cronn, R. C., Johnston, J. S., and Price, H. J. (2002). Feast and famine in plant genomes. Genetica 115, 37–47. Whitkus, R., Doebley, J., and Lee, M. (1992). Comparative genome mapping of sorghum and maize. Genetics 132, 1119–1130. Wilson, W. A., Harrington, S. E., Woodman, W. L., Lee, M., Sorrells, M. E., and McCough, S. R. (1999). Inferences on the genome structure of progenitor maize through comparative analysis of rice, maize and the domesticated Panicoids. Genetics 153, 453 –473. Wolfe, K. H. (2001). Yesterday’s polyploids and the mystery of polyploidization. Nat. Rev. Genet. 2, 333–341. Wolfe, K. H., and Shields, D. C. (1997). Molecular evidence for an ancient duplication of the entire yeast genome. Nature 387, 708 –713. Woodcock, D. W., and Manchester, C. A. (1998). Fossil cotton from the Salt Lake Crater area, O’ahu, Hawaii. Records of the Hawaii Biological Survey for 1997—Part 2. Notes. In “Bishop Museum Occasional Papers” (N. L. Evenhuis and S. E. Miller, Eds.), pp. 17– 19. Bishop Museum Press, Honolulu. Wright, R. J., Thaxton, P. M., El-Zik, K. M., and Paterson, A. H. (1998). D-subgenome bias of Xcm resistance genes in tetraploid Gossypium (cotton) suggests that polyploid formation has created novel avenues for evolution. Genetics 149, 1987–1996. Zhao, X. -P., Wing, R. A., and Paterson, A. H. (1995). Cloning and characterization of the majority of repetitive DNA in cotton (Gossypium L.). Genome 38, 1177–1188. Zhao, X. -P., Si, Y., Hanson, R. E., Crane, C. F., Price, H. J., Stelly, D. M., Wendel, J. F., and Paterson, A. H. (1998). Dispersed repetitive DNA has spread to new genomes since polyploid formation in cotton. Genome Res. 8, 479–492.
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS FOR ASSESSING QUALITY OF ACID SOILS Z. L. He,1,2 X. E. Yang,1 V. C. Baligar3 and D. V. Calvert2 1
Department of Resource Science, College of Natural Resource and Environmental Sciences, Zhejiang University, Huajiachi Campus, 310029 Hangzhou, People’s Republic of China 2 Indian River Research and Education Centre, Institute of Food and Agricultural Sciences, University of Florida, Fort Pierce, Florida 34945, USA 3 USDA – ARS Alternate Crops and System Research Laboratory, Beltsville Agricultural Research Center, Beltsville, Maryland 20704, USA
I. Introduction II. Acid Soil Distribution in the World III. Quality Characteristics of Acidic Soil A. Definition and Attributes of Soil Quality B. Quality Characteristics of Acidic Soils IV. Measurement of Microbiological and Biochemical Parameters in Acidic Soils A. Microbial Biomass Carbon, Nitrogen, and Phosphorus B. Microbial Turnover of Carbon, Nitrogen, and Phosphorus C. Microbial Community Structures D. Soil Enzyme Activity V. Microbiological and Biochemical Indicators of Acid Soil Quality A. Microbial Biomass B. Microbial Biomass Turnover C. Microbial Biomass-related Indicators D. Microbial Community Structure Indicators E. Enzyme Activities VI. Soil pH Versus Microbiological and Biochemical Indicators VII. Development of Acid Soil Quality Indexing Systems VIII. Limitations and Prospective Acknowledgments References
Acid soils play an important role in the production of world food and fiber. The majority of acid soils in the tropical and subtropical regions are highly weathered, subject to intensive cropping, and vulnerable to soil erosion. Degradation of these soils poses a great challenge to sustainable agriculture in these regions. An efficient indexing system using minimal physical, 89 Advances in Agronomy, Volume 78 Copyright q 2003 by Academic Press. All rights of reproduction in any form reserved 0065-2113/02$35.00
90
Z. L. HE ET AL. chemical, and biological parameters is immediately needed to assess and monitor the dynamics of soil quality under diversified farming systems and to improve agricultural practices and productivity. Various physical, chemical, and biological properties can be used to characterize soil quality. However, the number, activity, and diversity of microorganisms and the related biochemical processes are the most important components of soil quality, especially for the highly weathered acid soils, in which plant productivity is closely related to biological cycling. Several microbiological and biochemical parameters have been suggested as indicators of soil quality. They include: microbial biomass carbon (Cmic), nitrogen (Nmic), and phosphorus (Pmic) and their turnover rates; the microbial quotient (MQ) (Cmic/organic C ratio); basal respiration (qCO2); the microbial metabolic quotient (MMQ) (qCO2/Cmic); the ratio of microbial N over total Kjeldahl N; and enzyme activity. Recently, microbial diversity parameters such as community level physiological profile, phospholipid fatty acids, the ratio of Gram-negative/Gram-positive bacteria, the ratio of fungal/bacterial microorganisms, and free-living diazotrophic bacteria, etc., have been identified as important indicators of soil quality. All these microbiological and biochemical parameters have been shown to relate to soil productivity and respond to changes of land use, vegetation coverage, agricultural practices such as liming, fertilization, and tillage, as well as climate factors (temperature and rainfall). Current progress on measurement, interpretation, and potential application of the microbiological and biochemical indices in assessing quality, fertility, and sustainability of highly weathered acid soils are reviewed in this chapter. q 2003 Academic Press.
I. INTRODUCTION Acid soil is defined as a soil with a pH value , 7.0 (Soil Science Society of America, 1997). These include strongly acid soils with pH value , 5.0, and moderately acid soils with pH between 5.0 and 6.5 (Brady and Weil, 1996). The importance of acid soils to the world economy is well known (Sanchez, 1976; Von Uexkull and Mutert, 1995; Baligar and Ahlrichs, 1998). These soils constitute the most important soil resources and support more than one-half of the world population. Moreover, most acid soils are distributed in the developing countries where population growth is fast and the pressure of demands for food and fiber supply is increasing. On the other hand, most acid soils are characteristically low in fertility. They have poor physical, chemical, and biological properties and are low in nutrient availability. These soils have various constraints for crop production and are readily degraded when subjected to erosion, leaching, or contamination. Therefore, proper management of acid soils is of both socio-economic and ecological importance for tropical and subtropical regions. An efficient indexing system with minimal physical, chemical, and biological parameters is immediately needed to assess and monitor the dynamics of soil
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS
91
quality and to improve agricultural practices and productivity. Many physical, chemical, and biological properties can be used to characterize soil quality. The number, activity, and diversity of soil organisms and related biochemical processes are likely the most important components of soil quality for the highly weathered acid soils where production sustainability, environmental quality, and plant and animal health are closely related to biological cycling. This chapter is intended to provide a brief review on current progress in measuring, interpreting, and applying microbiological and biochemical indicators in assessing soil fertility and other qualities of highly weathered acid soils.
II. ACID SOIL DISTRIBUTION IN THE WORLD Acid soils occupy about 3.95 billion ha and account for 30% of the world’s ice-free land area (Von Uexkull and Mutert, 1995). They are distributed in all the continents except for Antarctica. The land area of acid soils for different continents or regions decreases in the following order: South America . North America . Africa . Asia . Europe . Australia/New Zealand . Central America, whereas the percentage of acid land area decreases in the order: Asia . Europe . Central America . North America ¼ Australia/New Zealand . Africa . South America (Table I). Soil acidity results from a spectrum of natural and anthropogenic sources and processes including: (1) acid parent materials; (2) extensive leaching of bases under humid climate conditions, especially in the tropical and subtropical regions; (3) imbalanced cycling of basic and acid substances and organic and inorganic acids produced from plant roots and decomposition of litters, for
Table I Distribution of Acid Soils in the World Distribution classes Acid land area ( £ 106 ha) Acid land area (%)b
Australia/ North Central South Global America America Africa Asiaa New Zealand America Europe 3950
37
917
659
725
239
662
391
30
35
14
22
40
30
30
37
Modified from Von Uexkull and Mutert (1995) and Sehgal et al. (1998). South and East Asia include China, India, Philippines, Vietnam, Bangladesh, and Sri Lanka only. b Ice-free land area of the globe. a
92
Z. L. HE ET AL.
example, acid soils are developed under coniferous forest or S-accumulating trees such as mangroves; (4) dry and wet acid deposits; (5) fertilization and irrigation; (6) mining activity; and (7) long-term legume cultivation. Consequently, acid soils vary greatly in their nature and properties. The dominant acid soils include Spodosols, Alfisols, Inceptisols and Histosols in the northern belt (cold and temperate regions) and Ultisols and Oxisols in the southern belt (tropical and subtropical regions) (Table II). Of the acid soils, the Ultisols and Oxisols and part of the Alfisols, which account for about 40% of the total acid land area of the world, are widely cultivated, and are most important to world agriculture. These soils have witnessed the birth and early civilization of humankind and have supported a large proportion of the world population for a long time. Being old and vulnerable to degradation, these acid soils often pose multiple soil constraints to crop production, such as nutrient deficiency and Al toxicity (Table II).
III. QUALITY CHARACTERISTICS OF ACIDIC SOILS A. DEFINITION AND ATTRIBUTES OF SOIL QUALITY Soil quality is defined as the capacity of the soil to function within the ecosystem boundaries to sustain biological productivity, maintain environmental quality, and promote plant and animal health (Soil Science Society of America, 1997). Soil quality is a measure of the condition of soil relative to the requirement of one or more species and/or to any human need or purpose (Johnson et al., 1997; Lal, 1997; Doran et al., 1999). Soil quality consists of physical, chemical, and biological components (Table III). Texture, depth of top soil and rooting zone, bulk density and infiltration, and water holding capacity and water retention characteristics are all important physical attributes of soil quality. Chemical components of soil quality include total organic carbon, pH, cation exchange capacity (CEC), electrical conductivity, and extractable nitrogen, phosphorus, and potassium (Table III). Most methods used for measuring physical and chemical attributes of soil quality are well established and are available in most standard soil testing laboratories. Efforts are needed to minimize data sets and to develop benchmark indexes for these soil quality components. Biological properties of soil quality include quantity, activity, and diversity of soil fauna (micro/macro/meso), microflora, and enzymes. These components of soil regulate microbiological and biochemical processes, carbon and nutrient cycling, and degradation of organic and inorganic pollutants, and affect physical and chemical properties of soils, thus influencing the direction of soil quality
Elements 6
Area ( £ 10 ha)
a
pH
a
Organic C (g kg )
Soil order
Soil group
Oxisols Ultisols
Ferralsol Acrisol
727 (18.4)b 864 (21.8)
4.5–5.5 3.6–5.6
1–20 1–16
Alfisols Andisols Spodosols
Planosol Andepts Podzol
255 (6.5) 34 (0.9) 415 (10.5)
6.3–7.1 4.3–6.0 3.6–4.9
1–16 2–72 2–34
Histosols Inceptisols Entisols Total
Histol Andepts Arensol
270 (6.8) 561 (14.2) 824 (20.9) 3950 (100)
4.5–5.1 4.0–5.8 5.5–6.8
313–372 3–82 0–14
Modified from Baligar and Ahlrichs (1998). Sanchez (1976) and Singer and Stephanie (2000). b The number inside the parenthesis is % of total acid soil area. a
21
Deficiency P, Ca, Mg, Mo N, P, Ca and most others Most nutrients P, micronutrients N, P, K, Ca, micronutrients K, Si, Cu K, Zn, Fe, Cu, Mn
Toxicity Al, Mn, Fe Al, Mn, Fe Al Al Al
Al, Mn, Fe
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS
Table II Fertility and Chemical Properties of Major Types of Acid Soil in the World
93
94
Z. L. HE ET AL.
Table III Major Components of Soil Quality in Relation to Fertility and Environment in Tropical and Subtropical Regions Soil quality components
Relationship to soil functions and conditions Physical
Texture Depth of soil Top soil and rooting Soil bulk density & Infiltration Water retention characteristics
Retention and transport of water, nutrients, and chemicals; modeling use, soil erosion and variability estimate Estimate of productivity potential and erosion Plant and animal production potential, nutrient use efficiency, and groundwater contamination Potential for leaching, productivity, and erosivity, depth and volume of root Available water storage, water conductivity, water use efficiency, and surface runoff; aeration Chemical
Total organic C
pH Cation exchange capacity Electrical conductivity Extractable N, P, and K
Related to soil fertility, health, and environmental capacity such as nutrient availability, biological activity, and contaminant degradation/inactivation; aggregation Growth and health of plant and soil organisms; nutrient availability and toxicity; essential to process modeling Related to nutrient availability and leaching potential Defines plant and microbial activity thresholds Plant-available nutrients and potential for loss from soil; productivity and environmental quality indicators Biological
Microbial biomass
Microbial catalytic potential and cycling and availability of C, N, P, and S; modeling, indicator of organic matter quality, soil fertility, and heavy metal contamination Macrofauna/arthropods/ Indicators of ecological stress or restoration, organic matter earthworm breakdown, humification, redistribution of soil organic matter Enzyme activity Nutrient cycling, mineralization/immobilization, heavy metal toxicity, soil degradation Soil respiration Measure of microbial activity (in some case, plants); process modeling, estimate of microbial biomass activity Potentially mineralizable N Soil productivity and N-supplying potential; process modeling, (anaerobic incubation) surrogate indicator of microbial biomass N Modified from Elliott et al. (1996) and Doran et al. (1999).
change (Pankhurst et al., 1995; Elliott et al., 1996). Therefore, biological properties of soil quality have received much attention from the scientific and regulatory community.
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS
95
B. QUALITY CHARACTERISTICS OF ACIDIC SOILS Most of the acid soils having a pH range 4 –6 contain low organic matter (, 20 g kg21) and low nutrient availability. Deficiency of P, Ca, Mg, Mo, N, K, and micronutrients is very common for these soils. Toxicity of Al, Mn, and Fe and acidity stress frequently poses the most important soil constraints for crop production in the acid soil regions (Baligar and Ahlrichs, 1998). Highly weathered acidic soils have unique quality characteristics that make them different from other types of soils. Rooting depth is often a limiting factor for crop production. These soils (Ultisols and Oxisols) generally have a thick subsurface layer (argillic or oxic horizon). Drought conditions or aluminum toxicity in the subsoils are often more common barriers to root development than shallow rocks or hardpans (Sanchez, 1976; He et al., 2001a). Most Oxisols and Ultisols consist of stable sand-size granules, called “psudosand.” These granules are microaggregates cemented by Fe and Al oxides, and organic matter (Lu et al., 1986). Water stability of microaggregates is related to the content of clay, Fe and Al oxides, and organic matter (Zhang et al., 1997). Clay and Fe and Al oxides play important roles in the formation of microaggregates, whereas organic matter, particularly microorganisms and humus, contribute more to macroaggregate structures (Ladd et al., 1996; Zhang et al., 1997; Chan and Heenan, 1999). Cultivation or water-submergence often causes structure degradation in these soils by accelerating mineralization of organic matter and by dissolving Fe oxides under reducing conditions (Lu et al., 1986). Water retention patterns of highly weathered acid soils are intermediate between sandy soils and soils dominated by 2:1 clay minerals. At high moisture content, when large pores between macroaggregates are filled with water, water movement in these soils is as fast as in a sandy soil. At low moisture content, water is tightly held within small pores between microaggregates as in a silicate clay. As a result, these soils generally have less water available to plants at the same moisture tension level, as in layer silicate-dominated clayey soils (Sanchez, 1976). Severe soil erosion and drought in the tropical and subtropical regions are more or less associated with these unique water retention properties of the highly weathered acidic soils. The dominant clay mineral in these soils is kaolinite, Fe and Al oxides or hydroxides are also prominent. These minerals carry both negative and positive charges, depending on the pH. Thus, they are often called variable charge minerals and the soils dominated by these minerals are called variable charge soils. The variable charge soils differ from the permanent charge soils, which are dominated by 2:1 layer silicate minerals such as montmorillonite. CEC is generally low in these acidic soils and variable charge minerals and organic matter contributes 50 –70% of the total CEC (Baligar and Ahlrichs, 1998). Exchangeable Al3þ and Hþ constitute a large proportion of the total exchangeable cations (Naþ, Kþ, Mg2þ, Ca2þ, Al3þ, Hþ) in the acidic variable charge soils.
96
Z. L. HE ET AL.
Liming and application of organic matter are effective approaches to increase CEC and base saturation in these soils (Sanchez, 1976). Organic matter content varies greatly in acidic soils in the tropical and subtropical regions, depending on vegetation, cultivation history, use of organic manures, and degree of soil erosion. Soils under natural or artificial forestry usually have relatively high organic matter content (3 –5%). Change of land use from forestry to food crop cultivation reduces organic matter content due to accelerated mineralization. Loss of surface soil caused by intensive soil erosion often causes a drastic decrease in organic matter (He et al., 2001a). Deficiencies of N, P, and K are common in acidic soils due to low organic matter content, depletion of nutrient-bearing minerals, and fixation of phosphate by Fe and Al oxides (Baligar and Ahlrichs, 1998). Fertilizers and biological cycling are likely the major sources of N, P, and K supply for these soils. There is a great variation in community diversity of organisms in tropical and subtropical soils (Coleman et al., 1993). Favorable temperature and rainfall in the tropical and subtropical regions enhance growth of micro- and macroorganisms and accelerate turnover and cycling of organic C and N, P, and S. A reduced supply of organic C and nutrients caused by deforestation, soil erosion, and intensive cultivation as well as soil stress conditions such as soil acidity, aluminum toxicity, and heavy metal contamination reduce biodiversity in the soils (Coleman et al., 1993; Brown et al., 1994). Soil acidification, in general, reduces the number and activities of microorganisms and macrofauna, especially bacteria. Fewer earthworms were found in soils under very acidic conditions (pH , 4.0) (Ohno, 2001). Fungi become dominant in microbial communities in most acidic soils, and acid-sensitive bacteria give way to acid or aluminum tolerant bacterial species (Kanazawa and Kunito, 1996; Yu et al., 2002). These acid or aluminum tolerant bacteria exhibited longer dormant periods (Yao, 2000; Yu et al., 2002). A sufficient supply of organic C maintained high total microbial biomass in tea-growing acidic soils (pH , 4.5), but culturable microorganisms present in these soils were much lower than in neutral soils with comparable soil properties and environmental conditions (Yu et al., 2002). The roles of microand macroorganisms in soil productivity, especially transformation and availability of nutrients, remains to be fully understood.
IV. MEASUREMENT OF MICROBIOLOGICAL AND BIOCHEMICAL PARAMETERS IN ACIDIC SOILS In recent years, relatively rapid progress has been made in quantification of microbial biomass and characterization of microbial community structure (Smith and Paul, 1990; Tunlid and White, 1994; Martens, 1995). A number of methods that have been proposed for measuring microbial biomass and microbial
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS
97
community structure are listed in Table IV. Most of these methods, however, have been developed for temperate soils with pH close to neutral, and these methods are more or less subject to limitations when they are applied to acid soils.
A. MICROBIAL BIOMASS CARBON, NITROGEN, AND PHOSPHORUS In recent years, some biochemical and physiological methods have been developed for estimating microbial biomass and the amounts of N, P, and S contained in the microbial biomass (Jenkinson and Powlson, 1976b; Anderson and Domsch, 1978; Brookes et al., 1982; Brookes et al., 1985b; Wu et al., 1994). These methods include chloroform fumigation – incubation (CFI), chloroform fumigation– extraction (CFE), and substrate induced respiration (SIR) and have been widely used to evaluate the influences of agricultural practices, land use, and contaminants on soil microorganisms and nutrient cycling (Martens, 1995; Dalal, 1998; Giller et al., 1998). The CFI for quantifying soil microbial biomass carbon (Cmic) was proposed by Jenkinson and his colleagues (Jenkinson, 1976; Jenkinson and Powlson, 1976a,b; Jenkinson et al., 1976; Powlson and Jenkinson, 1976) as a replacement of former tedious and inaccurate microscopic counting techniques (Martens, 1995). In CFI, a moist soil is fumigated with ethanol-free chloroform for 24 h; chloroform is then removed by repeated evacuation; the soil is reinoculated with a small amount of unfumigated soil and then incubated at a constant temperature (usually 22 or 258C) for 10 days at field capacity or 50% of its water holding capacity (about 0.01 MPa). An additional soil sample is retained unfumigated and used as a control. The CO2 evolved during incubation is measured by gas chromatography, as a continuous flow or by absorption in alkali followed by titrimetric, conductometric or colorimetric determination. As the net C mineralized as CO2 is only a proportion of the total microbial biomass C, a kC factor defined as the fraction of Cmic that is recovered by the CFI is used to calculate the Cmic. This procedure has been widely used for measuring Cmic in soils of different types or the same type but under different management practices since the soil’s development. A pre-incubation of at least 10 days is generally recommended, especially for dry and frozen soil samples (Martens, 1995). Fresh moist soil without pre-incubation is used if the measurement is intended for monitoring seasonal fluctuation of microbial biomass (He et al., 1997a). However, this procedure is subject to limitations with soils at pH values below 5. Low pH impedes development of bacterial populations in soil and thus, results in an invalid kC because of reduced mineralization of the killed microorganisms (Vance et al., 1987a). The CFE was an improvement on CFI (Vance et al., 1987b; Wu et al., 1990). The advantages of CFE over CFI are: (1) it is applicable to a wide range of soil types, including acid soils with pH below 5.0 (Vance et al., 1987b; Martikainen and Palojarvi, 1990); (2) it can be used for soils with newly added substrate and
Parameters
Methods and authors MC (classic method), CFI (Jenkinson and Powlson, 1976a,b), SIR (Anderson and Domsch, 1978), CFE (Vance et al., 1987b), ATP (Jenkinson and Oades, 1979), CFE-ninhydrin (Joergensen and Brookes, 1990)
N (Nmic)
CFI (Shen et al., 1984), CFE-total N (Brookes et al., 1985a,b)
P (Pmic)
CFE, using 0.5 M NaHCO3 (Brookes et al., 1982), resin (Kouno et al., 1995), or 0.025M HCl þ 0.03M NH4F as extractant (He et al., 1997a; Oberson et al., 1997) CFE (Wu et al., 1994)
S (Smic)
Microbial turnover Cmic Nmic
Pmic
CFE and labeling microorganisms in situ with 14C (Chen et al., 2001b; Van Veen et al., 1985) CFE and labeling microorganisms in situ with 15N (Van Veen et al., 1985; Yao et al., 1999) CFE and labeling microorganisms in situ with 32P (Chen, 2000)
Remarks The CFI is not suitable for soil with pH ,5.0 (Vance et al., 1987a). The SIR may underestimate Cmic in Oxisols (Feigl et al., 1995). The CFE þ TOC-analyzer is the most satisfactory method. Pre-incubation at 258C and 20.01 MPa water potential for 10 days is recommended for measuring Cmic as a soil quality indicator The CFE-total N is preferred for acid soils, with total N concentration in the extract determined by Kjeldahl digestion-plate diffusion method (Chen et al., 1997) The CFE with 0.025M HCl þ 0.03M NH4F is preferred to the 0.5 M NaHCO3 for acid soils, correction for P sorption is needed The CFE with 0.02 M KCl as extractant, SO4-S determined using ion chromatography, correction for S sorption might be needed for most clayey acid soils About 20 days for the labeled soil microbial biomass to reach maximum value. The decay of microbial biomass 14C follows the first-order reaction equation About 22 days for the labeled soil microbial biomass to reach maximum value. The decay of microbial biomass 15N follows the first-order reaction equation About 10 days for the labeled soil microbial biomass to reach maximum value. The decay of microbial biomass 32P follows the first-order reaction equation
Z. L. HE ET AL.
Microbial biomass C (Cmic)
98
Table IV Suitability of Methods for Measuring Soil Microbial Biomass and its Derivative Parameters in Acid Soils
Table IV (continued)
Microbial community structure (MCS)
Methods and authors Culture and plate count (classic method), BIOLOG (Haack et al., 1995), PLFAs (Zelles et al., 1995b), 16s RNA (Pettigrew and Sayler, 1986), FAME (Cavigelli et al., 1995)
Remarks BIOLOG method needs to be modified for measuring MCS in acid soils (Yao et al., 2000a); BIOLOG, PLFA, and other techniques may be used simultaneously to provide relatively complete information on MCS in acid soils
MC: microplate count; CFI: chloroform fumigation–incubation; CFE: chloroform fumigation –extraction; SIR: substrate induced respiration; ATP: adenosine 50 triple phosphate; PLFA: phospholipid fatty acid; RNA: ribonucleic acid; FAME: fatty acid methylester.
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS
Parameters
99
100
Z. L. HE ET AL.
Cmic measured by fumigation incubation (mg C kg 1)
for submerged soils (Bremer and Van Kessel, 1990; Inubushi et al., 1991); and (3) determination of the carbon flush from the fumigated soil is more accurate and convenient with the automated analyzer (Wu et al., 1990). However, the CFE method may have limitations when applied to peat soils (Kaiser et al., 1992). Also a re-calibration of the kec, which is used to convert C flush to microbial biomass carbon (Cmic), is essential for different types of soil (Ross, 1990). The SIR method was introduced by Anderson and Domsch (1978). CO2 respired at the maximum initial response following glucose addition is closely correlated with Cmic measured by a microscopic count, with 1 ml CO2 h21 equal to approximately 40 mg Cmic. This method has been widely used to estimate Cmic in soils. However, in neutral or alkaline soils, the solubility and retention of CO2 as HCO2 3 in the soil solution can strongly affect the results if the experimental system does not allow a continuous flushing of air through such alkaline soil samples (Martens, 1987). The SIR method was also reported to underestimate Cmic in Oxisols (Feigl et al., 1995). The CFI, CFE, and SIR methods were recently evaluated for measuring Cmic in acid red soils (Ultisols with pH from 4.2 to 5.7) (Yao and He, 1999). Microbial biomass C values determined by the three methods were highly correlated. Microbial biomass C estimated by the CFI were close to that obtained by the CFE (Fig. 1). The automated analyzer (AA) method and the oxidation– titration (OT) method were also evaluated for determining dissolved organic C in the K2SO4 800 y = 48 + 0.87
700
r2 =0.93 (P<0.01)
600
1:1 line
500 400 300 200 100 0 0
100
200
300
400
500
600
Cmic measured by fumigation extraction method
700
800
(mg C kg 1)
Figure 1 Microbial biomass C (Cmic) in acid soils, as measured by the fumigation–extraction and fumigation–incubation methods.
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 101
extractant (Yao and He, 1999). The AA is superior to the OT for samples with low concentrations of dissolved organic C, because AA has a lower detection limit and greater precision than OT (Yao and He, 1999). High variability in kC factors (0.2 – 0.6) has been reported for soils with different pH (Vance et al., 1987c; Sparling and West, 1989; Martens, 1995). A limitation of the CFI for measuring Cmic in acid soils is the deviation of kC value from those obtained with non-acid soils (Vance et al., 1987a; Martens, 1995). The kEC factors in the CFE method determined by Chen (2000), using eight acid red soils, ranged from 0.27 to 0.35, with a mean of 0.31, which is about 30% lower than the kEC value (0.45) proposed by Wu et al. (1990). This unconsistency is probably due to differences in microbial communities in different soils (Kanazawa and Kunito, 1996). Yao and He (1999) reported that the CFI and CFE were suitable for measuring Cmic in acid soils provided that a new kC factor is established. However, the CFE –AA method is recommended because of its convenience and greater reliability. Measurement of microbial biomass N (Nmic) is similar to that of Cmic. The CFE is suitable for measuring Nmic in acid soils provided that a new kEC value is developed based on the kC factor, microbial C/N ratio, and C/N ratio of K2SO4 extract determined using the acid soils (Shen et al., 1984; Chen et al., 1997; Yao, 2000). After extraction with 0.5 M K2SO4, either total N or ninhydrin-reactive N is determined to estimate the Nmic (Joergensen and Brookes, 1990; Inubushi et al., 1991). The total N concentration in the K2SO4 extract can be determined using either a Kjeldahl method (Vance et al., 1987b) or a plate-diffusion method after Kjeldahl digestion (Chen et al., 1997). Both methods provide comparable results (Chen et al., 1997). A chloroform fumigation– 0.5 M NaHCO3 extraction procedure proposed by Brookes et al. (1982) has been frequently used to measure soil microbial biomass P (Pmic). The 0.5 M NaHCO3 extraction is subject to low P recovery when applied to highly weathered acid soils due to strong adsorption of P by variablecharge minerals (Feigl et al., 1995; Chen et al., 2000). Replacement of 0.5 M NaHCO3 with 0.025 M HCl þ 0.03 M NH4F (Bray P1 reagent) improved P recovery and the reliability of the Pmic estimation (He et al., 1997a; Oberson et al., 1997). An anion exchange resin membrane was also suggested for better recovery of P in the measurement of Pmic (Kouno et al., 1995). Similar to kEC or kEN, the kEP is defined as the extractable proportion of Pmic by the fumigation – extraction method and is used to calculate Pmic from the difference in extractable P between fumigated and unfumigated soil samples. This factor influences estimation of Pmic and needs to be determined for each soil with different properties. Chen (2000) determined the kEP value for 11 red acid soils varying in physical, chemical, and biological properties by 32P-labeling of soil microbial biomass in situ. Both 0.5 M NaHCO3 and 0.025 M HCl þ 0.03 M NH4F were used to extract P from the fumigated and unfumigated soil samples. The 0.5 M NaHCO3 extraction provided a lower average kEP value (0.27) with greater variation (from 0.09 to 0.67) among the different soils, as compared with the 0.025 M HCl þ 0.03 M NH4F extraction
102
Z. L. HE ET AL.
(0.22 –0.50) (Chen, 2000). Therefore, the fumigation– 0.025 M HCl þ 0.03 M NH4F extraction appears to be superior to the fumigation – 0.5 M NaHCO3 extraction for determining Pmic in acid soils.
B. MICROBIAL TURNOVER OF CARBON, NITROGEN, AND PHOSPHORUS Soil microorganisms are primary driving forces for many chemical and biochemical processes and thus affect nutrient cycling, soil fertility, and global carbon change. Turnover of microbial biomass is a dynamic process, in which soil microbial biomass is continuously renewed in response to changing environmental conditions such as supply of energy sources, nutrients, and O2. Microbial turnover rates of carbon, nitrogen, and phosphorus are defined as the time required for synthesizing or decomposing the amount of microbial C, N, or P equivalent to the original “stand crop” at steady state. The turnover of microbial C, N, or P can be used to estimate the annual flux of C, N, or P in a soil (Brookes et al., 1984; Yao et al., 1999; Chen et al., 2002). Turnover of microbial biomass C, N, and P in soil can be determined by labeling the microorganisms in situ using 14C, 15N or 32P. The decay of 14C, 15N or 32P incorporated into microbial biomass and the change in microbial biomass C, N, and P are measured (Van Veen et al., 1987; Kouno et al., 1994). The decay of microbial 14C, 15N or 32P has been observed to follow a first-order reaction (Wu et al., 1990; Kouno et al., 1994; Yao et al., 1999). The turnover rate can be calculated based on the rate constant obtained by fitting the data to the first-order equation (Chen et al., 2001b). Several factors need to be carefully considered in the measurement and interpretation of microbial C, N, or P turnover in soil. (1) Starting time for measuring the decay of labeled microbial biomass. It generally takes about 7 –10 days to label soil microbial biomass in situ, depending on the pool size of the microbial biomass, soil type, and incubation conditions (Chen et al., 2002). The exact time to start measuring microbial turnover of C, N, or P should be the time when the 14C, 15N or 32P-labeled microbial biomass reaches its maximum. (2) Synthesized 14C, 15N or 32P from microbial metabolites. For a given cohort of microbial biomass, e.g., 14C-labeled microbial C, 15N-labeled microbial N or 32 P-labeled microbial P, the decay can be described by the first-order reaction Yt ¼ Y0 e2kt
ð1Þ
where Y0 is the initial amount of microbial C, N, or P in the cohort at time t ¼ 0; Yt is the remaining amount of microbial C, N, or P in the cohort after time t and k is the decay rate constant of the cohort. For this system, the turnover (T) can be expressed as: T ¼ 1=k:
ð2Þ
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 103
These equations are theoretically true for a cohort of biomass not replenished as individual organisms die. However, if Equation (1) is applied to labeled microbial 14C, 15N, or 32P present at the beginning and end of time t, an erroneous k occurs. This error occurs because the labeled microbial 14C, 15N, or 32P pool is being replenished by the decomposition of labeled metabolites during time t. Let b represent this erroneous value and then the true decay rate constant k is calculated after correction of the measured value for b. If the amount of labeled microbial 14C, 15N, or 32P that is synthesized during time t is known, we can subtract it from the total amount of labeled 14C, 15N, or 32P present at time t, and obtain k according to equation (1) and calculate T. The labeled microbial 14C, 15 N, or 32P synthesized during t can be calculated using the data obtained from the experiment (Chen et al., 2002b). Ignoring this factor could affect the estimation of microbial biomass C, N or P turnover. (3) Correcting laboratory-measured microbial turnover to field conditions. The laboratory-measured turnover of microbial biomass C, N, or P may not correlate to field conditions because of substrate amendments that enhance growth of microorganisms in the laboratory during labeling of microbial biomass and the difference in environmental conditions, particularly temperature, between laboratory and field conditions. Therefore, the measured turnover of microbial biomass C, N, or P needs to be corrected for the effect of substrate amendment and the difference in temperature before it is used to estimate the fluxes of C, N, or P through microbial biomass in a soil under field conditions (Jenkinson et al., 1987, Chen et al., 2002b). In addition, measurement of microbial C, N, or P turnover in acid soils is also subject to the limitations in the determination of microbial biomass C, N, or P (Yao et al., 1999; Chen et al., 2002b), as discussed in the previous section.
C. MICROBIAL COMMUNITY STRUCTURE Soil microbial community structure is an important indicator of soil quality (Elliott et al., 1996; Kelly and Tate, 1998; Waldrop et al., 2000). Several methods have been proposed for characterizing soil microbial community structure (Tunlid and White, 1994). Among all the methods, the ester-linked fatty acids in the phospholipids (PLFAs) are currently the most sensitive and useful chemical measures of microbial functional structure for both fungal and bacterial communities (Tunlid and White, 1994; Cavigelli et al., 1995; Zelles et al., 1995a,b). Analyses of monocultures and consortia of microorganisms isolated from the environment have shown that subsets of a microbial community can be identified by specific “signature” PLFAs (Table V). PLFAs method has been successfully used for characterizing microbial community structure in acid soils, as affected by land use and management (Bending et al., 2000; Waldrop et al.,
104
Z. L. HE ET AL.
2000; Yao et al., 2000b), heavy metal pollution, and soil degradation (Giller et al., 1998; Kelly et al., 1999). The BIOLOG method is mainly used to determine microbial community physiological profiles in soils and the environment (Tunlid and White, 1994). However, the BIOLOG method is subject to limitations when applied to acid soils (Yao et al., 2000a). Application of triphenyl-tetrazolium chloride (TTC) as designed in the original procedure (Haack et al., 1995) markedly decreased color development and resulted in a biocidal effect on soil microorganisms. The TTC in the BIOLOG method might affect the characterization of a soil microbial community. Therefore, removal of the TTC from BIOLOG plates is desirable for obtaining better results. The pH of the medium influenced utilization of carbon sources, thus affecting the characterization of the microbial communities. Yao et al. (2000a) determined microbial communities in eight red acid soils using microtitre plates containing 21 carbon sources at two pH levels (pH 4.7 and 7.0). For acid soils, a medium pH at 4.7 facilitated differentiation of the microbial communities as compared to a medium at pH 7.0, although the average utilization of the carbon sources in the plates of two pH levels followed the same pattern. This finding indicates that a characterization medium at different pHs may be required for soils with varying pH. For acid soils, a low pH plate can provide a better results than a high pH plate in the sole carbon source test.
D. SOIL ENZYME ACTIVITY Soil enzymes are proteins with catalytic properties owing to their power of specific activation that can cause biochemical reactions to proceed at faster rates (Tabatabai, 1994). Factors that affect protein denaturation, such as pH, ionic strength, temperature, and the presence or absence of inhibitors or activators have a marked influence on the catalytic function of enzymes. Sources of enzymes in soils are primarily the microbial biomass, although enzymes can also originate from plant and animal residues. An assay of soil enzyme activity involves measurement of a specific reaction (changes in concentration of the substrate/ reactant and/or the product) catalyzed by the enzyme while all other potential contributors to the reactions are eliminated or minimized. Therefore, factors that affect enzyme properties and reaction such as soil handling and storage, soil sterilization, and assay conditions have important influences on the assay of soil enzyme activity (Tabatabai, 1994; Gianfreda and Bollag, 1996). Air-drying of field soils at room temperature has been reported to cause an increase (Tabatabai and Bremner, 1970), a decrease (Tabatabai and Singh, 1976; Frankenberger and Tabatabai, 1980), or no change (Gianfreda and Bollag, 1996) in soil enzyme activities. Yao (2000) studied effects of air-drying on the activities of invertase, phosphatase, and catalase in eight acidic soils. Air-drying of field soils decreased
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 105 Table V Examples of Signature PLFAs for Microorganisms Eubacteria Common signature Sulfate reducer Methane-oxidizing bacteria Flavobacterium balustinum Francisella tularensis Actinomycetes Fungi
i15:0, a15:0, 15:0, 16:1w5, i17:0, 17:0, 18:1w7 10 Me16:0, Br17:1, 17:1w6 Type 1: 16:1w8c,16:1w8t, 16:1w5c Type 2: 18:1w8c, 18:1w8t, 18:1w6c i17:1w7, Br 2OH-15:0 24:1w5c, 22:1w13c, 20:1w11c 10 Me18:0 18:2w6, 18:3w6
a
Modified from Tulid and White (1994); fatty acids are designated as total number of carbon atoms: number of double bonds, with the position closest to the aliphatic (w) end indicated and using c for cis and t for trans. The prefixes “i”, “a”, and “Br” refer to iso, anteiso, and (ms) methyl-branching in unconfirmed positions.
the activities of invertase by 1 –4 times, and phosphatase by 1 –3 times, depending on soil properties. Catalase activity was much less affected by airdrying. Some soil fertility indices such as organic matter, total N, total P, and microbial biomass carbon were more closely correlated with enzyme activity measured with fresh field moist soil than with the air-dried soils (Yao, 2000). Similar observations on acid soils of the Appalachian region of the USA and the Cerrado region of Brazil have been reported (Baligar et al., 1991a,b, 1999; Baligar and Wright, 1991). The moisture content of fresh soils should be maintained at the same level as when collected, prior to assaying of soil enzyme activity. For more discussion on factors of soil enzyme activity measurement, readers are referred to the review papers by Tabatabai (1994) and Gianfreda and Bollag (1996).
V.
MICROBIOLOGICAL AND BIOCHEMICAL INDICATORS OF ACID SOIL QUALITY
A number of microbiological and biochemical parameters have been suggested as indicators of soil quality (Table VI). They include microbial biomass carbon (Cmic), nitrogen (Nmic), and phosphorus (Pmic) and their turnover rates, the MQ (Cmic/organic C ratio), basal respiration (qCO2), the MMQ (qCO2/ Cmic), the ratio of microbial N over total Kjeldahl N, and enzyme activity, etc. Recently, microbial diversity parameters such as community level physiological profile (CLPP) (Bending et al., 2000; Yao et al., 2000b), phospholipid fatty acids (PLFAs) (Kelly and Tate, 1998; Kanazawa and Berthelin, 1999), the ratio of Gram-negative/Gram-positive bacteria (Press et al., 1996), the ratio of fungal/ bacterial microorganisms (Waldrop et al., 2000), and free-living diazotrophic
106
Table VI Microbiological and Biochemical Indicators of Soil Quality Soil quality indexes Microbial biomass C (Cmic)
Microbial biomass N (Nmic) Microbial biomass P (Pmic)
The specific respiration rate (SRR) MMQ (SRR/Cmic)
The ratio of SRR to organic C Nc/Nk ratioa
References
Indicator of cropping, land use, and management history, organic matter change Land use and management effects on soil fertility, N availability Land use and management effects on soil fertility change, P availability Indicator of cropping, land use, and management history, organic matter change, soil pollution Indicator of environmental stress, measured together with Cmic and organic C Indicator of cropping, land use, and management history and organic matter change, soil pollution Indicator of soil pollution Indicator of soil degradation, soil pollution
Bolinder et al., 1999; Haynes and Tregurtha, 1999; Jordan et al., 1995; Yeates et al., 1997
Microbial community structure
Indicator of heavy metal contamination, soil erosion; fertility restoration, management, ecosystem function, etc.
Activities of soil enzymes (phosphatase, arylsulfatase, and dehydrogenase, etc.)
Indicator of cropping and management history and organic matter change, soil degradation and contamination, climate change, etc.
McCarty and Meisinger, 1997; He et al., 1997b; Yao and He, 1998, 1999 Brookes et al., 1984; Chen et al., 2000; He et al., 1997a; Dalal, 1998 Brookes, 1995; Dalal, 1998; Haynes and Tregurtha, 1999; Yeates et al., 1997 Chander and Brookes, 1993; Chandini and Parkinson, 2000; Killham, 1985 Brookes, 1995; Dalal, 1998; Haynes and Tregurtha, 1999; Killham, 1985; Insam et al., 1996 Valsecchi et al., 1995 Leiros et al., 1999; Trasar-Cepeda et al., 1998 Banerjee et al., 1997; Brookes, 1995; Kelly and Tate, 1998; Doran et al., 1999; Kanazawa and Berthelin, 1999; Bending et al., 2000; Yao et al., 2000a,b Acosta-Martinez and Tabatabai, 2000; Bandick and Dick, 1999; Bergstrom et al., 1998; Jordan et al., 1995; Wick et al., 1998
Z. L. HE ET AL.
MQ (Cmic/organic C)
Relationship to soil condition or function
Table VI (continued)
Free-living diazotrophic bacteria The G 2 /G þ bacteria ratio The fungal/bacterial ratio The ratio of direct counts/culturable counts a
Relationship to soil condition or function Indicator of soil pollution Organic waste effect and fertility change Land use and soil management Indicator of heavy metal contamination and soil degradation
References Mikanova et al., 1996 Press et al., 1996 Waldrop et al., 2000 Pepper and Rensing, 2000
Nc, calculated from Cmic, mineralized N, phosphomonoester, b-glucosidase, and urease activities; Nk, total N measured by the Kjeldahl method.
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 107
Soil quality indexes
108
Z. L. HE ET AL.
bacteria (Mikanova et al., 1996), etc. have also been identified as important indicators of soil quality. All these microbiological and biochemical parameters have shown to relate to nature and productivity of soils and respond to changes of land use, heavy metal concentrations, vegetation coverage, cropping system, agricultural practices such as liming, fertilization, and tillage as well as climatic factors (temperature and rainfall) (Elliott et al., 1996; Dalal, 1998).
A. MICROBIAL BIOMASS INDICATOR OF SOIL FERTILITY QUALITY The importance of microfauna and microflora to soil quality is well known (Smith and Paul, 1990; Brookes, 1995; Dalal, 1998). Soil microbial biomass serves as: (1) a labile source or an immediate sink of carbon, nitrogen, phosphorus, and sulfur (Diazravina et al., 1993; He et al., 1997a; Dalal, 1998; Chen et al., 1999a); (2) a driving force of nutrient transformation and pesticide degradation in soils (Smith and Paul, 1990); and (3) a cementing agent of soil aggregates (Ladd et al., 1996; Chan and Heenan, 1999). In addition, microbial biomass is sensitive enough to measure: (1) early changes due to different land use and management (Ghoshal and Singh, 1995; Biederbeck et al., 1996; Staben et al., 1997; Aslam et al., 1999); (2) soil erosion (Lal, 1997; Islam and Weil, 2000a,b); (3) fertility restoration of eroded land (Mao et al., 1992; Yao et al., 1998); (4) tillage and fertilization practices (Ghoshal and Singh, 1995; He et al., 1997a; McCarty and Meisinger, 1997; Bolinder et al., 1999); (5) crop rotations (Chan and Heenan, 1999); and (6) heavy metal contamination (Brookes 1995; Giller et al., 1998). Few studies have been conducted on acid soils. Being highly weathered and subjected to soil erosion, nutrients released from minerals in acid soils are very limited. Microbiological and biochemical processes are crucial for sustaining soil fertility in acid soils in tropical and subtropical regions (Jia and Insam, 1991; He et al., 1997b). Soil microbial biomass C and N were reported to be significantly higher in productive acid soils of Kenya’s Central Highlands (Murage et al., 2000) or in more fertile acid red soils of Southeast China (Yao et al., 1998; Yao and He, 1999). The relationship between soil microbial biomass and crop yield of acidic Ultisols was investigated in three long-term field experiments by Insam et al. (1991). On all three sites soybean (Glycine max L.) yield was significantly correlated with microbial biomass C. On two sites, there were positive correlations of Cmic with the yields of sorghum, rye and corn (Insam et al., 1991). The relationships between Cmic or Nmic and crop yield or plant uptake of N in acid soils with increasing fertility were examined in a greenhouse study (He et al., 1997a; Yao et al., 1998) and in field experiments (Chen, 2000). The tested soils varied greatly in fertility as shown by the difference in organic carbon, total
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 109
N and available N (Table VII). Microbial biomass C in the soils ranged from 20.2 to 425.8 mg kg21, accounting for on an average 1.9% of total organic C (Corg) (Table VIII). Soil no. 8 contained 20 times more Cmic than soil no. 1, which contained much less organic matter and nutrients due to severe erosion. Microbial biomass N (Nmic) varied from 4.0 to 52.6 mg kg21, accounting for, on the average, 2.9% of total N (NT) or 28.8% of available N (NA) in the soils (Table VIII). The Nmic of soil no. 8 was 11 times more than that of soil no. 1. The average C/N ratio of microbial biomass (7.2) was about 3.7 units lower than that of soil (10.9), indicating that microbial biomass N was potentially available to plants. The Cmic was highly correlated with Corg, NT, and NA, whereas the Nmic was highly correlated with Corg, Cmic, NT, and NA (Table IX) (He et al., 1997b). These results showed that both Cmic and Nmic were closely related to soil fertility in these highly weathered soils. Results from greenhouse pot experiments were in good agreement with the soil fertility measurements (Table X). Dry matter yield (DMY) of ryegrass and N uptake of the plants were highly correlated with Cmic, and Nmic (Table XI). Soil organic matter and total N content are generally considered the most important soil fertility properties related to crop yields (Olson, 1986). This study (He et al., 1997b) demonstrated that the correlation between ryegrass DMY and Cmic was better than that between the DMY and Corg, and the correlation between N uptake and Nmic was better than that between N uptake and NT. These results indicate that microbial biomass is an important available N pool in red soils. Because of strong adsorption and occlusion of phosphate by Fe and Al oxides, plant availability of P is generally low for most highly weathered acid soils (Lin, 1995). Microbial biomass P has been reported to make a significant contribution to plant-available P pools in soil (Brookes et al., 1984; Perrot and Sarathchandra, 1989; Srivastava and Singh, 1991; Joergensen et al., 1995; Chen et al., 1999b). Laboratory analyses and greenhouse experiments were conducted by Chen et al. (2000) to examine the relationships between plant P uptake, chemical indexes of P, and Pmic in red acid soils at different fertility levels. Microbial biomass P, ranging from 2.1 to 43 mg kg21 (Table VIII), was comparable with Bray-1 extractable P in the red soils (Table VII). The Pmic was significantly correlated with total P, or Bray-1 extractable P (Table IX). The significant relationship between Pmic and extractable P suggests that Pmic has a great potential in predicting the P-supplying ability in red soils. Greenhouse experiments showed close relationships among ryegrass DMY, plant P uptake or tissue P concentration and Pmic in red soils. These results imply that Pmic plays an important role in the availability of P to plants, and is a potential biological index of P availability in the red soils. The amount of P stored in soil microbial biomass is generally larger than the contents in the above-ground plant biomass (Perrot and Sarathchandra, 1989). The importance of organic matter to soil structure is well recognized. As a living fraction of organic matter, the critical role of microbial biomass in soil
110
Table VII Basic Properties of Red Acidic Soils of China
1 2 3 4 5 6 7 8
Vegetation coverage Eroded—no crop Wheat/cotton—3 year Citrus—4 year Citrus—7 year Citrus—12 year Rapeseed/rice—15 year Tea—30 year Forest—38 year
Soil typea Ultisols Entisols Ultisols Ultisols Ultisols Ultisols Ultisols Ultisols
Organic C (g kg21)
Total N (g kg21)
Avail. N (NH4-N þ NO3-N) (mg kg21)
Total P (mg kg21)
Bray-1 P (mg kg21)
Soil C/N
Soil C/P
6.0 4.8 6.0 5.0 4.6 5.1 4.9 5.8
1.7 5.1 5.3 15.1 18.2 20.5 27.4 34.3
0.19 0.49 0.55 1.79 1.93 1.98 2.15 2.97
32.2 59.8 63.7 113.0 116.2 130.2 170.8 193.0
300 310 240 860 1804 750 550 440
0.9 1.8 43.8 171 215 120 117 4.11
8.2 11.2 10.4 8.4 9.0 12.3 12.9 14.4
5.6 16.5 21.9 17.6 10.1 27.3 49.8 78.0
Modified from He et al. (1997b) and Chen et al. (2000). a Ultisols: clayey, kaolinitic thermic plithitic Aquult; Entisols: loamy, siliceous thermic typic Udorthents.
Z. L. HE ET AL.
Soil no.
pH (H2O)
Soil no.
Cmic (mg kg21)
1 2 3 4 5 6 7 8
20.2 ^ 2c 152.2 ^ 10 129.9 ^ 8 235.3 ^ 15 264.6 ^ 20 361.9 ^ 22 400.2 ^ 30 425.8 ^ 35
% of total org. Ca 1.2 3.2 2.6 1.6 1.6 1.9 1.5 1.5
Nmic (mg g21) 4.0 ^ 0.3 23.7 ^ 1.5 19.0 ^ 1.2 31.9 ^ 2.5 35.2 ^ 2.5 45.2 ^ 3.0 47.9 ^ 3.2 52.6 ^ 4.1
Modified from He et al. (1997b) and Chen et al. (2000). a Percentage of Cmic in total organic C. b Percentage of Nmic in total N or available N. c Mean ^ SE from three replications.
% of total Nb
Microbial C/N ratio
1.90 5.51 3.96 1.84 1.93 2.95 2.35 2.69
5.1 6.4 6.8 7.4 7.5 8.0 8.4 8.1
Pmic (mg kg21) 2.1 ^ 0.2 6.8 ^ 0.7 20.2 ^ 1.4 31.5 ^ 2.9 42.3 ^ 3.7 30.8 ^ 2.8 24.6 ^ 2.1 15.3 ^ 1.1
% of total P
MicrobialC/P ratio
0.7 2.2 8.4 3.7 2.4 4.1 4.5 3.5
9.6 22.4 6.4 7.5 6.3 11.8 16.3 27.8
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 111
Table VIII Microbial Biomass-C (Cmic), -N (Nmic), and -P (Pmic) in Red Soils with Variable Fertility Levels and Cropping Systems
112
Z. L. HE ET AL.
Table IX Correlation Coefficients (r ) between Microbial Biomass C, N, or P and Soil Fertility Indices
Corg Cmic NT NA Nmic PT Pmic Pextr
Corga
Cmic
NT
NA
Nmic
PT
Pmic
0.971**b 0.936* 0.996** 0.954** 0.264 0.416 0.268
0.903* 0.974** 0.996** 0.568 0.778* 0.670*
0.930* 0.898* 0.404 0.549 0.407
0.963** 0.247 0.471 0.296
0.289 0.514 0.344
0.836** 0.867**
0.944**
Modified from He et al. (1997b) and Chen et al. (2000). Corg: total organic C; Cmic: microbial biomass C; NT: total N; NA: available N, Nmic: microbial biomass N; PT: total P; Pmic: microbial biomass P; Pextr: extractable P by Bray-1 reagent. b * and ** indicate 5 and 1% significant levels, respectively. a
stability is increasingly evident (Ladd et al., 1996; Chan and Heenan, 1999). Chan and Heenan (1999) in Australia detected significant differences in the quality (composition) of soil organic carbon in the red earth soils from different crop rotations. Soil from a wheat/lupin rotation had the highest salt- and acidextractable carbon, and soil from a wheat/barley rotation had the most hot water extractable carbon and Cmic. The observed differences in aggregate stability were only significantly related to Cmic, which made up of 1.3 – 1.7% of the total organic carbon in the soils (Chan and Heenan, 1999).
INDICATOR OF SOIL CONTAMINATION Increasing disposal of sewage sludge, industrial by-products, and municipal wastes on land increases heavy metals in soils. Heavy metal pollution affects plant growth, causes contamination of food chains, and may impair functions of the soil, resulting in soil quality degradation. Microbial biomass, the living component of soil, can serve as an early indicator of heavy metal contamination (Brookes and McGrath, 1984; Chander and Brookes, 1993; Brookes, 1995; Valsecchi et al., 1995; Giller et al., 1998). Heavy metals decrease microbial biomass by directly killing or biochemically disabling organisms in soil. The amounts of Cmic in agricultural soils applied with sewage sludge or sewage sludge-containing composts were much smaller than soils receiving farmyard manure over the same period (Brookes and McGrath, 1984). The Cmic decreased with increasing amounts of EDTA-extractable nickel (Ni) and copper (Cu). These effects were detectable even after 20 years of application. A combination of zinc (Zn) and Cu at high concentrations had an additive adverse effect on the amounts
Soil no.
Ryegrass DMY (I)a (g pot21)
Plant N conc. (g kg21)
Plant N uptake (mg pot21)
Ryegrass DMY (II) (g pot21)
Plant P conc. (g kg21)
P uptake (mg pot21)
1 2 3 4 5 6 7 8
1.66 ^ 0.2b 3.50 ^ 0.4 2.17 ^ 0.2 4.05 ^ 0.4 7.33 ^ 0.6 7.82 ^ 0.8 9.62 ^ 1.0 7.88 ^ 0.8
1.37 ^ 0.08 1.41 ^ 0.08 1.58 ^ 0.13 1.59 ^ 0.14 1.69 ^ 0.16 1.38 ^ 0.11 1.72 ^ 0.15 2.01 ^ 0.20
22.7 ^ 1.7 49.4 ^ 3.2 34.3 ^ 2.6 64.4 ^ 5.9 123.9 ^ 10 107.9 ^ 9.0 165.5 ^ 15 158.4 ^ 16
1.25 ^ 0.1 1.02 ^ 0.07 1.61 ^ 0.15 1.58 ^ 0.13 2.84 ^ 0.17 2.08 ^ 0.16 1.38 ^ 0.11 1.96 ^ 0.09
1.36 ^ 0.06 1.27 ^ 0.04 3.24 ^ 0.12 5.01 ^ 0.15 6.18 ^ 0.21 2.97 ^ 0.13 4.33 ^ 0.13 1.61 ^ 0.08
1.70 ^ 0.13 1.30 ^ 0.11 5.22 ^ 0.61 7.92 ^ 0.77 17.6 ^ 1.65 6.18 ^ 0.44 5.98 ^ 0.35 3.16 ^ 0.17
Modified from Chen (2000) and Yao (2000). a Ryegrass DMY (I) from greenhouse pot experiment oriented to microbial biomass N and ryegrass DMY (II) from a separate greenhouse pot experiment oriented to microbial biomass P. b Mean value followed by the standard error from three replications.
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 113
Table X DMY, Plant N and P Concentrations and N and P Uptake of Ryegrass
114
Table XI Correlation Coefficients (r ) between Microbial Biomass C, N, or P and Plant Growth Parameters NT
NA
PT
Pextr
Cmic
Nmic
Pmic
Nuptake
Puptake
0.902**b 0.782* 0.962** 0.438 0.236 0.224
0.866** 0.756* 0.896* NDc ND ND
0.899* 0.778* 0.947* ND ND ND
ND ND ND 0.840** 0.811* 0.958**
ND ND ND 0.651 0.955** 0.891**
0.938** 0.649 0.938* 0.418 0.269 ND
0.923* 0.645 0.921* 0.430 ND ND
ND ND ND 0.785* 0.909** 0.891**
0.971** 0.719* ND ND ND ND
ND ND ND 0.844** 0.903** ND
Modified from Chen (2000) and Yao (2000). a Corg: total organic C; Cmic: microbial biomass C: NT: total N; NA: available N; Nmic: microbial biomass N; Nuptake: plant N uptake; DMY: dry matter yield of ryegrass. b * and ** indicate 5 and 1% significant levels, respectively. c ND ¼ not determined.
Z. L. HE ET AL.
DMY (I) Plant N conc. Nuptak DMY (II) Plant P conc. Puptake
Corga
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 115
of Cmic present (Chander and Brookes, 1993). At similar concentrations, Cu decreased Cmic by 50% more than did Zn. Addition of Cu to three red soils at rates of 500 mg kg21 lowered Cmic by 10 to . 40%, depending on soil texture (Yao, 2000). The toxic effect was more pronounced in sandy soil than in clayey soil, probably due to differences in adsorption and bioavailability of Cu between the two soils. However, changes in microbial biomass are often transient if soil pollution is caused by organic chemicals such as pesticides or herbicides, as sensitive populations are replaced by more stress-tolerant populations within a short time. Combining population measurement and microbial activities such as specific respiration or MQ, and MMQ are recommended for early detection of soil pollution.
B. MICROBIAL BIOMASS TURNOVER The turnover rates of Cmic, Nmic, Pmic can be good indicators of soil quality, as these parameters reflect the soil’s ability to recycle nutrients and energy and to buffer external changes (He, 1997). Annual fluxes of C, N, and P calculated by multiplying Cmic, Nmic, or Pmic with the respective turnover rate may provide a rough estimate of N, and P available to plants (Brookes et al., 1984; Kondon et al., 1989; Kouno et al., 1994; Diazravina et al., 1995). The turnover rates of Cmic, Nmic, and Pmic were related to size of microbial biomass, land use and management, and soil texture (Yao et al., 1998; Chen et al., 2002a,b). Yao et al. (1999) evaluated Nmic turnover rate in three acid soils with different textures. The turnover rates of Nmic were 250, 89, and 63 days, respectively, for a long-cultivated red clayey soil (no. 7), a recently cultivated red clayey soil (no. 3), and a red sandy soil (no. 2). The corresponding annual fluxes of nitrogen through microbial biomass were 69, 78, and 137 mg kg21, which is more than the amount of N in the standing crops. For the sandy soil, the annual flux of N through microbial biomass was about 2.5 times greater than extractable soil available N (Table VII). These findings agree with field observations that plant-availability of N was higher in the sandy soil than in the clayey soil (Yao et al., 1999). Obviously, microbial biomass serves as an important dynamic N source for plants in the red soils. The turnover rates of Pmic in the three red soils were 130, 190, and 217 days, respectively, 1.7– 2.8 times per year (Chen et al., 2002a,b). Pmic was 12.2, 17.8, and 31.5 mg kg21, respectively, for the three soils. Microbial biomass turnover provided a dynamic source of available phosphorus (P), 2– 3 times greater than the Pmic. The annual fluxes of P through microbial biomass in these soils could amount to 91 –146 kg ha21, 3 – 5 times greater than the amount of P annually removed by the harvested crops. The turnover rate of Pmic was greater in the sandy soil than in the two clayey soils. This is explained by the larger amount of Pmic and lower mineralization in the clayey
116
Z. L. HE ET AL.
soils, as compared to the sandy soil. Apparently, the turnover of Pmic plays an important role in the supply of plant-available P and in ecological cycling of P in subtropical red soils. Microbial biomass and its turnover rate are very promising indicators of soil quality. To date, there are no established benchmarks, critical or threshold values against which soil quality can be evaluated. Benchmark values will vary in different soil types and land use (Sparling, 1997), climate (Insam and Haselwandter, 1989), and vegetation (Martens, 1995). These differences can be reduced by comparing the derived microbial indices including MQ, respiratory quotient, and the ratio of microbial specific respiration rate/organic C, etc.
C. MICROBIAL BIOMASS-RELATED INDICATORS Other important microbial indicators of soil quality are MQ, basal respiration rate, MMQ, the ratio of basal respiration rate/total organic C (qCO2/total organic C, qCO2/Corg), and the ratio of direct counts/culturable counts, etc. MICROBIAL QUOTIENT Microbial quotient is defined as percentage of Cmic in the total Corg (Smith and Paul, 1990) and is a measure of soil organic matter quality (Gregorich et al., 2000). MQ was observed to be sensitive to land use and management (Haynes and Tregurtha, 1999; Islam and Weil, 2000b), ecosystem changes (Pinzari et al., 1998), soil fertility or soil productivity (Insam et al., 1991; Sparling, 1992; Chen, 2000), and heavy metal contamination (Chander and Brookes, 1993; Brookes, 1995; Valsecchi et al., 1995; Giller et al., 1998). Therefore, MQ can provide an early indication of soil quality changes. Haynes and Tregurtha (1999) compared the effects of intensive cultivation of vegetables and pasture on soil quality of a Typic Haplohumult. MQ decreased from 2.3 to 1.1% as soil organic matter content declined from 65 to 15 g C kg21 with prolonged vegetable cultivation. Conservative management or organic farming is a combination of reduced tillage, increased crop diversity, more perennial crops, increased crop residue return, increased soil fertility and/or increased application of organic amendments, as compared with conventional practices. Conservation management and organic systems consistently and markedly influenced soil quality by increasing soil microbial biomass and the MQ (Fliessbach and Mader, 2000; Islam and Weil, 2000b). The MQ was also found to increase with crop yield (Insam et al., 1991). The MQ in some Chinese red acid soils decreased with land use in the following order: fallow grassland ¼ arable upland . orchards (bamboo or citrus) . forest ¼ tea garden ¼ vegetable crops ¼ paddy (Chen, 2000).
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 117
The MQ was reported to decrease from 1.5% in soil that received uncontaminated sewage sludge to 1% in soil receiving Zn (457 mg kg21 soil) contaminated sludge and to 0.5% in soil receiving Cu (415 mg kg21 soil) contaminated sludge (Chander and Brookes, 1993), implying that the MQ is more sensitive to increased heavy metal concentrations than the Cmic itself. The decrease in MQ could be caused by decreased microbial biomass and/or partially disabled function of its ability to mineralize organic matter. As a result of decreased MQ, the turnover rate of organic matter decreased in soil with the application of heavy metal contaminated sewage sludge. Valsecchi et al. (1995) found that a high input of heavy metals through long-term effluent decreased the MQ more than one order of magnitude, i.e., from 4% in the slightly contaminated soil to 0.2% in the heavily contaminated soil. These results suggest that MQ can provide a very sensitive indicator of the adverse effects of increased heavy metal concentration on soil microbial biomass. BASAL RESPIRATION RATE Basal or maintenance respiration rate (qCO2, mg CO2 –C kg21 per day) is a measure of microbial activity and biomass (Nanipieri and Grego, 1990; Brookes, 1995). Insam et al. (1991) observed that the qCO2 was negatively correlated with soybean yield on Ultisols at all three experimental sites. This implies that if more C is lost by microbial respiration with less C input, more care must be taken to maintain organic C contents (Insam et al., 1991). Conservative management and organic farming resulted in a decreased qCO2, indicating reduced stress on the soil microbial communities (Liebig and Doran, 1999; Islam and Weil, 2000b), whereas natural forest brought under cultivation caused increased qCO2, implying increased stress on the soil microbial communities (Islam and Weil, 2000a,b). Increasing concentrations of heavy metals often increases the qCO2 and the ratios of qCO2/Cmic and qCO2/organic C. This means that heavy metal toxicity reduces the energy utilization efficiency of the microbial metabolic processes, which then require greater amounts of C for maintenance, thus reducing the quantity of C incorporated into the microbial biomass (Valsecchi et al., 1995).
MICROBIAL METABOLIC QUOTIENT The microbial metabolic quotient, also called specific respiration, is defined as respiratory CO2 released per unit microbial biomass (the ratio of qCO2/Cmic, % day21) (Insam et al., 1996; Dalal, 1998). A change of the MMQ may indicate: (1) changes in substrates that an unchanged microbial community uses; (2) a change of microbial community composition; (3) a change in both (1) and (2); (4) no
118
Z. L. HE ET AL.
changes of substrates or community, but a change in the physiological status of the community due to altered maintenance requirement (Insam et al., 1996). Another relevant soil quality indicator is the ratio of qCO2/total organic C (% day21). Both were reported to be sensitive to land use, management changes (Islam and Weil, 2000a), and/or heavy metal contamination (Brookes and McGrath, 1984; Valsecchi et al., 1995). The MMQ is lower with conservative management systems than the conventional systems (Islam and Weil, 2000b), indicating a higher utilization efficiency of energy. The MMQ was also found to significantly increase with increasing concentrations of heavy metals whereas the ratio of qCO2/Corg decreased in the soils contaminated with heavy metals (Valsecchi et al., 1995), indicating a state of microbial stress due to heavy metal toxicity. OTHERS Several other microbiological and biochemical indices have been proposed in recent years. The ratio of the total N calculated from biochemical properties (Nc) to the total N measured by the Kjeldahl method (Nk) was recommended for rapid evaluation of soil degradation (Leiros et al., 1999). The Nc/Nk can distinguish among biochemically balanced soils, soils in a transient state of high microbiological and biochemical activity and degraded soils. It is sensitive to the presence of pollutant and able to differentiate between prior and pollutioninduced soil degradation (Trasar-Cepeda et al., 2000a,b). Mikanova et al. (1996) found that the most sensitive indicator of the anthropogenic soil load is free-living diazotrophic bacteria. Populations were two to three orders of magnitude lower in soil samples from polluted sites (where average annual SO2 and flying dust concentrations exceeded 100 mg SO2 m23 and 100 mg dust m23 air in 1985 –1989) than in those from the control plots. The numbers of actinomycetes, particularly of cellulolytic ones and of oligotrophic bacteria were enhanced in soil samples from anthropogenically polluted areas. Recently, Pepper and Rensing (2000) pointed out that the ratio of direct counts/cultural counts of microbes is potentially more sensitive than Cmic as an indicator of soil degradation and pollution. These results imply that microbial community structure is one of the important indicators of soil quality.
D. MICROBIAL COMMUNITY STRUCTURE INDICATORS Microorganisms differ in their sensitivity to nutritional and environmental changes (Giller et al., 1998). Some populations, more tolerant to stress, can survive whereas more sensitive populations may disappear under changed conditions. Thus parameters of microbial community structure, which can be
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 119
assessed by various physiological, biochemical or molecular techniques, have been recommended as a biological indicator of soil quality or sustainability (Kennedy and Smith, 1995). Two microbiological measurements, which hold considerable promise as rapid methods for microbial diversity, are carbon source utilization patterns (BIOLOG method, Haack et al., 1995) and PLFAs profiles (Zelles et al., 1992, 1995b). These two are usually used to characterize and identify microbial community structure (Cavigelli et al., 1995; Zelles et al., 1995a; Waldrop et al., 2000). Microbial community structure can provide a sensitive reflection of soil quality under different land use and management or environmental conditions (Grayston and Campbell, 1996; Baath et al., 1998; Drijber et al., 2000). The conversion of a tropical forest to pineapple plantation agriculture increased the relative amounts of fungi and actinomycetes, and decreased the relative amount of Gram-positive bacterial biomarkers (Waldrop et al., 2000). Patterns of microbial substrate utilization and metabolic diversity were observed to be more sensitive to the effects of management (such as crop rotations, plough-in of vetch crop or ley treatment) than to organic matter and biomass pools and therefore, have a value as an early indicator of the impacts of management on soil biological properties, and hence soil quality (Bending et al., 2000). Press et al. (1996) investigated organic by-product effects on soil chemical properties and microbial communities and found that newsprint plus NH4NO3 resulted in a shift to more Gram-positive bacteria, while newsprint plus poultry litter resulted in a shift to more Gram-negative bacteria. Both N sources resulted in a reduction in Bacillus sp. population. Shifts in bacterial populations and changes in species richness (number of species detected) and evenness (relative abundance of each species) were induced by organic by-product addition. Shifts in the microbial community structure towards more Gram-negative organisms can benefit plant growth and may be useful as an indicator of soil quality (Press et al., 1996). Soil microbial community is sensitive to increased concentrations of heavy metals (Giller et al., 1998). An increase in heavy metals concentrations by sewage sludge application decreased biomarkers of actinomycetes, arbuscular mycorrhizal fungi, and total fungi, but increased the relative amount of bacteria (Kelly et al., 1999). Bacteria have been observed to be more resistant to high concentrations of heavy metals than other microbial populations (Angle et al., 1993; Kozdroj, 1995).
E. ENZYME ACTIVITIES Enzyme activity in soils results from the activity of accumulated enzymes and from enzymes in proliferating microorganisms. The accumulated enzymes in soils are regarded as enzymes present and active in a soil, in which no microbial
120
Z. L. HE ET AL.
proliferation occurs. Enzyme activity in soils is derived from free enzymes, such as exoenzymes released from living cells, endoenzymes released from disintegrating cells, and enzymes bound to cell constituents (Tabatabai and Fu, 1994). Phosphatase, urease, arylsulphatase, pyrophosphatase, and dehydrogenase enzyme activities are closely related to soil fertility properties such as organic C, pH, moisture content, and nutrient status (Baligar and Wright, 1991; Baligar et al., 1988, 1991a,b, 1999). Soil organic C appears to have the greatest influence on the level of enzyme activities in acid soils (Baligar et al., 1999). Each soil type was shown to have its own inherent level of enzyme activities for phosphatase, urease, arylsulphatase, pyrophosphatase, and dehydrogenase (Baligar et al., 1988, 1991a –c, 1999). Recently, more research has been directed to examining the relationship between enzyme activity and soil quality. Soil enzyme activity is reported to respond to changes in land use and management (Bergstrom et al., 1998; Wick et al., 1998; Bandick and Dick, 1999; Acosta-Martinez and Tabatabai, 2000; Carreira et al., 2000), soil degradation and fertility restoration (Trasar-Cepeda et al., 2000b; Yao et al., 2000b), and heavy metal pollution (Garcia-Gil et al., 2000; Huang and Shindo, 2000). Therefore, enzyme activity may serve as indices of soil quality. In general, response of soil enzyme activity to land use or management is associated with organic input. Enzyme activities were higher in continuous grass fields than in cultivated fields. In cultivated systems, activity was higher where cover crops or organic residues were added as compared to treatments without organic amendments (Bandick and Dick, 1999). Bergstrom et al. (1998) found that no tillage and previous cropping to forage increased activity of all enzymes (dehydrogenase, urease, glutaminase, phosphatase, arylsulfatase, and P-glucosidase). Enzyme activity response to tillage practices was not consistent for soils with relatively large total organic C. Phosphatase, urease, arylsulphatase, pyrophosphatase, and dehydrogenase enzyme activities in acid soils of the Appalachian region, USA declined with soil depth and such decline was directly related to reduction in soil organic matter content (Baligar et al., 1991a – c). Different soil enzymes may vary in their response to soil quality changes. Wick et al. (1998) investigated soil microbiological and biochemical properties in relation to soil quality under improved fallow management systems with senna (Senna siamea L.), leucaena (Leucaena leucophala L.), and pueraria (Pueraria phaseoloides ) on the severely degraded and non-degraded Alfisols in southwestern Nigeria. They observed that contrasting fallow management systems (alley cropping, live mulch, planted fallow, and controls in long-term experiments) at three sites differing in degree of soil degradation could be evaluated adequately by microbial biomass, alkaline phosphatase (AlkP), total N, b-glucosidase, and organic C. b-glucosidase was more sensitive to soil quality changes than total organic C. The AlkP test was more sensitive than microbial biomass in characterizing the effects of improved fallow management on site
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 121
degradation. However, the acid phosphatase (AcdP) and protease tests were not sensitive indicators for soil quality evaluations of long-term management trials (Wick et al., 1998). Yao (2000) studied soil enzyme activities (invertase, catalase, protease, urease, phosphatase, and peroxidase) and fertility properties of eight red acid soils of China. He found that the activity of invertase, catalase, urease, or phosphatase was correlated with total N, total P or organic C (except for catalase), but that of pretease or peroxidase had no relationship to any of the fertility properties. Enzyme activities measured using fresh soil were generally greater than those obtained using air-dried samples (Yao, 2000). Field moist soils stored at 48C recorded substantially higher phosphatase, urease, arylsulfatase, and dehydrogenase enzyme activities than air-dried soils of temperate hill land acid soils (Baligar et al., 1988, 1991a – c; Baligar and Wright, 1991). Since soil fertility indices were more closely related to enzyme activity in fresh soil than in air-dried soils, measurement of enzyme activity using fresh soil is recommended. Effects of soil pollution on enzyme activities are relatively complicated. The response of different enzymes to the same pollutant may vary greatly and the same enzyme may respond differently to different pollutants (Garcia-Gil et al., 2000; Trasar-Cepeda et al., 2000b). More studies are needed to understand the exact effects of pollutants on enzyme activities.
VI. SOIL pH VERSUS MICROBIOLOGICAL AND BIOCHEMICAL INDICATORS Soil pH, as a soil quality index, cannot be overemphasized, as it plays a critical role in regulating soil microbial community and activity, carbon and nutrient cycling, as well as many other soil physical and chemical processes. The effects of soil pH and liming on physical and chemical properties, especially nutrient transformation and availability have been extensively studied (Pearson and Adams, 1967; Haynes, 1984; Haynes and Swift, 1986). However, soil pH effects on microbiological and biochemical properties in relation to soil quality changes have not been well studied or understood, although increasing research efforts are now directed to this topic. Raising pH of acid soils reduces microbial biomass within a very short time (Chen, 2000), but then increases microbial biomass and enzyme activity in the long run (Badalucco et al., 1992). Chen (2000) examined the short term effects of pH changes on microbial biomass C (Cmic) and P (Pmic) in three acid soils (pH 4.6– 6.0) by adjusting the soil pH to 3, 4, 5, 6, 7, 8, and 8.5, respectively, using dilute HCl or NaOH. He measured the Cmic and Pmic after the treated soils were incubated for 7 days at room temperature. He observed that Cmic or Pmic was maximal at the original pH and significantly decreased at higher or lower pH. At extreme pH values, i.e., 3.0 at the acid end and 8.5 at the alkaline end,
122
Z. L. HE ET AL.
80 to . 90% of the microorganisms were killed (Fig. 2). Microorganisms in a soil with higher original pH (6.0) appeared to be more tolerant to pH changes than those in soils with lower original pH (pH 4.6 or 5.0) (Chen, 2000). This is probably because bacteria that dominated in the soil with a pH close to neutral are generally more tolerant to stress conditions than fungi that are dominant in acid soils (Badalucco et al., 1992; Neale et al., 1997; Giller et al., 1998). Liming acid soils improves soil physical and chemical properties and nutrient availability (Haynes, 1982, 1984; Haynes and Naidu, 1998). The beneficial effects may well be associated with liming-induced changes in microbiological and biochemical properties in the acid soils. Liming was reported to drastically reduce exchangeable Al, increase CO2 evolution, increase microbial biomass, Soil 1 (pH 6.0) Soil 2 (pH 4.6) Soil 3 (pH5.5)
500
Cmic (mg C kg 1)
400 300 200 100 0
50
Pmic (mg P kg 1)
40
30
20
10
0 2
3
4
5
6
7
8
9
Soil pH Figure 2 Effect of soil pH on microbial biomass C and P in three acid soils (measured at day 7 after pH adjustment and incubation); (From He et al., 2001b).
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 123
and increase MMQ (Badalucco et al., 1992). MQ (Cmic/Corg) increased and microbial C/N ratio decreased after liming, indicating that liming improves conditions for growth of microorganisms, especially bacteria and thus enhances microbial activity in acid soils. Badalucco et al. (1992) also observed that liming increased dehydrogenase activity but not phosphatase activity, because of a decrease in AcdP after the pH was raised. Acosta-Martinez and Tabatabai (2000) reported that liming acid soil from 4.9 to 6.9 had a profound influence on the activities of different enzymes. AcdP was negatively correlated with soil pH whereas the activities of all the other enzymes including a- and b-glucosidases, a- and b-galactosidases, amidase, arylamidase, urease, L -glutaminase, L asparaginase, L -aspartase, AlkP, phosphodiesterase, and arylsulfatase were positively correlated with soil pH. The Delta activity/Delta pH values ranged from 4.4 to 38.5 for the activities of the glucosidases, from 1.0 to 107 for amidohydrolases and arylamidase, 97 for AlkP, 39.4 for phosphodiesterase, and 11.2 for arylsulfatase. The value for AcdP was 2 35.0. These results support the view that soil pH is an important indicator of soil health and quality (Acosta-Martinez and Tabatabai, 2000). The differential effect of liming on alkaline and AcdP activities is interesting. Dick et al. (2000) investigated the potential of using AlkP and AcdP activities to determine the optimum soil pH for crop production and the amount of lime required to achieve this optimum. The ratio of AlkP/AcdP responded immediately to the changes in pH caused by CaCO3 additions, and an AlkP/ AcdP ratio of approximately 0.5 divided soils into those with appropriate pH adjustment and those still needing additional lime treatment. Many other factors may affect the AlkP/AcdP ratio in soils and thus more work is needed to validate the use of AlkP/AcdP ratio as a pH adjustment indicator or as liming criteria. Growing tea bushes over 20 years or more decreases soil pH and increases exchangeable Al due to tannic acid release from tea-plant roots (Pandey and Palni, 1997; Yu et al., 2002). The soil can be severely degraded after long periods of growth of tea plants, leading to a pH as low as 3 and poor physical and chemical properties (Pansombat et al., 1997; Yu et al., 2002). Recent studies indicate that soil microbial biomass, microbial activity, and microbial community structure are significantly changed in the degraded soils under tea bushes (Pansombat et al., 1997; Yao, 2000; Yu et al., 2002). Yu et al. (2002) investigated population distribution and community structure of microorganisms and the related ecological factors in rhizosphere soil of tea plants of 10, 40 and 90 years old, respectively. Populations of soil eutrophic and oligotrophic bacteria, fungi, actinomycetes and bacillus were measured by the plate-count method. Soil acidification and soil organic carbon, total soluble phenol and total nitrogen increased with the age of tea plants. Soil microbial biomass carbon in the tea plant rhizosphere increased significantly with the age of tea plants and there was a negative correlation between soil microbial biomass carbon and total colonyforming microorganisms on media plates. The culturable microbes decreased
124
Z. L. HE ET AL.
with the age of tea plants although total microbial biomass tended to increase. Yao et al. (2000b) observed that soils from the tea gardens were particularly distinctive in their microbial CLPP and microorganisms in these soils exhibited the lowest utilization of carbon. A large proportion of microbes in tea soils were probably in a dormant status with low metabolic activity. Yu et al. (2002) also found that soil C/N ratio and microbial C/N ratio increased with age of the tea plant, indicating a change in soil microbial community structure. Results on structural distributions of soil microbial communities show that percentages of eutrophic bacteria and fungi increased with age of the tea plant, whereas oligotrophic bacteria had a reverse trend. Kanazawa and Kunito (1996) showed that the number of Al-resistant microorganisms was largest in the strongly acid tea garden soil, followed by the forest soil and lowest in the upland soil. The number of Al-resistant microorganisms accounted for 12.5% of the total number of microorganisms in the tea garden soil, suggesting that the increase in Al-resistant microorganisms resulted from soil acidification. Fungi accounted for most of the high Al-resistant microorganisms. About one-half of the acid-tolerant fungi (grown at pH 3.0) from the tea garden soils consisted of highly Al-resistant microorganisms. Obviously, the interactions between pH and microorganisms in acid soils have a great impact on soil quality. The influence of soil acidity merits greater attention to identification of microbiological and biochemical indicators of soil quality. Multiple indices of microbiological and biochemical attributes need to be combined to evaluate soil quality changes. It is also important to integrate physical and chemical parameters with microbial parameters in developing a minimal set of soil quality indices.
VII. DEVELOPMENT OF ACID SOIL QUALITY INDEXING SYSTEMS Maintaining or enhancing soil quality is a key to sustaining the soil resources of the world. High quality soils will produce more food and fiber, and provide a better quality of life for the world’s growing population. Moreover, high quality soil will play a major role in stabilizing natural ecosystems and enhancing air and water quality. Based on soil’s functions, soil quality consists of the following three components: (1) production sustainability, (2) environmental quality, and (3) plant and animal health (Doran et al., 1999). The first component addresses the soil’s capability to supply plants with optimal nutrients, water, and physical conditions for growth, and to buffer against external changes, thus sustaining optimal conditions for bioproduction and environmental cleaning. The second component reflects the soil’s capacity to retain, disperse and transform chemical and/or biological materials and thus function as an environmental filter or buffer.
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 125
The third component is associated with the soil’s function in producing healthy food and fiber and in providing healthy habitats for plant and animals, as well as beneficial microorganisms. To evaluate acidic soil quality, the capacity of the soil to function needs to be measured using appropriate indicators. The most desirable attributes of an appropriate indicator include the following: (1) it measures one or more soil functions; (2) it is sensitive enough to measure changes due to disturbance, restoration, or changes in land use and management; (3) it provides benchmark, critical or threshold values; (4) it can be readily interpreted; and (5) it is costeffective. Currently great effort has been directed to characterizing different attributes of quality or identifying specific indicators or parameters that have the potential to address one or more components of soil quality. Many soil attributes including physical, chemical, biochemical, and microbiological parameters (Tables III and VI) have been suggested as indicators of soil quality changes, each assessing probably one or more specific functions of the soil, but none of them alone is comprehensive enough to cover all the components of soil quality. In addition, it may not be feasible or possible to develop benchmark values for each of the many proposed soil quality indicators. Therefore, it is essential to establish one or more indexing systems, with each index covering a number of inter-related, directly measurable parameters to better reflect the complex processes affecting soil quality and to compensate for the wide variations occurring in individual properties. Karlen and Stott (1994) presented a framework for evaluating site-specific changes in soil quality. In this approach, they define a high quality soil as one that: (1) accommodates water entry; (2) retains and supplies water to plants; (3) resists degradation; and (4) supports plant growth. They describe a procedure by which soil quality indicators are identified, assigned a priority or weight that reflects its relative importance, and scored using a systematic engineering approach. Karlen et al. (1994) also demonstrated the utility of this approach in discriminating changes in soil quality between long-term crop residue and tillage management practices. Wang and Gong (1998) used an information system approach to develop an indexing system for assessing acid soil quality. This approach was applied to evaluate quality changes of acid soils after 11 years of reclamation at the QianYan-Zhou experimental station, located in subtropical China. Changes in soil quality were assessed and analyzed for different land use effects. The Qian-YanZhou soil quality information system (QYZSQIS) was developed using ARC/ INFO and FOXBASE software. A relative soil quality index (RSQI) defined as fractions of soil quality indicators in the tested soils against a reference soil and its difference (DRSQI) before and after 11 years reclamation was established for comparing land use effects on soil quality indicators. The equation for calculating RSQI value is: RSQI(%) ¼ (SQI/SQIm) £ 100, where SQI is soil quality index and SQIm is the maximum value of SQI. The SQI is calculated from the equation:
126
Z. L. HE ET AL.
P SQI ¼ ðWi Ii Þ; where Wi is the weight assigned to each individual indicator, and Ii is the score assigned to each indicator class. An optimal soil in a specific region has a normalized RSQI of 100, but real soils have lower values, which indicate directly their distance from the optimal level. By computing RSQI values, soil quality in different regions can be compared even if they are measured using different systems, weightings, and classes. Similarly, the DRSQI can be used to quantify soil quality changes during a specific time period or between different regions. By integrating the QYZSQIS with databases of soil physical and chemical (including soil texture and depth, pH, organic C, total and available nutrients, and CEC) properties for different time periods, the system was able to quantify soil quality changes with time and space in small watersheds. The RSQI and the DRSQI provide a standard for evaluating spatial or temporal changes in soil quality. With this indexing system, it is possible to quantify the degree and direction of soil quality changes as affected by different land use and management, and intrinsic quality of the original soil (Wang and Gong, 1998). This indexing system could be useful for soil fertility quality evaluation. However, the assignment of weight to each indicator and the division of each indicator into different classes and subsequent scoring remain experiencedependent and merit further refinement. In addition, many microbiological and biochemical properties that are recently identified as important quality indicators of acid soils such as microbial biomass, microbial activity, microbial community structure, and enzyme activity, etc., need to be incorporated into this system. Another approach for developing an acid soil quality indexing system is to define a reference soil against which the quality change of soils can be compared or quantified. A native soil supporting climax vegetation that has undergone minimal anthropogenic disturbance is used as a high quality reference soil (Leiros et al., 2000). Trasar-Cepeda et al. (1998) examined three such native acid soils of Galicia (NW Spain) bearing Atlantic oak-wood as the climax vegetation and found that the native soils of Galicia exhibit a biochemical equilibrium such that total N can be defined as a function of five biochemical and microbiological parameters: total N( £ 1023) ¼ 0.38 microbial biomass C þ 1.40 N mineralization capacity þ 13.6 phosphomonoesterase þ 8.9 b-glucosidase þ 1.6 urease. The ratio of the calculated total N (Nc) to the total N measured by the Kjeldahl method (Nk) was proposed as an index of soil quality (Leiros et al., 1999). For these climax soils, the Nc/Nk ratio is 1.00. For disturbed soils, soil degradation is reflected by the Nc/Nk ratio, which differed more or less from 1.00 (Table XII). Here the values for the disturbed but unpolluted soil were 0.75 or 0.79 rather than 1.00, reflecting pre-pollution degradation of the soils at the two sites (Leiros et al., 1999). For the samples polluted by tanning effluent or landfill effluent, intense contamination is indicated by decreased Nc/Nk ratios, ranging from 0.15 to 0.28 in the former case and from 0.19 to 0.35 in the latter, significantly different from the Nc/Nk values in their corresponding unpolluted controls (Trasar-Cepeda et al., 2000a). These results together with some previous reports by the same research
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 127 Table XII Soil Total N Calculated from Biological Properties (Nc), Total N Measured by Kjaldhal Digestion Method (Nk), and the Nc/Nk Values of Soils as Affected by Disturbance and Contamination by Tanning Effluent or Landfill Effluent Soils
Nc (g kg21)
Nk (g kg21)
Nc/Nka
7.56 12.67 10.39
0.75 0.28 0.15
3.76 2.27 1.11
0.79 0.35 0.19
Tanning effluent Disturbed but non-polluted Disturbed and polluted Disturbed and severely polluted
5.69 3.56 1.59 Landfill effluent
Disturbed but non-polluted Disturbed and polluted Disturbed and severely polluted
4.75 6.53 5.83
Modified from Trasar-Cepeda et al. (2000a,b). a The Nc/Nk ratio ¼ 1.0 for undisturbed native soil under climax vegetation.
group (Leiros et al., 1999) indicate that this kind of indexing system is simple, but has the advantage of providing common criteria for comparing the degree of soil degradation at different sites and caused by different factors (pollution, land use or management) because all the examined soils can be ranked by a common single index (the Nc/Nk value). Moreover, some microbiological and biochemical properties of the soils have been taken into consideration in this system. More research is needed to test the applicability of the Nc/Nk ratio to other production or ecological systems. Gregorich et al. (1994) developed a concept of minimum data sets for assessing soil organic matter quality in agricultural soils. This approach is applicable to quality assessment of acid soils considering that, in addition to physical and chemical attributes, many microbiological and biochemical parameters are also critical for characterizing acid soil quality. As discussed above, soil quality has various components (fertility quality, environmental quality, and health quality) corresponding to different functions of the soil. Data sets are the tools necessary to provide a picture of soil quality. Components of the data set will change depending on the type and end users. Thus, the strategy of data sets is dynamic and flexible. The data sets become the means whereby interest groups or society can relate to, utilize, or evaluate soil for a specific reason or purpose. Based on this principle, all major physical, chemical, biochemical, and microbiological indicators of soil quality can be grouped as subsets of data under each component of the soil quality (Table XIII). One or more indexing systems or mathematical expressions may be developed from each minimum data set for evaluating the specified components of soil quality. For example, the indexing system of Wang and Gong (1998) focuses on soil fertility
128
Z. L. HE ET AL. Table XIII Some Minimum Data Sets for Estimating Soil Quality
Soil quality components
Minimum data sets
Production sustainability
Total organic C, N, microbial biomass C, N, and P and their turnover rates, mineralizable N, available P and K, pe-pH, MQ, depth of soil, soil texture, electrical conductivity (EC), microbial community structure (PLFA or CLPP) Microbial biomass C, basal respiration, MMQ, basal respiration/total organic C, methane emission potential, microbial community structure, enzyme activity, adsorption capacity for heavy metals or chemicals, pe-pH, infiltration, total and bioavailable amounts of heavy metals Total organic C and microbial biomass C, pe-pH, EC, bioavailable heavy metals and chemicals and their fluxes through food chain, earthworms, soil-borne pathogens, nematodes, microbial community structure, basal respiration
Environmental quality
Plant and animal health
quality, whereas the index developed by Leiros et al. (1999) is more directed to assessing soil contamination. However, if necessary, all the indices from each subsets of data can be integrated and processed using an GIS-database system to establish one or more comprehensive indexing systems for quantifying and monitoring quality changes of acid soils as a whole.
VIII.
LIMITATIONS AND PROSPECTIVE
Microbial biomass, activity, and community structure and the derived parameters such as MQ and MMQ, etc., are closely associated with some important functions in soil. These microbiological and biochemical properties have many applications in agricultural and environmental sciences such as diagnosis of nutrient availability, organic matter quality, soil degradation, and contamination. They have great potential as early indicators of soil quality changes. However, microbial biomass, activity, and community structure are affected by moisture, temperature, pH, soil matrix, and quantity and quality of carbon and nutrient supply. As an individual index of soil quality, each of them is subjected to certain limitations: (1) it is difficult to compare these values across climate, soil types, and land uses (Sparling, 1997); (2) routine standard and reliable procedures for measuring microbial biomass, activity, and community structures in soils are still lacking, which makes it difficult to interpret data from
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 129
different sources; and (3) the benchmark values of individual microbiological and biochemical indices for soil quality are currently not available and if available they may differ depending on soil type and land uses (Sparling, 1997), climate (Insam, 1990), and vegetation (Martens, 1995). However, these parameters can be combined, modified or integrated to establish some useful tools for assessing soil quality changes (Trasar-Cepeda et al., 1998; Wang and Gong, 1998). Future studies need to focus on the following: (1) Modification and standardization of current methods or techniques for measuring microbial biomass, activity, and diversity; so that they can be used for routine analysis and are applicable for different soils. (2) Develop rapid, accurate, relatively inexpensive, and portable soil quality test kits for assessing the soil microbiological and biochemical indicators under field conditions. (3) Application of molecular technologies such as 16sRNA-, 16sDNA-PCR techniques (Sayler et al., 1994) for determining microbial community and biomass in soils. (4) Application of information systems such as GIS-Database to establish soil quality indexes. (5) Development of minimum data sets to facilitate the establishment of the indexing systems. (6) Translation of soil quality research data into management tools for the end users such as farmers, growers, agricultural production agencies, environmental protection agencies, and decision-making organizations. As Doran et al. (1999) pointed out, producers and other managers of the land need practical tools and approaches to measure the effects of management on soil quality and health which enable them to “finetune” and determine the sustainability of their production approaches.
Acknowledgments This study was, in part, supported by the National Natural Science Foundation of China with an Outstanding Young Scientist Grant (Approval no. 40025104) to the senior author. A similar paper was presented at the 5th International PlantSoil Interactions at Low pH Symposium (PSILPH) and published in the proceedings of the symposium [Pages 114– 155 In: Integrated Management and Use of Acid Soils for Sustainable Production (Mart Farina, Mara de Villiers, Robin Barnard, and Mike Walters Eds.), ARC 2001, Pretoria, South Africa]. The authors thank M. C. DE Villiers, the Convener of the 5th International PSILPH Symposium, for consenting us to use the materials from the proceedings. The authors thank Dr R. M. Sonoda for his review of this manuscript. Florida Agricultural Experiment Station Journal Series No. R-08638.
130
Z. L. HE ET AL.
REFERENCES Acosta-Martinez, V., and Tabatabai, M. A. (2000). Enzyme activities in a limed agricultural soil. Biol. Fert. Soils 31, 85–91. Anderson, J. E., and Domsch, K. H. (1978). A physiological method for measurement of microbial biomass in soils. Soil Biol. Biochem. 10, 215– 221. Angle, J. S., Chaney, R. L., and Rhee, D. (1993). Bacterial resistance to heavy metals related to extractable and total metal concentrations in soil and media. Soil Biol. Biochem. 25, 1443– 1446. Aslam, T., Choudhary, M. A., and Saggar, S. (1999). Tillage impact on soil microbial biomass C, N, and P, earthworms and agronomy after two years of cropping following permanent pasture in New Zealand. Soil Tillage Res. 51, 103 –111. Baath, E., Diazravina, M., Frostegard, A., and Campbell, C. (1998). The effect of metal rich sludge amendments on the soil microbial community. Appl. Environ. Microbiol. 64, 238 –245. Badalucco, L., Grego, S., Dellorco, S., and Nannipieri, P. (1992). Effect of liming on some chemical, biochemical, and microbiological properties of acid soils under spruce (Picea-abies L.). Biol. Fert. Soils 14, 76–83. Baligar, V. C., and Ahlrichs, J. L. (1998). Nature and distribution of acid soils in the world. In “Proceedings of the Workshop to Develop a Strategy for Collaborative Research and Dissemination of Technology in Sustainable Crop Production in Acid Savannas and Other Problem Soils of the World” (R. E. Schaffert, Ed.), pp. 1–11. Purdue University Press, EastLansing, IN. Baligar, V. C., and Wright, R. J. (1991). Enzyme activities in Appalachian soils. 1. Arylsulfatase. Commun. Soil Sci. Plant Anal. 22, 305–314. Baligar, V. C., Wright, R. J., and Smedley, M. D. (1988). Acid phosphatase activity in soils of the Appalachian region. Soil Sci. Soc. Am. J. 52, 1612– 1616. Baligar, V. C., Staley, T. E., and Wright, R. J. (1991a). Enzyme activities in Appalachian soils. 2. Urease. Commun. Soil Sci. Plant Anal. 22, 315–322. Baligar, V. C., Wright, R. J., and Smedley, M. D. (1991b). Enzyme activities in Appalachian soils. 4. Dehydrogenase. Commun. Soil Sci. Plant Anal. 22, 1797–1804. Baligar, V. C., Wright, R. J., Fageria, N. K., and Pitta, G. V. E. (1999). Enzyme activities in Cerrado soils of Brazil. Commun. Soil Sci. Plant Anal. 30, 1551–1560. Bandick, A. K., and Dick, R. P. (1999). Field management effects on soil enzyme activities. Soil Biol. Biochem. 31, 1471–1479. Banerjee, M. R., Burton, D. L., and Depoe, S. (1997). Impact of sewage sludge application on soil biological characteristics. Agric. Ecosyst. Environ. 66, 241 –249. Bending, G. D., Putland, C., and Rayns, F. (2000). Changes in microbial community metabolism and labile organic matter fractions as early indicators of the impact of management on soil biological quality. Biol. Fert. Soils 31, 78–84. Bergstrom, D. W., Monreal, C. M., and King, D. J. (1998). Sensitivity of soil enzyme activities to conservation practices. Soil Sci. Soc. Am. J. 62, 1286– 1295. Biederbeck, V. O., Campbell, C. A., Ukrainetz, H., Curtin, D., and Bouman, O. T. (1996). Soil microbial and biochemical properties after ten years of fertilization with urea and anhydrous ammonia. Can. J. Soil Sci. 76, 7– 14. Bolinder, M. A., Angers, D. A., Gregorich, E. G., and Carter, M. R. (1999). The response of soil quality indicators to conservation management. Can. J. Soil Sci. 79, 37–45. Brady, N. C., and Weil, R. R. (Eds.) (1996). In “The Nature and Properties of Soils” (11th ed). Prentice Hall, Upper Saddle River, NJ. Bremer, E., and Van Kessel, C. (1990). Extractability of microbial 14C and 15N following addition of variable rates of labeled glucose and (NH4)2SO4 to soil. Soil Biol. Biochem. 22, 707–713.
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 131 Brookes, P. C. (1995). The use of microbial parameters in monitoring soil pollution by heavy metals. Biol. Fert. Soils 19, 269–279. Brookes, P. C., and McGrath, S. P. (1984). Effects of metal toxicity on the size of the soil microbial biomass. J. Soil Sci. 35, 341–346. Brookes, P. C., Powlson, D. S., and Jenkinson, D. S. (1982). Measurement of microbial biomass phosphorus in soil. Soil Biol. Biochem. 14, 319– 329. Brookes, P. C., Powlson, D. S., and Jenkinson, D. S. (1984). Phosphorus in the soil microbial biomass. Soil Biol. Biochem. 16, 169–175. Brookes, P. C., Kragt, J. F., Powlson, D. S., and Jenkinson, D. S. (1985a). Chloroform fumigation and the release of soil organic nitrogen: The effect of fumigation time and temperature. Soil Biol. Biochem. 17, 831– 835. Brookes, P. C., Landman, A., Pruden, G., and Jenkinson, D. S. (1985b). Chloroform fumigation and the release of soil organic nitrogen: A rapid direct extraction method to measure microbial biomass nitrogen in soils. Soil Biol. Biochem. 17, 837–842. Brown, S., Anderson, J. M., Woomer, P. L., Swift, M. J., and Barrios, E. (1994). Soil biological processes in tropical ecosystems. In “The Biological Management of Tropical Soil Fertility” (P. L. Woomer and M. J. Swift, Eds.), pp. 15–46. Wiley, Chichester, UK. Carreira, J. A., Garcia-Ruiz, R., Lietor, J., and Harrison, A. F. (2000). Changes in soil phosphatase activity and P transformation rates induced by application of N- and S-containing acid mist to a forest canopy. Soil Biol. Biochem. 32, 1857–1865. Cavigelli, M. A., Robertson, G. O., and Klug, M. J. (1995). Fatty acid methyl ester (FAME) profiles as measures of soil microbial community structure. Plant Soil 170, 99 –113. Chan, K. Y., and Heenan, D. P. (1999). Microbial-induced soil aggregate stability under different crop rotations. Soil Biol. Biochem. 30, 29–32. Chander, K., and Brookes, P. C. (1993). Residual effects of zinc, copper and nickel in sewage sludge on microbial biomass in a sandy loam. Soil Biol. Biochem. 25, 587–597. Chandini, M. T., and Parkinson, D. (2000). Microbial respiration, biomass, metabolic quotient and litter decomposition in a lodgepole pine forest floor amended with nitrogen and phosphorus fertilizers. Soil Biol. Biochem. 32, 59–66. Chen, G. (2000). Microbial biomass C and P, their turnover rates in red soils with relation to soil fertility sustainability. PhD Thesis. College of Resource and Environmental Sciences, Zhejiang University, Hangzhou. Chen, G., He, Z. L., Zhu, J., and Wilson, M. J. (1997). Fumigation–extraction method for the measurement of microbial biomass-N in red soils. Pedosphere 7, 87 –91. Chen, G., He, Z. L., and Yao, H. (1999a). Seasonal variation of microbial biomass in red soils. Acta Universitasis Zhejiangensis (Agric. Life Sci. Ed.) 25, 387–388. Chen, G., He, Z. L., and Huang, C. Y. (1999b). Relationship between microbial biomass phosphorus and other phosphorus forms in red soils. Acta Universitasis Zhejiangensis (Agric. Life Sci. Ed.) 25, 513–516. Chen, G., He, Z. L., and Huang, C. Y. (2000). Microbial biomass phosphorus and its significance in predicting P availability in red soils. Commun. Soil Sci. Plant Anal. 31, 655– 667. Chen, G., He, Z. L., and Huang, C. Y. (2002a). Relationships between microbial biomass and pools and plant-availability of phosphorus in red soils. Acta Pedologica Sinica 39, 152–160. Chen, G., He, Z. L., and Huang, C. Y. (2002b). Microbial biomass phosphorus turnover in variable charge soils in China. Commun. Soil Sci. Plant Anal. 33, 2101–2117. Coleman, D. C., Hendrix, P. F., Beare, M. H., Cheng, W. X., and Crossley, D. A., Jr. (1993). Microbial and faunal interactions as they affect soil organic matter dynamics in subtropical agroecosystems. In “Soil Biota, Nutrient Cycling, and Farming Systems” (M. C. Paoletti, W. Foissner and D. C. Coleman, Eds.), pp. 1 –14. Lewis Publishers, Boca Raton, FL. Dalal, R. C. (1998). Soil microbial biomass-what do the number really mean? Aust. J. Expp. Agric. 38, 645 –665.
132
Z. L. HE ET AL.
Diazravina, M., Acea, M. J., and Carballas, T. (1993). Microbial biomass and its contribution to nutrient concentrations in Forest soils. Soil Biol. Biochem. 25, 25–31. Diazravina, M., Acea, M. J., and Carballas, T. (1995). Seasonal changes in microbial biomass and nutrient flush in forest soils. Biol. Fert. Soils 19, 220 –226. Dick, W. A., Cheng, L., and Wang, P. (2000). Soil acid and alkaline phosphatase activity as pH adjustment indicators. Soil Biol. Biochem. 32, 1915– 1919. Doran, J. W., Alice, J. J., Arshad, M. A., and Gilley, J. E. (1999). Determinants of soil quality and health. In “Soil Quality and Soil Erosion” (R. Lal, Ed.), pp. 17–36. CRC Press, New York. Drijber, R. A., Doran, J. W., Pankhurst, A. M., and Lyon, D. J. (2000). Changes in soil microbial community structure with tillage under long-term wheat-fallow management. Soil Biol. Biochem. 32, 1419–1430. Elliott, L. F., Lynch, J. M., and Papendick, R. I. (1996). The microbial component of soil quality. Soil Biochemistry (G. Stotzky and J. M. Bollag, Eds.), vol. 9, pp. 1–22. Dekker, New York. Feigl, B. J., Sparling, G. P., Ross, D. J., and Cerri, C. C. (1995). Soil microbial biomass in Amazonian soils—evaluation of methods and estimates of pool sizes. Soil Biol. Biochem. 27, 1467–1472. Fliessbach, A., and Mader, P. (2000). Microbial biomass and size-density fractions differ between soils of organic and conventional agricultural systems. Soil Biol. Biochem. 32, 757–768. Frankenberger, W. T., Jr., and Tabatabai, M. A. (1980). Amidase activity in soils. I. Methods of assay. Soil Sci. Soc. Am. J. 44, 282 –287. Garcia-Gil, J. C., Plaza, C., Soler-Rovira, P., and Polo, A. (2000). Long-term effects of municipal solid waste compost application on soil enzyme activities and microbial biomass. Soil Biol. Biochem. 32, 1907–1913. Ghoshal, N., and Singh, K. P. (1995). Effects of farmyard manure and inorganic fertilizer on the dynamics of soil microbial biomass in a tropical dryland agroecosystem. Biol. Fert. Soils 19, 231–238. Gianfreda, L., and Bollag, J. M. (1996). Influence of natural and anthropogenic factors on enzyme activity in soil. Soil Biochemistry (J. M. Bollag and G. Stotzky, Eds.), vol. 9, pp. 123– 193. Dekker, New York. Giller, K. N., Witter, E., Steve, P., and McGrath, S. P. (1998). Toxicity of heavy metals to microorganisms and microbial processes in agricultural soils: A review. Soil Biol. Biochem. 30, 1389–1414. Grayston, S. J., and Campbell, C. D. (1996). Functional biodiversity of microbial communities in the rhizospheres of hybrid larch (Larix eurolepis ) and Sitka spruce (Picea sitchesis ). Tree Physiol. 16, 1031–1038. Gregorich, E. G., Carter, M. R., Angers, D. A., Monreal, C. M., and Ellert, B. (1994). Towards a minimum data set to assess soil organic matter quality in agricultural soils. Can. J. Soil Sci. 74, 367–385. Gregorich, E. G., Liang, B. C., Drury, C. F., Mackenzie, A. F., and McGill, W. B. (2000). Elucidation of the source and turnover of water soluble and microbial biomass carbon in agricultural soils. Soil Biol. Biochem. 32, 581– 587. Haack, S. K., Garchow, H., Klug, M. J., and Forney, L. J. (1995). Analysis of factors affecting the accuracy, reproducibility, and interpretation of microbial community carbon source utilization patterns. Appl. Environ. Microbiol. 61, 1458–1468. Haynes, R. J. (1982). Effects of liming on phosphate availability in acid soils—A critical review. Plant Soil 68, 289 –308. Haynes, R. J. (1984). Lime and phosphate in the soil–plant system. Adv. Agron. 37, 249–315. Haynes, R. J., and Naidu, R. (1998). Influence of lime, fertilizer and manure application on soil organic matter content and soil physical condition: A review. Nutr. Cycl. Agroecosyst. 51, 123–137. Haynes, R. J., and Swift, R. S. (1986). Effect of soil acidification and subsequent leaching on levels of extractable nutrients in a soil. Plant Soil 95, 327 –336.
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 133 Haynes, R. J., and Tregurtha, R. (1999). Effects of increasing periods under intensive arable vegetable production on biological, chemical and physical indices of soil quality. Biol. Fert. Soils 28, 259 –266. He, Z. L. (1997). Turnover of soil microbial biomass and its relation to nutrient cycling in agricultural system: A review. Tura 29, 61–69. He, Z. L., Wu, J., O’Donnell, A. G., and Syers, J. K. (1997). Seasonal responses in microbial biomass carbon, phosphorus and sulfur in soils under pasture. Biol. Fert. Soils 24, 421–428. He, Z. L., Yao, H., Chen, G., Zhu, J., and Huang, C. Y. (1997). Relationship of crop yield to microbial biomass in highly weathered soils of China. In “Plant Nutrition for Sustainable Food Production and Environment”. (T. Ando et al., Eds.), pp. 745– 746. Kluwer Academic, Tokyo. He, Z. L., Yang, X. E., and Baligar, V. C. (2001a). Increasing nutrient utilization and crop production in the red soil regions of China. Commun. Soil Sci. Plant Anal. 32, 1251–1263. He, Z. L., Yang, X. E., Yao, H., Chen, G., Calvert, D. V., and Baligar, V. C. (2001b). Microbiological and biochemical indices for assessing quality of acid soils. In “Proceedings of the 5th International Plant-Soil Interactions at Low pH Symposium”. (M. Farina, M. de Villiers, R. Barnard, M. Walters, Eds.), pp. 114 –155. ARC-Institute for Soil, Climate and Water, Pretoria, South Africa. Huang, Q., and Shindo, H. (2000). Effects of copper on the activity and kinetics of free and immobilized acid phosphatase. Soil Biol. Biochem. 32, 1885–1892. Insam, H. (1990). Are the soil microbial biomass and basal respiration governed by climatic regime? Soil Biol. Biochem. 22, 525–532. Insam, H., and Haselwandter, K. (1989). Metabolic quotient of soil microflora in relation to plant succession. Oecologia 79, 174–178. Insam, H., Mitchell, C. C., and Dormaar, J. F. (1991). Relationship of soil microbial biomass and activity with fertilization and crop yield of three ultisols. Soil Biol. Biochem. 23, 259–264. Insam, H., Hutchinson, T. C., and Reber, H. H. (1996). Effects of heavy metal stress on the metabolic quotient of the soil microflora. Soil Biol. Biochem. 28, 691–694. Inubushi, K., Brookes, P. C., and Jenkinson, D. S. (1991). Soil microbial biomass C, N, and ninhydrinN in aerobic and anaerobic soils measured by the fumigation–extraction method. Soil Biol. Biochem. 23, 737– 741. Islam, K. R., and Weil, R. R. (2000a). Land use effects on soil quality in a tropical forest ecosystem of Bangladesh. Agric. Ecosyst. Environ. 79, 9–16. Islam, K. R., and Weil, R. R. (2000b). Soil quality indicator properties in mid-Atlantic soils as influenced by conservative management. J. Soil Water Conserv. 55, 69 –78. Jenkinson, D. S. (1976). The effects of biocidal treatment on metabolism in soil. IV. The decomposition of fumigated organisms in soil. Soil Biol. Biochem. 8, 203 –208. Jenkinson, D. S., and Oades, J. M. (1979). A method for measuring adenosine triphosphate in soil. Soil Biol. Biochem. 9(11), 193–199. Jenkinson, D. S., and Powlson, D. S. (1976a). The effects of biocidal treatment on metabolism in soil. I. Fumigation with chloroform. Soil Biol. Biochem. 8, 167 –177. Jenkinson, D. S., and Powlson, D. S. (1976b). The effects of biocidal treatment on metabolism in soil. V. A method for measuring soil biomass. Soil Biol. Biochem. 8, 209 –213. Jenkinson, D. S., Powlson, D. S., and Wedderburn, R. W. M. (1976). The effects of biocidal treatment on metabolism in soil. III. The relationship between soil biovolume, measured by optical microscopy, and the flush of decomposition caused by fumigation. Soil Biol. Biochem. 8, 189–202. Jenkinson, D. S., Hart, P. B. S., Payner, J. H., and Parry, L. C. (1987). Modeling the turnover of organic matter in long-term experiments at Rothamsted. In “Soil Organic Matter Dynamics and Soil Productivity”. (J. I. Cooley, Ed.), INTECOL Ball Vol. 15, 1–8. Jia, C. Y., and Insam, H. (1991). Microbial biomass and relative contributions of bacteria and fungi in soil beneath tropical rain-forest, Hainan Island. J. Tropp. Ecol. 7, 385–393.
134
Z. L. HE ET AL.
Joergensen, R. G., and Brookes, P. C. (1990). Ninhydrin-reactive nitrogen measurements of microbial biomass in 0.5 M K2SO4 soil extracts. Soil Biol. Biochem. 22, 1023–1027. Joergensen, R. G., Kubler, H., Mwyer, B., and Wolters, V. (1995). Microbial biomass phosphorus in soils of beech (Fagus sylvatica L.) forest. Biol. Fert. Soils 19, 215–219. Johnson, D. L., Ambrose, S. H., Bassett, T. J., Bove, M. L., Crummey, D. E., Isaacson, J. S., Johnson, D. N., Lamb, P., Saul, M., and Winter-Nelson, A. E. (1997). Meaning of environmental terms. J. Environ. Qual. 26, 581–589. Jordan, D., Kremer, R. J., Bergfield, W. A., Kim, K. Y., and Cacnio, V. N. (1995). Evaluation of microbial methods as potential indicators of soil quality in historical agricultural fields. Biol. Fert. Soils 19, 297 –302. Kaiser, E. A., Mueller, T., Jeorgensen, R. G., Insam, H., and Heinemeyer, O. (1992). Evaluation of methods to estimate the soil microbial biomass and the relationship with soil texture and organic matter. Soil Biol. Biochem. 24, 675–683. Kanazawa, F. Z., and Berthelin, J. (1999). Characterization of effects of a long-term wastewater irrigation on soil quality by microbiological and biochemical parameters. J. Plant Nutr. Soil Sci. 162, 409 –413. Kanazawa, S., and Kunito, T. (1996). Preparation of pH 3.0 agar plate, enumeration of acid-tolerant, and Al-resistant microorganisms in acid soils. Soil Sci. Plant Nutr. 42, 165 –173. Karlen, D. L., and Stott, D. E. (1994). A framwork for evaluating physical and chemical indicators of soil quality. SSSA Special Publication 35. In “Defining Soil Quality for a Sustainable Environment” (J. W. Doran, D. C. Coleman, D. F. Bezdicek and B. A. Stewart, Eds.), pp. 53–72. SSSA, Madison, WI. Karlen, D. L., Wollenhaupt, N. C., Erbach, D. C., Berry, E. C., Swan, J. B., Eash, N. S., and Jordahl, J. L. (1994). Crop residue effects on soil quality following 10-years of no till corn. Soil Tillage Res. 31, 149 –167. Kelly, J. J., and Tate, R. L. (1998). Effects of heavy metal contamination and remediation on soil microbial communities in the vicinity of zinc smelter. J. Environ. Qual. 27, 609–617. Kelly, J. J., Haggblom, M., and Tate, R. L., III (1999). Effect of the land application of sewage sludge on soil heavy metal concentrations and soil microbial communities. Soil Biol. Biochem. 31, 1467–1470. Kennedy, A. C., and Smith, K. L. (1995). Soil microbial diversity and the sustainability of agricultural soils. Plant Soil 170, 78– 86. Killham, K. (1985). A physiological determination of the impact of environmental stress on the activity of microbial biomass. Environ. Pollut. 38, 204 –283. Kondon, H., Koyama, N., Tsuiki, M., Shiyomi, M., Katoh, K., and Kurashima, K. (1989). Mineralization of phosphorus from microbial biomass in soil. Bull. Natl Grassland Res. Inst. 42, 69 –75. Kouno, K., Lukito, H. P., Ando, T., and Brookes, P. C. (1994). Microbial biomass phosphorus dynamics in soil. Trans 15th World Cong. Soil Sci., Acapulco, Mexico 4b, 85–96. Kouno, K., Tuchiya, Y., and Ando, T. (1995). Measuremnt of soil microbial biomass phosphorus by an anion exchange membrane method. Soil Biol. Biochem. 27, 1353–1357. Kozdroj, J. (1995). Microbial response to single or successive soil contamination with Cd or Cu. Soil Biol. Biochem. 27, 1459–1465. Ladd, J. N., Foster, R. C., Nannipieri, P., and Oades, J. M. (1996). Soil structure and biological activity. In Soil Biochemistry. (G. Stotzky and J. M. Bollag Eds.), vol. 9, pp. 23– 78. Dekker, New York. Lal, R. (1997). Degradation and resilience of soils. Philos. Trans. R. Soc. Lond. Ser. B 352, 997– 1010. Leiros, M. C., Trasar-Cepeda, C., Garcia-Fernandez, F., and Gil-Sotres, F. (1999). Defining the validity of a biochemical index of soil quality. Biol. Fert. Soils 30, 140– 146. Leiros, M. C., Trasar-Cepeda, C., Seoane, S., and Gil-Sotres, F. (2000). Biochemical properties of acid soils under climax vegetation (Atlantic oakwood) in an area of the European temperate-humid zone (Galicia, NW Spain): General parameters. Soil Biol. Biochem. 32, 733–745.
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 135 Liebig, M. A., and Doran, J. W. (1999). Impact of organic production practices on soil quality indicators. J. Environ. Qual. 28, 1601–1609. Lin, B. (1995). Strategies for efficient use of chemical fertilizers in agriculture. In “Proceedings of China National Congress Soil Science”. (Soil Sci. Soc. China, Ed.), pp. 109–114. Academic Press, Beijing. Lu, J. G., Liu, G. Q., and He, Z. L. (1986). Stability of the structure of red earths and its significance in soil classification. Acta Pedologica Sinica 23, 212–219. Mao, D. M., Min, Y. W., Yu, L. L., Martens, R., and Insam, H. (1992). Effect of afforestation on microbial biomass and activity in soils of tropical China. Soil Biol. Biochem. 24, 865–872. Martens, R. (1987). Estimation of microbial biomass in soil by the respiration method —importance of Soil pH and flushing methods for the measurement of respired CO2. Soil Biol. Biochem. 19, 77 –81. Martens, R. (1995). Current methods for measuring microbial biomass C in soil: Potentials and limitations. Biol. Fert. Soils 19, 87– 99. Martikainen, P. J., and Palojarvi, A. (1990). Evaluation of the fumigation–extraction method for the determination of microbial C and N in a range of forest soils. Soil Biol. Biochem. 22, 792–802. McCarty, G. W., and Meisinger, J. J. (1997). Effects of N fertilizer treatments on biologically active N pools in soils under plow and no tillage. Biol. Fert. Soils 24, 406 –412. Mikanova, O., Novakova, J., Hanzlikova, A., Kubat, J., and Filip, Z. (1996). Effects of long-term fertilization and anthropogenic load on biomass and the incidence of some groups of microorganisms. Rostlinna Vyroba 42, 391–397. Murage, E. W., Karanja, N. K., Smithson, P. C., and Woomer, P. L. (2000). Diagnostic indicators of soil quality in productive and non-productive smallholders’ fields of Kenya’s Central Highlands. Agric. Ecosyst. Environ. 79, 1–8. Nanipieri, P., and Grego, S. (1990). Ecological significance of the biological activity in soil. In Soil Biochemistry (J. M. Bollag and G. Stotzky, Eds.), vol. 6, pp. 293–358. Dekker, New York. Neale, S. P., Shah, Z., and Adams, W. A. (1997). Changes in microbial biomass and nitrogen turnover in acidic organic soils following liming. Soil Biol. Biochem. 29, 1463–1474. Oberson, A., Friesen, D. K., Morel, C., and Tiessen, H. (1997). Determination of phosphorus released by chloroform fumigation from microbial biomass in high P sorbing tropical soils. Soil Biol. Biochem. 29, 1579–1583. Ohno, M. (2001). Sensitivity of a Japanese earthworm (Allolobophora japonica ) to soil acidity. Water Air Soil Pollut. 130, 1019–1024. Olson, S. R. (1986). The role of organic matter and ammonium in producing high corn yields. In “The Role of Organic Matter in Modern Agriculture” (Y. Chen and Y. Avnimelech, Eds.), pp. 29 –70. Martinus Nijhoff Publications, Dordrecht. Pandey, A., and Palni, L. M. S. (1997). Bacillus species: The dominant bacteria of the rhizosphere of established tea bushes. Microbiol. Res. 152, 357– 365. Pankhurst, C. E., Hawke, B. G., McDonald, H. J., Kirby, C. A., Buckerfield, J. C., Michelsen, O., Obrien, K. A., Gupta, V. V. S. R., and Doube, B. M. (1995). Evaluation of soil biological properties as potential bioindicators of soil health. Aust. J. Expp. Agric. 35, 1015–1028. Pansombat, K., Kanazawa, S., and Horiguchi, T. (1997). Microbial ecology in tea soils. 2. soil protense activity. Soil Sci. Plant Nutr. 43, 431–438. Pearson, R. W., and Adams, F. (1967). In “Soil Acidity and Liming”. ASA Publisher, Madison, WI. Pepper, I. L., and Rensing, C (2000). Microbial indicators of soil quality. The Annual Meetings of ASA–CSA–SSSA, Minneapolis, Abstract 266. Perrot, K. W., and Sarathchandra, S. U. (1989). Phosphorus in the microbial biomass of New Zealand soils under established pasture. NZ J. Agric. Res. 32, 409– 413. Pettigrew, C., and Sayler, G. S. (1986). The use of DNA: DNA colony hybridization in the rapid isolation of 4-chloro-biphenyl degradative bacterial phenotypes. J. Microbiol. Meth. 5, 205 –213.
136
Z. L. HE ET AL.
Pinzari, F., Trichera, A., Benedetti, A., and Sequi, P. (1998). Defining soil quality in Mediterranean forest systems: Microbial biomass activity. Fresenius Environ. Bull. 7, 447 –457. Powlson, D. S., and Jenkinson, D. S. (1976). The effects of biocidal treatment on metabolism in soil. II. Gamma irradiation, autoclaving, air-drying and fumigation. Soil Biol. Biochem. 8, 179– 188. Press, C. M., Mahaffee, W. F., Edwards, J. H., and Kloepper, J. W. (1996). Organic by-product effects on soil chemical properties and microbial communities. Compos. Sci. Util. 4, 70 –80. Ross, D. J. (1990). Estimation of soil microbial C by a fumigation–extraction method: Influence of seasons, soils and calibration with the fumigation –incubation procedure. Soil Biol. Biochem. 22, 295–300. Sanchez, P. A. (1976). In “Properties and Management of Soils in the Tropics”. Wiley, New York. Sayler, G. S., Nikbakht, K., Fleming, J. T., and Packard, J. (1994). Application of molecular techniques to soil biochemistry. Soil Biochemistry (J. M. Bollag and G. Stotzky, Eds.), vol. 8, pp. 131–171. Dekker, New York. Sehgal, J., Blum, W. E., and Gajbhiye, K. S. (1998). In “Red and Lateritic Soils of the World”. A.A. Balkema, Rotterdam, Netherlands. Shen, S. M., Pruden, G., and Jenkinson, D. S. (1984). Mineralization and immobilization of nitrogen in fumigated soil and the measurement of microbial biomass nitrogen. Soil Biol. Biochem. 16, 437–444. Singer, M. J., and Stephanie, E. (2000). Soil quality. In “Handbook of Soil Science” (M. E. Sumner Ed.), pp. G271 –G297. CRC Press, Boca Raton, FL. Smith, J. L., and Paul, E. A. (1990). The significance of soil microbial biomass estimations. In Soil Biochemistry. J. M. Bollag and G. Stotzky, Eds.), vol. 6, pp. 359 –396. Dekker, New York. Soil Science Society of America, (Eds), (1997). In “Glossary of Soil Science Terms”. SSSA, Madison, WI. Sparling, G. P. (1992). Ratio of microbial biomass carbon to soil organic carbon as a sensitive indicator of changes in soil organic matter. Aust. J. Soil Res. 30, 195 –207. Sparling, G. P. (1997). Soil microbial biomass, activity and nutrient cycling as indicators of soil health. In “Biological Indicators of Soil Health” (C. E. Pankhurst, B. M. Doube and V. V. S. R. Gupta, Eds.), pp. 97–119. CAB International, Wallingford, UK. Sparling, G. P., and West, A. W. (1989). Importance of soil water content when estimating soil microbial biomass C, N, and P by the fumigation–extraction methods. Soil Biol. Biochem. 21, 245–253. Srivastava, S. C., and Singh, J. S. (1991). Microbial C, N and P in dry tropical forest soils: Effects of alternate land-uses and nutrient flux. Soil Biol. Biochem. 23, 117–124. Staben, M. L., Bezdicek, D. F., Smith, J. L., and Fauci, M. F. (1997). Assessment of soil quality in conservation reserve program and wheat-fallow soils. Soil Sci. Soc. Am. J. 61, 124–130. Tabatabai, M. A. (1994). Soil enzymes. In “Methods of Soil Analysis. Part 2. Microbiological and Biochemical Properties” (R. W. Weaver, S. Angle and P. S. Bottomley, Eds.), pp. 775– 826. SSSA, Madison, WI. Tabatabai, M. A., and Bremner, J. M. (1970). Factors affecting soil arylsulfatase activity. Soil Sci. Soc. Am Proc. 34, 427–429. Tabatabai, M. A., and Fu, M. (1994). Extraction of enzymes from soils. In Soil Biochemistry (J. M. Bollag and G. Stotzky, Eds.), vol. 8, pp. 197–227. Dekker, New York. Tabatabai, M. A., and Singh, B. B. (1976). Rhodanese activity of soils. Soil Sci. Soc. Am. J. 40, 381–385. Trasar-Cepeda, C., Leiros, M. C., Gil-Sotres, F., and Seoane, S. (1998). Towards a biochemical quality index for soils: An expression relating several biological and biochemical properties. Biol. Fert. Soils 26, 100– 106. Trasar-Cepeda, C., Leiros, M. C., and Gil-Sotres, F. (2000a). Biochemical properties of acid soils under climax vegetation (Atlantic oakwood) in an area of the European temperate–humid zone (Galicia, NW Spain): Specific parameters. Soil Biol. Biochem. 32, 747– 755.
MICROBIOLOGICAL AND BIOCHEMICAL INDEXING SYSTEMS 137 Trasar-Cepeda, C., Leiros, M. C., Seoane, S., and Gil-Sotres, F. (2000b). Limitations of soil enzymes as indicators of soil pollution. Soil Biol. Biochem. 32, 1867–1875. Tulid, A., and White, D. C. (1994). Biochemical analysis of biomass, community structure, nutritional status, and metabolic activity of microbial communities in soil. Soil Biochemistry. (J. M. Bollag and G. Stotzky, Eds.), vol. 8, pp. 229–263. Dekker, New York. Valsecchi, G., Gigliotti, C., and Farini, A. (1995). Microbial biomass, activity, and organic matter accumulation in soils contaminated with heavy metals. Biol. Fert. Soils 20, 253–259. Vance, E. D., Brookes, P. C., and Jenkinson, D. S. (1987a). Microbial biomass measurement in forest soils: The use of the chloroform fumigation–incubation method in strongly acid soils. Soil Biol. Biochem. 19, 697– 702. Vance, E. D., Brookes, P. C., and Jenkinson, D. S. (1987b). An extraction method for measuring soil microbial biomass C. Soil Biol. Biochem. 19, 703–707. Vance, E. D., Brookes, P. C., and, Jenkinson, D. S. (1987c). Microbial biomass measurements in forest soils: Determination of kC values and tests of hypotheses to explain the failure of the chloroform fumigation–incubation method in acid soils. Soil Biol. Biochem. 19, 689 –696. 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 incubated with 14C-glucose and 15N-(NH4)2SO4 under different moisture regimes. Soil Biol. Biochem. 17, 747– 756. Van Veen, J. A., Ladd, J. N., and Martin, J. K. (1987). Turnover of carbon, nitrogen and phosphorus through the microbial biomass in soils incubated with 14C-, 15N-, and 32P-labelled bacterial cells. Soil Biol. Biochem. 19, 559–565. Von Uexkull, H. P., and Mutert, E. (1995). Global extent, development and economic impact of acid soils. Plant Soil 171, 1–15. Waldrop, M. P., Balser, T. C., and Firestone, M. K. (2000). Linking microbial community composition to function in a tropical soil. Soil Biol. Biochem. 32, 1837– 1846. Wang, X. J., and Gong, Z. T. (1998). Assessment and analysis of soil quality changes after eleven years of reclamation in subtropical China. Geoderma 81, 339–355. Wick, B., Kuhne, R. F., and Vlek, P. L. G. (1998). Soil microbiological parameters as indicators of soil quality under improved fallow management systems in south-western Nigeria. Plant Soil 202, 97 –107. Wu, J., Joergensen, R. G., Pommerening, B., Chaussod, R., and Brookes, P. C. (1990). Measurement of soil microbial biomass C by fumigation–extraction—an automated procedure. Soil Biol. Biochem. 22, 1167–1169. Wu, J., O’Donnell, A. G., He, Z. L., and Syers, J. K. (1994). Measurement of soil microbial biomass sulfur by fumigation–extraction method. Soil Biol. Biochem. 26, 117– 125. Yao, H. Y (2000). Microbiological characterization of red soils with varying fertility and quality status. PhD Thesis. College of Resource and Environmental Sciences, Zhejiang University, Hangzhou. Yao, H., and He, Z. L. (1998). Rationalization of enhanced utilization of applied chemical N fertilizer by ryegrass through organic carbon addition. Acta Agriculturae Universitatis Zhejingensis 24, 617 –618. Yao, H., and He, Z. L. (1999). Microbial biomass in red soils and its significance in fertility sustainability of plant–soil system. Acta Ecologica Sinica 10, 725–728. Yao, H., He, Z. L., Chen, G., Zhu, J., Huang, C. Y., and Wilson, M. J. (1998). Plant availability of nitrogen in relation to microbial biomass turnover in red soils of China. Proc. 16th World Cong. Soil Sci., Montpellier, France, Reg. No. 269, Sympp. 40, 1–8. Yao, H., He, Z. L., and Huang, C. (1999). Turnover period of microbial biomass nitrogen in red soils and its significance in soil fertility evaluation. Acta Pedologica Sinica 36, 387 –394. Yao, H., He, Z. L., Campbell, C. D., and Wilson, M. J. (2000a). Some limitations of BIOLOG method for determining microbial community structure in red soils. Pedosphere 10, 37–44.
138
Z. L. HE ET AL.
Yao, H., He, Z. L., Wilson, M. J., and Campbell, C. D. (2000b). Microbial biomass and community structure in a sequence of soils with increasing fertility and changing land use. Microbial Ecol. 40, 223 –237. Yeates, G. W., Saggar, S., and Daly, B. K. (1997). Soil microbial C, N, and P, and microfaunal populations under Pinus radiata and grazed pasture land-use system. Pedobiologica 41, 549–565. Yu, S., He, Z. L., Zhang, R., Chen, G., and Huang, C. (2002). Ecological characteristics of soil microbes in the tea bush rhizosphere in a red soil. Acta Ecologica Sinica (in press). Zelles, L., Bai, Q. Y., Beck, T., and Beese, F. (1992). Signature fatty acids in phospholipid and lipopolysaccharides as indicators of microbial biomass and community structure in agricultural soils. Soil Biol. Biochem. 24, 317 –323. Zelles, L., Bai, Q. Y., Rackwitz, R., Chadwick, D., and Beese, F. (1995). Determination of phospholipid- and lipopolysaccharide-derived fatty acids as an estimate of microbial biomass and community structures in soils. Biol. Fert. Soils 19, 115–123. Zelles, L., Rackwitz, R., Bai, Q. Y., Beck, T., and Beese, F. (1995). Discrimination of microbial diversity by fatty acid profiles of phospholipids and lipopolysaccharides in differently cultivated soils. Plant Soil 170, 115– 122. Zhang, M., He, Z. L., Chen, G., Huang, C., and Wilson, M. J. (1997). Effect of land use on the formation and water-stability of aggregates in red soils. Acta Pedologica Sinica 34, 359 –366.
DEVELOPMENT OF ACIDIC SUBSURFACE LAYERS OF SOIL UNDER VARIOUS MANAGEMENT SYSTEMS Keryn I. Paul,1,2 A. Scott Black1 and Mark K. Conyers3 1
School of Agriculture Charles Sturt University Locked Bag 588, Wagga Wagga, NSW 2678, Australia 2 CSIRO Forestry and Forest Products, P.O. Box E4008, Kingston, ACT 2604, Australia 3 Agricultural Institute NSW Agriculture, PMB, Wagga Wagga, NSW 2650, Australia
I. Introduction II. Widespread Occurrence of Acidic Subsurface Layers III. Detrimental Effects of Acidic Subsurface Layers on Agricultural Production A. Water and Nutrient Limitations due to Poor Root Growth B. Suppression of N Mineralisation C. Poor Root Nodulation of Legumes D. Poor Growth Response to Topdressing of P Fertiliser E. Poor Growth Response to Lime Application IV. Rate of Development of Acidic Subsurface Layers V. Cause of Development of Acidic Subsurface Layers A. Plant N Uptake B. Plant Residue Return C. Mn Reduction and Oxidation D. Urine Excretion from Grazing Stock E. Soil pH Buffering Capacity VI. Environmental Factors Affecting the Difference in pH between Surface and Subsurface Layers A. Soil Fertility B. Initial Soil pH C. Rainfall and Fluctuations in Soil Moisture Content D. Earthworm Populations VII. Management Factors Affecting the Difference in pH Between Surface and Subsurface Layers A. Agricultural Production B. Plant Species Grown C. Productivity and the Quantity of Plant Residues Added to the Soil Surface D. Minimum Soil Disturbance and Tillage 187 Advances in Agronomy, Volume 78 Copyright q 2003 by Academic Press. All rights of reproduction in any form reserved 0065-2113/02$35.00
188
K. I. PAUL ET AL. E. Fertiliser Application F. Lime Application VIII. Management Implications IX. Conclusions References
I. INTRODUCTION Soil acidification is one of the most prominent threats to agricultural sustainability (Von Uexkull and Mutert, 1995). Decisions for management of acidic soils should account for not only the lateral variations in pH, but also the corresponding vertical variations. Numerous studies have shown that within the surface 15 cm of moderately acidic soils under crops or pastures, pH decreases with depth such that the subsurface layer at about 5 –15 cm depth is more acidic than the surface 5 cm of soil (Fig. 1). We describe how acidic subsurface layers enhance the detrimental effect of soil acidity on agricultural production, how quickly they develop and why, and what environmental and management factors influence them. Implications for management are then discussed.
Soil pH 3.8
4.0
4.2
4.4
4.6
4.8
5.0
5.2
Soil depth (cm)
0
5
10
15
20 Figure 1 An example of acidic subsurface layers observed under an annual pasture (B), directdrilled wheat – lupin rotation (K), conventionally cultivated wheat – lupin rotation (O), and conventionally cultivated continuous wheat (W). Data were taken from Conyers et al. (1996, 1997).
ACIDIC SUBSURFACE LAYERS OF SOIL
189
II. WIDESPREAD OCCURRENCE OF ACIDIC SUBSURFACE LAYERS Numerous workers have measured soil pH in 2 cm (Conyers and Scott, 1989; McLaughlin et al., 1990; Evans et al., 1998; Purnomo et al., 2000a), 2.5 cm (Bromfield et al., 1983b; Baker et al., 1999; Matowo et al., 1999; Paul et al., 2001a) and 5 cm (Conyers et al., 1996; Pinkerton and Simpson, 1986b) depth intervals within moderately acidic surface soils under pastures and under crops with a range of tillage regimes. These studies have consistently shown that the lowest soil pH occurred within subsurface soil layers at a depth between 5 and 15 cm. There is also evidence of acidic subsurface layers in sandy soils that have been sampled in 10 cm depth increments (Dolling and Porter, 1994; Dolling et al., 1994; Dolling, 1995). These studies showed that the lowest pH often occurred at 10 –20 cm depth. Table I is a global collation of available results demonstrating the difference in pH between the surface and subsurface layers of unlimed soils. These results show that acidic subsurface layers are found not only under pastures and crops, but also under bare soil and forests, and that the magnitude of the pH decrease between surface and subsurface soils layers is up to 1.4 units under crops and pastures, up to 1.0 units in bare soils, and up to 0.9 units under forests. There were only a few sites where pH remained relatively constant or increased through surface soil depth (Table I). These exceptions mainly occurred where NHþ 4 -based fertilisers were applied to the soil surface (Blevins et al., 1983; Mahler and Harder, 1984; Pierce et al., 1994; Pikul and Aase, 1995; Bowman and Halvorson, 1998; Thompson and Whitney, 2000).
III. DETRIMENTAL EFFECTS OF ACIDIC SUBSURFACE LAYERS ON AGRICULTURAL PRODUCTION The development of acidic subsurface layers may exacerbate the detrimental effects of soil acidification on agricultural production due to induced water and nutrient deficiencies, and poor responses to fertiliser and lime application.
A. WATER AND NUTRIENT LIMITATIONS DUE TO POOR ROOT GROWTH Acidic subsurface layers often have high concentrations of Al and low concentrations of Ca (Pinkerton and Simpson, 1986b). Pinkerton and Simpson (1986a) noted that this was associated with poor root growth of wheat, rape, lucerne and phalaris at 10 –40 cm depth. Stunting and thickening of roots was
190
K. I. PAUL ET AL.
Table I Surface Soil Depth through which pH was Observed to Decrease under Bare Soil or Fallow, Virgin Bushland, Pasture, Crop and Annual Pasture Rotation, and Under Crops which were Direct Drilled, Reduced Tilled, Conventionally Cultivated, or Under Forests
Soil depth (cm)
Observed pH decrease
No.
Location
1 1 15 1 4 1 1 2 5 7 1 3 4 1 3 1 1
NSW, Australia Victoria, Australia Southeastern Australia NSW, Australia NSW, Australia NSW, Australia NSW, Australia NSW, Australia NSW, Australia Southeastern Australia NSW, Australia NSW, Australia NSW, Australia Brazil Western Australia Western Australia North-east New Zealand
Reference
Pasture 0– 5 0– 6 0– 6 0– 15 0– 10 0– 10 0– 10 0– 10 0– 10 0– 10 0– 10 0– 12 0– 15 0– 17.5 0– 20 0– 20 0– 20
0.3 0.81 0.40a 0.93b 0.17–0.99 0.5 0.2– 0.7 0.82 0.66–1.43 0.1– 0.8 0.82 0.77–1.09b 0.51–1.33 20.15 0.53–0.65 0.1 0.45
Williams (1980) Richardson and Simpson (1988) McLaughlin et al. (1990) Paul et al. (2001a) Bromfield et al. (1983b) Pinkerton and Simpson (1986b) Cornish (1987) Purnomo and Black (1994) Young et al. (1995) Baker et al. (1999) Fettell and Gill (1995) Conyers and Scott (1989) Conyers et al. (1997) Burle et al. (1997) Loss et al. (1993) Dolling et al. (1994) Springett (1984)
Conventionally cultivated crops 0– 15 0– 7.5 0– 10 0– 10 0– 10 0– 10 0– 10 0– 10 0– 12.5 0– 15 0– 15 0– 15 0– 15 0– 15 0– 18 0– 30 0– 15 0– 15 0– 15 0– 15 0– 17.5 0– 20 0– 20
b
0.71 0.05–0.18 0.2 0.20–0.45 0.63 0.22–0.51b 0.8 0.03–0.18 0.3– 0.4 21.7 0–0.05 0–0.20 20.05a 0.1 1.01–1.21 0.9– 1.2 0.63 0.25–0.63 0.1– 0.2 0.39–0.77 20.05 to 0.81b 20.20 to 0.60 0.05–0.35
1 6 1 3 1 4 1 4 2 2 9 4 4 1 2 4 1 3 2 2 8 3 19
NSW, Australia Oklahoma, USA NSW, Australia NSW, Australia Victoria, Australia NSW, Australia NSW, Australia NSW, Australia Canada Montana, USA Idaho, USA Kentucky, USA Kansas, USA Mitchigan, USA Western Australia Illinois, USA NSW, Australia NSW, Australia Kentucky, USA NSW, Australia Brazil Western Australia Western Australia
Paul et al. (2001a) Jacobsen and Westerman (1991) Cornish (1987) Conyers et al. (1996) Slattery et al. (1998) Evans et al. (1998) Purnomo et al. (2000b) Fettell and Gill (1995) Grant and Bailey (1994) Pikul and Aase (1995) Mahler and Harder (1984) Blevins et al. (1983) Thompson and Whitney (2000) Pierce et al. (1994) Jarvis and Robson (1983) Hussain et al. (1999) Chan et al. (1992) Chan and Heenan (1993) Ismail et al. (1994) Chan and Heenan (1996) Burle et al. (1997) Loss et al. (1993) Dolling et al. (1994)
ACIDIC SUBSURFACE LAYERS OF SOIL
191
Table I (continued)
Soil depth (cm)
Observed pH decrease
No.
Location
0–20 0–20 0–30 0–20 0–30 0–20 0–20
0.5–0.9a 0.4–0.8a 20.3 to 0.5 0.1–0.2 0.2–1.0 0.5–0.7 NS
45 158 2 2 12 3 14
0–7.5 0–10 0–10 0–15 0–15 0–15 0–15 0–20 0–30
20.09 to 0.32 0.35–0.45 0.50–1.44 20.2a 20.16 to 0.10 20.1 to 0.30 0.78 0.22a 20.2 to 0.4
6 2 1 4 3 9 1 17 2
Oklahoma, USA NSW, Australia NSW, Australia Kansas, USA Colorado, USA Idaho, USA NSW, Australia North Carolina, USA Ohio, USA
0–7.5 0–10 0–10 0–10 0–10 0–10 0–10 0–12.5 0–15 0–15 0–15 0–15 0–15 0–15 0–20 0–20 0–30 0–30 0–30
0–0.28 0.4 0.50–0.75 1.44 0.5–1.1 0.3–0.7 0.14–0.20 0.1–0.9 1.15 20.05 to 0.35 20.20 to 0.20 20.9 21.2 20.5a 0.29a 0.8–1.0 0–1.3 20.5 to 0.4 1.0–1.2
6 1 2 1 1 1 2 2 1 9 4 1 1 4 13 3 12 2 2
Oklahoma, USA NSW, Australia NSW, Australia NSW, Australia NSW, Australia Victoria, Australia NSW, Australia Canada NSW, Australia Idaho, USA Kentucky, USA Michigan, USA Montana, USA Kansas, USA North Carolina, USA Kansas, USA West Virginia, USA Ohio, USA Illinois, USA
0–10 0–20 0–20 0–25 0–30 0–15
20.08 to 0.10 20.20 to 0.61 20.4 0.30 0.80–0.91 20.2
3 3 6 1 2 1
Western Australia Western Australia Western Australia ACT, Australia Utah, USA ACT, Australia
Western Australia Western Australia Ohio, USA Alberta, Canada West Virginia, USA Kansas, USA North Carolina, USA
Reference Dolling and Porter (1994) Dolling (1995) Dick (1983) Arshad et al. (1999) Staley and Boyer (1997) Matowo et al. (1999) Crozier et al. (1999)
Reduced-tilled crops Jacobsen and Westerman (1991) Conyers et al. (1996) Purnomo et al. (2000a,b) Thompson and Whitney (2000) Bowman and Halvorson (1998) Mahler and Harder (1984) Chan et al. (1992) Crozier et al. (1999) Dick (1983)
Direct-drilled crops Jacobsen and Westerman (1991) Cornish (1987) Conyers et al. (1996) Purnomo et al. (2000a) Purnomo et al. (2000b) Coventry and Slattery (1991) Fettell and Gill (1995) Grant and Bailey (1994) Chan et al. (1992) Mahler and Harder (1984) Blevins et al. (1983) Pierce et al. (1994) Pikul and Aase (1995) Thompson and Whitney (2000) Crozier et al. (1999) Matowo et al. (1999) Staley and Boyer (1997) Dick (1983) Hussain et al. (1999)
Forest or bushland O’Connell and Rance (1999) Loss et al. (1993) Dolling and Porter (1994) Benson et al. (1992) Van Miegroet et al. (2000) Keith et al. (1997) (continued on next page)
192
K. I. PAUL ET AL.
Table I (continued)
Soil depth (cm)
Observed pH decrease
No.
Location
Reference
Bare soil 0– 5 0– 15 0– 17.5 0– 18 a b
0.1 0.28 a 0.20 a 0.40–0.95
1 1 1 2
NSW, Australia NSW, Australia Brazil Western Australia
Williams (1980) Paul et al. (2001a) Burle et al. (1997) Jarvis and Robson (1983)
Mean value for sites. Soil was initially mixed.
associated with slow root penetration of soil at this depth. Slow and poor root growth attributable to acidic subsurface layers will limit the potential for water extraction and make plants susceptible to drought (Pinkerton and Simpson, 1986a; Cregan, 1992; Ferrufino et al., 2000). The surface of the soil is frequently exposed to intermittent drying. This may result in a rapid suppression of root growth near the surface (Engels et al., 1994). It has also been demonstrated that the concentration of plant nutrients N, P, Ca and Mg is greatest within the surface 2.5 cm of soil (Conyers and Scott, 1989; McLaughlin et al., 1990; Young et al., 1995). Pinkerton and Simpson (1986b) found that when the surface layer dries out, the N and P contained in this layer were no longer available to plant roots. This, together with poor root growth in the acidic subsurface layer, induced deficiencies of N and P in subterranean clover, lucerne, wheat and oats.
B. SUPPRESSION OF N MINERALISATION There is evidence (Young et al., 1995; Purnomo et al., 2000b; Paul et al., 2001c) that low pH contributes to the suppression of mineral N production within the acidic subsurface layer. Net N mineralisation was correlated to pH when soil from depth intervals with heterogeneous properties was separated. Therefore, both the production of and plant root access to mineral N may be limited as a result of the presence of an acidic subsurface layer.
C. POOR ROOT NODULATION OF LEGUMES Nodulation of legume roots appear to be suppressed by acidic subsurface layers. The highest population of Rhizobium trifolii, and quantity of subterranean clover root nodules, were mainly found in the more favorable pH regions of the
ACIDIC SUBSURFACE LAYERS OF SOIL
193
surface 4 cm rather than in the acidic subsurface soil layers (Richardson et al., 1985, 1988; Richardson and Simpson, 1988).
D. POOR GROWTH RESPONSE TO TOPDRESSING OF P FERTILISER McLaughlin et al. (1990) suggested that acidic subsurface layers may have significant effects on yield responses to P fertilisers. Purnomo and Black (1994) investigated P fertiliser application in surface soils with acidic subsurface layers. They found that broadcast application of rock phosphate was found to be inappropriate since the relatively high pH of the surface layer may limit its dissolution. However, mixing rock phosphate into the acidic layer achieved growth responses similar to that from superphosphate.
E. POOR GROWTH RESPONSE TO LIME APPLICATION In most cropping systems, lime is ploughed into the soil. In long-term pastures where cultivation is not possible, lime may be applied to the soil surface. However in southern Australia, it has been shown that even with cultivation, incorporation of significant amounts of lime beyond 8– 10 cm depth is unlikely where scarifiers or shallow-disc or -tined instruments are commonly used (Conyers and Scott, 1989; Dolling et al., 1991; Smith et al., 1994; Scott et al., 1997, 1999). In such instances it is also probably not realistic to incorporate lime to a depth of greater than 10 cm because suitable machinery is seldom available, and there may be a detrimental effect on soil structure and nutrient loss due to erosion. Results obtained by Scott et al. (1999) showed that conventional lime application had little effect on the plant growth limitations imposed by the presence of acidic subsurface layers. This is discussed further in Section VII.
IV. RATE OF DEVELOPMENT OF ACIDIC SUBSURFACE LAYERS Acidic subsurface layers were found to have developed within five years under pasture (Conyers and Scott, 1989) and various crop rotations (Burle et al., 1997, Fig. 2a). Field studies which have monitored changes in the pH of surface soil layers during the first season following surface soil mixing (7 – 8 cm depth using a rotary hoe, Evans et al., 1998; Paul et al., 2001a) showed that these acidic subsurface layers develop during the late spring and summer period.
194 K. I. PAUL ET AL. Figure 2 The pH of surface soil layers in an initially mixed soil (dashed line) and after (a) five (solid symbols) or 10 years (open symbols) under a cloverSpergula/Maize rotation (W, X) or a Cajanus-Maize (V, S) rotation in an initially mixed soil (Burle et al., 1997), (b) seven months in unplanted (O) and oat (K) plots in hay amended soil (Evans et al., 1998), and (c) after nine months under clover (B) and wheat (X) (Paul et al., 2001a).
ACIDIC SUBSURFACE LAYERS OF SOIL
195
In the study by Evans et al. (1998), pH changes were monitored in soil layers amended with ammonium sulfate, sucrose or lucerne hay. They reported that there was no acidic subsurface layer following soil mixing in early winter. However by early summer, a pHCa (pH measured in a CaCl2 extract of soil) decrease of up to 0.51 units was observed from the 0– 2 cm to the 8– 10 cm layer. Acidic subsurface layers developed due to an acidification of all soil layers except in the surface 2 cm of soil. The greatest acidification of the 8 –10 cm layer was observed in the hay-amended soil (Fig. 2b). Paul et al. (2001a) monitored the change in pHCa within surface soil layers following mixing in fallow, wheat and subterranean clover plots in autumn. This resulted in a uniform pHCa through 0 – 7.5 cm depth that was maintained under fallow. However by summer, the decrease in pHCa from the 0– 2.5 cm to the 5 – 7.5 cm layer was up to 0.20 and 0.75 units under wheat and clover, respectively (Fig. 2c). The development of acidic subsurface layers was most pronounced under clover during the late spring and early summer period. The relatively low pH of the subsurface layer was generally maintained during the second season after soil mixing. A pot experiment (Black, 1992) and field study (Condon, pers. commun.) showed that acidic subsurface layers develop within about two months in urineinfluenced soil since pH decreases at depths of 2– 6 cm but changed little within the surface 2 cm of soil.
V. CAUSE OF DEVELOPMENT OF ACIDIC SUBSURFACE LAYERS There are a number of factors contributing to the development of acidic subsurface layers, many of which have only recently been demonstrated.
A. PLANT N UPTAKE A plant uptake of excess charge as cations over anions results in net Hþ excretion to the rhizosphere, whereas uptake of excess charge as anions over 2 cations results in net alkaline efflux (as either OH2, HCO2 3 or RCOO ) into the rhizosphere. Plant uptake of N has a predominant impact on plant maintenance of electroneutrality because N is normally: (1) taken up by plants in greater concentrations than most other elements, and (2) can be absorbed in forms of þ varying charge (NO2 3 , NH4 or N2). Therefore, in order to maintain internal electroneutrality, plants may excrete net alkali if NO2 3 absorption exceeds 83% of the N taken up. Net acid excretion from plant roots occurs when plant NHþ 4 absorption exceeds 17% of the N taken up (Nye, 1987).
196
K. I. PAUL ET AL.
The influence of plant N uptake on soil pH rests solely on the fact that plants must maintain an overall internal electroneutrality. It is unrelated to where the plant controls its internal electroneutrality. This appears to have been misunderstood by some authors who have suggested that the effect of plant N uptake on soil pH was dependent on the relative activity of nitrate reductase enzyme in the root and shoot tissues (Smiley, 1974; Bolan et al., 1991). Also, some authors (Tang et al., 2000) have incorrectly stated that plant roots extrude Hþ when absorbing cations, and extrude OH2 when absorbing anions. In fact, plant root Hþ/OH2 excretion is dependent on the overall net excess cation/anion uptake. Non-legume plants generally utilise NO2 3 as their main source of N, with the exception of urine patches. Uptake of N by these plants will therefore generally result in a net alkali excretion from their roots (Bagayoko et al., 2000). In contrast, legumes may fix 64 – 99% of their N (Ellington, 1988), and therefore take up a large excess of cations, resulting in a net acid excretion from their roots (Israel and Jackson, 1978; Nye, 1981). For a range of legumes, it has been shown that total acid production is highly correlated with total N2 fixation, and the accumulation of excess cations (McLay et al., 1997, 1998a,b; Tang et al., 1997; Tang, 1998). There is some evidence that legumes increase subsurface acidification. Coventry and Slattery (1991) noted that compared with continuous wheat, soil pH was 0.7 and 1.0 units lower at 5– 10 cm depth under wheat – lupin and continuous lupin rotations, respectively. Others have also observed that relative to soil under a continuous wheat crop, there was an increase in acidification at 5– 20 cm depth when lupin was grown continuously or in rotation with wheat (Chan and Heenan, 1993; Dolling, 1995; Conyers et al., 1996; Slattery et al., 1998). It has been suggested (Loss et al., 1993; Dolling, 1995) that this may be attributed to a relatively large excess cation uptake and consequent Hþ excretion from the roots of legumes that were presumably most active at a depth of 5 –20 cm. Further evidence of acidification of the subsurface soil layer by growth of legumes was provided by Burle et al. (1997). They studied 10 cropping rotations over 10 years in an initially deep ploughed soil. They found that the greatest soil pH decreases over time occurred at depths of 2.5– 7.5 and 7.5– 17.5 cm, and under subterranean clover (Trifolium subterraneum ) and pigeonpea (Canjanus spp.); pH in these depths decreased more than under fallow/maize (Zea mays ) rotations (Fig. 3a). Similarly, Paul et al. (2001a) found that in contrast to fallow soil or soil under wheat, subsurface (2.5 –10 cm) acidification was evident in spring under rapidly growing clover (Fig. 3b). As there were numerous active nodules on the clover roots, and the total N content of clover shoots was up to threefold higher than that of wheat shoots, it was suggested that clover supplemented NO2 3 uptake with N2 fixation and this resulted in a net Hþ excretion by clover roots within the subsurface layer. In contrast, pHCa within 0 –7.5 cm soil under wheat was 0.50 units higher than that under fallow and this was likely to have been associated with net OH2 excretion from wheat roots in response to NO2 3 uptake.
ACIDIC SUBSURFACE LAYERS OF SOIL Soil pH
Soil depth ( cm)
4.5 0
4.7
4.9
5.1
197
Soil pH 5.3
4.8
5.2
5.6
6.0
6.4
5
10 a
b
15 Figure 3 The pHCa of surface soil layers under (a) continuous maize (V) or clover-maize rotation (S) (Burle et al., 1997), and (b) under clover (B), fallow (X), and wheat (O) the first spring after soil mixing (Paul et al., 2001a).
In summary, there is growing evidence that uptake of N by plants contributes to the development of acidic subsurface layers of soil. The excess cation uptake by N2-fixing legumes results in subsurface acidification during plant growth. In contrast, the excess anion uptake of non-legumes in nitrifying soils appears to result in the alkalinisation of the soil surface.
B. PLANT RESIDUE RETURN Confusion appears to exist in some papers over the influence of C and N transformations on soil pH following the addition of plant residues to soil. The addition of organic C from plant residues generally leads to an initial increase in soil pH through both the association of organic anions (their pK being higher than the soil pH) and the biological oxidation of these anions to CO2. Although in some studies the complexation of Al by organic anions is considered to be the cause of increased soil pH, it is best to ascribe the loss of Al from solution to the increase in soil pH that arises from the association and oxidation of organic anions. In the long term, decomposition of C (glucose) polymers such as cellulose and starch produces humic acids. These large C polymers produce a range of carboxylic and related functional groups that release Hþ and concurrently increase ECEC. The initial release of organic N as NHþ 4 (mineralisation) 2 consumes Hþ, thus raising soil pH. Subsequently, the oxidation of NHþ 4 (to NO3 ) þ releases 2H , thereby lowering the soil pH. If the oxidation of mineralised NHþ 4 is complete, the net result is an acidified soil. Therefore, the net effect of adding plant material on soil pH depends on the following: (1) relative rates of association/dissociation, oxidation and gradual humification of C, (2) relative
198
K. I. PAUL ET AL.
Amended - unamended soil pH
rates of N mineralisation followed by nitrification, and (3) the time period over which the soil pH is monitored. Short-term laboratory incubation studies have demonstrated that addition of organic anions (Yan et al., 1996a) or plant material (Hoyt and Turner, 1975; Yan et al., 1996b; Tang et al., 1999a; Paul et al., 2001b) to moderately acidic soils initially results in a rapid pH rise (due to organic anion association and oxidation) followed by a more gradual pH decline (due to N mineralisation and subsequent nitrification) (Fig. 4). However relative to unamended soil, it has consistently been demonstrated that during incubation of acidic (pHCa , 5.7) soils, there is a net pH increase following the application of plant material (Hoyt and Turner, 1975; Ritchie and Dolling, 1985; Tang et al., 1999a; Tang and Yu, 1999; Paul et al., 2001b). It has been shown that the initial pH increase following plant residue (Paul et al., 2001b) or organic anion (Yan et al., 1996a) addition was correlated to CO2 evolution, and thus microbial activity. However, the increase in pH following the addition of six different plant residues (lupin (Lupinus angustifolius ), field pea (Pisum sativum ), pink serredalla (Ornothopus sativus ), bisserrula (Biserrula pelecinus ), clover (Thrifolium subterraneum ) and wheat (Triticum aestivum ) shoots) to two contrasting soils (initial pHCa 3.98 and 5.06) was 0.1 –1.78 units (average 0.65 units) less in sterilised (fumigation with chloroform) than nonsterilised soil (Tang and Yu, 1999). Also, a majority (55 – 100%) of the soil pH increase in non-sterilised soil occurs within the first hour (Paul et al., 2001b) or 18 h (Yan et al., 1996b; Tang et al., 1999a; Tang and Yu, 1999) of plant residue addition. These results indicate that both association and biological oxidation contributes to the soil pH increase following addition of organic anions to soil.
1.0 0.8 0.6 0.4 0.2 0.0 0
20
40
60
80
100
120
Incubation days
Figure 4 Difference in pH between unamended soil and soil amended with 3.08 g DM kg21 of clover shoots (W, X) or lupin leaves (V, S), 15 g DM kg21 of field beans (K), and 12.5 g DM kg21 of wheat shoots (O) or spring sampled clover shoots (A) and summer sampled cover shoots (B). Data taken from Yan et al. (1996b), Tang et al. (1999a), and Paul et al. (2001b).
ACIDIC SUBSURFACE LAYERS OF SOIL
199
The association and oxidation of organic anions from plant shoot residues may occur predominantly at the soil surface since (1) that is where they are added, (2) they are rapidly utilised by soil microbes, and (3) most heterotrophic acitivity occurs at the soil surface (Purnomo et al., 2000b). Organic anions may lose 80 – 90% of their C within 3 –6 months after addition to soil (Jenkinson and Ayanaba, 1977). Since the liming effect of plant shoots is much greater than that of plant roots (McLay et al., 1997; Tang et al., 1999a), alkalinity added from root slough and decomposition is less likely to influence the development of acidic subsurface layers. Paul et al. (2001a) provided direct evidence from field experiments that the return of plant shoot components to the soil surface was the predominant cause of development of gradients in acidity in an initially mixed surface soil. During late spring and early summer in fallow plots, the surface 10 cm of soil tended to acidify due to nitrification, whereas the magnitude of the pHCa decreases between surface and subsurface layers was maintained under wheat and increased under clover. In a subsequent 119 day soil incubation experiment (Paul et al., 2001b), the simulated addition of wheat and clover residues observed in the field (i.e., 12.5 g DM kg21) resulted in final 0 –2.5 cm soil pHKCl (pH measured in a KCl extract of soil) which was significantly 0.21 – 0.45 units higher than in unamended soil. It was, therefore, concluded that over late spring and summer in the field, the return of 11– 14 g DM kg21 of wheat or clover residues to the surface 2.5 cm accounted for the average 0.45 or 0.61 unit higher pHCa in 0 – 2.5 cm soil under wheat or clover plots, respectively, than in fallow plots (Paul et al., 2001a). The increased magnitude of pH decrease through surface soil depth in response to plant residue return was further demonstrated by the gradual increase of 0 – 2.5 cm soil pH as plants senesce during late spring and early summer (Fig. 5). In the surface 20 cm of soil in a wheat plot, Poss et al. (1995) observed that over one season, alkalinity released from crop residues was not counterbalanced by the net acidity produced by mineralisation and subsequent nitrification of organic N. Paul et al. (2001b) also demonstrated that within the uppermost soil layer (0 – 2.5 cm), mineralisation and subsequent nitrification of the added organic N only partly counterbalanced the rapid initial alkalinisation resulting from the oxidation of added organic anions. In fact, acidic subsurface layers that developed following plant residue return were maintained during the subsequent growing season. Others have also found that in the field, plant residue return resulted in a net increase in soil pH. For example, Hafner et al. (1993) obtained evidence that application of millet straw (2 – 4 t ha21 yr21 for five years) resulted in a 0.5– 0.7 units greater soil (0 –20 cm) pHKCl than in unamended soil. In Utah, van Miegroet et al. (2000) noted that the decrease in soil pHH2 O (pH measured in a water extract of soil) through 0 – 30 cm depth was 0.80– 0.91 units under forest canopy, but only 0.27 –0.38 in a nearby exposed soil, and was attributable to the
200
K. I. PAUL ET AL. Soil pH 5.0
5.2
5.4
5.6
5.8
6.0
6.2
6.4
Soil depth (cm)
0
5
10
15 Figure 5 The pHCa of surface soil layers in spring under growing clover (X), in summer under senescing clover (B), and in the following autumn under senesced clover (O). Data from Paul et al. (2001a).
fact that under forest canopy the pH within the surface 5 cm of soil was significantly higher than in exposed soil.
C. MN REDUCTION AND OXIDATION It has been shown (Conyers et al., 1997; Paul et al., 2001a) that in moderately acidic soils in southeastern Australia, extractable Mn accumulates at the soil surface during the late spring, the early summer period. It was previously suggested (Evans et al., 1998) that in acidic soil (pHCa , 4.5) with a relatively high concentration of extractable Mn (290 mg kg21), one of the causes of relatively low pH within subsurface layers was the alkalisation of the surface 2 cm in response to the reduction of Mn. However in surface soils (0 –10 cm) that are only moderately acidic (pHCa 5.0– 6.0), net Mn reduction may account for less than 2% of the absolute surface soil Hþ change (Paul et al., 2001a).
D. URINE EXCRETION FROM GRAZING STOCK There is evidence (Black, 1992) that urine deposition from grazing stock may decrease the subsurface (2 – 8 cm) pHCa by up to 1.2 units relative to the unamended soil. Condon et al. (2000) monitored changes in the pHCa profile in urea-amended pots planted with ryegrass. They found that pH changes were attributable to N transformations and were restricted to the surface 8 cm of soil.
ACIDIC SUBSURFACE LAYERS OF SOIL
201
By day 9 the soil (0 – 8 cm) pH had significantly increased due to hydrolysis of applied urea, while subsequent acidification of this soil (day 9 –26) was attributable to NH3 volatilisation and nitrification. Then between days 26 and 65, the soil pH increased in the surface 6 cm of soil, presumably due to plant uptake of NO2 3 . Thus, by day 65 the pHCa within the 0– 6 cm layer was up to half a unit greater than that within the 6– 8 cm layer.
E. SOIL pH BUFFERING CAPACITY Numerous workers (Jarvis and Robson, 1983; Bromfield et al., 1983a; Chan et al., 1992; Purnomo and Black, 1994; Purnomo et al., 2000a,b) have observed that organic C content and CEC decreased through depth in moderately acidic surface soils. It has been shown that each 1% increase in soil organic matter can increase soil pH buffering capacity (pHBC) by 30– 40 mmol Hþ kg21 soil (Chan et al., 1992; Slattery et al., 1998). These observations are consistent with the decrease in pHBC with depth through the surface 20 cm of soil (Bromfield et al., 1983a; Ridley et al., 1990; Loss et al., 1993; Dolling et al., 1994). These results suggest that in unmixed soils with a pronounced decline of pHBC through 0 – 15 cm depth, subsurface layers will acidify to a greater extent than the soil surface for a given input of Hþ.
VI. ENVIRONMENTAL FACTORS AFFECTING THE DIFFERENCE IN pH BETWEEN SURFACE AND SUBSURFACE LAYERS A. SOIL FERTILITY It is likely that the mineral N content of the soil will influence the extent to which legumes fix N and thus contribute to subsurface acidification (Paul et al., 2001a). Also, through its effects on productivity, soil fertility may influence the quantity of plant residue return, and thus the net alkalinisation of the soil surface.
B. INITIAL SOIL pH The association of organic anions within plant residues is dependent on soil pH. The lower the initial soil pH, the greater is the extent of organic anion association (Ritchie and Dolling, 1985; Paul et al., 2001b). Studying five soils differing in initial pHCa (3.60 –5.58), Tang and Yu (1999) found that adding
202
K. I. PAUL ET AL.
plant materials increased the pHCa in soils with initial pH between 3.60 and 4.54, but varied (depending on the residue type) in soils with an initial pH of 5.06 and 5.58. In contrast to organic anion association, net N mineralisation and subsequent nitrification is suppressed by low soil pH (Purnomo et al., 2000b). Tang and Yu (1999) found that when the incubation time of plant residue amended soils increased from 35 to 100 days, the amount of net alkalinity produced decreased in the soils with the initial pHCa above 4.5, and increased in the soils with pHCa below 4.0. They suggested that this was because nitrification was lower in the lower pH soils. Similarly, during the 119 day incubation of 0 – 2.5 cm (initial pHKCl 5.72 units) and 7.5 – 10 cm (initial pHKCl 5.32 units) soil, Paul et al. (2001b) observed that the addition of 12.5 g DM kg21 of plant material resulted in a significant 0.13 – 0.25 unit greater net alkalisation in 7.5– 10 cm than in 0– 2.5 cm soil. This was attributed to the 5– 12% greater in release of alkalinity, and the 39 – 51 mg N kg21 less net N nitrified in the more acidic soil. These results suggest that moderately acidic soils are most prone to the development of acidic subsurface layers in response to the addition of plant residues to the soil surface. This is consistent with observations that acidic subsurface layers occur throughout regions where surface soils are moderately acidic. In highly acidic surface soils, plant productivity and thus residue return may be suppressed for most species.
C. RAINFALL AND FLUCTUATIONS IN SOIL MOISTURE CONTENT Rainfall has a substantial influence on plant productivity and the quantity of plant residues returned to the surface of the soil. It follows that alkalinisation of the soil surface due to plant residue return will be more pronounced in high than low rainfall regions. Further, Paul et al. (2001a) investigated the changes in the pH of soil layers within the surface 15 cm of soil over two consecutive seasons. They found that as a result of drought and minimal plant residue return, the magnitude of pH decrease between the surface (0 –2.5 cm) and subsurface layers (2.5 – 10 cm) during the second season was less than that observed during the first season. Plant shoot residues are returned to the surface of the soil. The soil surface is exposed to frequent drying and rewetting cycles of much greater magnitude than experienced in the deeper soil layers (Campbell and Biederbeck, 1976). Jager and Bruins (1975) observed suppressed acidification with intermittent drying of an acidic soil (pHH2O 5.2), with an acidification of only 0.2 units in moist –dry soil (one cycle of 3-days dry, 4-weeks moist at 308C) but 0.3 units in continually moist soil. After 20 such moist – dry cycles, acidification was only 0.4 units in moist – dry treated soil, but 0.8 units in the continually moist soil. Paul et al. (1999) observed that after 28 days of incubation of 0 – 2 cm soil, acidification was up to 0.26 units less in moist –dry (7-days dry, 7-days moist at 208C) than in
ACIDIC SUBSURFACE LAYERS OF SOIL
203
moist-treated soils. Similarly, 119 days after the addition of wheat or clover residues to 0– 2.5 cm soil at similar rates to that observed in the field, Paul et al. (2001b) noted that soil pHKCl was up to 0.15 units higher in moist – dry (7-days dry, 7-days moist at 208C) relative to moist-treated soils. In moderately acidic soils (pHCa 5.0 – 6.0), acidification rates may be suppressed on exposure to moist – dry cycles due to an increase in Hþ consumption in response to a twofold enhancement of net Mn reduction (Paul et al., 1999). However, the main reason for the lower acidification rates on exposure to moist –dry cycles is a decrease in Hþ production in response to a 4 – 49% suppression of net N mineralisation and subsequent nitrification in soils from relatively dry and highly variable climate (van Veen et al., 1985; van Gestel et al., 1991, 1992; Paul et al., 1999, 2001b). This was attributed to the adaptation of soil microbes to climatic conditions, and a resultant small release of available substrate from biomass killed on drying. It should be noted that in surface soils sampled from relatively humid environments, mineralisation is enhanced on exposure to intermittent drying (Birch, 1958, 1960; Soulides and Allison, 1961; Seneviratne and Wild, 1985; van Gestel et al., 1993), probably because soil microbes are not adapted to moist – dry cycles in these environments. It is likely that in such soils, net N mineralisation and subsequent nitrification and acidification may increase in response to moist – dry cycles. In an arable region of southern Australia (annual rainfall 550 mm), Paul et al. (2001c) noted that the relatively high rates of 0– 2.5 cm soil net N mineralisation, nitrification and acidification observed during laboratory incubation were not observed under senescing plants in the field (Paul et al., 2001a). It was therefore concluded that in this environment, the potential for 0 –2.5 cm soil acidification may not be realised under senescing plants in the field due to the prolonged dry periods and exposure of 0 – 2.5 cm soil to moist – dry cycles. This contributed to the maintenance of the gradient in soil pH.
D. EARTHWORM POPULATIONS Although there are some regions where acidic subsurface layers are common, such as in the arable regions of southeastern Australia, there are other higherrainfall regions in Brazil and the USA where they are not observed (Table I). This disparity between observations may depend on the species and population density of earthworms because (1) earthworm populations have been observed to range from only 1 to 120 g m22 within the surface 10 or 20 cm under crops and pastures in southeastern Australia (Rovira et al., 1987; Haines and Uren, 1990; Thompson, 1992; Doube et al., 1994), whereas they reach populations of the order of 444 g m22 to 30 cm depth in more humid environments (Barnes and Ellis, 1979), and (2) populations of at least 385 g m22 are required to mix surface
204
K. I. PAUL ET AL.
soils (Baker et al., 1999). Furthermore, agricultural soils in southeastern Australia are dominated by shallow burrowing earthworm species such as Aporrectodea caliginosa and A. trapezoides (Baker et al., 1999). During 1984, the low rainfall and lack of earthworms in the Otago uplands of New Zealand results in limited mixing of the surface and subsurface strata. Deep burrowers such as A. longa are rare in some arable and pasture soils of Australia and New Zealand, and the introduction of these species in these locations enhanced the burial of lime and helped to reduce subsurface acidity (Springett, 1984; Baker et al., 1999).
VII. MANAGEMENT FACTORS AFFECTING THE DIFFERENCE IN pH BETWEEN SURFACE AND SUBSURFACE LAYERS A. AGRICULTURAL PRODUCTION There tends to be an increase in the relative high rates of acidification within the subsurface following the conversion of virgin soils to soils used for agricultural production. Under cereal-annual pasture rotation, the magnitude of pH decrease between the surface and subsurface layers increases with years since clearing, particularly in acidic soils (Dolling et al., 1994; Dolling and Porter, 1994, Fig. 6). Jarvis and Robson (1983) found pHCa decreases of 0.65 –1.06 units through 0 –12 cm depth in soils that had a 17 – 26 year history of crop and pasture
Soil pH 3.8
4.0
4.2
4.4
4.6
4.8
Soil pH 5.0
5.2 3.8
4.0
4.2
4.4
4.6
4.8
5.0
5.2
5.4
0
Soil depth (cm)
a
b
10
20
30
40
Figure 6 Effect of depth and years since clearing ((W) mean uncleared sites; (X) 8–12 years; (B) 50 years; (O) 74 –105 years) on pH within (a) moderately acidic sandy duplex soils (Dolling et al., 1994) and (b) very acidic deep yellow sands (Dolling and Porter, 1994) and under cereal-annual pasture rotations in Western Australia.
ACIDIC SUBSURFACE LAYERS OF SOIL
205
production. In contrast, under adjacent virgin soils, pHCa decreases of only 0.40– 0.83 units were observed through 0 –12 cm depth. Others (Williams, 1980; Bromfield et al., 1983a; Dolling, 1995) have obtained results indicating that there is little or no evidence of acidic subsurface layers in virgin soils (Table I). In two of the three sites studied by Loss et al. (1993), the magnitude of pHKCl decrease through surface soil depth was greater under pasture or cropping systems than under adjacent native bush.
B. PLANT SPECIES GROWN Acidic subsurface layers are evident in surface soils under pastures and crops, regardless of the extent of soil disturbance during tillage (Table I). Available data also indicate that there is no consistent difference in the magnitude of pH decrease through 0 – 10 cm depth under crop and grazed pasture systems (Cornish, 1987; Cornish and Lymbery, 1987). Furthermore, in wheat-grazed pasture rotations, increasing the length of the pasture phase made no consistent difference in the magnitude of pH decrease through 0– 20 cm depth (Loss et al., 1993; Dolling et al., 1994). However, some results suggest that lupin growth (Coventry and Slattery, 1991; Chan and Heenan, 1993; Conyers et al., 1996), or subterranean clover growth (Paul et al., 2001a), results in greater subsurface acidification than that of non-legumes such as wheat. In contrast, De Maria et al. (1999) found that there was little difference in the extent of decline in soil pH between 0– 5 and 5– 10 cm depth in soil under soybean compared with soil under corn. Effects of non-legume versus legume, and discrepancies between species of legumes, are related to differences in the N source and thus variation in the extent to which legume species acidify the soil due to Hþ excretion by roots. Studying Hþ excretion by a range of grain legumes grown in a nutrient solution, McLay et al. (1997) found that chickpea was the most acidifying, while field pea was the least acidifying species (Table II). Tang et al. (1997) measured Hþ excretion from N2-fixing pasture legume species using a solution culture assay. They concluded that clover species often excreted greater amounts of Hþ per unit biomass than medics, which, in turn, excreted more Hþ than serradella species (Table II). Similar results were obtained by comparing Hþ excretion by pasture legume species in a pot experiment (Tang et al., 1998a). However, the relative Hþ excretion of serradella species were greater than that observed in the nutrient solution. The smallest Hþ excretion was observed under Biserrula (Biserrula pelecinus ). In moderately acidic soil, the pH increase observed after plant material incorporation was strongly correlated to the plant material organic anion content (Noble et al., 1996; Yan et al., 1996b; Tang et al., 1999a; Tang and Yu, 1999; Paul et al., 2001b). Since legumes tend to maintain a higher organic anion content
206
K. I. PAUL ET AL.
Table II Specific H1 Production of Grain and Pasture Legumes Grown in Nutrient Solution (From McLay et al., 1997; Tang et al., 1997)
Grain legume species Pisum sativum (Field pea) Lupinus albus (White lupin) Vicia faba (Faba bean) Lens culinaris (Lentil) Lupinus pilosus (Pilosus) Lupinus luteus (Yellow lupin) Vicia sativa (Common vetch) Lathyrus sativus (Grasspea) Lupinus angustifolius (Narrow-leafed lupin) Cicer arietinum (Chickpea) a b
Specific Hþ productiona (cmol(þ) kg21) 77 84 85 89 100 102 105 106 119 136
Pasture legume species
Specific Hþ productionb (cmol(þ ) kg21).
Ornithopus sativus Ornithopus compressus Medicago truncatula Medicago polymorpha Medicago murex Trifolium subterraneum Trifolium vesiculosum Trifolium subterraneum Trifolium balansae Trifolium glomeratum Trifolium subterraneum Trifolium tomentosum
143 186 186 188 209 170 180 188 190 211 214 265
During 0– 35 days after visible nodules, LSD ¼ 16 at P , 0.05. During 40– 61 days, LSD ¼ 53 at P , 0.05.
because of their excess cation uptake (Pierre and Banwart, 1973; Jarvis and Robson, 1983; Slattery et al., 1991; Tang and Yu, 1999; Paul et al., 2001a,b), their incorporation into soil tends to result in a greater increase of pH than observed with incorporation of non-legumes (Table III). Bessho and Bell (1992) observed that the leaves of tropical legumes were approximately twice as effective as a cereal straw in raising an acidic surface soil (0 –15 cm) pHCa at any given rate of dry matter addition between 4 and 64 g kg21. Similarly, Hue and Amien (1989) found that the application of leaves of cowpea and leucanea induced a greater increase in soil pH than grass leaves. Similarly, over two seasons in the field (Paul et al., 2001a), and during laboratory incubation (Paul et al., 2001b), it was shown that alkalisation of 0– 2.5 cm soil was 0.19 –0.61 pHCa units greater with addition of subterranean clover than addition of wheat residues. Tang and Yu (1999) found that after 35 days of incubation of five different soils, alkalinity production was greatest with field pea, followed by chickpea, lupin shoot, faba bean, lupin leaves, serradella, biserrula, tagasate leaves, and least with clover shoots, yellow lupin and wheat straw. In summary, compared to cereal crop systems, legume-based pasture systems result in both greater subsurface (2.5 –10 cm) acidification due to root Hþ excretion, and greater surface (0 – 2.5 cm) alkalisation due to plant residue return. These factors accounted for the average 0.26 unit larger magnitude of pHCa
Table III Net Increase in pH in Acidic and Moderately Acidic Soils after Incubation with Ground Plant Shoot/Leaf Materials for Between 28 and 307 days, at Addition Rates of Between 3 and 30 g DM kg21 Soil
Plant material incorporated into soil
,5.0
.5.0
Non-legumes Triticum aestivum (wheat) Zea mays (maize)
20.05 to 0.15
20.05 to 0.22 0.10–0.35
Tang and Yu (1999); Paul et al. (2001b)a Yan et al. (1996b)a
0.10 to 0.39 0.04 to 0.32 0.24 to 1.71 0.19 to 2.00 0.47 to 1.88 0.06 to 1.55 0.25 to 1.32 20.06 to 1.43 0.09 to 0.72 20.03 to 0.33
1.34 20.13 to 0.46 20.15 to 0.61 20.42 to 0.76 0.04–0.48 20.19 to 0.43 0.03–0.25 20.23 to 0.21 0.09 to 0.16 20.03 to 0.05
Hoyt and Turner (1975); Ritchie and Dolling (1985)b Tang and Yu (1999); Paul et al. (2001b)a Tang and Yu (1999); Tang et al. (1999a)a Tang and Yu (1999)a Tang and Yu (1999)a Yan et al. (1996b); Tang and Yu (1999)a Tang and Yu (1999)a Tang and Yu (1999)a Tang and Yu (1999)a Tang and Yu (1999)a
Legumes Medicago sativa (lucerne) Trifolium subterranean (clover) Lupinus angustifolius (lupin) Pisum sativum (field pea) Cicer arietinum (chickpea) Vicia faba (faba bean) Ornothopus sativus (pink serredalla) Biserrula pelecinus (Biserrula) Chamaecytisus palmensis (tagasare) Lupinus luteus (yellow lupin)
Reference
ACIDIC SUBSURFACE LAYERS OF SOIL
Net increase in pH (initial soil pHCa)
a
Net pH increase relative to unamended soil pH. Net pH increase relative to initial soil pH.
b
207
208
K. I. PAUL ET AL.
decrease between surface and subsurface layers in clover than in wheat plots over late spring to autumn (Paul et al., 2001a, Fig. 2b).
C. PRODUCTIVITY AND THE QUANTITY OF PLANT RESIDUES ADDED TO THE SOIL SURFACE During laboratory incubation of bulked surface soils, many workers (Barrow, 1960; Hue and Amien, 1989; Bessho and Bell, 1992; Tang et al., 1998a) noted that alkalisation increased with increased plant material addition rates between 0 and 20 g DM kg21. Over a 119-day incubation period, Paul et al. (2001b) observed that 0– 2.5 cm soil pHKCl increased by 0.02 and 0.03 units for each additional g DM kg21 of wheat and clover material, respectively. This was mainly attributed to the fact that the proportion of alkalinity released was unaffected by the quantity of plant material added. In contrast, the proportion of organic nitrogen mineralised and subsequently nitrified decreased with increased quantity of plant material added. The pHBC of soil also increased with plant material addition rate. However, as there was a linear increase in pHKCl with increased plant material addition rate, the increase in pHBC was small compared to the alkali added. It may be concluded that not only the type of plant residue, but also the quantity of the plant residue return, has a substantial influence on soil alkalisation. This is particularly evident for plant residues of high ash alkalinity. Consequently, plant residue addition rates will have a greater influence on acidic subsurface layers in legume-based pasture systems than in cereal cropping systems.
D. MINIMUM SOIL DISTURBANCE AND TILLAGE Due to the adoption of minimum soil disturbance agricultural techniques such as direct-drilling, the surface soil may remain relatively unmixed for a number of consecutive seasons. Under crops, direct-drilling has been found to result in a greater pH decrease between the surface and subsurface layers relative to conventional cultivation (Groffman, 1984; Haynes and Knight, 1989; Chan et al., 1992; Conyers et al., 1996; Matowo et al., 1999; De Maria et al., 1999; Crozier et al., 1999; Purnomo et al., 2000b), whereas others have found that the level of tillage had little or no effect on pH decreases through surface soil depth (Dick, 1983; Jacobsen and Westerman, 1991; Ismail et al., 1994; Staley and Boyer, 1997; Hussain et al., 1999), or that the influence of tillage on acidic subsurface layers was dependent on the soil-type (Grant and Bailey, 1994). Discrepancies between the effects of tillage operations on acidic subsurface layers may be attributable to the extent of mixing of the surface soil with tillage.
ACIDIC SUBSURFACE LAYERS OF SOIL
209
E. FERTILISER APPLICATION The broadcast or surface application of NHþ 4 -based fertilisers tends to minimise the development of acidic subsurface layers (Jacobsen and Westerman, 1991; Dolling et al., 1994; Ismail et al., 1994; Conyers et al., 1996; Staley and Boyer, 1997; Bowman and Halvorson, 1998; Matowo et al., 1999; Thompson and Whitney, 2000; Tang et al., 2000). For example it has been found (Chan and Heenan, 1993) that in response to the addition of 100 kg N ha21 as urea, the magnitude of pHCa decrease through 0 –15 cm depth was 0.20 units less than that observed in the unfertilised control. Similarly, Fettell and Gill (1995) found that 14 years of addition of ammonium nitrate to wheat resulted in a 0.30 unit smaller decrease in pHCa between 0 –2.5 and 5 –10 cm depth than in unfertilised plots. Where there has been a long history of NHþ 4 -based fertiliser application to the soil surface, soil pH may increase with increasing depth (Mahler and Harder, 1984). This is likely to be attributable to greater acidification of the surface layers because of two main processes. Firstly, nitrification at the soil surface will increase following topdressing with NHþ 4 . Secondly, there may initially be a greater excess cation uptake, and thus more Hþ excretion from roots, in NHþ 4treated soil. In contrast, it is expected that the surface application of NO2 3 -based fertilisers may increase development of acidic subsurface layers (Grant and Bailey, 1994). In a pot experiment, Tang et al. (1999b) showed that for a range of legume species, the release of Hþ from plant roots declined with increasing rate of NO2 3 addition. Tang et al. (2000) found that acidification of the surface 5 cm of soil 2 was significantly less in NO2 3 -treated than non-NO3 -treated soil under clover or 2 lupin. This was presumably attributable to NO3 uptake occurring predominantly within the surface 5 cm, and a lower excess cation uptake, and thus less Hþ excretion, in NO2 3 -treated soil.
F. LIME APPLICATION There is evidence that surface application of lime enhances the development of gradients in pH. Conyers and Scott (1989) found that the decreases in soil pHCa through 0 – 12 cm depth was 1.17 – 1.51 units five years after the incorporation of 2 t ha21 of lime with a rotary hoe (set to a 10 cm nominal depth). In adjacent plots of unamended soil, pHCa decreases through 0– 12 cm depth were only 0.77 –1.09 units. Whereas to be expected, Smith et al. (1994) noted that six months after 2.5 t ha21 of lime was applied to the surface of a pasture, the magnitude of pHCa decrease through 0– 10 cm depth increased from less than 0.2 units to over 1.0 unit. The application of lime may be essential to maintain productivity in acidifying soils. However, as the movement of relatively insoluble lime down the soil profile
210
K. I. PAUL ET AL.
Figure 7 Soil pHCa one year (a) and eight years (b) after the application of 0 t ha21 (white bars), 2 t ha21 (black bars), or 8 t ha21 (gray bars) of lime and incorporation to a depth of 10 cm (black and black striped bars) or 20 cm (gray and gray striped bars). Data were derived from Scott et al. (1997).
is minimal in the absence of numerous deep burrowing earthworms, lime may need to be mechanically mixed through the surface soil, rather than surface applied, where possible. However, Scott et al. (1997) found that at agriculturally realistic rates (2 t ha21) of lime incorporation into the 0 –10 cm layer, no significant effect on pHCa of the 10 –20 cm soil was detected even after eight years. To ameliorate acidic subsurface layers eight years after the application of lime, 8 t ha21 needed to be mixed within the surface 20 cm (Fig. 7).
VIII. MANAGEMENT IMPLICATIONS Genotypes may be selected to minimise the detrimental effects of acidic subsurface layers on plant growth. Ferrufino et al. (2000) found that there was scope for selecting soybean genotypes that were more tolerant to subsurface acidity, and could therefore possibly gain better access to water and nutrients. Bagayoko et al. (2000) found evidence that the root-induced 0.3– 2.0 unit increase of pH in the rhizoplane of millet, cowpea and sorghum led to large increases in nutrient availability (P, Ca and Mg), and decreases levels of exchangeable Al, close to the roots. They concluded that there was an important role of root-induced increase of pH for crops to cope with acidity-induced nutrient deficiency and Al stress. Tang et al. (2000) proposed that uptake of leached NO2 3 by plant roots may reduce subsoil acidification. Similarly, Noble et al. (1996) proposed that the production of Ca(NO3)2 through the addition of lime and NHþ 4 -based fertilisers, its subsequent leaching into the subsoil, and the preferential uptake of NO2 3 offered a feasible and practical method of ameliorating subsoil acidity in
ACIDIC SUBSURFACE LAYERS OF SOIL
211
sugarcane production. However, this procedure would result in net soil acidification as a result of nitrification of the applied NHþ 4 , and it is unlikely that NO2 3 would be leached to only the acidic subsurface layers and no further. Although heavy topdressing with lime may ameliorate subsurface acidification by allowing movement of excess alkali, high rates may have adverse effects on lupin crops in the rotation (McLay et al., 1994). Scott et al. (1999) concluded that the best strategy was to apply large quantities of lime to the surface 10 cm of soil and to grow acidity tolerant cultivars while waiting for the effect of amendment to move down the soil profile. Smith et al. (1994) investigated the depth to which materials more soluble than lime may ameliorate subsurface soil acidity. They noted that after 18 months, surface application of gypsum and phosphogypsum (at 2.5 t ha21) had no effect on soil pHCa at 0– 5 cm depth, and gypsum decreased pHCa at 5– 15 cm depth. Consequently, to date there remains no effective economic alternative to lime in the amelioration of acidic subsurface soil layers in soils with a high cation exchange capacity. However, in soils with a very low cation exchange capacity, or an anion exchange capacity, gypsum may successfully increase the pH of the acidic subsurface layer (Farina and Channon, 1998). Earthworms can greatly enhance the burial of lime through their burrowing activities, probably by rain water washing lime down surface-venting burrows (Baker et al., 1999). At seven pasture sites that had subsurface layers (2.5 – 10 cm) with a pHCa 0.1– 0.8 units lower than the surface soil (0 –2.5 cm), introduction of three exotic lumbricid earthworms buried lime into the surface layer such that the pHCa at 2.5 – 5 and 5 –10 cm depth increased by up to 0.6 and 0.2 units, respectively (Baker et al., 1999). Perhaps the best means to decrease the extent of acidic subsurface layers is occasional surface soil mixing. In crop – pasture rotations, surface soil mixing may need to be conducted after a few years under productive legume-based pasture. This could be done prior to the commencement of a cropping phase and may be associated with the incorporation of lime. Surface soil mixing will improve plant nutrient availability in two main ways. First, plant root growth and nodulation may improve due to the disruption of the acidic subsurface soil layer. Secondly, added substrate from plant residue return will be mixed throughout the surface soil. Total surface soil net N mineralisation, and thus mineral N availability, will be enhanced when organic N substrate is not predominantly restricted to the frequently dry surface 2.5 cm of soil.
IX. CONCLUSIONS The relatively low pH within subsurface layers is largely attributable to acidification at 5– 10 cm depth due to legume root Hþ excretion, and the
212
K. I. PAUL ET AL.
alkalinisation of the surface 2.5 cm of soil due to plant residue return. Net reduction of Mn, urine excretion by grazing stock and the decline of pHBC through surface soil depth may also contribute to the development of acidic subsurface layers. The magnitude of pH decrease through depth within the surface soil will be particularly large where productive legume species are grown in undisturbed acidic soils which are exposed to moist – dry cycles. Although recent work has largely clarified the main processes leading to the development of acidic subsurface layers, it seems that species selection, encouraging lumbricid earthworm populations and occasional surface soil mixing may be the only feasible way to minimise the magnitude of decrease in pH though surface soil depth.
REFERENCES Arshad, M. A., Franzluebbers, A. J., and Gill, K. S. (1999). Soil Till. Res. 50, 47 –53. Bagayoko, M., Alvey, S., Neumann, G., and Buerkert, A. (2000). Plant Soil 225, 117 –127. Baker, G. H., Carter, P. J., and Barrett, V. J. (1999). Aust. J. Soil Res. 37, 831– 845. Barnes, B. T., and Ellis, F. B. (1979). J. Soil Sci. 30, 669–679. Barrow, N. J. (1960). Aust. J. Soil Res. 11, 331–338. Benson, M. L., Landsberg, J. J., and Borough, C. J. (1992). Forest Ecol. Mgmt 52, 1 –16. Bessho, T., and Bell, C. (1992). Plant Soil 140, 183–196. Birch, H. F. (1958). Plant and Soil. 10, 9–30. Black, A. S. (1992). Aust. J. Soil Res. 30, 989 –999. Blevins, R. L., Thomas, G. W., Smith, M. S., Frye, W. W., and Cornelius, P. L. (1983). Soil Till. Res. 3, 135– 145. Bolan, N. S., Hedley, M. J., and White, R. E. (1991). Plant Soil 134, 53–63. Bowman, R. A., and Halvorson, A. D. (1998). Soil Sci. 163, 241–247. Bromfield, S. M., Cumming, R. W., David, D. J., and Williams, C. H. (1983). Aust. J. Expl. Agric. 23, 181–191. Bromfield, S. M., Cumming, R. W., David, D. J., and Williams, C. H. (1983). Aust. J. Expp. Agric. Animal Husb. 23, 192–200. Burle, M. L., Mielniczuk, J., and Focchi, S. (1997). Plant Soil 190, 309–316. Campbell, C. A., and Biederbeck, V. O. (1976). Can. J. Soil Sci. 56, 293– 310. Chan, K. Y., and Heenan, D. P. (1993). Aust. J. Agric. Res. 44, 1971–1984. Chan, K. Y., and Heenan, D. P. (1996). Aust. J. Expl. Agric. 36, 539–543. Chan, K. Y., Roberts, W. P., and Heenan, D. P. (1992). Aust. J. Soil Res. 30, 71 –83. Condon, J. R., Black, A. S., and Conyers, M. K. (2000). In “Soil 2000: New Horizons for a New Century.” (J. A. Adams and A. K. Metherell Eds.), pp. 63 –65. Lincoln University, New Zealand. Conyers, M. K., and Scott, B. J. (1989). Aust. J. Expp. Agric. 29, 201–207. Conyers, M. K., Heenan, D. P., Poile, G. J., Cullis, B. R., and Helyar, K. R. (1996). Soil Till. Res. 37, 127–141. Conyers, M. K., Uren, N. C., Helyar, K. R., Poile, G. J., and Cullis, B. R. (1997). Aust. J. Soil Res. 35, 1115–1129. Cornish, P. S. (1987). Aust. J. Agric. Res. 38, 775–789. Cornish, P. S., and Lymbery, J. R. (1987). Aust. J. Expl. Agric. 27, 869– 880. Coventry, D. R., and Slattery, W. J. (1991). Aust. J. Agric. Res. 42, 391–397. Cregan, P. D (1992). Growth of pasture legumes on acid soils. MAgrSc Thesis, The University of Melbourne.
ACIDIC SUBSURFACE LAYERS OF SOIL
213
Crozier, C. R., Naderman, G. C., Tuckers, M. R., and Sugg, R. E. (1999). Commun. Soil Sci. Plant Anal. 30, 65–74. De Maria, I. C., Nnabude, P. C., and de Castro, O. M. (1999). Soil Till. Res. 51, 71 –79. Dick, W. A. (1983). Soil Sci. Soc. Am. J. 47, 102–107. Dolling, P. J. (1995). Aust. J. Expp. Agric. 35, 753–763. Dolling, P. J., and Porter, W. M. (1994). Aust. J. Expp. Agric. 34, 1155–1164. Dolling, P. J., Porter, W. M., and Robson, A. D. (1991). Aust. J. Expp. Agric. 31, 803–810. Dolling, P. J., Porter, W. M., and Rowland, I. C. (1994). Aust. J. Expp. Agric. 34, 1165–1172. Doube, B. M., Buckerfield, J. C., and Kirkegaard, J. A. (1994). Aust. J. Agric. Res. 45, 1587–1600. Ellington, A. (1988). Nitrogen Cycling in Temperate Agricultural Systems. (P. E. Bacon, J. Evans, R. R. Storrier and A. C. Taylor Eds.), vol. 1, pp. 58–75. Australian Society of Soil Science, Riverina. Engels, C., Mollenkopf, M., and Marschner, H. (1994). Z. Pflanzenernahr. Bodenk 157, 139 –144. Evans, C. M., Conyers, M. K., Black, A. S., and Poile, G. J. (1998). Aust. J. Soil Res. 36, 641–653. Farina, M. P. W., and Channon, P. (1998). Soil Sci. Soc. Am. J. 52, 175 –180. Ferrufino, A., Smyth, T. J., Israel, D. W., and Carter, T. E. (2000). Crop Sci. 40, 413 –421. Fettell, N. A., and Gill, H. S. (1995). Aust. J. Expp. Agric. 35, 923– 928. Grant, C. A., and Bailey, L. D. (1994). Can. J. Soil Sci. 74, 307 –314. Groffman, P. M. (1984). Soil Sci. Soc. Am. J. 49, 329–334. Hafner, H., George, E., Bationo, A., and Marschner, H. (1993). Plant Soil 150, 117–127. Haynes, R. J., and Knight, T. L. (1989). Soil Till. Res. 14, 197 –208. Hoyt, P. B., and Turner, R. C. (1975). Soil Sci. 119, 227 –237. Hue, N. V., and Amien, I. (1989). Commun. Soil Plant Anal. 20, 1499– 1511. Hussain, I., Olson, K. R., and Ebelhar, S. A. (1999). Soil Sci. Soc. Am. J. 63, 1335– 1341. Ismail, I., Blevins, R. L., and Frye, W. W. (1994). Soil Sci. Soc. Am. J. 58, 193–198. Israel, D. W., and Jackson, W. A. (1978). In “Mineral Nutrition of Legumes in Tropical and Sub-tropical Soils”. (C. S. Andrew and E. J. Kamprath Eds.), pp. 113–129. CSRIO Australia, Melbourne. Jacobsen, J. S., and Westerman, R. L. (1991). Commun. Soil Sci. Plant Anal. 22, 1335– 1346. Jager, G., and Bruins, E. H. (1975). Soil Biol. Biochem. 7, 153– 159. Jarvis, S. C., and Robson, A. D. (1983). Aust. J. Agric. Res. 34, 341–353. Jenkinson, D. S., and Ayanaba, A. (1977). Soil Sci. Soc. Am. J. 41, 912–915. Keith, H., Jacobsen, K. L., and Raison, R. J. (1997). Plant Soil 190, 127– 141. Loss, S. P., Ritchie, G. S. P., and Robson, A. D. (1993). Aust. J. Expp. Agric. 33, 457–464. Mahler, R. L., and Harder, R. W. (1984). Soil Sci. 137, 52 –60. Matowo, P. R., Pierzynski, G. M., Whitney, D., and Lamond, R. E. (1999). Soil Till. Res. 50, 11–19. McLaughlin, M. J., Baker, T. G., James, T. R., and Rundle, J. A. (1990). Aust. J. Soil Res. 28, 371 –385. McLay, C. D. A., Ritchie, G. S. P., and Porter, W. M. (1994, 1997). Aust. J. Soil Res. 32, 846–935. McLay, C. D. A., Barton, L., and Tang, C. (1997). Aust. J. Agric. Res. 48, 1025–1032. Noble, A. D., Zenneck, I., and Randall, P. J. (1996). Plant Soil 179, 293–303. Nye, P. H. (1981). Plant Soil 61, 7– 26. Nye, P. H. (1987). Adv. Plant Nutr. 2, 129 –153. O’Connell, A. M., and Rance, S. J. (1999). Soil Biol. Biochem. 31, 1943–1951. Paul, K. I., Black, A. S., and Conyers, M. K. (1999). Aust. J. Soil Res. 37, 1057–1072. Paul, K. I., Black, A. S., and Conyers, M. K. (2001). Biol. Fertil. Soils 33, 41 –52. Paul, K. I., Black, A. S., and Conyers, M. K. (2001). Biol. Fertil. Soils 33, 75 –82. Paul, K. I., Black, A. S., Conyers, M. K (2001). Plant Soil, 234, 239 –246. Pierce, F. J., Fortin, M. C., and Staton, M. J. (1994). Soil Sci. Soc. Am. J. 58, 1782–1787. Pierre, W. H., and Banwart, W. L. (1973). Agron. J. 65, 91–96. Pinkerton, A., and Simpson, J. R. (1986). Aust. J. Expp. Agric. 26, 107 –113. Pinkerton, A., and Simpson, J. R. (1986). Aust. J. Expp. Agric. 26, 681 –689.
214
K. I. PAUL ET AL.
Pikul, J. L., and Aase, J. K. (1995). Agron. J. 87, 656–662. Poss, R., Smith, C. J., Dunin, F. X., and Angus, J. F. (1995). Plant Soil 177, 85 –100. Purnomo, E., and Black, A. S. (1994). Fertil. Res. 39, 77 –82. Purnomo, E., Black, A. S., Smith, C. J., and Conyers, M. K. (2000). Aust. J. Soil Res. 38, 129 –140. Purnomo, E., Black, A. S., and Conyers, M. K. (2000). Aust. J. Soil Res. 38, 643–652. Richardson, A. E., and Simpson, R. J. (1988). Soil Biol. Biochem. 20, 431–438. Richardson, A. E., James, G. S., and Simpson, R. J. (1985). In “Proceedings of the 3rd Australian Agronomy Conference”. (J. J. Yates Ed.), . Australian Society of Agronomy, Parkville, Vic. Richardson, A. E., Henderson, G. S., and Simpson, R. J. (1988). Soil Biol. Biochem. 20, 439–445. Ridley, A. M., Helyar, K. R., and Cowling, A. (1990). Aust. J. Expp. Agric. 30, 529–537. Ritchie, G. S. P., and Dolling, S. E. (1985). Aust. J. Soil Res. 23, 569–576. Rovira, A. D., Smettem, K. R. J., and Lee, K. E. (1987). Aust. J. Agric. Res. 38, 829–834. Scott, B. J., Conyers, M. K., Polie, G. J., and Cullis, B. R. (1997). Aust. J. Agric. Res. 48, 834– 854. Scott, B. J., Conyers, M. K., Polie, G. J., and Cullis, B. R. (1999). Aust. J. Expp. Agric. 39, 849 –856. Slattery, W. J., Ridley, A. M., and Windsor, S. M. (1991). Aust. J. Expp. Agric. 31, 321–324. Slattery, W. J., Edwards, D. G., Bell, L. C., Coventry, D. R., and Helyar, K. R. (1998). Aust. J. Soil Res. 36, 273 –290. Smiley, R. W. (1974). Soil Sci. Soc. Am. Proc. 38, 795–799. Smith, C. J., Peoples, M. B., Keerthisinghe, G., James, T. R., and Garden, D. L. (1994). Aust. J. Soil Res. 32, 995 –1008. Springett, J. R. (1984). In “Ecological Interactions in Soil”. (A. H. Fitter, D. Atkinson, D. J. Read and M. B. Usher Eds.), pp. 339– 405. Blackwell Scientific Publications, Oxford. Staley, T. E., and Boyer, D. G. (1997). Soil Till. Res. 42, 115 –126. Tang, C. (1998). Plant Soil 199, 275 –282. Tang, C., and Yu, Q. (1999). Plant Soil 215, 29–38. Tang, C., McLay, C. D. A., and Barton, L. (1997). Aust. J. Expp. Agric. 37, 563–570. Tang, C., Barton, L., and Raphael, C. (1998). Aust. J. Agric. Res. 49, 53– 58. Tang, C., Fang, R. Y., and Raphael, C. (1998). Aust. J. Agric. Res. 49, 657 –664. Tang, C., Sparling, G. P., McLay, C. D. A., and Raphael, C. (1999). Aust. J. Soil Res. 37, 561– 573. Tang, C., Unkovich, M. J., and Bowden, J. W. (1999). New Phytol. 149, 513– 521. Tang, C., Raphael, C., Rengel, Z., and Bowden, J. W. (2000). Aust. J. Soil Res. 38, 837–849. Thompson, J. P. (1992). Soil Till. Res. 22, 339 –361. Thompson, C. A., and Whitney, D. A. (2000). Commun. Soil Sci. Plant Anal. 31, 241 –257. Van Miegroet, H., Hysell, M. T., and Johnson, A. D. (2000). Soil Sci. Soc. Am. J. 64, 1515–1525. van Veen, J. A., Ladd, J. N., and Amato, M. (1985). Soil Biol. Biochem. 17, 747 –756. van Gestel, M., Ladd, J. N., and Amato, M. (1991). Soil Biol. Biochem. 23, 313 –322. Von Uexkull, H. R., and Mutert, E. (1995). In “Plant and Soil Interactions at Low pH”. (R. A. Date, N. J. Grundon, G. E. Rayment and M. E. Probert Eds.), pp. 5–19. Kluwer Academic Publishers, Netherlands. Williams, C. H. (1980). Aust. J. Expp. Agric. Animal Husb. 20, 561–567. Yan, F., Schubert, S., and Mengel, K. (1996). Soil Biol. Biochem. 28, 617–624. Yan, F., Schubert, S., and Mengel, K. (1996). Biol. Fertil. Soils 23, 236– 242. Young, S. R., Black, A. S., and Conyers, M. K. (1995). In “Plant and Soil Interactions at Low pH”. (R. A. Date, N. J. Grundon, G. E. Rayment and M. E. Probert Eds.), pp. 111–115. Kluwer Academic Publishers, Netherlands.
SOIL ACIDIFICATION AND LIMING INTERACTIONS WITH NUTRIENT AND HEAVY METAL TRANSFORMATION AND BIOAVAILABILITY Nanthi S. Bolan,1 Domy C. Adriano2 and Denis Curtin3 1
Fertilizer and Lime Research Centre, Massey University, Palmerston North, New Zealand 2 University of Georgia, Savannah River Ecology Laboratory, Drawer E, Aiken, South Carolina 29802, USA 3 Crop and Food, Christchurch, New Zealand
I. Introduction II. Processes of Acid Generation in Soils A. Natural Ecosystems B. Managed Ecosystems III. Effect of Soil Acidity on Nutrient and Heavy Metal Transformation in Soils A. Plant Nutrients B. Heavy Metals IV. Amelioration of Soil Acidity Through Liming A. Liming Materials B. Effects of Liming V. Lime, Nutrient and Heavy Metal Interactions A. Plant Nutrients B. Heavy Metals VI. Conclusions and Future Research Needs References
“No other single chemical soil characteristic is more important in determining the chemical environment of higher plants and soil microbes than the pH. There are few reactions involving any component of the soil or of its biological inhabitants that are not sensitive to soil pH. This sensitivity must be recognized in any soil-management system.” “Lime is truly a foundation for much of modern humid-region agriculture. Knowing how pH is controlled, how it influences the supply and availability of essential plant nutrients as well as toxic elements, how it affects higher plants and human beings, and how it can be ameliorated is essential for the conservation and sustainable management of soils throughout the world.” (Brady and Weil, 1999) 215 Advances in Agronomy, Volume 78 Copyright q 2003 by Academic Press. All rights of reproduction in any form reserved 0065-2113/02$35.00
216
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN Under areas where rainfall exceeds evapotranspiration, soil acidification is an ongoing natural process, which can either be accelerated by the activity of plants, animals and humans or can be impeded by careful management practices. In areas affected by industrial activities, soil acidification is caused by acid drainage from pyrite oxidation and also from acid precipitation. In areas that remain unaffected by industrial pollution, soil acidification in managed ecosystems is mainly caused by the release of protons (Hþ) during the transformation and cycling of carbon (C), nitrogen (N) and sulfur (S). Just like in managed ecosystems, soil acidification in natural ecosystems caused by acid drainage and acid precipitation can have adverse impacts where soils have low pH buffering capacity. Liming is the most common management practice aimed at neutralizing the acid produced, thereby overcoming the adverse impacts of soil acidification. This review brings together fundamental aspects of soil acidification and recent developments on the implications of liming in relation to soil processes, particularly nutrient and heavy metal transformation and bioavailability in soils. The article first outlines the various soil, plant and microbial processes that generate acid (protons; Hþ ions) both under natural and managed ecosystems. It then discusses the effects of soil acidity on soil chemical and biological properties. The effect of liming to overcome the problems associated with soil acidity is examined in relation to the transformation of nutrient ions and heavy metals. The practical implications of liming to overcome heavy metal toxicity have been discussed in relation to the adsorption, leaching and phytoavailability of these metal ions. Future research should aim to focus on the development of methods to quantify lime-enhanced (im)mobilization of nutrient ions and heavy metals in soils and to explore further the role of liming in remediating contaminated soils. q 2003 Academic Press.
I. INTRODUCTION Soil acidification is a natural process, which can either be accelerated by the activity of plants, animals and humans or can be impeded by sound management practices. Industrial and mining activzities can lead to soil acidification through acid drains resulting primarily from pyrite oxidation and from acid precipitation caused by the emission of sulfur (SOx) and nitrogen (NOx) gases. In areas that remain unaffected by industrial pollution, soil acidification is mainly caused by the release of Hþ ions during the transformation and cycling of C, N and S in managed ecosystems. Soil acidification can exert itself in several ways: (a) increase of soil acidity or decrease of pH, (b) decrease of base saturation, (c) unbalanced availability of elements in the root environment, or (d) decrease of the acid neutralizing capacity (ANC) of the soil (van Breemen, 1991). Soil acidification caused by these processes can have adverse impacts where soils are unable to buffer against pH decrease. For example, in parts of North America and Europe, soil acidification caused by acid precipitation has resulted in forest
SOIL ACIDIFICATION AND LIMING INTERACTIONS
217
decline (Binns, 1988; Longhurst, 1991) and in some parts of Australia, continuous legume cultivation and inappropriate N fertilizer use have generated sufficient soil acidity that wheat cultivation has had to be abandoned due to aluminium (Al) and manganese (Mn) toxicity (Dolling and Porter, 1994; Mason et al., 1994). Soil acidity causes detrimental effects to both plants and soil organisms (Robson and Abbott, 1989; Runge and Rode, 1991). In excessively acid soils (pH , 4.0) many plants do not grow well. The activities of soil organisms generally are reduced, resulting in the inhibition of biological N fixation by legumes and decomposition of organic matter. In acid soils, the concentrations of Al and Mn become toxic to plant growth. Likewise soil acidification enhances the mobilization of metals in soils, resulting in increased uptake by plants. Some of these toxic metals subsequently reach the food chain through plant products and grazing animals. Phosphorus (P) and molybdenum (Mo) may become insoluble and unavailable, and low pH may indicate the deficiency of basic essential cations, such as calcium (Ca) and magnesium (Mg) in soils (Ritchie, 1989; Sumner et al., 1991). Historically, liming is the most common management practice used to neutralize the acid produced in the soil and to overcome the problems associated with soil acidification. Most plants grow well at a pH range of 5.5 –6.5 and liming is aimed to maintain the pH at this range. Liming enhances the physical, chemical and biological properties of soil through its direct effect on the amelioration of soil acidity and its indirect effect on the mobilization of plant nutrients, immobilization of toxic heavy metals, and improvements in soil structure and hydraulic conductivity (Ks). In variable charge soils, liming can be used as a management tool to manipulate the surface charge, thereby controlling the reactions of nutrient ions and heavy metal cations. Liming has been shown to provide optimum conditions for a suite of biological activities that include N fixation and mineralization of N, P and S in soils (Haynes and Naidu, 1998; Bolan et al., 1999a). Although the effect of acidification on chemical reactions of soils in relation to plant growth has been reviewed by a number of workers (e.g., Reuss, 1986; Sumner et al., 1991), no comprehensive review on the effect of soil acidification in relations to environmental contamination has been presented. Similarly, the effect of liming has often been examined in relation to overcoming nutrient deficiencies and metal toxicity (primarily Al and Mn) for optimum plant growth. For example, Ritchie (1989) reviewed the reactions of Al and Mn, which are considered to be the major growth liming factors resulting from soil acidification. Similarly, the interaction of lime and P has been reviewed by Haynes (1984). Recently liming materials and stabilized alkaline by-products are increasingly being used to immobilize heavy metals in soils receiving industrial and domestic waste sludges (Dick et al., 2000; Adriano, 2001). However, no comprehensive review on liming interactions with the transformation and (im)mobilization of nutrient ions and heavy metals in
218
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
soils in relation to sustainable production and environmental protection has been published. Liming is increasingly being viewed as a management tool in enhancing the natural attenuation of nutrient and heavy metal contaminants in soils (Adriano, 2001). The present review aims to integrate fundamental aspects of soil acidification and recent developments on the implications of liming in relation to biogeochemical processes, particularly its interactions with nutrients and heavy metals in soils. The review first discusses the various processes that generate acid (protons; Hþ ions) from acid precipitation and pyrite oxidation, resulting from industrial activities, and during the cycling of C, N and S in managed farming systems. The detrimental effects of soil acidity on plant growth and microbial functions, and the beneficial effects of liming to overcome the problems associated with soil acidity are examined in relation to the transformation and bioavailability of nutrients and heavy metals. More specifically, the development of methods to quantify lime-enhanced mobilization of nutrient ions and immobilization of heavy metals in soils and to explore the role of liming in remediating contaminated soils are discussed further.
II. PROCESSES OF ACID GENERATION IN SOILS Understanding the processes involved in acid generation in soils and determining the underpinning causes are precursors to the implementation of preventive measures to reduce acid input to soils. These processes can be broadly grouped into two categories: (i) those occurring under natural ecosystems through industrial activities, and (ii) those occurring under managed ecosystems through farming activities. The various reactions involved in these processes are given in Table I.
A. NATURAL ECOSYSTEMS The two most important acid generating processes under natural ecosystems resulting from industrial activities occur with acid drainage through pyrite oxidation, and acid precipitation (Longhurst, 1991; Evangelou, 1995). While the first process occurs at local level, the second process leads to acidification of sites further away from the origin of acid precipitation and hence warrants international collaborative effort to overcome the problems associated with this process (Reuss, 1986).
SOIL ACIDIFICATION AND LIMING INTERACTIONS
219
Table I Proton Generation and Consumption Processes in Acid Precipitation, Pyrite Oxidation and C, N and S Biogeochemical Cycles
Process
Reaction equation
Hþ (molc mol21)
Eq. No
Acid precipitation Oxidation of sulfur dioxide
2SO2 þ O2 ! 2SO3
Hydrolyis of sulfur trioxide
SO3 þ H2 O ! H2 SO4 ! SO4 þ 2Hþ
Photochemicial oxidation of nitric oxide Hydrolyis of nitrogen dioxide
O3 þ NO ! N2 Oþ O2
2NO2 þ H2 O ! HNO3 þ HNO2 ! NO3 þ Hþ
0
1
þ2
2
0
3
þ1
4
Pyrite oxidation Pyrite oxidation by oxygen
þ 2FeS2 þ 7O2 þ 2H2 O ! 2Fe2þ þ 4SO2þ 4 þ 4H
þ2
5
Ferrous iron oxidation
4Fe2þ þ O2 þ 4Hþ ! 4Fe3þ þ 2H2 O
21
6
Ferric iron precipitation
Fe3þ þ 3H2 O ! FeðOHÞ3 þ 3Hþ
þ3
7
Pyrite oxidation by ferric iron
þ FeS2 þ 14Fe3þ þ H2 O ! 15Fe2þ þ 2SO2þ 4 þ 16H
8
Carbon cycle Dissolution of carbon dioxide
CO2þ H2 O ! H2 CO3 ! Hþ þ HCO2 3
þ1
9
Synthesis of organic acid
Organic C ! RCOOH ! RCOO2 þ Hþ
þ1
10
N fixation
2N2 þ 2H2 O þ 4R·OH ! 4R·NH2 þ 3O2
0
11
Mineralization of organic N
RNH2 þ Hþ þ H2 O ! R·OH þ NHþ 4
21
12
2 ðNH2 Þ2 CO þ 3H2 O ! 2NHþ 4 þ 2OH þ CO2
21
13
þ NHþ 4 þ R·OH ! R·NH2 þ H2 O þ H
þ1
14
2 NHþ 4 þ OH ! NH3 " þH2 O
þ1
15
Nitrogen cycle
Urea hydrolysis Ammonium assimilation Ammonia volatilization
(continued on next page)
220
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
Table I (continued)
Process
Reaction equation
Eq. No
þ2
16
Nitrification
NHþ 4
Nitrate assimilation
þ 2 2 NO2 3 þ 8H þ 8e ! NH3 þ 2H2 O þ OH
21
17
Denitrification
þ 4NO2 3 þ 4H ! 2N2 þ 5O2 þ 2H2 O
21
18
þ 2O2 !
NO2 3
þ
Hþ (molc mol21)
þ H2 O þ 2H
Sulfur cycle Mineralization of organic S
þ 2Organic S þ 3O2 þ 2H2 O ! 2SO22 4 þ 4H
þ2
19
Assimilation of sulfate
þ 2 2 SO22 4 þ 8H þ 8e ! SH2 þ 2H2 O þ 2OH
22
20
Oxidation of S0
þ 2S0 þ 2H2 O þ 3O2 ! 2SO22 4 þ 4H
þ2
21
1.
ACID DRAINAGE
Acid drainage has various anthropogenic and natural sources, but the most extensive and widely known source originates from mining coal and various sulfide-rich metal ores, including copper, gold, lead and silver. Other human activities that contribute to acid drainage include various forms of land disturbances, such as industrial and residential development and farming (e.g., rice cultivation). Generally, strong acid forming processes in nature involve exposure of metal-sulfides enriched with heavy metals or metalloids to atmospheric air, which leads to oxidation and acidification of water bodies and their subsequent enrichment with total dissolved solids, including metals and metalloids. Other contributors to acid drainage production include mineral processing, manufacturing or recycling of batteries, electronics, wood pulp, paper and heavy steel industries, such as manufacturing of cars or heavy equipment, tanneries and textile manufacturing, food processing, and waste disposal/management industries. Oxidation of pyrite is considered as the major source of acid drainage prompting intensive efforts to minimize the production of pyrite –borne acid drainage and mitigation measures to alleviate its environmental impacts (Evangelou and Zhang, 1995; Frazer, 2001). Pyrite is a mineral commonly associated with coal and various metal ores as well as mine deltas, wetlands, and rice fields. Often, pyrite becomes exposed to the atmosphere through various human activities including, mining, land development for agricultural purposes, and construction of highways, tunnels,
SOIL ACIDIFICATION AND LIMING INTERACTIONS
221
airports and dams. Pyrite exposure to the atmosphere leads to its oxidation and the subsequent production of extremely acidic drainage typically enriched with iron (Fe), Mn, Al and sulfate (SO22 4 ) and other heavy metals. The acidic drainage caused by pyrite oxidation not only acts as a source of heavy metals associated with the parent minerals but also serves as an agent for the solubilization and the transport of heavy metals in the soils through which it drains (Evangelou, 1995; Mays and Edwards, 2001). Pyrite oxidation is mediated by biological and electrochemical reactions, some of which are given in Table I. Reactions (1) and (3) indicate that Fe3þ and O2 are the major sources of electron acceptor for pyrite oxidation. At low pH (, 4.5), Fe3þ oxidizes pyrite much more rapidly than does O2 and more rapidly than O2 oxidizing Fe2þ to Fe3þ (Evangelou and Zhang, 1995). For this reason, Reaction (6) is considered to be the rate-limiting step in pyrite oxidation. Reaction (7) is a readily reversible dissolution/precipitation reaction taking place at pH as low as 3, and is a major step in the release of acid to the environment. Abiotic oxidation of Fe2þ with O2 is pH sensitive and is extremely slow in very acidic solution. The mechanisms of pyrite oxidation mediated by microorganisms are grouped into (a) direct metabolic reactions, and (b) indirect metabolic reactions (Evangelou and Zhang, 1995). Direct metabolic reactions require physical contact between the bacteria and pyrite particles, while indirect metabolic reactions do not require such physical contact. During the indirect metabolic reactions, the bacteria typified by Thiobacillus ferrooxidans and T. thiooxidans oxidize Fe2þ, thereby regenerating the Fe3þ required for the chemical oxidation of pyrite. While T. ferrooixdans is active in the oxidation of both Fe2þ and sulfides, T. thiooxidans is involved only in the oxidation of sulfides. In the presence of T. ferrooxidans, Fe2þ oxidation is very rapid in acidic conditions. At neutral to alkaline pH, there is minimal bacterial involvement in pyrite oxidation and in such environment O2 becomes a more important pyrite oxidant than Fe3þ. Under most conditions, however Fe3þ plays a major role in pyrite oxidation and O2 is involved in the oxidation of Fe2þ (Evangelou and Zhang, 1995; Benner et al., 2000; Frazer, 2001).
2. ACID PRECIPITATION Carbon dioxide combines with water in the air to form a dilute solution of carbonic acid (H2CO3) at about pH 5.6. For this reason, acid precipitation is arbitrarily defined as precipitation with a pH below 5.6. The acidity of rain, snow and atmospheric particulates that fall upon much of the world appears to have increased significantly over the last four decades. In addition to natural sources of acid precipitation that results from geological weathering, volcanic eruption,
222
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
anaerobic decomposition of organic matter, air-borne sea salt sprays, and lightning, most of the increased acid precipitation burden has been attributed to consumption of fossil fuels, especially coal. Major anthropogenic sources include combustion of fossil fuels, smelting of ores, exhausts from internal combustion engines, and N fertilization of agricultural and forest lands (Binns, 1988; Longhurst, 1991). Widespread occurrence of acid precipitation and dry deposition results in large part from industrial emissions of SOx and NOx (Longhurst, 1991). These gases are transformed in the atmosphere to sulfuric and nitric acids (Table I), transported over long distances and deposited on vegetation, soils, surface water, and building materials. While majority of the NOx emissions are local/ natural origin, SOx emissions are often transboundary in nature (Table II). The average annual ratio of sulfuric acid to nitric acid is about 2:1 in North America, but nitric acid is becoming progressively more important because of the installation of flue gas desulfurization (FGD) systems in coal-fired power stations (Dick et al., 2000). Acidification of lakes and streams in North America and Europe has altered their trace metal chemistry by (Adriano, 2001): (i) increasing total metal concentrations; (ii) shifting the speciation of dissolved metals toward free aqueous ions, typically the species most toxic to aquatic biota; and (iii) reducing particulate metal concentration in favor of higher dissolved levels. The impact of acidification on terrestrial ecosystems is less straightforward; forest health may be one casualty, but that could also be affected by increasing tropospheric ozone. Where
Table II Annual Emission of Sulfur Dioxide and Nitrogen Oxide from Various Countries Sulfur dioxide deposition (%)
Emissions (103 tonnes) Country USSR USA China Poland Germany Canada United Kingdom Czechoslovakia France Finland Sweden Norway
Sulfur dioxide
Nitrogen oxide
Foreign
Domestic
24 000 20 800 18 000 4300 6400 3727 3540 3040 1845 370 272 100
2930 19 400 4130 840 2900 1785 2900 1100 1693 250 305 215
32
53
2 52 45 50 12 56 32 55 58 63
90 42 48 50 79 37 54 26 18 8
SOIL ACIDIFICATION AND LIMING INTERACTIONS
223
however, acidic emissions have been reduced and where mitigating measures have been employed there is some evidence that affected ecosystems can recover (Longhurst, 1991).
B. MANAGED ECOSYSTEMS The most significant Hþ and hydroxyl ion (OH2) generating processes occur during the biogeochemical cycling of C, N and S (Table I). Although these processes occur both under natural and managed ecosystems, under the latter system, these processes are accelerated by the activities of humans through intensive land-based crop and animal production. In the case of the C cycle, dissolution of CO2 to form H2CO3 in soil solution and synthesis and dissociation of carboxylic acids produced by plants and microorganisms are the two main sources of Hþ ions. The assimilation of CO2 into carboxylic acids (including amino and fatty acids) in higher plants indirectly acidifies the soil explored by their roots. In the case of the N and S cycles, mineralization and oxidation of organic N and S result in the production of Hþ ions. However, this will be balanced by OH2 22 generated through uptake and assimilation of NO2 3 –N and SO4 – S by plant and 22 2 microorganisms. Leaching of SO4 and NO3 with a charge-balancing basic cation (Ca, Mg, potassium (K), or sodium (Na)) rather than the Hþ ions generated during oxidation results in permanent acidity remaining in the soil. This will be reflected in a decrease in pH in soils with low pH buffering capacity. The processes involved in the generation of Hþ and OH2 ions during C, N and S cycling in soils can be grouped into two main categories: plant-induced—the uptake and assimilation of C, N and S; and soil-induced—the transformation of C, N and S in soils. These processes can have a bearing on the extent of bioavailability of certain plant nutrients, and the natural attenuation of heavy metals in the root zone (Adriano, 2001). In managed ecosystems used for agricultural production, regular fertilizer use is one of the major contributors of soil acidification.
1.
PLANT-INDUCED PROCESSES
CARBON ASSIMILATION. In higher plants, C is first assimilated as carbohydrates during the photosynthetic process. The subsequent metabolism of the photosynthates results in the synthesis of organic acids, such as malic and oxalic acids. At the cytoplasmic pH of the plants (pH 7.2– 7.4), some of the carboxyl groups of simple acids, amino acids, proteins and more complex structural carbohydrates (e.g., pectin) dissociate to produce Hþ ions (Raven,
224
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
1985). The excess Hþ ions are disposed of by neutralization resulting from decarboxylation, by transport into the vacuole, or by transport via the phloem into the roots and thence into the soil solution. Excretion of Hþ into the surrounding aqueous medium is the usual means of pH regulation in aquatic plants. In the case of terrestrial plants, some species counteract the change in cytoplasm pH by excreting Hþ ions into the soil solution, at the same time taking in a basic nutrient cation to balance the charge (Felle, 1988). UPTAKE AND ASSIMILATION OF NITROGEN. Plants utilize N in three main þ forms—as an anion (nitrate, NO2 3 ), as a cation (ammonium, NH4 ), or as a neutral N2 molecule (N2 fixation) (Marschner, 1995). Depending upon the form of N taken up and the mechanism of assimilation in the plant, excesses of cation or anion uptake may occur. To maintain charge balance during the uptake process, Hþ, OH2 or bicarbonate (HCO2 3 ) ions must pass out of the root into the surrounding soil. The Hþ ions may be derived from the dissociation of organic acids within the cell, and OH2 and HCO2 3 ions from the decarboxylation of organic acid anions. It has been shown that while the uptake of NHþ 4 and N2 fixation results in a net release of Hþ ions, uptake of NO2 3 can result in a net release of OH2 ions (Haynes, 1990). (i) Nitrogen fixation. In the case of N2 fixation, the neutral N2 can be assimilated into protein and no charge imbalance is generated across the soil/root interface. Many legumes, however, commonly export Hþ ions into their rhizosphere when actively fixing N2 (Haynes, 1983; Liu et al., 1989). Part of the Hþ ions generated within the legume root comes from the dissociation of the carboxyl groups of amino acids. The acidity generated by legume fixation of N2 has been found to be equivalent to the excess uptake of cations over anions by the plant and to vary from 0.2 to 0.7 mol Hþ mol21 of fixed N (Helyar, 1976; Bolan et al., 1991; de Klein et al., 1997). The amount of Hþ ions released during N2 fixation is a function of C assimilation and therefore depends mainly on the form and amount of amino acids and organic acids synthesized within the plant (Raven, 1985). Some tropical legumes apparently do not acidify their rhizosphere as much as do temperate legumes when actively fixing N2 (Israel and Jackson, 1978; Tang et al., 1997). Part of the reason for this is that their N assimilation products appear to be ureides (allantoin and allantoic acid) that have high pKa values (e.g., allantoin pKa 8.96) and are therefore unlikely to dissociate and donate Hþ ions under cytoplasmic and xylem pHs. Thus many tropical legumes accumulate less cations than do temperate legumes (Andrew and Johnson, 1976). The ability of N fixing plants to accumulate metal cations can be considered as a means of enhancing phytoremediation of metal-contaminated soils. (ii) Ammonium and nitrate assimilation. When NHþ 4 assimilation occurs in þ roots, deprotonation of NHþ releases one mole of H per mole of NHþ 4 4 (Eq. þ (14)). Additional small amounts of H ions are generated during the assimilation
SOIL ACIDIFICATION AND LIMING INTERACTIONS
225
of the deprotonation product, R·NH2 into amino acids and proteins which have isoelectric pHs lower than the cytoplasmic pH (Raven, 1985). When plants take 2 þ up N in the form of NO2 3 ion, the NO3 ion is first reduced to NH4 , which is 2 subsequently assimilated into amino acids. When NO3 is reduced in roots, for þ 2 every mole of NO2 3 reduced to NH4 one mole of OH ion is produced (Eq. (17)). þ When NH4 is assimilated into amino acids, small amounts of Hþ ions are produced through the dissociation of the carboxyl groups of amino acids. The net excess of OH2 ions can either be excreted into the rooting medium or can be neutralized by Hþ ions produced during the dissociation of organic acid (Raven, 1985). UPTAKE AND ASSIMILATION OF SULFUR. Sulfate is assimilated into Scontaining amino acids (cysteine, cystine and methionine) in the form of sulfydryl (–SH) group (Saggar et al., 1998). This reduction process is similar to 2 for each mole of S NO2 3 assimilation and produces two net moles of OH assimilated (Eq. (20)). Upon deprotonation of sulfydryl-containing amino acids, two Hþ ions are generated for each mole of – SH oxidized to SO22 4 . Since plants require 10 times less S than N, assimilation of SO22 has only a small effect on Hþ 4 balance in plants, and likewise deprotonation of S-containing proteins contributes little to Hþ generation in soils.
2.
SOIL-INDUCED PROCESSES
DECOMPOSITION OF ORGANIC MATTER. As microorganisms decompose soil organic matter, they respire CO2, which dissolves in water to form H2CO3 (Eq. (9)). The continuous production of CO2 through soil and root respiration increases the concentration of CO2 in the soil air space, so the extent of soil acidity from this source is considerably greater than that from CO2 dissolved in rainwater. However, acidic soil solutions at around pH 5 hold very little CO2. Thus, respiration is unlikely to cause soil pH to drop below 5. Soil microorganisms produce organic acids when decomposing plant litter that is rich in organic compounds but low in basic cations (Eq. (10)). A number of low-molecular-weight organic acids have been isolated from soils (Uren, 2001). Depending upon the nature of the plants growing in a particular soil, different amounts and diversity of organic acids are generated from the litter. In general, forest soils that have thick litter layer tend to be more acidic than grassland soils. Further, the litter from conifers tends to produce more organic acids when decomposed than the leaf fall from deciduous woodlands (Parfitt et al., 1997). TRANSFORMATION OF NITROGEN. Nitrification and ammonia (NH3) volatilization processes result in the release of Hþ ions. Both heterotrophic and
226
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
autotrophic microorganisms are involved first, in the conversion of organic forms þ 2 of N to NHþ 4 –N (ammonification), and subsequent oxidation of NH4 to NO3 (nitrification). While the ammonification process results in the release of OH2 ions, the nitrification process results in the release of Hþ ions. Combined ammonification (Eq. (12)) and nitrification (Eq. (16)) of organic N compounds, including urea, in theory generate one net mole of Hþ for every mole of N transformed. Ammonium ions in an alkaline medium dissociate into gaseous NH3, which is subject to volatilization loss (Eq. (15)). During NH3 volatilization, the pH of the soil decreases due to the consumption of OH2 ions (or release of Hþ ions) as NHþ 4 is converted to NH3. Ammonia volatilization occurs when the soil pH is high (. 7.5). When N is applied in the form of urea (e.g., urea fertilizer and animal excreta) the initial increase in soil pH through the ammonification process is likely to result in NH3 volatilization. TRANSFORMATION OF SULFUR. In aerobic soils, large proportion of S is present in organic form in roots, undecomposed litter and humified organic matter. Sulfur in soil organic matter and plant litter is mainly present as sulfydryl ( –SH) groups in proteins, nucleic acids and sulfolipids and bonded directly to C (Saggar et al., 1998). Protons are produced during the mineralization and subsequent oxidation of S in soil organic matter (Eq. (19)). As soil bacteria and fungi grow on plant litter and soil organic matter rich in C and poor in S, soil solution SO22 4 may be immobilized. In this case Eq. (19) in Table I is reversed is assimilated to microbial and becomes a Hþ-consuming reaction as SO22 4 protein. In periodic anaerobic conditions that occur following aerobic generation of SO22 4 , O2 concentration may be depleted by rapidly growing bacteria. Some as a terminal electron acceptor for bacteria have the capacity to use SO22 4 fermentation. The result is Hþ consumption as SO22 4 is reduced along a chain of intermediate compounds to H2S. It is common for H2S to react with metal ions to precipitate as metal sulfides (Cowling et al., 1992). This process is a Hþconsuming process. However, when these metal sulfides are reoxidized, they generate Hþ and acidify the soil (acid drainage). This sequence of reactions are common in soils used for lowland rice cultivation and results in a phytophillic neutral pH when waterlogged, but can create low phytotoxic pH when the soil is in the aerobic state (Li et al., 2001).
3.
FERTILIZER USE AND SOIL ACIDIFICATION
Fertilizer application in managed ecosystems used for agricultural production is a major contributor to soil acidification and it is important to understand
SOIL ACIDIFICATION AND LIMING INTERACTIONS
227
Table III Nutrient Content and Acidity Equivalent of Various Fertilizers Nutrient content (% w/w) Fertilizer
Chemical formula
N
P
K
S
Ammonium sulfate Ammonium chloride Ammonium nitrate Diammonium phosphate Monoammonium phosphate Urea Potassium nitrate Calcium nitrate Sodium nitrate Nitrogen fixation Single superphosphate Triple superphosphate North Carolina phosphate rock Calcium sulfate Potassium sulfate Elemental sulfur (S0)
(NH4)2SO4 NH4Cl NH4NO3 (NH4)2HPO4 NH4H2PO4 CONH2CO KNO3 Ca(NO3)2 NaNO3 – Ca(H2PO4)2. CaSO4.2H2O Ca(H2PO4)2 Ca10(CO3)x(PO4)62xF2 CaSO4.2H2O K2SO4 S0
21 26 33 18 11 46 14 14 16 – 0 0 0 0 0 0
0 0 0 20 21 0 0 0 0 – 10 18 13 0 0 0
0 0 0 0 0 0 39 0 0 – 0 0 0 0 50 0
24 0 0 0 0 0 0 0 0 – 12 1 0 18 18 100
Acidity a equivalent 110 93 60 74 55 79 223 250 229 70–250 8 15 250 257 264 310
a Acidity equivalent is the number of parts by weight of pure lime (calcium carbonate) required to neutralize the acidity caused by 100 parts of the fertilizer. Negative values indicate the liming value (kg CaCO3/100 kg) of the fertilizer.
the mechanisms involved in the acidifying effects of different fertilizers. The effects on soil acidification of various fertilizers used as sources of the major nutrients N, P, and S are discussed in the following section. The nutrient contents and the acidifying effects of the most common fertilizers used in agricultural production are presented in Table III. NITROGEN FERTILIZERS. Nitrogen is derived from fertilizer and manure applications and also from biological fixation of the atmospheric N by leguminous plants. Fixation of atmospheric N and the subsequent leaching of NO2 3 formed from the mineralization of the fixed N result in soil acidification, the extent of which depends on the amount and the fate of N fixed. It has been shown that in high producing pastures, approximately 400 –550 kg CaCO3 ha21 is required annually to neutralize the acidity produced by these processes (Bolan et al., 1991; de Klein et al., 1997). Application of N fertilizers, such as urea and ammonium sulfate to soils produces Hþ by two processes: nitrification (Eq. (16)) and NO2 3 leaching. Part of the Hþ produced is neutralized by OH2 released by the plants during
228
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
2 the subsequent uptake of the NO2 3 ions (Eq. (17)). The negatively charged NO3 ions carry positively charged basic cations (Ca, K, Mg and Na) in order to maintain the electric charge on the soil particles. The depletion of these basic cations during the leaching of NO2 3 ions accelerates the acidification process. In cases where ammonium sulfate is used the transport of both the SO22 4 and the NO2 3 ions in the soil causes greater depletion of basic cations. With urea, the 2 initial conversion of amide N to NHþ 4 –N (ammonification) releases OH ions þ (Eq. (13)) which neutralize part of the H produced during the subsequent 2 oxidation of NHþ 4 ions to NO3 ions, which explains why urea-based N fertilizers are less acidifying than the NHþ 4 -based fertilizers (Table III).
PHOSPHATE FERTILIZERS. Superphosphates are the most common phosphate fertilizers used and monocalcium phosphate (MCP) is the principal P component present in superphosphate fertilizers. Dissolution of MCP in soils results in the formation of dicalcium phosphate with a release of phosphoric acid close to the fertilizer granules. Phosphoric acid subsequently dissociates into phosphate þ þ 2 (H2PO2 4 ) and H ions. Part of H is subsequently neutralized by the OH ions 2 released during the adsorption of the H2PO4 ions by soil particles. Since the 2 H2PO2 4 ions are strongly adsorbed by most soils, H2PO4 -induced leaching of basic cations is unlikely to occur. In legume-based pasture and crop production systems, phosphate fertilizers are added mainly to promote the N fixation by the legumes by overcoming soil deficiency of P. It is important to point out that irrespective of P fertilizer source, application of P to legume-based systems promotes N fixation, thereby indirectly causing soil acidification. The amount of acidity produced indirectly by N fixation depends mainly on the extent of NO2 3 leaching and is higher than that produced directly by the dissolution of MCP in superphosphate fertilizer granules (Table III). In recent times, increasing amounts of phosphate rocks (PRs), such as North Carolina Phosphate rock (NCPR) are added directly to soils as a source of P. Unlike superphosphate fertilizers, PRs neutralize acids, thus adding some liming value. The liming action of PRs is discussed in Section IV. SULFATE FERTILIZERS. Gypsum, a component of single superphosphate, and S0 are the most common sources of S in pasture soils. Since gypsum contains both ions, leaching of SO22 is unlikely to completely deplete the Ca2þ and SO22 4 4 2þ added Ca . Elemental sulfur (S0) is frequently used as an acidifying agent, slowrelease S fertilizer, or in a finely divided form as a fungicide. When S0 is added to þ soils, it is oxidized to sulfuric acid which dissociates into SO22 4 ions and H ions 2 (Eq. (21)). In some soils, part of the acidity is neutralized by the OH ions released during the ligand-exchange adsorption of SO22 4 . Increases in soil pH due to the ligand-exchange adsorption of SO22 , commonly referred to as “self-liming 4
SOIL ACIDIFICATION AND LIMING INTERACTIONS
229
effect,” has often been observed in soils rich in variable charge components, such as Fe and Al oxides (Reeve and Sumner, 1972).
CHANGES IN SOIL pH DUE TO FERTILIZER ADDITION. When plants are grown in unbuffered nutrient medium supplied with different forms of N, the release of Hþ and OH2 ions alters the pH of the medium. In soils, however, acidification is effected either by a decrease in soil pH or by a decrease in ANC. The extent of soil acidification depends mainly on the pH buffering capacity of soils. Soil constituents, such as organic matter, Fe and Al oxides and CaCO3 (in calcareous soils) contribute to the pH buffering capacity of soils. Thus soils vary in their pH buffering capacity. For example, New Zealand soils generally have higher buffering capacity than Australian soils. In Table IV, the amount of acidity produced and the number of years required to reduce the pH by one unit for the top 7.5 cm of two soils (with different pH buffering capacities) due to fertilizer additions are presented. It is obvious that continuous application of ammonium sulfate and ammonium phosphates is likely to acidify the soils more quickly, particularly in soils with low pH buffering capacity. Although S0 can be used as an acidifying agent, at the normal application rate of 30 kg S ha21, the acidity produced by the oxidation of S0 is negligible when compared to the ability of most New Zealand soils to resist pH change. Similarly the acidity produced from the dissolution of superphosphate is unlikely to cause any significant measurable change in the pH of the bulk soil.
Table IV Acidifying Effects of Various Fertilizers Number of years required to reduce the pH by one unitc Fertilizer Ammonium sulfate (AS) Diammonium phosphate (DAP) Urea Single superphosphate (SSP) Triple superphosphate (TSP) Elemental sulfur (S0)
Acidity equivalenta
Acidity produced (kmol Hþ ha21)b
Tokomaru
Egmont
110 74 79 8 15 310
2.60 2.06 0.86 0.48 0.50 1.55
8 10 25 45 43 14
26 33 78 140 135 43
kg CaCO3100 kg21 fertilizer. AS, DAP and Urea added at the rate of 25 kg N ha21 year21; SSP and TSP at the rate of 30 kg P ha21 year21; and S0 at the rate of 30 kg S ha21 year21. c pH buffering capacity (kmol Hþ ha21) ¼ 21.7 and 67.5 for the Tokomaru and the Egmont soil, respectively. a b
230
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
III. EFFECT OF SOIL ACIDITY ON NUTRIENT AND HEAVY METAL TRANSFORMATION IN SOILS Acidification affects the transformation and biogeochemical cycling of both nutrients and heavy metals through its effect on the physical, chemical and biological characteristics of soils. According to Adriano (2001), the pH can be viewed as the master variable of all the driving factors because it can affect the surface charge and subsequent adsorption of solutes by variable charge soil components, such as layer silicate clays, organic matter, and oxides of Fe and Al. In addition to the effect on the sorption of metal cations and anions in soils, it also influences metal speciation, complexation of metals with organic matter, precipitation/dissolution reactions, redox reactions, mobility and leaching, dispersion of colloids, and the eventual bioavailability of trace metals. A number of soil physical and chemical properties are controlled by the nature and the amount of surface charge and the variation of surface charge with soil solution characteristics. The surface reactions of charged particles are essential to the biogeochemical cycling of nutrients and pollutants, and the pathway of detoxification of the latter when present at hazardous concentrations (Sposito, 1984; Sparks, 1995; Bolan et al., 1999a). It is important to first understand the effect of acidification on surface charge so that the surface charge can be manipulated to take advantage of solid phase interactions relating to the movement of nutrient and pollutant ions in soils, the degradation of pesticides and the decontamination of soils. Soil solution pH is one of the major factors controlling surface properties of variable charge components (Sposito, 1984; Barrow, 1985; Sparks, 1986). pH affects the surface charge through the supply of Hþ for adsorption onto the metal oxides and the dissociation of the functional groups in the soil organic matter. An increase in pH increases the net negative charge (often referred to as cation exchange capacity or CEC) and a decrease in pH increases the net positive charge (often referred to as anion exchange capacity or AEC) (Singh and Uehara, 1986). Thus, change in surface charge is a major reason for the effect of pH on anion and cation adsorption. Lower pH values also elevate the concentration of Al in soil solution, taking up a greater proportion of the cation exchange sites and reducing base saturation. On the other hand, precipitation of Al at high pH values could block the exchange sites with positively-charged Al hydrous oxides, leading to a decrease in CEC; when these soils are acidified, Al is resolubilized from the negatively charged soil particle coatings, exposing more negatively charged surface leading to an increase in CEC (Ritchie, 1989). Sumner et al. (1991) wrote a comprehensive review on nutrient status and toxicities in acid soils. In the following section, a brief discussion on the direct effects of acidification on the transformation of nutrients and heavy metals is given. The beneficial effects of liming to neutralize the acidity in relation to
SOIL ACIDIFICATION AND LIMING INTERACTIONS
231
the mobilization of nutrients and the immobilization of heavy metals will be discussed in detail in Section V (Tables VII – IX).
A. PLANT NUTRIENTS 1.
PRIMARY NUTRIENTS
NITROGEN. Acidity, by virtue of governing the type, number and activity of microorganisms, regulates the rate of organic matter mineralization, thereby reducing the number of simple organic molecules available for further decomposition and eventually rendering N and other constituent elements (P and S) soluble (Alexander, 1977). In highly acid conditions, organic matter accumulates giving rise to vast storehouse of nutrients that can be exploited by liming. Nitrification is markedly reduced below pH 6 and is undetectable below pH 4.5 (Alexander, 1977). On the other hand, ammonification reactions are insensitive to acidity over a range of pH typified in agricultural soils, resulting in the accumulation of NHþ 4 – N. Thus, the effect of acidity on the soluble N status of most agricultural soils is limited to nitrification, implying that for plants which 2 use both NHþ 4 and NO3 , acidity plays little role in determining the availability of N. For certain crops unable to use NHþ 4 – N, acidification can result in restricted þ uptake of N or even NHþ 4 toxicity. Use of NH4 –N by both higher plants and by certain groups of algae usually will result in even higher Hþ accumulation exacerbating acidification (Marschner, 1995). Acidity has a deleterious effect on the symbiotic relationship between rhizobia and legumes, and generally in soils with pH below 6, poor nodulation and N fixation result. Several physiological reasons have been attributed to this phenomenon including: (i) inhibition of infection of legume roots by nodule bacteria, decreasing nodule formation; (ii) inhibition of nitrogenase enzyme activity in the nodule due to modification of the nitrogenase iron protein; (iii) decrease in bacterial membrane potential and the inhibition of the leghaemoglobin; and (iv) decrease in the supply of photosynthate to the rhizobium due to the poor supply of major nutrients, such as P. The inhibitory effect of acidity on biological N fixation has also been attributed to the poor supply of Mo and Ca which are essential for N fixation. Thus, when nutrient deficiencies, especially Ca and Mo are overcome in acid soils, biological N fixation can be improved (Munns et al., 1977; Coventry et al., 1985; Unkovich et al., 1996). PHOSPHORUS. Much has already been reported on the effect of acidity on the solubility of soil P (Haynes, 1984). A decrease in soil pH initially increases the concentration of Fe and Al in soil solution thereby increasing the adsorption/ precipitation of P. For example, when water-soluble P fertilizers, such single and
232
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
triple superphosphates are added to soils, the Hþ ions produced from the dissolution reactions (Eq. (22)) reduce the pH around the fertilizer granules to a very low level (pH , 2). This dissolves the Fe and Al compounds in the soil resulting in the adsorption/precipitation of P. CaðH2 PO4 Þ2 þ H2 O ! CaHPO4 ·H2 O þ H3 PO4
ð22Þ
In variable charge soils, a decrease in pH increases the AEC, increasing the retention of P (Bolan et al., 1999a). As a consequence, liming acid soils results in the release of P for plant uptake. This effect is often referred to as “P sparing effect” of lime which will be discussed later (see Section V). However at very low pH, solubilization of P compounds in soils results in an increase in the concentration of P in soil solution. It has often observed acidification caused by continuous use of NHþ 4 – N results in the release of P for plant uptake (Soon and Miller, 1977). Studying anion adsorption at a range of pH values, Hingston (1981) obtained a relationship between the apparent Langmuir maxima for a range of anions and pH. This was termed the “adsorption envelope” and the apparent maxima in the envelope were found at the pKa values for anions with conjugate acids. A good linear relationship was obtained between points of inflection in the adsorption envelope and pKa values for conjugate acids. Based on this he demonstrated the sharp decrease in H2PO2 4 adsorption at pH values above 6.8, which coincided with the pKa value of orthophosphate. POTASSIUM. Under acid conditions, weathering liberates K from micaceous and feldspar minerals enhancing it to enter the soluble and exchangeable pools (Barshad and Kishk, 1970). However, in variable charge soils, increasing acidity decreases CEC reducing the ability of the soil to retain K, resulting in more soil solution K. This solution K would then be prone to leaching (Blue and Ferrer, 1986; Alibrahim et al., 1988).
2.
SECONDARY NUTRIENTS
SULFUR. In many soils, organic matter is the main source of S and since mineralization of organic matter is affected by acidity, the release of S for plant uptake decreases with increasing acidity. Further, in highly weathered acid soils, SO22 4 is adsorbed by sesquioxide surfaces, precipitated as an Al –OH – SO4 type mineral, such as alunite and basulminite and/or held as simple exchangeable anions on positively charged sites on sesquioxides under acid conditions (Marsh et al., 1987). As subsoil of most highly weathered soils are acid, they represent a potential storehouse for SO22 4 –S against leaching, and provided roots are able to enter these zones, this S should contribute significantly to plant requirements.
SOIL ACIDIFICATION AND LIMING INTERACTIONS
233
CALCIUM AND MAGNESIUM. One of the major consequences of acidification is the decline in basic cations, such as Ca and Mg, leading to deficiency of these cations for plant growth. In acid soils, most of the Ca present would exist in soluble form, but both soluble and exchangeable Ca decreases with decreasing soil pH (Haynes and Ludecke, 1981). Further more at low pH, the bioavailability of Ca is retarded by high concentration of Al. With increasing soil acidification, decreasing amounts of Mg remain in exchangeable form due to reduction in variable charge, and more is present in solution, liable to leaching losses. Also since Mg is a poor competitor with Al and Ca for the exchange sites, it tends to accumulate in the solution phase and is therefore prone to leaching (Edmeades et al., 1985; Myers et al., 1988).
3.
ESSENTIAL TRACE ELEMENTS
In general, the solubility and phytoavailability of metals are inversely related to soil pH. In very acid soils, Cu deficiency is likely to occur due to reduced retention. Since Cu is complexed with organic matter, the slow rate of decomposition of organic matter in acid soils decreases the release of Cu (Cavallaro and McBride, 1980; Jeffery and Uren, 1983). However, Cu toxicities have been reported on acid soils receiving repeated application of Cu pesticides (Alva et al., 1995; Adriano, 2001). Temminghoff (1998) observed that a decrease in soil pH increased the free Cu and Cu bound by the fulvic fraction, while Cu bound by the humic fraction decreased. At pH 3.9, about 30% Cu in solution was bound by dissolved organic carbon (DOC), whereas at pH 6.6, Cu-DOC composed of . 99%. Zinc (Zn) activity increases rapidly with decreasing pH, indicating that Zn nutritional problems are seldom encountered in soils at pH value below 5.5 provided they contain sufficient Zn. The pH-dependent solubility of Zn in soils is governed by a complex mixture of mechanisms including adsorption on sesquioxides, co-precipitation with Al, and complexation with organic matter (Barrow, 1986; Shuman, 1986; Stahl and James, 1991). Hesterberg et al. (1993) modeled changes in the solubility of some trace elements in soil as a result of acidification and found Zn, Cd, and Al solubilities increased exponentially with decreasing pH and Ca concentration. One of the important effects of acidification is the increase in Mn toxicity in plants (Shuman, 1986; Ritchie, 1989; Sumner et al., 1991; Patra and Mohanty, 1994). A decrease in pH results in increasing levels of soluble Mn and below pH 6, Mn can be expected to become soluble in toxic quantities which can adversely affect the growth of sensitive crops (Jones and Fox, 1978). The toxic levels of Mn can be reduced by the addition of organic matter through chelation (Foy, 1984). In addition to being pH sensitive, Mn is also very sensitive to changes in redox
234
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
potential. The effect of redox conditions is much greater than that of acidity in determining the level of soluble Mn in soils. Because of low solubility of iron (Fe) even under very acid conditions (Lindsay, 1971), redox reactions are likely to be of greater importance in the rhizosphere in solubilizing sufficient Fe to meet the requirements of plants (Marschner, 1995). Under acid conditions, boron (B) occurs in solution as the uncharged H3BO3 molecule and is strongly adsorbed by soil containing sesquioxides, organic matter and soluble Al as the pH is raised. Thus, B availability is increased with decreasing pH (Bingham et al., 1971). Unlike other anions, Mo is highly insoluble in low pH conditions and hence becomes less available in acid soils. The beneficial effect of liming on Mo availability is discussed in Section V. In general, enhanced solubility of certain metal cations is an important consequence of soil acidification. The resulting toxicity of those elements, which are essential to plant growth, appears to be partly due to a nutrient imbalance brought about by abnormal accumulation in plant tissue (Sumner et al., 1991).
B. HEAVY METALS One of the major consequences of soil acidification is the increase in concentration of Al and Mn which are highly toxic to plant growth. One of the primary aims of liming soils for agricultural production is to decrease the concentration of these elements. While Mn toxicity is related directly to the metabolic requirements of plants, the effect of Al toxicity appears to be largely manifested as malformation and malfunction of the root system, a syndrome which is exacerbated by low levels of solution Ca in acid soils (Hechtbuchholz and Foy, 1981). Acidification affects the transformation of heavy metal ions through: (a) modification of surface charge in variable charge soils; (b) altering the speciation of metals; and (c) influencing the reduction and oxidation reactions of the metals. In most countries, cadmium (Cd) has been identified as the major heavy metal reaching the food chain through animal transfer in pastoral agriculture. This is one of the main reasons why this element has been studied extensively in relation to soil and plant factors affecting its bioavailability. It has been observed that the adsorption of Cd increased with decreasing pH (Tiller et al., 1979; Basta and Tabatabai, 1992; Naidu et al., 1994; Bolan et al., 1999b). Three reasons have been advanced for this phenomenon: firstly, in variable-charge soils, an increase in pH causes an increase in surface negative charge resulting in an increase in cation adsorption (Naidu et al., 1994). Secondly, an increase in soil pH is likely to result in the formation of hydroxy species of metal cations which are adsorbed preferentially over the metal cation. Naidu et al. (1994) observed that CdOHþ species are formed above pH 8 which have a greater affinity for adsorption sites
SOIL ACIDIFICATION AND LIMING INTERACTIONS
235
than just Cd2þ. Thirdly, precipitation of Cd as Cd(OH)2 is likely to result in greater retention at pH above 10 (Naidu et al., 1994). Attempts have been made to relate the pH-induced increases in surface charge to Cd adsorption by variable charge soils (Boekhold et al., 1993; Naidu et al., 1994; Bolan et al., 1999b). For example, Bolan et al. (1999b) observed that approximately 50% of the pH-induced increase in surface negative charge in variable charge soils was occupied by Cd. The remaining surface negative charge was presumed to be occupied by the Hþ and Kþ ions, added in acid and alkali solutions to alter the soil pH. This indicates that the increased Cd adsorption with increasing pH is attributable to increasing negative charge. Similarly, Naidu et al. (1994) demonstrated that the effect of ionic strength on Cd adsorption operates through its effect on surface charge. The effect of pH on metal sorption has also been related to the exchange of Hþ for the metal ions. On this basis, Christensen (1984) and Boekhold et al. (1993) modified the Freundlich equation to account for the effect of pH on Cd sorption in soils (Eq. (23)). S ¼ Kf Cn ðHþ Þm
ð23Þ
The exponent, m is considered as a stoichiometric coefficient indicating relative replacement ratio of Hþ by Cd2þ (moles Hþ replaced by one mole of Cd). A range of m values ranging from 0.5 to 1.8 (Boekhold et al., 1993; Naidu et al., 1994; Filius et al., 1998) have been obtained for Cd adsorption in soils, indicating that depending on the soil and solution composition, varying amounts of Hþ are released per unit Cd sorbed. Kinniburgh and Jackson (1981) tabulated Hþ release values for a number of metal ions on a range of adsorbents illustrating that the value of m can vary widely and is generally high for Fe and Al oxides. Filius et al. (1998) observed that while a decrease in soil pH decreased the adsorption of Cd it caused the opposite effect on desorption. With increasing pH, the sorption/desorption isotherms became less continuous in the transition zone from adsorption to desorption. The equilibrium solution concentration at which zero sorption – desorption occurred (called nut point) decreased with increasing pH, indicating that even at low solution concentration adsorption continued to occur at high pH values. For example, at the lowest pH (4.68) the soil sample released 50 mmol Cd kg21 soil at an equilibrium Cd concentration of 0.1 mM, but at the same concentration, the soil with the highest pH (6.81) was still adsorbing Cd from the solution. Generally with increasing pH, increasing amount of irreversibly bound Cd occupies specific sorption sites whereby the proportion of Cd bound reversibly to non-specific exchange sites becomes less pronounced. In general, Cd uptake by plants increases with decreasing pH. For example, higher Cd concentrations were obtained for lettuce and Swiss chard on acid soils
236
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
(pH 4.8 – 5.7) than on calcareous soils (pH 7.4– 7.8) (Mahler et al., 1978). Consequently, it is recommended that soil pH be maintained at pH 6.5 or greater in land receiving biosolids containing Cd (Adriano, 2001). Acidification affects the leaching and residence time of many trace elements. Taylor (1975) observed that the amounts of trace elements released from the mor soils of Sweden increased with decreasing pH of the acid precipitation. Approximately 85% of the total Cd was released at pH 2.8 and the estimated length of time for a 10% decrease through leaching in the total concentration of Cd in the mor horizon would be 1.7 years at pH 2.8, 4 – 5 years at pH 3.2, and 20 years at pH 4.2. The effect of soil acidity on the adsorption of metalloids, such as arsenic (As) and selenium (Se) is manifested through two interacting factors—the increasing negative surface potential on the plane of adsorption and increasing amount of negatively charged ionic species present in soil solution. For example, while the first factor results in a decrease in As(V) adsorption, the latter factor is likely to cause the opposite effect. Thus, the effect of pH on As(V) adsorption has been shown to be dependent on the nature of the mineral surface. In soils with low oxides content, increasing the pH had little effect on As(V) adsorption, while in highly oxidic soils, adsorption decreased with increasing pH (Smith et al., 1998). In general, adsorption of As(V) decreases with increasing pH. In contrast to As(V), however, adsorption of As(III) tends to increase with increasing pH (Adriano, 2001). Soil acidification affects the solubility of Cr through its effect on adsorption/precipitation and oxidation/reduction reactions (Bartlett, 1991; James, 1996). While the adsorption of Cr(VI) in soil increases with decreasing pH, the adsorption of Cr(III) decreases (Bartlett and Kimble, 1976). Similarly, while the reduction of Cr(VI) to Cr(III) (a Hþ consumption reaction) increases with decreasing soil pH (James, 1996), the oxidation of Cr(III) to Cr(VI) (a Hþ donation reaction) decreases (Bartlett and James, 1979). The reduction of Cr(VI) to Cr(III) occurs readily in most soils due to the presence of organic matter (Eq. (24)); whereas the oxidation of Cr(III) to Cr(VI) requires the presence of oxidized Mn in the soil as an electron acceptor for the reaction to proceed (Bartlett and James, 1979). 2Cr2 O7 þ 3C0 þ 16Hþ ! 4Cr3þ þ 3CO2 þ 8H2 O
ð24Þ
In general, with the exception of Se and Mo, trace elements are more soluble in soils at low pH due to the dissolution of the carbonates, phosphates, and other solid phases. Low pH also lowers the CEC of organic matter and mineral surfaces, thereby weakening the sorption of metals to specific adsorption sites.
SOIL ACIDIFICATION AND LIMING INTERACTIONS
237
IV. AMELIORATION OF SOIL ACIDITY THROUGH LIMING Three approaches may be taken to minimize the rate of acidification and its subsequent negative impacts on sustainable agricultural production and environmental protection (Gregan et al., 1989; Helyar, 1990): (i) reduce the amount of Hþ ions generated, (ii) reduce the uncoupling of Hþ and OH2 ions generation processes, and (iii) neutralize the acid produced. In the case of acidity caused by industrial activities, the rate of acid generation can be retarded by reducing NOx and SOx gaseous emissions and by reducing the oxidation of pyrite (Longhurst, 1991; Frazer, 2001). But in the case of managed ecosystems used for agricultural production, the rate of acid generation can be altered by selecting the fertilizer type that produce less acid, selecting plant species that do not accumulate cation excesses, and reducing the losses of C, N and S from the system (Gregan et al., 1989). Traditionally liming is the most common practice used to overcome the impact of soil acidification. However, an integrated approach involving liming, cultural practices and plant tolerance will probably be necessary, particularly where the acidification potential is high and its effect likely to extend into the subsoil. These approaches are discussed in detail by Bolan and Hedley (2001) which is not within the scope of this review.
A. LIMING MATERIALS A range of liming materials are available, which vary in their ability to neutralize the acidity. These include calcite (CaCO3), burnt lime (CaO), slaked lime (Ca(OH)2), dolomite (CaMg(CO3)2) and slag (CaSiO3). The acid neutralizing value of liming materials is expressed in terms of calcium carbonate equivalent (CCE), defined as the acid neutralizing capacity of a liming material expressed as a weight percentage of pure CaCO3 (Table V). A neutralizing value . 100 indicates greater efficiency of the material relative to pure CaCO3. The amount of liming material required to rectify soil acidity depends on the neutralizing value of the liming material and pH buffering capacity of the soil. Recently the potential value of other Ca-containing compounds in overcoming the problems associated with acidification has been evaluated (Dick et al., 2000). Some of these materials include PRs, FGD gypsum, fluidized bed boiler ash, fly ash, and lime stabilized organic composts. Increasing amounts of PRs are added directly to soils mainly as a source of P. Unlike soluble P fertilizers, such as superphosphates, PRs can also have a liming value in addition to supplying P and Ca. The liming action of PRs can occur through two processes. Firstly, most PRs contain some free CaCO3 which
238
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN Table V Neutralizing Value of Liming Materials Neutralizing valuea
Liming material
Chemical formula
Burnt lime Slaked lime Dolomite Lime Slag Phosphogypsum Mined gypsum FGD gypsum (The Netherlands) FGD gypsum (USA) FBA (New Zealand) Coal fly ash Alkaline biosolid
CaO Ca(OH)2 CaMg(CO3)2 CaCO3 CaSiO3 CaSO42·H2O CaSO42·H2O CaSO42·H2O
179 136 109 100 86 0.33 12.4 0.42
CaSO42·H2O Variable
0.1 65 Variable Variable
a
Reference Brady and Weil (1999) Brady and Weil (1999) Brady and Weil (1999) Brady and Weil (1999) Bolan et al. (1991) Bolan et al. (1991) Bolan et al. (1991) Bolan et al. (1991) Wang et al. (1999) Dick et al. (2000) Basta (2000)
Expressed as a weight percentage of pure CaCO3.
itself can act as a liming agent. Secondly, the dissolution process of the P mineral component (i.e., apatite) in soils consumes Hþ, thereby reducing the soil acidity. It is estimated that every 1 kg of P dissolved from PRs generates a liming value equivalent to 3.2 kg CaCO3. From the amounts of P and free CaCO3 present in the PR it may be possible to calculate its total liming value. For example, a tonne of North Carolina Phosphate Rock (NCPR) which contains 13.1% P and 11.7% free CaCO3 can have a potential liming value of 536 kg CaCO3 (132 kg free CaCO3 þ 3.2 £ 132 ¼ 419 kg CaCO3 upon dissolution). The liming values of various PRs are presented in Table VI, which ranges from 450 to 560 kg CaCO3 tonne21 of PR and most of the liming value in PRs is derived from the dissolution of the apatite. While the free CaCO3 in PRs dissolves reasonably fast providing a small amount of immediate liming value, the apatite dissolves at a variable but generally slower rate providing liming value over a longer period of time. Certain unreactive PRs, such as Christmas Island PR, Nauru PR, and Duchess PR also have significant amounts of potential liming value, but since they are unlikely to dissolve in soils, there is no benefit from adding these PRs either as a P source or as a liming material. Alkaline stabilized biosolids are increasingly being used as an agricultural lime substitute, soil amendment and surrogate soil. Alkaline stabilization of biosolid utilizes a combination of high pH, heat, and drying to kill pathogens and stabilize organic matter. A range of alkaline materials are used for this purpose, including, cement kiln dust, lime kiln dust, lime, limestone, alkaline coal fly ash, FGD, other coal burning ashes, and wood ash (Basta, 2000). Logan and Harrison (1995) examined the value of a commercial alkaline stabilized biosolid product called “N-Viro” soil as a soil substitute. N-Viro is
SOIL ACIDIFICATION AND LIMING INTERACTIONS
239
Table VI Theoretical Liming Value of Phosphate Rocks
Phosphate rock North Carolina phosphate rock Sechura phosphate rock Gafsa phosphate rock Chatham rise phosphorite Arad phosphate rock Youssafia phosphate rock Khourigba phosphate rock Egyptian phosphate rock Jordan phosphate rock Nauru phosphate rock Christmas Island Duchess phosphate rock
Total P (% w/w)
Free CaCO3 (% w/w)
Liming value (kg CaCO3 Mg21)a
13.1 13.1 13.4 8.9 14.1 13.8 14.4 13.0 13.4 15.6 16.4 13.5
11.7 5.1 7.1 27.6 8.2 5.4 6.1 4.9 7.7 4.1 2.1 1.8
536 470 500 560 533 495 520 465 505 540 545 450
a
Amount of acidity neutralized, expressed as an equivalent weight of pure CaCO3 per tonne (Mg) of phosphate rock, through dissolution of apatite, and free CaCO3.
produced by heat treatment of a mixture of cement kiln dust and municipal sewage sludge. Addition of this material improved the physical and chemical properties of a degraded mine soil. Such alkaline materials are effective in reducing the acidity produced during the nitrification of NHþ 4 in biosolids, thereby reducing the bioavailability of heavy metals in biosolid-amended soils (Brown et al., 1977; Basta, 2000; Dinel et al., 2000). To minimize metal mobility and bioavailability in biosolid-amended soils, the USEPA recommends the application of alkaline-stabilized biosolids and other liming agents to increase the soil pH to 6.5 or greater. In France, application of lime combined with organic matter has been used for more than 30 years to reduce Cu phytotoxicity in vineyards receiving regular application of Cu-based fungicides (Mench et al., 1994).
B. EFFECTS OF LIMING Liming enhances the physical, chemical and biological characteristics of soil through its direct effect on the amelioration of soil acidity and through its indirect effects on the mobilization of plant nutrients, immobilization of toxic heavy metals, and the improvements in soil structure and Ks (Haynes and Naidu, 1998).
240
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
1.
PHYSICAL AMELIORATION
Dispersion and flocculation of colloid particles are often manifested through changes in surface potential and charge densities. Thus manipulation of particlecharge density assists management of dispersive soils. Such charge manipulations has often been obtained by the use of inorganic salts including liming materials (Bolan et al., 1999a). Dispersion is caused by mutual repulsion of soil particles because of surface charge. If the repulsive forces are predominant, the soil becomes dispersed and virtually unmanageable in an agronomic sense. It is the balance of attractive and repulsive forces that determines whether a soil is flocculated or dispersed. Factors which affect the surface charge of soil particles determine the extent of dispersion, and include electrolyte concentration of the soil solution, the valence of the dominant cation occupying the exchange sites, and pH. pH can influence the dispersion, and the sensitivity of Ks to pH change depends on the quantity of variable charge minerals and organic matter present in the soil (Chiang et al., 1987). Liming influences flocculation/dispersion through its effect on soil pH and Ca concentration in soil solution. Bolan et al. (1996) observed that the effect of pH on dispersion varied between the Na- and the Ca-saturated soils. At the same value of net charge, the Ca2þ-saturated soils exhibited less dispersion than the Na-saturated soils. This can partly be explained by the increased surface charge screening mechanism of Ca than Na. It has often been observed that the dispersion of clay decreases as the percentage Ca-saturation increases, which has been related to the decrease in charge density (Rengasamy, 1983). The thickness of the diffuse-double layer (DDL) in Ca-saturated soil is likely to be smaller than that of the Na-saturated soil (Bolan et al., 1996). As the DDL becomes smaller, the soil particles are attracted to each other resulting in increased flocculation and greater Ks (Rengasamy, 1983). Liming has often been shown to improve soil structure and Ks of soils. The Ca in the liming materials helps in the formation of soil aggregates, thereby improving soil structure (Chan and Heenan, 1998). The lime-induced improvement in aggregate stability and Ks is manifested through the effect of liming on dispersion and flocculation of soil particles.
2.
CHEMICAL AMELIORATION
The primary purpose of liming arable soils is to overcome the chemical problems associated with soil acidity that include high concentrations of acid ions (Hþ and Al3þ) and toxic elements (Mn2þ), and low concentrations of basic cations (Ca and Mg) and other nutrient ions, such as Mo and P. In relation to environmental pollution, soil acidity enhances the solubility and the subsequent
SOIL ACIDIFICATION AND LIMING INTERACTIONS
241
bioavailability of heavy metals. The hydrolysis of the basic cations in lime produces OH2 ions which neutralize the Hþ ions, thereby decreasing the activity and bioavailability of Al and Mn. But liming also increases the solubility of Mo and P, thereby increasing their availability. Lime provides the basic nutrient cations (Ca and Mg), and also reduces the solubility of heavy metals, thereby minimizing their bioavailability and mobility in soils. Thus lime can interact with nutrient and heavy metal transformation and bioavailability, as discussed in detail in Section V. Elevation in pH due to addition of lime results in the precipitation of exchangeable Al (with the vacated sites being mostly occupied by Ca) and increases the negative charge or CEC. In many soils, the increase in negative charge is mainly due to dissociation of Hþ from weakly acidic functional groups of organic matter (Thomas and Hargrove, 1984; Curtin et al., 1996). It has been estimated that raising pH by one unit increases the CEC of soil organic matter by about 30 cmol (þ ) kg21 (Helling et al., 1964). The CEC of the soil mineral component is generally far less pH-dependent than that of soil organic matter. For example, the CEC of soil clay may only increase by as little as 3 or 4 cmol (þ ) kg21 per pH unit (Helling et al., 1964; Curtin et al., 1996). However, the pH-dependence of mineral CEC can vary considerably depending on the nature of the component minerals. Mineral constituents that dissociate Hþ when lime is added include hydroxy-Al polymers associated with the surfaces of phyllosilicate minerals, amorphous and short-range ordered aluminosilicates, and ruptured surfaces of silicates and oxides (Thomas and Hargrove, 1984). Thus, mineral soils whose CEC varies substantially with pH (sometimes described as variable charge soils ) generally are high in allophonic materials, hydrous oxides of Al and Fe, or 1:1 layer silicates, such as koalinite and halloysite.
3.
BIOLOGICAL AMELIORATION
Liming has been shown to provide optimum conditions for a number of biological activities that include N fixation, and mineralization of N, P and S in soils. The enhanced mineralization of these nutrient ions is likely to cause an increase in their concentration in soil solution for plant uptake and for leaching (Lyngstad, 1992; Arnold et al., 1994; Neale et al., 1997). Nitrogen fixing bacteria in legume plants require Ca, hence liming is likely to enhance N fixation (Muchovej et al., 1986). Liming is often recommended for the successful colonization of earthworm in pasture soils. The lime-induced increase in earthworm activity may influence the soil structure and macroporosity through the release of polysaccharide and the burrowing activity of earthworm (Springett and Syers, 1984).
242
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
Liming has been shown to cause short-term increases in microbial biomass and soil enzyme activity (Haynes and Swift, 1988). Increased microbial activity and the subsequent production of extracellular polysaccharides which act as a binding agent can increase soil aggregate stability. Although an increase in pH through liming is likely to enhance the solubilization of soil organic matter, an increase in the coagulation of DOC can be expected with a decrease in negative charge of DOC. The charge on the DOC is influenced by the nature and the valency of the cation. Calcium is found to be very effective in decreasing the negative charge on DOC, which may be a reason for the decrease in the leaching of DOC with liming (Temminghoff, 1998).
V. LIME, NUTRIENT AND HEAVY METAL INTERACTIONS Liming influences the transformation and uptake of nutrients and heavy metals by plants through its direct effect on the neutralization of soil acidity and its indirect effect on the physical, chemical and biological characteristics of soils. The processes involved in the liming-induced transformation and phytoavailability of nutrients and heavy metal ions are given in Tables VII–IX.
A. PLANT NUTRIENTS 1. PRIMARY NUTRIENTS NITROGEN. Liming has often been shown to enhance the mineralization of organic matter, thereby releasing inorganic plant nutrients such as N, S and P to soil solution. Unless these nutrients are actively taken up plants they are liable for leaching losses. Lime-induced greening of plant leaves is often associated with enhanced availability of N (Ogata and Caldwell, 1960; Nyborg and Hoyt, 1978). Nitrate, being weakly retained by soil particles, is subjected to leaching. Liming exacerbates the leaching of NO2 3 because an increase in pH through liming decreases the positive surface charges in variable charge soils. Further, liming also provides basic cations (Ca and Mg) as a companion ion for NO2 3 leaching (Tinsley, 1973; Adams, 1986). Liming affects both the chemical and microbial transformation of N in soils. In general, NHþ 4 – N is nitrified more rapidly on addition of lime due to an increase in the activity of microorganisms involved in nitrification (Lyngstad, 1992; Puttanna et al., 1999). The efficiency of nitrification inhibitors decreases with the addition of lime. This is probably due to an increase in nitrifier activity and also due to an increase in general microbial activity (Slangen and Kerkhoff, 1984),
Table VII Selected References on the Effects of Liming on the Transformation and Plant Uptake of Primary and Secondary Nutrients in Soils Effects
Reference
Nitrogen
Increased mineralization/nitrification
Gardner et al. (1965), Nyborg and Hoyt (1978), Edmeades et al. (1981), Carter (1986), Curtin and Smillie (1986), Klemmedson et al. (1989), Clay et al. (1993), Stevens and Laughlin (1996), Unkovich et al. (1996), and Curtin et al. (1998a) Winter et al. (1981), Sommer and Ersboll (1996), and Howard and Essington (1998) Munns et al. (1977), Coventry et al. (1985), Peoples et al. (1995), Unkovich et al. (1996), and Raychaudhuri et al. (1998) Ogata and Caldwell (1960), Rosolem and Caires (1998), and Raychaudhuri et al. (1998) Ogata and Caldwell (1960) and Bailey (1995)
Increased ammonium volatilization Increased biological N fixation Increase in plant uptake Decrease in plant uptake Phosphorous
Increased mineralization of organic P Increased adsorption/precipitation Increase in plant uptake Decrease in plant uptake
Potassium
Sulfur
Increase in adsorption
Trasarcepeda et al. (1991), Condron et al. (1993), and Fernandes and Coutinho (1999) Badora and Filipek (1988), Naidu et al. (1990), Holford et al. (1994), Agbenin (1996), Mongia et al. (1997), and Fransson et al. (1999) Lucas and Blue (1973), Friesen et al. (1980b), Hemphill et al. (1982), Bhella and Wilcox (1989), and Raychaudhuri et al. (1998) Wahab and Shah (1952) and Amarasiri and Olsen (1973)
Decrease in adsorption Increase in plant uptake Decrease in plant uptake
Magdoff and Bartlett (1980), Blue and Ferrer (1986), and Alibrahim et al. (1988) Bolan et al. (1999a) Abraham et al. (1980) and Mason et al.(1994) Magdoff and Bartlett (1980)
Increased mineralization of organic S
Bolan et al. (1988) and Valeur et al. (2000)
243
(continued on next page)
SOIL ACIDIFICATION AND LIMING INTERACTIONS
Nutrient
244
Table VII (continued) Nutrient
Magnesium
Reference
Decreased adsorption Increased leaching Increase in plant uptake Decrease in plant uptake
Marsh et al. (1987) and Bolan et al. (1988) Korentajer et al. (1983) and Bolan et al. (1988) Martini and Mutters (1984) Korentajer et al. (1983) and Kemper and Sorensen (1984)
Increase in exchangeable and solution Ca Increase in plant uptake
Curtin and Smillie (1983), Edmeades et al. (1983), and Curtin and Smillie (1995) Crouchley (1981) and Bhella and Wilcox (1989)
Increased adsorption/fixation Increase in plant uptake Decrease in plant uptake
Grove et al. (1981), Riggs et al. (1995a) and Wheeler (1997) Bhella and Wilcox (1989) and Mason et al. (1994) Crouchley (1981)
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
Calcium
Effects
Table VIII Selected References on the Effects of Liming on the Transformation and Plant Uptake of Trace Element Nutrients in Soils Effects
Reference
Copper
Increased adsorption Increased organic matter complexation Increase in plant uptake Decrease in plant uptake
Msaky and Calvet (1990), De Vera and Pocsidio (1998), and Federer and Sticher (1999) McBride and Bouldin (1984) and Hodgson et al. (1966) Adcock et al. (1999) Bates (1971), John and van Laerhoven (1976), Anderson and Siman (1991), Preer et al. (1995), Fang and Wong (1999), and Tyler and Olsson (2001)
Zinc
Increased adsorption Increase in plant uptake Decrease in plant uptake
Friesen et al. (1980a), Shuman (1986), Harter (1991), and Federer and Sticher (1999) Adcock et al. (1999) Lee and Craddock (1969), MacLean et al. (1972), John and van Laerhoven (1976), Chang et al. (1982), Pepper et al. (1983), Sanders et al. (1987), Bhella and Wilcox (1989), Anderson and Siman (1991), Preer et al. (1995), Brallier et al. (1996), Chowdhury et al. (1997), Fang and Wong (1999), Ye et al. (1999), Oste et al. (2001), and Tyler and Olsson (2001)
Cobalt
Increased adsorption Decrease in plant uptake
Backes et al. (1995) Deram et al. (2000) and Tyler and Olsson (2001)
Boron
Increased adsorption Increase in plant uptake Decrease in plant uptake
Bingham et al. (1970) and Parker and Gardner (1982) Bingham et al. (1970), Peterson and Newman (1976) and Su et al. (1994) Bingham et al. (1970), Gupta (1972), Bartlett and Picarelli (1973), Gupta and MacLeod (1973), Peterson and Newman (1976), Blamey and Chapman (1979), Brown (1979), Lipsett et al. (1979), Haddad and Kaldor (1982), Hemphill et al. (1982), Su et al. (1994), and Tyler and Olsson (2001)
Selenium
Increased bimethylation Increase in plant uptake Decrease in plant uptake
Frakenberger and Karlson (1994) Cary et al. (1967), Cary and Allaway (1969), and Gupta et al. (1982) Gupta and Winter (1975) and Tyler and Olsson (2001)
245
(continued on next page)
SOIL ACIDIFICATION AND LIMING INTERACTIONS
Nutrient
246
Table VIII (continued) Effects
Reference
Iron
Decrease in plant uptake
Patra and Mohanty (1994) and Franzen and Richardson (2000)
Molybdenum
Decreased adsorption Increase in plant uptake
Robinson et al. (1951) Robinson et al. (1951), Vlek and Lindsay (1977), Gupta (1979), Gupta and Kunelius (1980), Crouchley (1981), Coventry et al. (1985), Coventry et al. (1987), Mason et al. (1994), and Wheeler (1998)
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
Nutrient
Table IX Selected References on the Effects of Liming on the Transformation and Plant Uptake of Toxic Heavy Metals in Soils Effects
Reference
Aluminium
Increased adsorption
MacLean et al. (1972), Badora and Filipek (1988), Hochman et al. (1992), Unkovich et al. (1996), Mongia et al. (1997), and Mora et al. (1999) MacLean et al. (1972), Haynes and Ludecke (1981), Edmeades et al. (1983), Wang et al. (1999), and Tyler and Olsson (2001)
Decrease in plant uptake Manganese
Cadmium
MacLean et al. (1972) and Jauregui and Reisenauer (1982) Jauregui and Reisenauer (1982)
Increased adsorption/ precipitation
MacLean (1976), Estan et al. (1987), He and Singh (1994), Brallier et al. (1996), Hooda and Alloway (1996), Filius et al. (1998), Krebs et al. (1998), Federer and Sticher (1999), and Gray et al. (1999) Filius et al. (1998) Li et al. (1996) Lagerwerf (1971), John and van Laerhoven (1976), Chaney et al. (1977), Bingham et al. (1979), Hortenstine and Webber (1981), Adriano et al. (1982), Pepper et al. (1983), Albasel and Cottenie (1985), Anderson and Siman (1991), Jackson and Alloway (1991), Sparrow et al. (1993), He and Singh (1994), Preer et al. (1995), Brallier et al. (1996), Han and Lee (1996), Hooda and Alloway (1996), Oliver et al. (1996), Maier et al. (1997), Redente and Richards (1997), Sparrow and Salardini (1997), Krebs et al. (1998), Singh and Myhr (1998), Vasseur et al. (1998), Fang and Wong (1999), Fernandes et al. (1999), Gray et al. (1999), Lehoczky et al. (2000), Oste et al. (2001), and Tyler and Olsson (2001) (continued on next page)
Decreased desorption Increase in plant uptake Decrease in plant uptake
Godo and Reisenauer (1980), Haynes and Ludecke (1981), Edmeades et al. (1983), Adcock et al. (1999), and Tyler and Olsson (2001)
247
Increased adsorption Increased precipitation as manganocalcite Decrease in plant uptake
SOIL ACIDIFICATION AND LIMING INTERACTIONS
Metal
248
Table IX (continued) Effects
Reference
Arsenic
Increased immobilization Decreased adsorption Increase in plant uptake Decrease in plant uptake
Bothe and Brown (1999) Jones et al. (1997) Heeraman et al. (2001) Jiang and Singh (1994) and Tyler and Olsson (2001)
Mercury
Increased reduction to Hg0 Decreased reduction of Hgo Increased adsorption Increased uptake Decreased uptake
Frear and Dills (1967) and Landa (1978) Alberts et al. (1974) Farrah and Pickering (1978) Heeraman et al. (2001) Tyler and Olsson (2001)
Lead
Increased adsorption Increase in hydroxy species Decrease in plant uptake
Chlopecka et al. (1996) Roy et al. (1993) and Chlopecka et al. (1996) Cox and Rains (1972), John and van Laerhoven (1972), Zimdahl and Foster (1976), Reddy and Patrick (1977), Ye et al. (1999), and Tyler and Olsson (2001)
Nickel
Increased adsorption/precipitation at high pH Decrease in plant uptake
Pratt et al. (1964)
Increased Cr(III) oxidation Decreased Cr(VI) adsorption Increased Cr(III) adsorption Decrease in plant uptake
Bartlett and James (1979) Bartlett and James (1979) and Bolan and Thiagarajan (2001) Bolan and Thiagarajan (2001) Bolan and Thiagarajan (2001) and Tyler and Olsson (2001)
Chromium
Bisessar (1989), Brallier et al. (1996), and Tyler and Olsson (2001)
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
Metal
SOIL ACIDIFICATION AND LIMING INTERACTIONS
249
resulting in rapid biodegradation of nitrification inhibitors. All these processes 2 result in an increase in NO2 3 concentration, leading to NO3 leaching, especially in the absence of active plant uptake. Although an increase in pH due to liming is likely to enhance the volatilization losses of NH3 gas, the enhanced effect of liming on NH3 volatilization has not been documented under field conditions (Winter et al., 1981; Sommer and Ersboll, 1996; Howard and Essington, 1998). PHOSPHORUS. A reason commonly given for liming-induced improvement in plant growth in acid soils was to increase P availability. It is well established that, in strongly acid soils, Al toxicity can have a substantial inhibitory effect on the uptake and translocation of P (Chen and Barber, 1990). Haynes (1984) noted that the effects of liming on Al toxicity and P deficiency could be difficult to delineate. In soils high in exchangeable and soluble Al, liming may increase plant P uptake by decreasing Al, rather than by increasing P availability per se. This may be due to improved root growth where Al toxicity is alleviated, allowing a greater volume of soil to be explored (Friesen et al., 1980b). Interactions between P and Al in acid soils were already covered in detail by Haynes (1984). In situations where Al toxicity is not a factor, the impact of liming on P availability is not clear-cut. It has become apparent that generalizations on the direction (positive or negative) of the response of available P to lime can be misleading (Haynes, 1984). Lack of a consistent response can be attributed to the fact that liming may change several factors that regulate the concentration of P in the soil solution (Barrow, 1984). Some of these changes tend to increase soluble P concentration whereas others can have the reverse effect. In many soils, adsorption –desorption reactions regulate the concentration of P in the soil solution. Barrow (1984) suggested that several major factors influence the relations between pH and P adsorption. When pH is increased, the proportion of the divalent phosphate ion (HPO22 4 ), the P species considered to be adsorbed, also increased. This change in phosphate speciation promotes adsorption but at the same time, surface electrostatic potential becomes more negative as pH increases thus lowering the AEC. The resultant effect of these two competing tendencies has a large bearing on whether P adsorption could be altered by liming. The pH-dependence of surface potential is sensitive to factors, such as exchangeable cation composition and ionic strength of the soil solution (Barrow, 1984), both of which change when lime is applied. Other possible effects of liming on P availability include the precipitation of P as calcium phosphate, often cited as the cause of increased P retention as pH approaches 7 (Naidu et al.,1990). Liming may accelerate the rate of organic P mineralization due to increased rates of microbial activity, but the practical significance of this effect remains unclear because of the difficulty in measuring P mineralization rates. At a longterm experimental site in New Zealand, Condron and Goh (1989) attributed declines in organic P in the 0 –7.5 cm soil layer between 1971 and 1974 to increased mineralization as a result of liming in 1972.
250
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
The net effect on P availability of the various soil chemical and biochemical changes upon liming is difficult to predict. Approaches to measuring effects of lime on P availability include studies of its effects on P adsorption (Naidu et al., 1990; Holford et al., 1994; Agbenin, 1996), P release to extractants (Rhue and Hensel, 1983; Sorn-Srivchai et al., 1984; Curtin and Syers, 2001) and P uptake by plants (Mansell et al., 1984; Fernandes and Coutinho, 1999). Studies using chemical extractants have shown that the response of extractable P to lime is influenced by the nature of the extractant. In a fertilized soil in which lime increased soil solution P from 0.29 mg l21 (pH 4.7) to 0.67 mg l21 (pH 6.2), Curtin and Smillie (1984) observed that acidic extractants gave similar values suggesting that liming had little effect on P availability whereas alkaline extractants (0.1 M NaOH) indicated that lime indeed decreased available P. There is evidence that the commonly-used Olsen bicarbonate test may give artificially low values for limed soil because of precipitation of calcium phosphate resulting from a combination of high pH (, 8.5) and soluble Ca in the extract (Sorn-Srivchai et al., 1984). Recently, Curtin and Syers (2001) also confirmed that Olsen P tends to decrease when pH was raised; however, water-extractable P also decreased by liming, suggesting that the lime-induced decreases in Olsen P were due to increased P adsorption rather than to Ca –P precipitation. It appears then that the least ambiguous way of evaluating the effect of liming on the solubility of P is by measuring the levels of P in soil solution. While this approach has been successfully used in some studies (Adams et al., 1982; Curtin and Smillie, 1984), procedures for the displacement of soil solution are too laborious for routine use in soil testing laboratories. There have been relatively few studies in which the so-called P-sparing effect of lime (Tillman and Syers, 1982) has been quantified under field conditions. The P-sparing effect of lime is a term which is specific to a situation in which lime increases the actual availability of P to the plant. It excludes situations where plant P uptake is increased as a result of elimination of Al (or Mn) toxicity by liming. Mansell et al. (1984) evaluated the results of 25 field experiments that investigated the effects of P and lime on pasture production in New Zealand for evidence of a P-sparing effect of lime. Only 11 of the trials gave evidence of a Psparing effect and only 4 of these showed effects large enough to be of practical significance (i.e., 11– 20 kg P ha21year21 in some sedimentary-type soils at low pH (, 5.5) at the normal liming rates). An increase in the availability of P can only be expected on soils which have accumulated some reserve (in this case organic P or inorganic “fixed” P) of this element. Unfortunately, field situations where liming might result in worthwhile reductions in fertilizer P requirement is not predictable. Mansell et al. (1984) concluded that, because of the scarcity and unpredictability of important P-sparing effects, it would seem unwise to recommend to farmers to reduce fertilizer P inputs after lime has been applied. Therefore liming should not be practiced with the expectation that fertilizer P input can be reduced in the long term.
SOIL ACIDIFICATION AND LIMING INTERACTIONS
251
In summary, the information on lime-phosphate interactions that has been published since the topic was last reviewed in Advance in Agronomy (Haynes, 1984) further suggests that, once pH is high enough to eliminate Al/Mn toxicity (, pH 5), liming will neither have a large nor consistent effect on the efficiency of use of soil or applied P. POTASSIUM. Comprehensive reviews have been published in which the forms and availability of soil K have been described (e.g., Syers, 1998). Assessment of the plant-availability of K is somewhat more complex than that of Ca or Mg because non-exchangeable form of the element (i.e., K fixed by micaceous minerals and K in the lattice of K-containing primary minerals) can supply significant amount of the K for crop uptake (Syers, 1998). Effects of lime on soil K availability to plants are not well documented however. In theory, a number of processes that control the concentration of K in the soil solution could be influenced by liming. Liming could alter the equilibrium between soil solution K and exchangeable K due to increases in CEC and removal of Al from exchange sites or because of competition for exchange sites with lime-derived Ca. In laboratory studies, the concentration of K in soil solution decreased after liming due to increased K adsorption (Curtin and Smillie, 1983). Although the ratio of Ca to K in soil solution can increase substantially when soil is limed (Curtin and Smillie, 1995), there is no evidence that this antagonistic effect of Ca reduces the uptake of K by plants. The efficiency of fertilizer K use in variable charge soils may improve after liming because increased CEC may result in less leaching of K. It is possible that, by reducing the intensity of acid weathering, liming could slow down the release of non-exchangeable K. Measurement made 17 years after liming showed that exchangeable K decreased from 1.0 to 0.7 cmol (þ ) kg21 as pH increased from 4.5 to 6.1 (Curtin and Smillie, 1995). Such a trend could be due to increased plant uptake of K or possibly insignificant release of nonexchangeable K when the soil was limed. Since liming also increases the concentration of Ca in soil solution, the adsorption of cations, such as K, can be affected. For example, in batch experiments, the decrease in K adsorption induced by liming was attributed to the increase in Ca concentration in soil solution (Galindo and Bingham, 1977) and to a decrease in charge density (Goedert et al., 1975). Thus, in the absence of competition from Ca, the increased negative charge at higher pH-induced by liming is likely to result in an increase in K retention. There is clearly a need for greater understanding of K-lime interactions. As recently pointed out by Syers (1998) “the extent to which K fertilizer application rates should be varied as soil pH varies requires further evaluation.”
252
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
2.
SECONDARY NUTRIENTS
SULFUR. Addition of lime has often been observed to increase the concentration of anions, such as SO22 4 in soil solution (Bolan et al., 1988) and several reasons have been proposed to explain this (Korentajer et al., 1983). These include: (i) SO22 4 mineralized from soil organic matter by microorganisms growing in a more favorable pH environment; (ii) SO22 4 released from organic matter by chemical hydrolysis; (iii) adsorbed SO22 released from the soil surface; and (iv) SO22 4 4 released from sparingly soluble Fe and Al hydroxy sulfates, which become more soluble at higher pH. After liming an acid soil, Elkins and Ensminger (1971) observed an increase in the SO22 4 concentration in the soil solution with a corresponding increase in the S content of soybean plants grown in the soil. However, Bolan et al. (1988) indicated that in the absence of active uptake by plants, any SO22 4 released into soil solution by liming is susceptible to leaching and may be lost to subsoil horizons. Because general aspects of S biogeochemical cycling in the soil – plant system have already been reviewed elsewhere (e.g., Saggar et al., 1998) the focus here will be on ways in which liming affects the processes that regulate S cycling. The two processes that are most likely to be affected by liming include S adsorption/ desorption (Bolan et al., 1988) and S mineralization (Grego et al., 2000). Although the precise mechanism by which SO22 is adsorbed by soil 4 components is still arguable, it is clear that SO22 4 retention decreases rapidly with increasing pH. As a general rule, many soils adsorb little, if any, SO22 4 once pH exceeds about 6 (Marsh et al., 1987; Bolan et al., 1988). Liming of variable charge soils has often been shown to decrease the retention of anions, such as SO22 (Marsh et al., 1987; Bolan et al., 1988) and HPO22 4 4 (Naidu et al., 1990) and increase the retention of cations, such as nutrient ions (Adams, 1984) and heavy metals (Helmke and Naidu, 1996). In general, liming invariably increases soil pH, thereby decreasing the positive charge (i.e., AEC) and hence the adsorption of SO22 and HPO22 4 4 ; whereas an increase in Ca concentration through liming can increase anion retention. Calcium-induced anion adsorption by variable charge soils has been reported. For example, Bolan et al. (1993) observed that the adsorption of SO22 ions by variable-charge soils was higher in the presence of Ca than 4 K. Various hypotheses have been postulated for the increase in SO22 4 adsorption in the presence of Ca. Firstly, the increase in SO22 adsorption in the presence of 4 Ca has been related to the formation of a surface complex between the anion and Ca. This involves coordination of one Ca to two adsorbed anion groups, reducing the repulsive force between two adjacent anion groups, thereby enhancing further adsorption. Secondly, increased adsorption of SO22 4 at higher levels of Ca addition has been attributed to precipitation reactions occurring at high pH (. 7.0). Thirdly, specific adsorption of Ca by hydrous oxides has been
SOIL ACIDIFICATION AND LIMING INTERACTIONS
253
shown to increase the positive charge on the surface, thereby increasing the adsorption of anions. Bolan et al. (1993) observed that the increase in positive charge with Ca adsorption accounted for most of the increase in SO22 4 adsorption at low levels of Ca (, 0.002 M) in solution. The role of positive charge in SO22 adsorption by soils has been well documented (Marsh et al., 4 1987). Soil column studies have demonstrated that SO22 4 adsorption capacity of the soil is a major factor determining leaching of SO4 –S. Leaching of SO22 4 from a New Zealand soil with high S retention capacity increased several-fold when pH was increased from 4.7 to 7.0 by application of CaCO3 (Bolan et al., 1988). The lime-induced leaching of SO22 4 was attributed mainly to the large decrease in SO22 4 adsorption that occurred when pH was raised and, possibly, to greater S mineralization in the limed soil. In situations where S supply is marginal, enhanced SO22 4 leaching after liming could lead to S deficiency. Decreases in the uptake of S by corn (Zea mays L.) from two limed soils which were subjected to leaching under green house conditions provided some evidence that leachinginduced S losses may lead to S deficiencies in plants grown in limed soils (Korentajer et al., 1983). Unfortunately, similar data for field-grown crops are lacking. Several authors allude to the possibility that liming may accelerate the mineralization of organic S (Korentajer et al., 1983; Bolan et al., 1988), though the significance of this effect is open to question. Recent reviews of organic S transformations in cultivated and grassland soils (Saggar et al., 1998) do not list pH amongst the factors affecting S mineralization. Under field conditions, very little is known about S mineralization rates and there is no documented evidence that S mineralization rates are pH-sensitive. Concern over the effects on S nutrition in crops impacted by low or declining atmospheric inputs in some countries has renewed the interest in soil S, particularly in the 1980s (McGrath and Zhao, 1995). Much work has been conducted on lime-sulfate interaction and leaching of SO22 in simplified soil 4 systems (e.g., repacked columns containing homogeneous soil). However, more information is needed to develop ways of predicting the effects of lime on S leaching under field conditions. Sulfate leaching models that incorporate soil pH as a driving variable would be a useful advance on existing, empirical SO22 4 leaching indices (McGrath and Zhao, 1995). CALCIUM AND MAGNESIUM. Liming materials supply Ca and Mg to soil and one of the primary purposes of liming is to overcome the deficiency of basic cations. An increase in pH through liming increases the net negative charge, thereby increasing the adsorption of cations; whereas an increase in Ca in soil solution through liming is likely to decrease the adsorption of other cations. Therefore the resultant effect of liming on the adsorption of cations depends largely on the concentration of Ca in soil solution. Under natural leaching
254
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
conditions where most of the Ca is lost from soil solution, liming of soils may not necessarily cause increased leaching of subsequently added K and Mg (Goedert et al., 1975; Grove et al., 1981). It is possible, however, that liming a soil may lead to displacement of other cations already present in the soil, hence inducing leaching if there is a water flux (Edmeades, 1982). Increasing CEC without Mg addition reduces Mg saturation and such limeinduced Mg deficiencies can be quite striking. As soil pH is raised to the vicinity of 5.5, Mg has been shown to precipitate possibly as a mixture of Al, Mg double hydroxide or as poorly ordered Mg silicate, rendering it less soluble and available to plants (Grove et al., 1981). Plants derive their nutrients including the cations (Ca, Mg, and K) directly from the soil solution. The concentrations of Ca, Mg, and K in the soil solution are determined primarily by cation exchange equilibria (Curtin and Smillie, 1995). Addition of lime can alter the solution concentrations of these cations as a consequence of: (a) the input of Ca (and Mg in the case of dolomitic lime), and (b) pH-induced changes in the extent and nature of the cation exchange complex, which may shift the equilibrium between solution and exchangeable cations. Addition of lime usually increases the contribution of organic matter to CEC (Helling et al., 1964; Curtin et al., 1998a). As organic and mineral exchange sites differ considerably in their affinity or selectivity for cations (Baes and Bloom, 1988; Suarez and Simunek, 1997), changes in the relative proportions of organic and mineral sites may have some effect on the distribution of cations between the exchange and solution phases of soil. Several studies have demonstrated that organic matter exhibits a preference for Ca over Mg. For example, Suarez and Simunek (1997) and Curtin et al. (1998b) reported a selectivity constant for Ca – Mg exchange of about 4 for organic matter compared to , 1 (i.e., little or no preference) for smectite clay. An increase in the proportion of organic sites following liming might be expected to lead to greater selectivity for Ca over Mg but experimental results have been inconsistent. Edmeades and Judd (1980) reported that liming New Zealand soils generally increased selectivity for Ca over Mg. In contrast, Curtin et al. (1998b) concluded that Ca – Mg selectivity was insignificantly affected by pH in Canadian prairie soils. Information of the effects of pH on K selectivity is sparse. Data for a sandy loam from South Carolina indicated increased selectivity for Ca relative to K when pH was increased under laboratory conditions (Rhue and Mansell, 1988). Negative charge sites activated when soils are limed are mostly occupied by Ca with minimal effects on the levels of exchangeable Mg (Hochman et al., 1992). However, there have been frequent reports of decreases in exchangeable Mg following the use of calcitic limestone (Grove et al., 1981; Edmeades et al., 1985; Myers et al., 1988; Riggs et al., 1995a). This phenomenon has been associated mainly with highly weathered soils (Oxisols and Ultisols), but it has also been observed in less weathered soils in Great Britain (Riggs et al., 1995a). The magnitude of the effect may be large enough to be of some concern in
SOIL ACIDIFICATION AND LIMING INTERACTIONS
255
relation to Mg supply to plants and grazing animals. In a laboratory study, for example, raising pH of Iowa soils (initial pH 3.8 – 5.1) to near neutrality reduced exchangeable Mg by 17 – 34% of that initially present (Myers et al., 1988). Riggs et al. (1995a) showed that the decrease in exchangeable Mg in British soils treated with calcitic lime was directly proportional to the initial exchangeable Al content. Adsorption of soluble Mg on to newly-precipitated hydroxy-Al polymers (a process referred to as Mg fixation ) has been proposed as a mechanism responsible for lime-induced reduction in exchangeable Mg (Grove et al., 1981). There is evidence that lime-induced reductions in exchangeable Mg may occur even in low Al soils where Mg fixation is unlikely (Edmeades et al., 1985). The concentration of Mg in soil solution can either increase or decrease after liming (Curtin and Smillie, 1983). Increased concentrations of soluble Mg can be attributed to displacement of exchangeable Mg by limeCa. In an open system where leaching can occur, this may result in limeinduced loss of Mg from the soil, as suggested by Edmeades et al. (1985). Decreases in soil solution Mg following liming (Curtin and Smillie, 1986) could be due either to adsorption of Mg by newly activated cation exchange sites or to Mg fixation. Changes in the concentration of soluble Mg as a result of liming are usually small compared with the large increase in the solution concentration of Ca. Thus, the ratio of Mg-to-Ca in the soil solution is invariably decreased by liming (Curtin and Smillie, 1995). Even when soil available Mg is not affected by liming, plant Mg may be depressed because of the antagonistic effect of Ca on Mg uptake (Edmeades et al., 1983). Decreased herbage Mg associated with low soil Mg/Ca ratios is believed to be a factor causing increased incidence of hypomagnesemia (grass tetany) in animals grazing limed pastures (Thomson, 1982). Calcium-induced Mg deficiency has also been proposed as a possible cause of negative plant yield responses to lime. Carron (1991) observed Ca-induced Mg deficiency in potgrown clover (Trifolium repens L.) when the ratio of exchangeable Ca to exchangeable Mg exceeded 20. Awareness of the potential for Mg deficiency in plants and animals after use of calcitic lime has stimulated an interest in use of dolomitic limestone. Although the use of dolomitic limestone is more likely to provide adequate levels of Mg for plants and grazing animals than is calcitic limestone, neither exchangeable Mg nor the ratio of exchangeable Mg to Ca adequately predicted Mg availability to pot-grown ryegrass (Lolium perenne E.) (Riggs et al., 1995b).
3.
TRACE ELEMENTS
With the exception of Mo, plant availability of most other trace element nutrients decreases with liming mainly due to decrease in the concentration of
256
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
these elements in soil solution. For example, most problems with Fe nutrition are encountered when pH is raised by liming, resulting in a sufficient depression of Fe solubility to limit uptake in crops (Marschner, 1995). This phenomenon is often referred to as “lime-induced iron chlorosis.” The effect of pH . 6 in lowering free metal ion activities in soils has been attributed to the increase in pH-dependent surface charge on oxides of Fe, Al, and Mn (Stahl and James, 1991), chelation by organic matter, or precipitation of metal hydroxides (Lindsay, 1971). At pH above 7.0 the bioavailability of Zn to crops is substantially reduced. Severe Zn deficiencies are often associated with alkaline and calcareous soils; in these soils, acidification of root zone may prove an efficient method to increase the bioavailability of Zn to plants (Fenn et al., 1990). The effect of pH on the activity of Zn in solution in naturally acid soils is found to decrease with increasing pH. The gradual decrease in Zn activity with increasing pH is attributed to increasing CEC (Shuman, 1986). Similarly, Stahl and James (1991) observed that increasing surface charge due to liming increased Zn retention. In general, both the CEC and the total amount of Zn removed from soil solution increased with liming. On the other hand, strong complexation of Cu by soil organic matter is believed to be an important factor in explaining why Cu deficiencies are not as prevalent as Zn deficiencies in limed soils, even though the two cations show similar diminution in solubility with increasing pH (Lindsay, 1971). Precipitation of Cu contaminated industrial waste is usually achieved using lime or sodium hydroxide (caustic). Precipitation as cupric oxide, which is very effective between pH 9 and 10.3 using lime seems to offer distinct advantages with respect to cost and handling. For example, residual concentration of 0.2– 1.1 mg l21 has been achieved for Cu in the timber treatment effluent using lime precipitation (Patterson, 1985). Unlike most other trace elements, bioavailability of Mo in soils is greatest under alkaline pH than under acidic condition. Liming acid soils often helps to correct Mo deficiency; liming may substitute for Mo fertilization by releasing Mo from soils into forms readily bioavailable for plant uptake. Conversely, plant response to application of Mo under field conditions is more effective on acid soils (Adriano, 2001). However, liming could only increase the amount of plantavailable Mo on soils which have a reserve of Mo. There are some soils in New Zealand (e.g., some highly weathered “gumland” soils of Northland and deep acid peats in the Waikato) that have no reserve Mo and hence liming is ineffective in raising available Mo. In such cases Mo must be applied as a fertilizer.
B. HEAVY METALS Liming is increasingly being accepted as an important management tool in reducing the toxicity of heavy metals in soils. In addition to the traditional
SOIL ACIDIFICATION AND LIMING INTERACTIONS
257
agricultural lime, a large number of studies have examined the potential value of other liming materials as immobilizing agents in reducing the bioavailability of a range of heavy metals in soils (Table IX). In this regard Cd contamination of agricultural soils is of particular concern because this metal reaches the food chain through regular use of Cd containing fertilizer materials, such as single superphosphates. Also it remains mobile even at about neutral pH. Low soil pH and soils of low CEC induce a greater uptake of soluble Cd by plants. The Cd in soils can be immobilized by increasing the soil pH through the addition of liming materials. One benefit arises from the antagonistic effect from Ca added through liming, which may serve to depress Cd uptake by competing for exchange sites at the root surface. Limited Cd uptake may also arise from increases in Cd adsorption caused by increases in pH that induce increases in negative charge (Temminghoff, 1998). However, adsorption may decrease with an increase in Ca concentration due to a decrease in activity coefficient, increase of inorganic complexation and increase in Ca competition. The resultant effect of liming on Cd adsorption and uptake depends on the relative change in pH and Ca concentration in soil solution. Liming, as part of the normal cultural practices, has often been shown to reduce the concentration of Cd and other heavy metals in the edible parts of a number of crops (Table IX). Addition of other alkaline materials such as coal fly ash has also been shown the decrease the Cd contents of plants (Lagerwerf, 1971; Adriano et al., 1982). In these cases, the effect of liming materials in decreasing Cd uptake by plants has been attributed to both decreased mobility of Cd in soils and to the competition between Ca and Cd ions on the root surface. It is also possible that above pH 7, solubility and uptake of Cd can be enhanced due to facilitated complexation of Cd with humic or organic acids (Naidu and Harter, 1998). Arsenic can be precipitated using sodium sulfides, lime, ferric sulfate, ferric chloride and alum. The primary mechanism of As removal is by precipitation as hydroxide. Lime has been considered as a better and cheaper treatment chemical for As in industrial wastewater. Arsenic in industrial waste, which also contains other heavy metals in solution, can be concurrently co-precipitated upon precipitation of the heavy metals. The effect of liming soils on As mobility has been rather inconsistent. Calcium from lime forms calcium arsenate (Ca(AsO4)2 and because the solubility product constant of this compound is greater than that for Fe and Al arsenates, the role of Ca in the sorption process of As is not as clear cut as the role of Fe and Al. For this reason, liming is not practiced widely to overcome As toxicity in soils, although liming has been shown to reduce the concentration of water soluble As in soils (Jones et al., 1997). Recently, Heeraman et al. (2001) have observed that liming of acid mine-soils contaminated with Hg and As has resulted in an increase in the uptake of these elements because of the lime-induced improvement in plant growth. In a limed loamy soil treated with HgCl2, reduction of Hg salts to Hg8
258
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
increased, resulting in increased volatilization losses (Frear and Dills, 1967; Landa, 1978). Manganese uptake by plants is more closely related to the soil pH than is the uptake rate of any other micronutrient (Marschner, 1995). Marked increases in the bioavailability of Mn should be expected when soil pH decreases below 5.5. Manganese toxicity in acidic soils can be easily alleviated through adsorption and precipitation reactions by lime application at rates sufficient to raise the soil pH to about 6.5 (Jauregui and Reisenauer, 1982). Phytotoxicity of nickel (Ni) can be alleviated rather easily by increasing the levels of Ca through liming; Ca competes with Ni for plant uptake, thereby reducing the uptake of Ni. Thus liming serpentine soils containing toxic levels of Ni and agricultural soils treated with alkaline stabilized biosolid helps to overcome the phytotoxicity of Ni (Bolton, 1975). Selenium (Se) is quite bioavailable in well-aerated, alkaline soils where it occurs primarily as selenates. In acid soils selenite is formed which is sparingly soluble and generally unavailable to plants. In general, liming the soil could enhance the uptake of Se by plants (Adriano, 2001). The uptake of lead (Pb) is often found to decrease with liming, which is attributed to increased adsorption/precipitation at high pH, and competition between Pb and other cations for uptake (Cox and Rains, 1972). Basta and Tabatabai (1992) observed positive correlation between Pb sorption by soils and soil pH. Calcium addition through liming causes an inhibition of the translocation of Pb from root to shoot. Removal of Cr (III) from industrial effluent is achieved using lime or magnesium oxide to precipitate as chromic hydroxide. Precipitation is reported to be most effective at pH 8.5– 9.5, due to the low solubility of chromic hydroxide in that range (Patterson, 1985). This method will decrease Cr concentration to very low levels of Cr and hence precipitation systems are very widely accepted by major tanneries. Recently, Bolan and Thiagarajan (2001) examined the effect of liming materials on the adsorption and plant availability of Cr(V) and Cr(III) species. Addition of liming materials to soils increased the retention of Cr(III) but had the opposite effect on the retention of Cr(VI). The liming materials were found to be effective in reducing the phytotoxicity of Cr(III) but not Cr(VI). Addition of the liming materials decreased the concentration of the soluble Cr(III), the main reason for the decrease in the phytotoxicity of Cr(III) in the presence of liming material.
VI. CONCLUSIONS AND FUTURE RESEARCH NEEDS Soil acidification is a natural process, but it can be accelerated by certain industrial and farming activities. Pyrite oxidation causing acid drainage and acid
SOIL ACIDIFICATION AND LIMING INTERACTIONS
259
precipitation through the emission of SOx and NOx gases are the two most important industrial-borne processes of acidification. However, under managed farming systems, most of the acidity is generated from fertilizer application and from C, N and S biogeochemical cycles in soils. Soil acidification reduces the availability of major plant nutrients and induces the solubilization of metals that can lead to phytoaccumulation. Various liming materials are used to neutralize the acidity, there by overcoming the problems associated with acidification. The primary objective of liming soils used for agricultural production is to reduce the concentrations of Al and Mn, which are more soluble in acid soils and can be phytotoxic. Liming also enhances the rate of decomposition of organic matter and the subsequent transformation and release of nutrient ions from it. For example, liming has been identified for the enhanced leaching of NO2 3 in cultivated soils and the consequential contamination of ground water. This is attributed to increased mineralization and nitrification of organic N and the supply of co-cation (Ca) for NO2 3 movement in soils. Liming has been shown to reduce the amount of P fertilizer required to boost yield in some soils. This reduction in P requirements results directly from an increased solubilization of soil P and its subsequent uptake and/or indirectly from an increase in P uptake due to reduced Al and Mn toxicity. This is termed as limeinduced P-sparing effect that has been observed at normal liming rates to generate 11– 20 kg P ha21 year21 in some sedimentary-type soils having low pH (, 5.5). Liming is increasingly being practiced as a management tool to immobilize heavy metals in soils, biosolids and mine tailings, thereby reducing their availability for plant uptake and leaching to ground water. In fact, alkalinization of biosolids is commonly practiced to enhance their immobilization potential for heavy metals (Hsiau and Lo, 1998). Several reasons have been attributed to the lime-induced immobilization of heavy metals: increases in negative charge (CEC) in variable charge soils; formation of strongly-bound hydroxy metal species; precipitation of metals as hydroxides; and sequestration due to enhanced microbial activity. However, in soils with low cation exchange capacity, liming may increase the plant availability of heavy metals due to the exchange of limeborne Ca with the heavy metal ions and subsequent increase in their concentration in soil solution. The net effect of liming on heavy metal transformation in these soils largely depends on the extent of pH change and Ca release from the liming material. Lime is used to enhance the natural attenuation of metals in contaminated soils. Alkaline stabilized biosoilds are now becoming increasingly popular as a liming material because of the complementary mitigating effect from organic matter in immobilizing metals in sols. Lime-induced mobilization of nutrient ions and immobilization of heavy metals are important in sustainable agricultural production and soil environmental protection. Future research should aim to focus on the development of methods to quantify lime-enhanced mobilization of
260
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
nutrient ions and lime-induced immobilization of heavy metals in soils under field conditions and to explore further the role of liming in remediating contaminated soils. This will entail elucidation of the role of liming in various biogeochemical processes in the rhizosphere and the direct effects on microbial consortium and root-mycorrhizal association.
REFERENCES Abraham, C. T., Sreedharan, C., and Pillai, G. R. (1980). Effect of nitrogen and lime on forage quality of dinnath grass. Forage Res. 6, 95–98. Adams, F. (1984). In “Soil Acidity and Liming”. Soil Sci. Soc. Am, Madison, WI. Adams, S. N. (1986). The interaction between liming and forms of nitrogen-fertilizer on established grassland. J. Agri. Sci. 106, 509– 513. Adams, J. F., Adams, F., and Odom, J. W. (1982). Interaction of phosphorus rates and soil pH on soybean yield and soil solution composition of two phosphorus-sufficient Ultisols. Soil Sci. Soc. Am. J. 46, 323 –328. Adcock, C. W., Abaye, A. O., and Alley, M. M. (1999). Subsurface liming effects on cotton development and root growth: A greenhouse study. Commun. Soil Sci. Plant Anal. 30, 317– 327. Adriano, D. C. (2001). Biogeochemistry, Bioavailability and Risks of Metals. In “Trace Elements in Terrestrial Environments” (2nd ed.). Springer, New York. Adriano, D. C., Page, A. L., Elseewi, A. A., and Chang, A. C. (1982). Cadmium availability to sudangrass grown on soils amended with sewage sludge and fly ash. J. Environ. Qual. 11, 197–203. Agbenin, J. O. (1996). Phosphorus sorption by three cultivated savanna Alfisols as influenced by pH. Fert. Res. 44, 107– 112. Albasel, N., and Cottenie, A. (1985). Heavy-metals uptake from contaminated soils as affected by peat, lime, and chelates. Soil. Sci. Soc. Am. J. 49, 386–390. Alberts, J. J., Schindle, J. E., Miller, R. W., and Nutter, D. E. (1974). Elemental mercury evolution mediated by humic acid. Science 184, 895– 896. Alexander, M. (1977). In “Introduction to Soil Microbiology” (2nd ed.). Wiley, New York. Alibrahim, A., Sayag, D. R., and Andre, L. E. (1988). Effect of liming on the fixation and release of ammonium and potassium by a trioctahedral vermiculite. Agrochimica 32, 229–235. Alva, A. K., Graham, J. H., C, A., and Anderson, C. A. (1995). Soil and copper effects on young ‘Hamlin’ orange trees. Soil Sci. Soc. Am. J. 59, 481–487. Amarasiri, S. I., and Olsen, S. R. (1973). Liming as related to solubility of P and plant growth in an acid tropical soil. Soil Sci. Soc. Am. Proc. 37, 716–721. Anderson, A., and Siman, G. (1991). Levels of Cd and some other trace-elements in soils and crops as influenced by lime and fertilizer level. Acta. Agri. Scand. 41, 3–11. Andrew, C. S., and Johnson, A. D. (1976). Effect of calcium, pH and nitrogen on the growth and chemical composition of some tropical and temperate pasture legumes. II. Chemical composition (calcium, nitrogen, potassium, magnesium, sodium and phosphorus). Aust. J. Agric. Res. 27, 625–636. Arnold, G., Vanbeusichem, M. L., and van Diest, A. (1994). Nitrogen mineralization and Hþ transfers in a scots pine (Pinus sylvestris L) forest soil as affected by liming. Plant Soil 161, 209–218. Backes, E. A., Mclaren, R. G., Rate, A. W., and Swift, R. S. (1995). Kinetics of cadmium and cobalt desorption from iron and manganese oxides. Soil Sci. Soc. Am. J. 59, 778– 785. Badora, A., and Filipek, T. (1988). An assessment of rehabilitation of strongly acidic sandy soils. Adv. Geol. 31, 681–688.
SOIL ACIDIFICATION AND LIMING INTERACTIONS
261
Baes, A. U., and Bloom, P. R. (1988). Exchange of alkaline–earth cations in soil organic-matter. Soil Sci. 146, 6 –14. Bailey, J. S. (1995). Liming and nitrogen efficiency—some effects of increased calcium supply and increased soil-pH on nitrogen recovery by perennial ryegrass. Commun. Soil. Sci. Plant Anal. 26, 1233–1246. Barrow, N. J. (1984). Modeling the effects of pH on phosphate sorption by soils.. J. Soil Sci. 35, 283 –297. Barrow, N. J. (1985). Reactions of anions and cations with variable-charge soils. Adv. Agron. 38, 183 –230. Barrow, N. J. (1986). Testing a mechanistic model. IV. Describing the effects of pH on zinc retention by soils. J. Soil Sci. 37, 295– 302. Barshad, I., and Kishk, F. M. (1970). Factors affecting potassium fixation and cation exchange capacities of soil vermiculite clays. Clays Clay Miner. 18, 127–131. Bartlett, R. J. (1991). Chromium cycling in soils and water: Links, gaps, and methods. Environ. Health Perspect. 92, 17–24. Bartlett, R. J., and James, B. R. (1979). Behavior of chromium in soils: III. Oxidation. J. Environ. Qual. 8, 31 –35. Bartlett, R. J., and Kimble, J. M. (1976). Behavior of chromium in soils: II. Hexavalent forms. J. Environ. Qual. 5, 383– 386. Bartlett, R. J., and Picarelli, C. J. (1973). Availability of boron and phosphorus as affected by liming an acid potato soil. Soil Sci. 116, 77– 83. Basta, N. T. (2000). Examples of case studies of beneficial reuse of municipal by-products. In “Land application of agricultural, industrial, and municipal by-products” (W. A. Dick, Ed.), pp. 481 –504. Soil Sci. Soc. Am, Madison, WI. Basta, N. T., and Tabatabai, M. A. (1992). Effect of cropping systems on adsorption of metals by soils. II. Effect of pH. Soil Sci. 153, 195–204. Bates, T. E. (1971). Factors affecting critical nutrient concentrations in plants and their evaluation— review. Soil Sci. 112, 116 –120. Benner, S. G., Gould, W. D., and Blowes, D. W. (2000). Microbial populations associated with the generation and treatment of acid mine drainage. Chem. Geol. 169, 435–448. Bhella, H. S., and Wilcox, G. E. (1989). Lime and N influence soil acidity, nutritional status, vegetative growth, and yield of muskmelon. J. Am. Soc. Hort. Sci. 114, 606–610. Bingham, F. T., Arkley, R. J., Coleman, N. T., and Bradford, G. R. (1970). Characteristics of high boron soils in western kern county. Hilgardia 40, 193–204. Bingham, F. T., Page, A. L., Coleman, N. T., and Flach, K. (1971). Boron desorption characteristics of selected amorphous soils from Mexico and Hawaii. Soil Sci. Soc. Am. Proc. 35, 546 –550. Bingham, F. T., Page, A. L., Mitchell, G. A., and Strong, J. E. (1979). Effects of liming an acid soil amended with sewage sludge enriched with Cd, Cu, Ni, and Zn on yield and Cd content of wheatgrain. J. Environ. Qual. 8, 202–207. Binns, W. (1988). Vegetation and soils. Report Number 18, The Watt Committee on Energy. In “Air Pollution, Acid Rain and the Environment” (K. Mellanby, Ed.), pp. 13–22. Elsevier Applied Science, New York. Bisessar, S. (1989). Effects of lime on nickel uptake and toxicity in celery grown on muck soil contaminated by a nickel refinery. Sci. Total Environ. 84, 83– 90. Blamey, F. P. C., and Chapman, J. (1979). Soil amelioration effects on boron uptake by sunflowers. Commun. Soil Sci. Plant Anal. 10, 1057–1066. Blue, W. G., and Ferrer, M. S. (1986). Potassium retention in a sandy Florida spodosol with variable organic-matter concentrations and liming histories. Soil Crop Sci. Soc. Fl. 45, 90– 94. Boekhold, A. E., Temminghoff, E. J. M., and van der Zee, S. E. A. T. M. (1993). Influence of electrolyte composition and pH on cadmium sorption by an acid soil. J. Soil Sci. 44, 85–96.
262
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
Bolan, N. S., and Hedley, M. J. (2001). Role of carbon, nitrogen and sulfur cycles in soil acidification. In “Handbook of Soil Acidity” (Z. Rengel, Ed.), Marcel Dekker, New York. Bolan, N. S., and Thiagarajan, S. (2001). Retention and plant availability of chromium in soils as affected by lime and organic amendments. Aust. J. Soil Res. 39, 1091–1103. Bolan, N. S., Syers, J. K., Tillman, R. W., and Scotter, D. R. (1988). Effect of liming and phosphate additions on sulfate leaching in soils. J. Soil Sci. 39, 493–504. Bolan, N. S., Hedley, M. J., and White, R. E. (1991). Processes of soil acidification during nitrogen cycling with emphasis on legume based pastures. Plant Soil 134, 53–63. Bolan, N. S., Syers, J. K., and Sumner, M. E. (1993). Calcium induced sulfate adsorption by soil. Soil Sci. Soc. Am. J. 57, 691– 696. Bolan, N. S., Syers, J. K., and Sumner, M. E. (1996). Origin of the effect of pH on the saturated hydraulic conductivity of soils. Commun. Soil Sci. Plant Anal. 27, 2265–2278. Bolan, N. S., Naidu, R., Syers, J. K., and Tillman, R. W. (1999a). Surface charge and solute interactions in soils. Adv. Agron. 67, 88 –141. Bolan, N. S., Naidu, R., Tillman, R. W., and Syers, J. K. (1999b). Effect of anion sorption on cadmium sorption by soils. Aust J. Soil Res. 37, 445 –460. Bolton, J. (1975). Liming effects on toxicity to perennial ryegrass of a sewage sludge contaminated with zinc, nickel, copper and chromium. Environ. Pollut. 9, 295 –304. Bothe, J. V., and Brown, P. W. (1999). Arsenic immobilization by calcium arsenate formation. Environ. Sci. Technol. 33, 3806–3811. Brady, N. C., and Weil, R. R. (1999). In “Nature and Properties of Soils”. Prentice-Hall, Englewood Cliffs, NJ. Brallier, S., Harrison, R. B., Henry, C. L., and Dongsen, X. (1996). Liming effects on availability of Cd, Cu, Ni and Zn in a soil amended with sewage sludge 16 years previously. Water Air Soil Pollut. 86, 195– 206. van Breemen, N. (1991). Soil acidification and alkalisation. In “Soil Acidity” (B. Ulrich, and M. E. Sumner, Eds.), pp. 1– 7. Springer, New York. Brown, J. C. (1979). Role of calcium in micronutrient stresses of plants. Commun. Soil Sci. Plant Anal. 10, 459 –472. Brown, S., Chaney, R., and Angle, J. S. (1977). Subsurface liming and metal movement in soils amended with lime-stabilized biosolids. J. Environ. Qual. 26, 724 –732. Carron, R. A. (1991). Calcium magnesium imbalance in clovers: A negative yield response to liming. Plant Soil 134, 107–114. Carter, M. R. (1986). Microbial biomass and mineralisable nitrogen in Solonetzic soils: Influence of gypsum and lime amendments. Soil Biol. Biochem. 18, 531–537. Cary, E. E., and Allaway, W. H. (1969). Stability of different forms of selenium applied to lowselenium soils. Soil Sci. Soc. Am. Proc. 33, 571– 574. Cary, E. E., Wieczorek, G. A., and Allaway, W. H. (1967). Reactions of selenite–selenium added to soils that produce low-selenium forages. Soil Sci. Soc. Am. Proc. 31, 21 –26. Cavallaro, N., and McBride, M. B. (1980). Activities of Cu2þ and Cd2þ in soil solutions as affected by pH. Soil Sci. Soc. Am. J. 44, 729 –732. Chan, K. Y., and Heenan, D. P. (1998). Effect of lime (CaCO3) application on soil structural stability of a red earth. Aust. J. Soil Res. 36, 73 –86. Chaney, W. R., Strickland, R. C., and Lamoreaux, R. J. (1977). Phytotoxicity of cadmium inhibited by lime. Plant Soil 47, 275 –278. Chang, A. C., Page, A. L., Foster, K. W., and Jones, T. E. (1982). A comparison of cadmium and zinc accumulation by 4 cultivars of barley grown in sludge-amended soils. J. Environ. Qual. 11, 409–412. Chen, J. H., and Barber, S. A. (1990). Effect of liming and adding phosphate on predicted phosphorus uptake by maize on acid soils of 3 soil orders. Soil Sci. 150, 844 –850.
SOIL ACIDIFICATION AND LIMING INTERACTIONS
263
Chiang, S. C., Radcliffe, D. E., Miller, W. P., and Newman, K. D. (1987). Hydraulic conductivity of three south-eastern soils as affected by sodium, electrolyte concentration, and pH. Soil Sci. Soc. Am. J. 51, 1293–1299. Chlopecka, A., Bacon, J. R., Wilson, M. J., and Kay, J. (1996). Forms of cadmium, lead, and zinc in contaminated soils from southwest Poland. J. Environ. Qual. 25, 69– 79. Chowdhury, A. K., McLaren, R. G., and Swift, R. S. (1997). Effects of phosphate and lime applications on pasture Zn status. NZ J. Agric. Res. 40, 417 –424. Christensen, T. H. (1984). Cadmium soil sorption at low concentrations: Effect of time, cadmium loading, pH and calcium. Water, Air Soil Pollut. 21, 105 –114. Clay, D. E., Clapp, C. E., Dowdy, R. H., and Molina, J. A. E. (1993). Mineralisation of nitrogen in fertiliser-acidified lime-amended soils. Biol. Fert. Soils 15, 249–252. Condron, L. M., and Goh, K. M. (1989). Effects of long-term phosphatic fertilizer applications on amounts and forms of phosphorus in soils under irrigated pasture in New-Zealand. J. Soil Sci. 40, 383 –395. Condron, L. M., Tiessen, H., Trasarcepeda, C., Moir, J. O., and Stewart, J. W. B. (1993). Effects of liming on organic-matter decomposition and phosphorus extractability in an acid humic ranker soil from northwest Spain. Biol. Fert. Soils 15, 279–284. Coventry, D. R., Hirth, J. R., Reeves, T. G., and Burnett, V. F. (1985). Growth and nitrogen fixation by subterranean clover in response to inoculation, molybdenum application and soil amendment with lime. Soil Biol. Biochem. 17, 791–796. Coventry, D. R., Hirth, J. R., and Flung, K. K. H. (1987). Nutritional restraints on subterranean clover grown on acid soil used for crop–pasture rotation. Aust. J. Agric. Res. 38, 163 –176. Cowling, S. J., Gardner, M. J., and Hunt, D. T. J. (1992). Removal of heavy-metals from sewage by sulfide precipitation—thermodynamic calculations and tests on a pilot-scale anaerobic reactor. Environ. Technol. 13, 281–291. Cox, W. J., and Rains, D. W. (1972). Effect of lime on lead uptake by five plant species. J. Environ. Qual. 1, 167 –169. Crouchley, G. (1981). Guano from gannets but lime for the hills. NZ J. Agric. 142, 37 –42. Curtin, D., and Smillie, G. W. (1983). Soil solution composition as affected by liming and incubation. Soil Sci. Soc. Am. J. 47, 701–707. Curtin, D., and Smillie, G. W. (1984). Influence of liming on soluble and labile P in fertilized soil. Commun. Soil Sci. Plant Anal. 15, 177 –188. Curtin, D., and Smillie, G. W. (1986). Effects of liming on soil chemical characteristics and grass growth in laboratory and long-term field-amended soils. 1. Soil chemistry. Plant Soil 95, 15–22. Curtin, D., and Smillie, G. W. (1995). Effects of incubation and pH on soil solution and exchangeable cation ratios. Soil Sci. Soc. Am. J. 59, 1006–1011. Curtin, D., and Syers, J. K. (2001). Lime-induced changes in indices of soil phosphate availability. Soil Sci. Soc. Am. J. 65, 147–152. Curtin, D., Campbell, C. A., and Messer, D. (1996). Prediction of titratable acidity and soil sensitivity to pH change. J. Environ. Qual. 25, 1280–1284. Curtin, D., Campbell, C. A., and Jalil, A. (1998a). Effects of acidity on mineralization: pH-dependence of organic matter mineralization in weakly acidic soils. Soil Biol. Biochem. 30, 57–64. Curtin, D., Selles, F., and Steppuhn, H. (1998b). Estimating calcium–magnesium selectivity in smectitic soils from organic matter and texture. Soil Sci. Soc. Am. J. 62, 1280–1285. Deram, A., Petit, D., Robinson, B., Brooks, R., Gregg, P., and Van Halluwyn, C. (2000). Natural and induced heavy-metal accumulation by Arrhenatherum elatius: Implications for phytoremediation. Commun. Soil Sci. Plant Anal. 31, 413–421. Dick, W. A., Stehouwer, R., Bingham, J. M., Wolfe, W. E., Hao, Y., Adriano, D. C., Beeghly, J. H., and Haefner, R. J. (2000). Beneficial uses of flue gad desulphurisation by-products: Examples and case studies of land application. In “Land Application of Agricultural, Industrial, and Municipal By-products” (W. A. Dick, Ed.), pp. 505–536. Soil Sci. Soc. Am, Madison, WI.
264
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
Dinel, H., Pare, T., Schnitzer, M., and Pelzer, N. (2000). Direct land application of cement kiln dustand lime-sanitized biosolids: Extractability of trace metals and organic matter quality. Geoderma 96, 307 –320. Dolling, P. J., and Porter, W. M. (1994). Acidification rates in the central wheatbelt of Western Australia. 1. On a deep yellow sand. Aust. J. Expp. Agric. 34, 1155–1164. Edmeades, D. C. (1982). Effects of lime on effective cation exchange capacity and exchangeable cations on a range of New Zealand soils. NZ J. Agric. Res. 25, 27–33. Edmeades, D. C., and Judd, M. J. (1980). The effects of lime on the magnesium status and equilibria in some New Zealand soils. Soil Sci. 129, 156–161. Edmeades, D. C., Judd, M., and Sarathchandra, S. U. (1981). The effect of lime on nitrogen mineralization as measured by grass growth. Plant Soil 60, 177 –186. Edmeades, D. C., Smart, C. E., Wheeler, D. M., and Rys, G. (1983). Effects of lime on the chemical composition of ryegrass and white clover. NZ J. Agric. Res. 26, 473–481. Edmeades, D. C., Wheeler, D. M., and Crouchley, G. (1985). Effects of liming on soil magnesium on some soils in New Zealand. Commun. Soil Sci. Plant Anal. 16, 727– 739. Elkins, D. M., and Ensminger, L. E. (1971). Effect of soil pH on the availability of adsorbed sulphate. Soil Sci. Soc. Am. Proc. 35, 931–934. Estan, M. T., Bolarin, M. C., and Guillen, M. G. (1987). Adsorption of cadmium in lime soils— estimating parameters of various isotherms using nonlinear-regression. Agrochimica 31, 500–509. Evangelou, V. P. (1995). Acid Mine Drainage, Surface Chemistry, Molecular Oxidation Mechanisms, Microbial Role, Kinetics, Control, Ameliorates, Limitations, Microencapsulation. In “Pyrite oxidation and its control”. CRC/Lewis Press, Boca Raton, FL. Evangelou, V. P., and Zhang, Y. L. (1995). A review: Pyrite oxidation mechanisms and acid mine drainage prevention. Critical Rev. Environ. Sci. Technol. 252, 141–199. Fang, M., and Wong, J. W. C. (1999). Effects of lime amendment on availability of heavy metals and maturation in sewage sludge composting. Environ. Pollut. 106, 83– 89. Farrah, H., and Pickering, W. F. (1978). Sorption of mercury species by clay-minerals. Water Air Soil Pollut. 9, 23–31. Federer, P., and Sticher, H. (1999). Behavior of copper, zinc and cadmium in heavily polluted lime soil. Chimia 45, 228–232. Felle, H. (1988). Short-term pH regulations in plants. Physiol. Plant. 74, 583 –591. Fenn, L. B., Malstrom, H. L., Riley, T., and Horst, G. L. (1990). Acidification of calcareous soils improves zinc-absorption of pecan trees. J. Am. Soc. Hort. Sci. 115, 741 –744. Fernandes, M. L. V., and Coutinho, J. F. (1999). Effect of liming and phosphate application on Sudan grass growth and phosphorus availability in two temperate acid soils. Commun. Soil Sci. Plant Anal. 30, 855 –871. Fernandes, M. L., Abreu, M. M., Calouro, F., and Vaz, M. C. (1999). Effect of liming and cadmium application in an acid soil on cadmium availability to Sudan grass. Commun. Soil Sci. Plant Anal. 30, 1051–1062. Filius, A., Streck, T., and Richter, J. (1998). Cadmium sorption and desorption in limed topsoils as influenced by pH: Isotherms and simulated leaching. J. Environ. Qual. 27, 12 –18. Foy, C. D. (1984). Adaptation of plants to mineral stress in problem soils. CIBA F. Sympp. 102, 20–31. Frakenberger, W. T., Jr, and Karlson, U. (1994). Microbial volatilization of selenium from soils and sediments. In “Selenium in Environment” (W. T. Frankenberger, and S. Benson, Eds.), pp. 369 –388. Marcel Dekker, New York. Fransson, A. M., Bergkvist, B., and Tyler, G. (1999). Phosphorus solubility in an acid forest soil as influenced by form of applied phosphorus and liming. Scand. J. Forest Sci. 14, 538–544. Franzen, D. W., and Richardson, J. L. (2000). Soil factors affecting iron chlorosis of soybean in the red river valley of North Dakota and Minnesota. J. Plant Nutr. 23, 67–78.
SOIL ACIDIFICATION AND LIMING INTERACTIONS
265
Frazer, L. (2001). Probing the depths of a solution for acid mine drainage. Environ. Health Perspect. 109, 486–489. Friesen, D. K., Miller, M. H., and Juo, A. S. R. (1980). Liming and lime –phosphorus–zinc interactions in two Nigerian Ultisols. I. Interactions in the soil. Soil Sci. Soc. Am. J. 44, 1221–1226. Friesen, D. K., Miller, M. H., and Juo, A. S. R. (1980). Liming and lime –phosphorus–zinc interactions in two Nigerian Ultisols. II. Effects on maize root and shoot growth. Soil Sci. Soc. Am. J. 44, 1227–1232. Galindo, G.G., and Bingham, F. T. (1977). Homovalent and heterovalent cation exchange equilibria in soils with variable surface charge. Soil Sci. Soc. Am. Proc. 41, 883– 886. Gardner, E. H., Derics, I. A., and Basaraba, J. (1965). Mineralization of nitrogen in some Fraser valley soils as affected by liming. Can. J. Soil. Sci. 45, 355– 360. Godo, G. H., and Reisenauer, H. M. (1980). Plant effects on soil manganese availability. Soil Sci. Soc. Am. J. 44, 993–995. Goedert, W. J., Corey, R. B., and Syers, J. K. (1975). Lime effects on potassium equilibria in soils of Rio Grande Do Sul, Brazil. Soil Sci. 120, 107– 111. Gray, C. W., McLaren, R. G., Roberts, A. H. C., and Condron, L. M. (1999). Effect of soil pH on cadmium phytoavailability in some New Zealand soils. NZ J. Crop Hort. 27, 169–179. Gregan, P. D., Hirth, J. R., and Conyers, M. K. (1989). Amelioration of soil acidity by liming and other amendments. In “Soil Acidity and Plant Growth” (A. D. Robson, Ed.), pp. 205–264. Academic Press, New York. Grego, S., Moscatelli, M. C., Marinari, S., and Badalucco, L. (2000). The influence of liming and natural acidification on chemical and biological processes of an Italian forest soil. Agrochimica 44, 161–170. Grove, J. H., Sumner, M. E., and Syers, J. K. (1981). Effect of lime on exchangeable magnesium in variable charge soils. Soil Sci. Soc. Am. J. 45, 497–500. Gupta, U. C. (1972). Effects of Boron and lime on boron concentration and growth of forage legumes under greenhouse conditions. Commun. Soil Sci. Plant Anal. 3, 355–365. Gupta, U. C. (1979). Effect of methods of application and residual effect of molybdenum on the Mo concentration and yield of forages on podzol soils. Can. J. Soil Sci. 59, 183–189. Gupta, U. C., and Kunelius, H. T. (1980). Effect of molybdenum, lime and cropping on yield and molybdenum concentration of grass– legume mixtures and cabbage. Can. J. Plant Sci. 60, 113 –120. Gupta, U. C., and MacLeod, J. A. (1973). A boron nutrition and growth of timothy as affected by soil pH. Commun. Soil Sci. Plant Anal. 4, 389 –395. Gupta, U. C., and Winter, K. A. (1975). Selenium content of soils and crops and effects of lime and sulfur on plant selenium. Can. J. Soil Sci. 55, 161 –166. Gupta, U. G., McRae, K. B., and Winter, K. A. (1982). Effect of applied selenium on the selenium content of barley and forages and soil selenium depletion rates. Can. J. Soil Sci. 62, 145 –154. Haddad, K. S., and Kaldor, C. J. (1982). Effect of parent material, natural available soil boron, and applied boron and lime on the growth and chemical composition of lucerne on some acidic soils of the Central Tablelands of New South Wales. Aust. J. Expt. Agric. Ani. Husb. 22, 317 –323. Han, D. H., and Lee, J. H. (1996). Effects of liming on uptake of lead and cadmium by Raphanus sativa. Arch. Environ. Cont. Toxicol. 31, 488 –493. Harter, R. D. (1991). Micronutrient adsorption– desorption reactions in soils. SSSA. In “Micronutrients in Agriculture” (J. J. Mortvedt,, F. R. Cox,, L. M. Shuman, and R. M. Welch, Eds.), (2nd ed.). 4, pp. 59–88. Soil Sci. Soc. Am, Madison, WI. Haynes, R. J. (1983). Soil acidification induced by leguminous crops. Grass Forage Sci. 38, 1–11. Haynes, R. J. (1984). Lime and phosphate in the soil–plant system. Adv. Agron. 37, 249– 467.
266
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
Haynes, R. J. (1990). Active ion uptake and maintenance of cation–anion balance: A critical examination of their role in regulating rhizosphere pH. Plant Soil 126, 247– 264. Haynes, R. J., and Ludecke, T. E. (1981). Effect of lime and phosphorus applications on concentrations of available nutrients and on P, Al and Mn uptake by 2 pasture legumes in an acid soil. Plant Soil 62, 117–128. Haynes, R. J., and Naidu, R. (1998). Influence of lime, fertilizer and manure applications on soil organic matter content and soil physical conditions: A review. Nutr. Cycl. Agroecosyst. 51, 123–137. Haynes, R. J., and Swift, R. S. (1988). Effects of lime and phosphate additions on changes in enzymeactivities, microbial biomass and levels of extractable nitrogen, sulfur and phosphorus in an acid soil. Biol. Fert. Soils 6, 153–158. He, Q. B., and Singh, B. R. (1994). Plant availability of cadmium in soils. 2. Factors related to the extractability and plant uptake of cadmium in cultivated soils. Acta. Agric. Scand. 43, 142– 150. Hechtbuchholz, C., and Foy, C. D. (1981). Effect of aluminium toxicity on root morphology of barley. Plant Soil 63, 93–95. Heeraman, D. A., Claassen, V. P., and Zasoski, R. J. (2001). Interaction of lime, organic matter and fertilizer on growth and uptake of arsenic and mercury by Zorro fescue (Vulpia myuros L. ). Plant Soil 234, 215–231. Helling, C. S., Chesters, G., and Corey, R. B. (1964). Contribution of organic matter and clay to soil cation exchange capacity as affected by pH of the saturating solution. Soil Sci. Soc. Am. J. 28, 517–520. Helmke, P. A., and Naidu, R. (1996). Fate of contaminants in the soil environment: Metal contaminants. In “Contaminants and the Soil Environment in the Australasia –Pacific Region” (R. Naidu,, R. S. Kookana,, D. Oliver,, S. Rogers, and M. J. McLaughlin, Eds.), pp. 69– 94. Kluwer Academic Publishers, Dordrecht, The Netherlands. Helyar, K. R. (1976). Nitrogen cycling and soil acidification. J. Aust. Inst. Agric. Sci. 42, 217 –221. Helyar, K. R. (1990). The management of acid soils. In “Plant– Soil Interactions at Low pH” (R. J. Baligar,, V. C. Baligar, and R. P. Murrmann, Eds.), pp. 365 –382. Kluwer Academic Publishers, Dordrecht, The Netherlands. Hemphill, D. D., Weber, M. S., Jr., and Jackson, T. L. (1982). Table beet yield and boron deficiency as influenced by lime, nitrogen and boron. Soil Sci. Soc. Am. J. 46, 1190–1192. Hesterberg, D., Bril, J., and Delcastilho, P. (1993). Thermodynamic modeling of zinc, cadmium, and copper solubilities in a manured, acidic loamy-sand topsoil. J. Environ. Qual. 22, 681 –688. Hingston, F. J. (1981). A review of anion adsorption. In “Adsorption of Inorganics at Solid– Liquid Interfaces” (M. A. Anderson, and A. J. Robin, Eds.), pp. 51 –90. Ann Arbor Science, Ann Arbor, MI. Hochman, Z., Edmeades, D. C., and White, E. (1992). Changes in effective CEC and exchangeable aluminium with soil pH in lime-amended soil. Aust. J. Soil Res. 30, 177 –187. Hodgson, J. F., Lindsay, W. L., and Trierwei, J. F. (1966). Micronutrient cation complexing in soil solution. 2. Complexing of zinc and copper in displaced solution from calcareous soils. Soil Sci. Soc. Am. J. 30, 723–726. Holford, I. C. R., Schweitzer, B. E., and Crocker, G. J. (1994). Long-term effects of lime on soil phosphorus solubility and sorption in eight soils. Aust. J. Soil Sci. 12, 795 –803. Hooda, P. S., and Alloway, B. J. (1996). The effect of liming on heavy metal concentrations in wheat, carrots and spinach grown on previously sludge-applied soils. J. Agric. Sci. 127, 289–294. Hortenstine, C. C., and Webber, L. R. (1981). Growth and cadmium uptake by spinach and corn in an acid typic hapludalf as affected by cadmium, peat, and lime additions. Soil Cropp. Sci. Soc. Fl. 40, 40–44. Howard, D. D., and Essington, M. E. (1998). Effects of surface-applied limestone on the efficiency of urea-containing nitrogen sources for no-till corn. Agron. J. 90, 523–528.
SOIL ACIDIFICATION AND LIMING INTERACTIONS
267
Hsiau, P. C., and Lo, S. L. (1998). Fractionation and leachability of Cu in lime-treated sewage sludge. Water Res. 32, 1103–1108. Israel, D. W., and Jackson, W. A. (1978). The influence of nitrogen nutrition on ion uptake and translocation by leguminous plants. In “Mineral Nutrition of Legumes in Tropical and Subtropical Soil” (C. S. Andrew, and E. J. Kamprath, Eds.), pp. 113 –129. CSIRO, Melbourne. Jackson, A. P., and Alloway, B. J. (1991). The transfer of cadmium from sewage-sludge amended soils into the edible components of food crops. Water Air Soil Pollut. 57, 873 –881. James, B. R. (1996). The challenge of remediating chromium contaminated soil. Environ. Sci. Technol. 30, 248–251. Jauregui, M. A., and Reisenauer, H. M. (1982). Calcium-carbonate and manganese-dioxide as regulators of available manganese and iron. Soil Sci. 134, 105–110. Jeffery, J. J., and Uren, N. C. (1983). Copper and zinc species in the soil solution and the effects of soil pH. Aust. J. Soil Res. 21, 479– 488. Jiang, Q. Q., and Singh, B. R. (1994). Effect of different forms and sources of arsenic on crop yield and arsenic concentration. Water Air Soil Pollut. 74, 321–343. John, M. K., and van Laerhoven, C. J. (1972). Lead uptake by lettuce and oats as affected by lime, nitrogen and sources of lead. J. Environ. Qual. 1, 172 –199. John, M. K., and Van Laerhoven, C. J. (1976). Effects of sewage sludge composition, application rate, and lime regime on plant availability of heavy-metals. J. Environ. Qual. 5, 246–251. Jones, J. P., and Fox, R. L. (1978). Phosphorus nutrition of plants influenced by manganese and aluminium uptake from an oxisol. Soil Sci. 126, 230 –236. Jones, C. A., Inskeep, W. P., and Neuman, D. R. (1997). Arsenic transport in contaminated mine tailings following liming. J. Environ. Qual. 26, 433–439. Kemper, D. W., and Sorensen, R. C. (1984). Comparative effects of nitrogen and sulfur fertilization and liming on 3 crops grown on 4 soils. Agron. J. 66, 92–97. Kinniburgh, D. G., and Jackson, M. L. (1981). Cation adsorption by hydrous metal oxides and clay. In “Adsorption of Inorganics at Solid–Liquid Interfaces” (M. A. Anderson, and A. J. Rubin, Eds.), pp. 91–160. Ann Arbor Science, Ann Arbor, MI. de Klein, C. A. M., Monaghan, R. M., and Sinclair, A. G. (1997). Soil acidification: A provisional model for New Zealand pastoral systems. NZ J. Agric. Res. 40, 541–557. Klemmedson, J. O., Rehfuess, K. E., Makeschin, F., and Rodenkirchen, H. (1989). Nitrogen mineralisation in lime- and gypsum-amended substrates from ameliorated acid forest soils. Soil Sci. 147, 55 –63. Korentajer, L., Byrnes, H. B., and Hellums, T. D. (1983). The effect of liming and leaching on sulfur supplying capacity of soils. Soil Sci. Soc. Am. J. 47, 525–530. Krebs, R., Gupta, S. K., Furrer, G., and Schulin, R. (1998). Solubility and plant uptake of metals with and without liming of sludge-amended soils. J. Environ. Qual. 27, 18–23. Lagerwerf, J. V. (1971). Uptake of cadmium, lead and zinc by radish from soil and air. Soil Sci. 111, 129 –134. Landa, E. R. (1978). Microbial aspects of volatile loss of applied mercury(ii) from soils. J. Environ. Qual. 7, 84 –86. Lee, C. R., and Craddock, G. R. (1969). Factors affecting plant growth in high-zinc medium. 2. influence of soil treatments on growth of soybeans on strongly acid soil containing zinc from peach sprays. Agron. J. 61, 565– 567. Lehoczky, E., Marth, P., Szabados, I., and Szomolanyi, A. (2000). The cadmium uptake by lettuce on contaminated soils as influenced by liming. Commun. Soil Sci. Plant Anal. 31, 2433–2438. Li, Y. M., Chaney, R. L., Schneiter, A. A., and Johnson, B. L. (1996). Effect of field limestone applications on cadmium content of sunflower (Helianthus annuus L. ) leaves and kernels. Plant Soil 180, 297– 302. Li, S. T., Lin, B., and Zhou, W. (2001). Soil organic sulfur mineralization in the presence of growing plants under aerobic or waterlogged conditions. Soil Biol. Biochem. 33, 721 –727.
268
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
Lindsay, W. L. (1971). In “Chemical Equilibria in Soils”. Wiley, New York. Lipsett, J., Pinkerton, A., and David, D. J. (1979). Boron deficiency as a factor in the reclamation by liming of a soil contaminated by mine waste. Environ. Pollut. 20, 231– 240. Liu, W. C., Lund, L. J., and Page, A. L. (1989). Acidity produced by leguminous plants through symbiotic di-nitrogen fixation. J. Environ. Qual. 18, 529 –534. Logan, T. J., and Harrison, B. J. (1995). Physical characteristics of alkaline stabilised sewage sludge (N-Viro Soil) and their effects on soil physical properties. J. Environ. Qual. 24, 153–164. Longhurst, J. W. S. (1991). In “Acid Deposition: Origin, Impacts and Abatment Strategies”. Springer, New York. Lucas, L. N., and Blue, W. G. (1973). Effects of lime and phosphorus on selected alluvial entisols from eastern Costa Rica. 2. Forage plant responses. Tropp. Agri. 50, 63–74. Lyngstad, I. (1992). Effect of liming on mineralization of soil-nitrogen as measured by plant uptake and nitrogen released during incubation. Plant Soil 144, 247–253. MacLean, A. J. (1976). Cadmium in different plant species and its availability in soils as influenced by organic-matter and additions of lime, P, Cd and Zn. Can. J. Soil Sci. 56, 129–138. MacLean, A. J., Halstead, R. L., and Finn, B. J. (1972). Effects of lime on extractable aluminium and other soil properties and on barley and alfalfa grown in pot tests. Can. J. Soil Sci. 52, 427– 438. Magdoff, F. R., and Bartlett, R. J. (1980). Effect of liming acid soils on potassium availability. Soil Sci. 129, 12 –14. Mahler, R. J., Bingham, F. T., and Page, A. L. (1978). Cadmium-enriched sewage sludge application to acid and calcareous soils—effect on yield and cadmium uptake by lettuce and chard. J. Environ. Qual. 7, 274–281. Maier, N. A., McLaughlin, M. J., Heap, M., Butt, M., Smart, M. K., and Williams, C. M. J. (1997). Effect of current-season application of calcitic lime on soil pH, yield and cadmium concentration in potato (Solanum tuberosum L. ) tubers. Nutr. Cycl. Agroecosys. 47, 29–40. Mansell, G. P., Pringle, R. M., Edmeades, D. C., and Shannon, P. W. (1984). Effects of lime on pasture production on soils of the North Island of New Zealand. 3. Interaction of lime with phosphorus. NZ J. Agric. Res. 27, 363 –369. Marschner, H. (1995). In “Mineral Nutrition of Higher Plants” (2nd ed.). Academic Press, London. Marsh, K. B., Tillman, R. W., and Syers, J. K. (1987). Charge relationships of sulfate sorption by soils. Soil Sci. Soc. Am. J. 51, 318 –323. Martini, J. A., and Mutters, R. G. (1984). Effect of liming and fertilization on sulfur availability, mobility, and uptake in cultivated soils of South Carolina. Soil Sci. 138, 403–410. Mason, M. G., Porter, W. M., and Cox, W. J. (1994). Effect of an acidifying N fertiliser and lime on soil pH and wheat yields. 2. Plant response. Aust. J. Expt. Agric. 34, 247 –253. Mays, P. A., and Edwards, G. S. (2001). Comparison of heavy metal accumulation in a natural wetland and constructed wetlands receiving acid mine drainage. Ecol Engng 18, 251 –252. McBride, M. B., and Bouldin, D. R. (1984). Long-term reactions of copper(ii) in a contaminated calcareous soil. Soil Sci. Soc. Am. J. 48, 56–59. McGrath, S. P., and Zhao, F. J. (1995). A risk assessment of sulphur deficiency in cereals using soils and atmospheric deposition data. Soil Use Mgmt 11, 110–114. Mench, M. J., Didier, V. L., Loeffler, 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. Mongia, A. D., Singh, N. T., Mandal, L. N., and Guha, A. (1997). Effect of soil water regimes and liming on chemical changes and nutrient transformations in acid and acid sulphate soils. J. Indian Soc. Soil Sci. 45, 344– 348. Mora, M. L., Baeza, G., Pizarro, C., and Demanet, R. (1999). Effect of calcitic and dolomitic lime on physicochemical properties of a Chilean Andisol. Commun. Soil Sci. Plant Anal. 30, 339–427. Msaky, J. J., and Calvet, R. (1990). Adsorption behavior of copper and zinc in soils—influence of pH on adsorption characteristics. Soil Sci. 150, 513–522.
SOIL ACIDIFICATION AND LIMING INTERACTIONS
269
Muchovej, R. M. C., Borges, A. C., Novias, R. F., and Thiebaut, J. T. L. (1986). Effect of liming levels and Ca–Mg ratios on yield, nitrogen-content and nodulation of soybeans grown in acid cerrado. J. Soil Sci. 37, 235–240. Munns, D. N., Fox, R. L., and Koch, B. L. (1977). Influence of lime on nitrogen fixation by tropical and temperate legumes. Plant Soil 46, 591–601. Myers, J. A., McLean, E. O., and Bingham, J. M. (1988). Reductions in exchangeable magnesium with liming acid Ohio soils. Soil Sci. Soc. Am. J. 52, 131–136. Naidu, R., and Harter, R. D. (1998). Effectiveness of different organic ligands on sorption and extractability of cadmium by soils. Soil Sci. Soc. Am. J. 62, 644–650. Naidu, R., Syers, J. K., Tillman, R. W., and Kirkman, J. H. (1990). Effect of liming on phosphate sorption by acid soils. J. Soil Sci. 41, 165–175. Naidu, R., Bolan, N. S., Kookona, R. S., and Tiller, K. G. (1994). Ionic-strength and pH effects on the sorption of cadmium and the surface charge of soils. Eur. J Soil Sci. 45, 419–429. Neale, S. P., Shah, Z., and Adams, W. A. (1997). Changes in microbial biomass and nitrogen turnover in acidic organic soils following liming. Soil Biol. Biochem. 29, 1463–1474. Nyborg, M., and Hoyt, P. B. (1978). Effects of soil acidity and liming on mineralization of soil nitrogen. Can. J. Soil Sci. 58, 331 –338. Ogata, G., and Caldwell, A. C. (1960). Nitrate content of soils and nitrogen content of oat plants as affected by rates of liming. Agron. J. 52, 65–68. Oliver, D. P., Tiller, K. G., Conyers, M. K., Slattery, W. J., Alston, A. M., and Merry, R. H. (1996). Effectiveness of liming to minimise uptake of cadmium by wheat and barley grain grown in the field. Aust. J. Agric. Res. 47, 1181–1193. Oste, L. A., Dolfing, J., Ma, W. C., and Lexmond, T. M. (2001). Effect of beringite on cadmium and zinc uptake by plants and earthworms: More than a liming effect? Environ. Toxicol. Chem. 20, 1339–1345. Parfitt, R. L., Percival, H. J., Dahlgren, H. A., and Hill, L. F. (1997). Soil and soil solution chemistry under pasture and radiata pine in New Zealand. Plant Soil 91, 279–290. Parker, D. R., and Gardner, E. H. (1982). Factors affecting the mobility and plant availability of boron in some western Oregon soils. Soil Sci. Soc. Am. J. 46, 573–578. Patra, B. N., and Mohanty, S. K. (1994). Effect of nutrients and liming on changes in pH, redox potential, and uptake of iron and manganese by wetland rice in iron-toxic soil. Biol. Fert. Soils 17, 285 –288. Patterson, W. J. (1985). In “Industrial Wastewater Treatment Technology” (2nd ed.). Butterworth Publishers, Boston. Peoples, M. B., Lilley, D. M., Burnett, V. F., Ridley, A. M., and Garden, D. L. (1995). Effects of surface application of lime and superphosphate to acid soils on growth and N2 fixation by subterranean clover in mixed pasture swards. Soil Biol. Biochem. 27, 663–671. Pepper, I. L., Bezdicek, D. F., Baker, A. S., and Sims, J. M. (1983). Silage corn uptake of sludgeapplied zinc and cadmium as affected by soil-pH. J Environ. Qual. 12, 270–275. Peterson, L. A., and Newman, R. C. (1976). Influence of soil-pH on availability of added boron. Soil Sci. Soc. Am. J. 40, 280–282. Pratt, P. F., Blair, F. L., and McLean, G. W. (1964). Reactions of phosphate with soluble and exchangeable nickel. Soil Sci. Soc. Am. Proc. 28, 363–365. Preer, J. R., Abdi, A. N., Sekhon, H. S., and Murchison, G. B. (1995). Metals in urban gardens—effect of lime and sludge. J. Environ. Sci. Heal. A 30, 2041– 2056. Puttanna, K., Gowda, N. M. N., and Rao, E. V. S. P. (1999). Effect of concentration, temperature, moisture, liming and organic matter on the efficacy of the nitrification inhibitors benzotriazole, onitrophenol, m-nitroaniline and dicyandiamide. Nutr. Cycl. Agroecosys. 54, 251–257. Raven, J. A. (1985). pH regulation in plants. Sci. Prog. 69, 495–509. Raychaudhuri, M., Kumar, K., and Raychaudhuri, S. (1998). Response of soybean to lime and phosphorus on an Ultisol of Manipur. J. Indian Soc. Soil Sci. 46, 628–632.
270
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
Reddy, C. N., and Patrick, W. H., Jr. (1977). Effect of redox potential and pH on uptake of cadmium and lead by rice plants. J. Environ. Qual. 6, 259 –262. Redente, E. F., and Richards, J. L. (1997). Effects of lime and fertilizer amendments on plant growth in smelter impacted soils in Montana. Arid Soil Res. Rehab. 11, 353 –366. Reeve, N. G., and Sumner, M. E. (1972). Amelioration of subsoil acidity in Natal Oxisols by leaching of surface applied amendments. Agrochemophysica 4, 1–5. Rengasamy, P. (1983). Clay dispersion in relation to the changes in the electrolyte composition of dialyzed red–brown earths. J. Soil Sci. 34, 123 –132. Reuss, J. O. (1986). In “Acid deposition and the acidification of soil and water”. Springer, New York. Rhue, R. D., and Hensel, D. R. (1983). The effect of lime on the availability of residual phosphorus and its extractability by dilute acid. Soil Sci. Soc. Am. J. 47, 266 –270. Rhue, R. D., and Mansell, R. S. (1988). The effect of pH on sodium–calcium and potassium–calcium exchange selectivity for Cecil soil. Soil Sci. Soc. Am. J. 52, 641–647. Riggs, K. S., Rimmer, D. L., and Syers, J. K. (1995a). Effect of liming on the chemical composition of soils. J. Sci. Food Agric. 69, 159 –167. Riggs, K. S., Syers, J. K., Rimmer, D. L., and Sumner, M. E. (1995b). Effect of liming on calcium and magnesium concentrations in herbage. Effect of liming on the chemical composition of soils. J. Sci. Food Agric. 69, 169– 174. Ritchie, G. S. P. (1989). The chemical behaviour of aluminium, hydrogen ad manganese in acid soils. In “Soil Acidity and Plant Growth” (A. D. Robson, Ed.), pp. 1–60. Academic Press, New York. Robinson, W. O., Edgington, G., Armiger, W. H., and Breen, A. V. (1951). Availability of molybdenum as influenced by liming. Soil Sci. 72, 267 –274. Robson, A. D., and Abbott, L. K. (1989). The effect of soil acidity on microbial activity in soils. In “Soil Acidity and Plant Growth” (A. D. Robson, Ed.), pp. 139 –166. Academic Press, New York. Rosolem, C. A., and Caires, E. F. (1998). Yield and nitrogen uptake of peanuts as affected by lime, cobalt and molybdenum. J. Plant Nutr. 21, 827–835. Roy, W. R., Krapac, I. G., and Steele, J. D. (1993). Soil processes and chemical-transport—sorption of cadmium and lead by clays from municipal incinerator ash water suspensions. J. Environ. Qual. 22, 537 –543. Runge, M., and Rode, M. W. (1991). Effects of soil acidity on plant associations. In “Soil Acidity” (B. Ulrich, and M. E. Sumner, Eds.), pp. 183–202. Springer, New York. Saggar, S., Hedley, M. J., and Phimsarn, S. (1998). Dynamics of sulfur transformations in grazed pastures. In “Sulfur in the Environment” (D. G. Maynard, Ed.), pp. 45 –93. Marcel Dekker, New York. Sanders, J. R., McGrath, S. P., and Adams, T. M. (1987). Zinc, copper and nickel concentrations in soil extracts and crops grown on 4 soils treated with metal-loaded sewage sludges. Environ. Pollut. 44, 193 –210. Shuman, L. M. (1986). Effect of ionic strength and anions on zinc adsorption by two soils. Soil Sci. Soc. Am. J. 50, 1438–1442. Singh, B. R., and Myhr, K. (1998). Cadmium uptake by barley as affected by Cd sources and pH levels. Geoderma 84, 185–194. Singh, U., and Uehara, G. (1986). Electrochemistry of the double layer: Principles and applications to soils. In “Soil Physical Chemistry” (D. L. Sparks, Ed.), pp. 1 –38. CRC Press, Boca Raton, FL. Slangen, J. H. G., and Kerkhoff, P. (1984). Nitrification inhibitors in agriculture and horticulture—a literature-review. Fert. Res. 5, 1–76. Smith, E., Naidu, R., and Alston, A. M. (1998). Arsenic in the soil environment: A review. Adv. Agron. 64, 149 –195. Sommer, S. G., and Ersboll, A. K. (1996). Effect of air flow rate, lime amendments, and chemical soil properties on the volatilization of ammonia from fertilizers applied to sandy soils. Biol. Fert. Soils 21, 53 –60.
SOIL ACIDIFICATION AND LIMING INTERACTIONS
271
2 Soon, Y. K., and Miller, M. H. (1977). Changes in the rhizosphere due to NHþ 4 and NO3 fertilization and phosphorus uptake by corn seedlings (Zea mays L. ). Soil Sci. Soc. Am. Proc. 41, 77–80. Sorn-Srivchai, P., Tillman, R. W., Syers, J. K., and Cornforth, I. C. (1984). Effect of soil pH on Olsen bicarbonate phosphate values. J. Sci. Food Agric. 35, 1169– 1172. Sparks, D. L. (1986). In “Soil Physical Chemistry”. CRC Press, Boca Raton, FL. Sparks, D. L. (1995). In “Environmental Soil Chemistry”. Academic Press, New York. Sparrow, L. A., and Salardini, A. A. (1997). Effects of residues of lime and phosphorus fertilizer on cadmium uptake and yield of potatoes and carrots. J. Plant Nutr. 20, 1333–1349. Sparrow, L. A., Salardini, A. A., and Bishop, A. C. (1993). Field studies of cadmium in potatoes (Solanum Tuberosum L. ).1. Effects of lime and phosphorus on cv russet burbankaust. J. Agric. Res. 44, 845–853. Sposito, G. (1984). In “The Surface Chemistry of Soils”. Oxford University Press, New York. Springett, J. A., and Syers, J. K. (1984). Effect of pH and calcium content of soil on earthworm cast production in the laboratory. Soil Biol. Biochem. 16, 185 –189. Stahl, R. S., and James, B. R. (1991). Zinc sorption by B horizon soils as a function of pH. Soil Sci. Soc. Am. J. 55, 1592–1597. Stevens, R. J., and Laughlin, R. J. (1996). Effects of lime and N fertiliser on two sward types over a ten-year period. J. Agric. Sci. 127, 451– 461. Su, C., Evans, L. J., Bates, T. E., and Spiers, G. A. (1994). Extractable soil boron and alfalfa uptake— calcium-carbonate effects on acid soil. Soil Sci. Soc. Am. J. 58, 1145–1150. Suarez, D. L., and Simunek, J. (1997). UNSATCHEM: Unsaturated water and solute transport model with equilibrium and kinetic chemistry. Soil Sci. Soc. Am. J. 61, 1633–1646. Sumner, M. E., Fey, M. V., and Noble, A. D. (1991). Nutrient status and toxicity problems in acid soils. In “Soil Acidity” (B. Ulrich, and M. E. Sumner, Eds.), pp. 149–182. Springer, New York. Syers, J. K. (1998). Proceedings of the Fertiliser Society, UK. In “Soil and plant potassium in agriculture”. Tang, C., McLay, C. D. A., and Barton, L. (1997). Comparison of the potential proton excretion of twelve pasture legumes grown in nutrient solution. Aust. J. Expp. Agric. 37, 563–570. Taylor, G. (1975). Leaching rates of heavy metal ions in forest soils. Water Air Soil Pollut. 9, 137 –148. Temminghoff, E (1998). Chemical speciation of heavy metals in sandy soils in relation to availability and mobility. PhD Thesis, Wageningen Agricultural University, Wageningen, The Netherlands. Thomas, G. W., and Hargrove, W. L. (1984). The chemistry of soil acidity. Agron. Monogr. 12. In “Soil Acidity and Liming” (F. Adams, Ed.), pp. 3–56. American Society of Agronomy, Madison, WI. Thomson, N. A. (1982). Lime and dairy production. Proc. NZ Grasslands Assoc. 43, 93–103. Tiller, K. G., Nayyar, V. K., and Clayton, P. M. (1979). Specific and non-specific adsorption of Cd by soil clays as influenced by zinc and calcium. Aust. J. Soil Res. 17, 17–28. Tillman, R. W., and Syers, J. K. (1982). Potential of lime for improving the efficiency of phosphate fertilizer use. NZ J. Agric. Sci. 16, 90–92. Tinsley, J. (1973). Leaching of nitrogen and other soil constituents in a podzol under scots pine following nitrogen fertilizer and liming treatments. J. Sci. Food Agri. 24, 483–483. Trasarcepeda, M. C., Carballas, T., Gilsotres, F., and Deblas, E. (1991). Liming and the phosphataseactivity and mineralization of phosphorus in an andic soil. Soil Biol. Biochem. 23, 209 –215. Tyler, G., and Olsson, T. (2001). Plant uptake of major and minor mineral elements as influenced by soil acidity and liming. Plant Soil 230, 307– 321. Unkovich, M. J., Sanford, P., and Pate, J. S. (1996). Nodulation and N fixation by subterranean clover in acid soils as influenced by lime application, toxic Al, soil mineral N, and competition from annual rye grass. Soil Biol. Biochem. 28, 639–648.
272
N. S. BOLAN, D. C. ADRIANO AND D. CURTIN
Uren, N. C. (2001). Types, amounts, and possible functions of compounds released into the rhizosphere by soil-grown plants. In “The Rhizosphere: Biochemistry, and Organic Substances at the Soil –Plant Interface” (R. Pinton, Z. Varanini, and P. Nannipieri, Eds.), pp. 22 –32. Marcel Dekker, New York. Valeur, I., Andersson, S., and Nilsson, S. I. (2000). Calcium content of liming material and its effect on sulphur release in a coniferous forest soil. Biogeochemistry 50, 1– 20. Vasseur, L., Fortin, M. J., and Cyr, J. (1998). Clover and cress as indicator species of impacts from limed sewage sludge and landfill wastewater land application. Sci. Total Environ. 217, 231– 239. de Vera, M. P., and Pocsidio, G. N. (1998). Potential protective effect of calcium carbonate as liming agent against copper toxicity in the African tilapia Oreochromis mossambicus. Sci. Total Environ. 214, 193 –202. Vlek, P. L. G., and Lindsay, W. L. (1977). Molybdenum contamination in Colorado pasture soils. In “Molybdenum in Environment” (W. R. Chappel, and K. K. Paterson, Eds.), pp. 619–650. Marcel Dekker, New York. Wahab, A., and Shah, Z. H. (1952). Effect of liming on the absorption of calcium, phosphorus and nitrogen by sorghum plant in alkaline soils. Agron. J. 44, 621–624. Wang, H. L., Hedley, M. J., Bolan, N. S., and Horne, D. J. (1999). The influence of surface incorporated lime and gypsiferous by-products on surface and subsurface soil acidity. II. Root growth and agronomic implications. Aust. J. Soil Res. 37, 181–190. Wheeler, D. M. (1997). Temporal changes in some soil chemical properties at four depths following the surface application of lime. NZ J. Agric. Res. 40, 309–316. Wheeler, D. M. (1998). Investigation into the mechanisms causing lime responses in a grass/clover pasture on a clay loam soil. NZ J. Agric. Res. 41, 497– 515. Winter, K. T., Whitney, D. A., Kissel, D. E., and Ferguson, R. B. (1981). Ammonia volatilization from lime urea ammonium-nitrate suspensions before and after soil application. Soil Sci. Soc. Am. J. 45, 1124–1128. Ye, Z. H., Wong, J. W. C., Wong, M. H., Lan, C. Y., and Baker, A. J. M. (1999). Lime and pig manure as ameliorants for revegetating lead/zinc mine tailings: A greenhouse study. Biores. Technol. 69, 35–43. Zimdahl, R. L., and Foster, I. M. (1976). Influence of applied phosphorus, manure, or lime on uptake of lead from soil. J. Environ. Qual. 5, 31 –34.
Index A acid drainage, 220– 1 acid precipitation, 221–3 acid soils, 89 –138 see also acid subsurface layers; liming; soil acidification characteristics and properties, 92 –6 distribution, 91– 2 enzyme activity, 104–5, 119 –21 fertility and chemical properties, 93–4, 108 –12 microbial biomass estimation, 97 –102 microbial biomass turnover, 115–16 microbial community structure, 103–4, 118 –19 nutrient turnover, 102– 3 occurrence, 90 pH value, 121– 4 quality indexing systems, 124–8 quality indicators, 105 –21 acidic subsurface layers, 187–214 see also soil acidification; soil pH causes, 195 –201 detrimental effects, 189–93 development rate, 193–5 environmental factors, 201–4 management factors and implications, 204 –12 occurrence, 189 poor growth, 189 –92, 193 surface mixing, 211 agricultural production on acid soils, 204 –5, 208 Alfisols, 92 allopolyploid cotton, 141–2, 149 –50 formation and diversification, 158–61 parentage, 161 –5 phylogeny, 151–2, 167, 171 polyphyly, 164 aluminium, 247, 249 amelioration of acid soils, 237–42 animal feed, 46, 68–70 anion exclusion, 24 arsenic, 236, 248, 257–8 attached microbes, see microbial attachment
B bacteria see also microbial attachment ammonium oxidation, 13 chemical substrate, 14 –17 electrostatics, 19–31 hydrophobic effects, 19–26, 30 microfibrillar structures, 31–2 physical substrate, 11 –13 basal respiration rate, 117 BATH (bacterial adherence to hydrocarbons) method, 24–5 biochemical soil quality indicators, 105 –21 biodegradation, 12 BIOLOG method, 104 biological amelioration, 241–2 biological properties of soils, 92 –4 biomass, see microbial biomass breeding of corn, 47– 8, 49–50 evaluation of crosses, 61–6 GEM activities and results, 58–67 GEM project, 51–4 GEM protocol, 59–60 C cadmium, 234 –6, 247, 257 calcium, 233, 244, 253– 5 carbon, microbial, 97 –103, 108 –9, 111–12, 114–15 carbon cycle, 219, 223 –4 cation exchange capacity, (CEC), 6, 9–10 CFE, see chloroform fumigation–extraction CFI, see chloroform fumigation–incubation charge of organic particles, 9 –10, 17 see also electrostatics chemical amelioration, 240–1 chemical distribution of microbes, 7–11 chemical properties of soils, 92, 93 China, red soils, 110 chloroform fumigation–extraction (CFE), 97–101 chloroform fumigation–incubation (CFI), 97, 98, 100 chromatography, 24–5 chromium, 236, 248, 258
273
274
INDEX
chromosomes see also genomes evolution (Gossypium), 148–50 stabilization, 168 clay fractions, 5–6 clover, 199, 200 see also legumes coatings, 6, 10–11 mineral, 27–9 organic, 29 –31 collaboration GEM project, 51–8 LAMP project, 46, 48, 49–50, 51, 81– 2 contamination of soils, 112 copper, 233, 245, 256 corn see also germplasm enhancement of maize breeding activities and results, 47–8, 49–50, 58–67 enhancement need, 46–9 European corn borer resistance, 75–7 genetic vulnerability, 46–7 germplasm, 47 –9 grain composition, 68–72 oil quality, 72–4, 75 starch quality, 70–2, 73 –4, 78 corn oil, 72 –4, 75 corn starch, 70– 2, 73–4, 78 cotton, 139 –86 see also allopolyploid cotton; diploid cotton; Gossypium; poplyploid cotton cultivation, 140 –1 fibers, 139, 140, 176 –8 origin of tribe, 142–4 Cretaceous, Gossypium origin, 153–4, 158–9 crops see also corn; legumes; maize; wheat soil depth and pH, 190– 1, 194, 197, 200 subsurface acidification, 204 –5, 208 cultivation of cotton, 140–1 cytoplasmic capture, Gossypium, 156– 7, 164 D Derjaguin– Landau–Verwey–Overbeck (DLVO) theory, 22 differential scanning calorimetry (DSC), corn, 71–2, 73–4 diploid cotton, 141 –2, 147–8, 151–2, 161–2, 166–9, 171 dispersal of cotton, 144–8, 155
DLVO theory, see Derjaguin–Landau– Verwey–Overbeck theory DNA sequence data, Gossypium, 150 –4, 159– 60 drainage, acid, 220–1 DSC, see differential scanning calorimetry duplication genes, 169 –73 genomes, 165–7 E E. coli cells adhesion, 5 earthworms, 96, 203– 4, 211 ecological consequences of polyploidization, 175– 6 ecosystems managed, 223–9 natural, 218–23 electrostatics, 19–31 DLVO theory, 22 hydrophobic effects, 19 –26 mineral coatings, 27 –9 organic coatings, 29–31 potential, 22 primary mineral differences, 26–7 repulsion, 24 energy content of corn, 68–9 environmental factors, soil acidification, 201–4 enzyme activity, acid soils, 104–5, 119–21 European corn borer resistance, 75–7 evolutionary history of cotton, 139–86 allopolyploids, 158 –65, 171 diploids, 171 polyploids, 165 –75 exotic germplasm, 54 –5 F farming activities, 223–9 fauna, 96, 203–4, 211 see also bacteria fertility, see soil fertility fertilizers, 209, 226–9, 259 see also liming; soil pH fibers, cotton, 139, 140, 176–8 fimbriae, 31 foodstuff, 46, 68–70 free-living microbes, 17–19 fumigation techniques, 97– 101 funding, GEM project, 58
INDEX G gas–water interface (GWI), 33 –5 GEM, see germplasm enhancement of maize gene interactions, Gossypium, 155 –8, 169– 73 genealogy, Gossypium, 150–3 genetic base of corn, 46– 8 genomes (cotton), 141–2 composition, 149–50 doubling cycles, 165 –7 evolution, 151 –3, 173 –5 interactions, 169–73 parentage, 161 –5 size variation, 148–9 genotypes, 210 germplasm enhancement of maize (GEM), 45 –87 bank accessions, 48 breeding activities and results, 58–67 concept expansion, 81–3 European corn borer resistance, 75– 7 funding mechanism, 58 grain composition, 68– 72 international collaboration, 49–50 need for, 46 –9 objective, 54–5 organization, 55–8 public cooperator findings, 79–80 public and private interaction, 51–4 research projects, 52– 3, 56–7 success factors, 80– 1 value-added trait analysis, 67 –8 wet milling efficiency, 77– 9 Gossypieae tribe, 142–4 Gossypium, 139– 86 alloploid origin, 158–65 characters, 143 chromosomal evolution, 148–50 chromosome stabilization, 168 diploids, 147–8, 151 –2, 161 –2, 166–9, 171 divergence events, 153–4, 158–61 emergence and diversification, 144– 8, 155 G. barbadense, 140– 2, 150, 152, 161–2, 164 –5, 172 G. gossipioides, 164 –5 G. hirsutum, 140–2, 150, 152, 172 G. raimondii, 161 –2, 164– 5 genealogy, 150– 3 genic and genomic interactions, 155 –8, 169 –73 genome composition, 149 –50
275
genome doubling, 165– 7 genome evolution, 151 –3, 173–5 genome size variation, 148–9 hybridization, 155–8 phylogeny, 150–4 polyploid evolution, 165–75 recombination, 168 –9 grain see also corn; maize; wheat composition, 68– 72 legumes, 206 moisture values, 62–5 grazing stock, urine excretion, 200–1 growth, 189 –92, 193, 217 see also agricultural production; breeding of corn; crops GWI, see gas– water interface H heavy metals contamination, 112 liming effects, 247 –8, 256– 8, 259 soil acidification, 234–6 HIC, see hydrophobic interaction chromatography humus 9 see also organic matter hybridization, Gossypium, 155– 8 hydrophobic effects, 19–26, 30 hydrophobic interaction chromatography (HIC), 24–5 I immersion of microbes, 3 indexing systems, see acid soils indicators of soil quality, 105–21 contamination, 112 –15 fertility, 108–12 microbial biomass-related, 116 –21 microbial community structure, 118–19 pH value, 121 –4 industrial activities, 218–23 international collaboration, 46, 49–50, 51, 81–2 iron, 234, 246, 256 L Latin American Maize Project (LAMP), 46, 48, 49–50, 81 –2
276
INDEX
lead, 248, 258 legumes poor root nodulation, 192–3 subsurface acidification, 196 –7, 199, 200, 205–7 liming, 122 –3, 259 amelioration of acid soils, 239 –42 heavy metal interactions, 247–8, 256 –8 materials, 237 –9 nutrient interactions, 242–56 poor growth response, 193 soil pH, 209 –10, 211 uses, 217 –18 M magnesium, 233, 244, 253–5 maize enhancement, 45– 87 see also germplasm enhancement of maize breeders, 51 –4 germplasm, 47 –9 grain composition, 68–72 LAMP project, 46, 48, 49–50, 51, 81– 2 need for, 46–9 managed ecosystems, 223 –9 manganese, 200, 233–4, 247, 258 mercury, 248 metal oxide coatings, 28–9 metals, 243, 244–6 see also heavy metals; trace elements microbial attachment, 1–43 appendages and cements, 31–2 chemical distribution, 7 –11 chemical substrate, 14–17 electrostatics, 19 –31 importance, 17–19 particle-size distribution, 4–7 physical substrate, 11–13 reasons, 2–3 saturated v. unsaturated conditions, 33–5 microbial biomass estimation, 97 –102 microbial metabolic quotient, 117–18 microbial quotient, 116 –17 nutrient turnover, 102–3 related indicators, 116 –21 soil contamination indicator, 112–15 soil fertility indicator, 108–12 turnover, 115 –16 microbial community structure, 103 –4, 118–19 microfibrillar structures, 31 –2
mineral coatings, 27 –9 mineral particles chemical substrate, 14–15 distribution, 7–8 microbial attachment, 26– 7 organic-coated, 10 –11 moisture, see rainfall; water molybdenum, 246, 256 N natural ecosystems, 218–23 nickel, 248, 258 nitrification, 202 nitrifying bacteria, 13 nitrogen, 219 fertilizers, 227–8 liming effects, 242–3 microbial, 101, 102 –3, 108– 9, 111–12, 114– 16 mineralisation suppression, 192 plant uptake and soil pH, 195 –7 soil acidity effects, 231 transformation, 225–6 uptake and assimilation, 224–5 non-reversible attachment, 19 nutrients, 10 acidic layers, 189–92 enhancement, 11–13 fertilizers, 227 liming interactions, 242–56 microbial biomass estimation, 97– 102 microbial biomass fertility, 108–12 microbial turnover, 102– 3 soil acidification, 230 –4 O oil quality, corn, 72–4, 75 organic coatings, 29–31 organic matter see also plant residues acid soils, 96 charge, 7, 9–10 composition and properties, 8–10 decomposition, 225 humus, 9 microbial attachment, 15– 17 timing effects, 254 organic-coated mineral particles, 10–11 organisms, see bacteria; fauna
INDEX Oxisols, 92, 95 see also acid soils P particle attachment, see microbial attachment; mineral particles; organic matter particle-size distribution, 4–7 pasture, 190, 205 –8 pH, see soil pH phosphate fertilizers, 228, 259 phosphate rocks (PRs), 237–9 phosphorus fertilizer, 193 liming effects, 238, 243, 249–51 microbial, 101– 3, 111–12, 114–16 P-sparing effect, 250 soil acidity effects, 231 –2 phylogeny allopolyploids, 158 –65, 171 diploids, 171 Gossypium, 143–4, 150–4 microbes, 18 polyphyly, 164 physical amelioration of soil, 240 plants acification processes, 223 –5 growth, 189– 92, 193, 217 nitrogen uptake, 195– 7 nutrients, 231–4, 242–56 residues, 197–200, 202, 208 species on acid soils, 205–8, 210 Pleistocene, 159– 60 polyphyly, 164 polyploid cotton, 139–86, 141 ecological consequences, 175 –6 evolution, 165 –75 fiber, 176–8 genomic composition, 149–50 recombination, 168 –9 taxonomic diversity, 150 potassium, 232, 243, 251 precipitation, 202–3, 221 –3 primary nutrients, 231–2, 242 –51 private sector, see public/private cooperation protein, corn, 68–9 protocol, corn breeding, 59–60 proton generation, 219–20 PRs, see phosphate rocks public/private cooperation GEM project, 51 –8
277
GEM research and results, 74–80 GEM success factors, 80–2 US agricultural research, 50– 1 pyrite, 14–15, 219–21, 258 –9 Q quality, see soil quality R rainfall, acid, 202–3, 221–3 red soils, 110– 11 research GEM projects, 52–3, 56 –7, 74–80 US agriculture, 50–1 value-added trait, 68 residues, see plant residues reversible attachment, 19 roots, 189–93 S saturation, 33– 5 secondary nutrients, 232–3, 242–3, 252–8 selenium, 236, 245, 258 silicates, 7 –8 size distribution, soil particles, 4–7 soil acidification, 215–72 see also acid soils; acidic subsurface layers; liming acid drainage, 220 –1 acid precipitation, 219, 221 –3 amelioration through liming, 237–42 causes and effects, 216–18 fertilizer use, 226 –9 heavy metal transformation, 234 –6 lime, nutrient and heavy metal interactions, 242–58 pasture, 190, 205–8 plant-induced processes, 223–5 primary nutrients, 230–2, 242–51 secondary nutrients, 232–3, 242–3, 252 –8 soil-induced processes, 225–6 trace elements, 233–4, 245 –6 soil disturbance, 208 soil enzyme activity, 104– 5, 119–21 soil fertility, 201 acid soils, 93 –4 indicators, 108–12 liming effects, 238 –42
278
INDEX
soil moisture, see soil water soil particles see also mineral particles; organic matter microbial attachment, 1–43 size distribution, 4–7 soil pH, 188 –9 see also acid soils; soil acidification acid soils, 121–4 buffering capacity, 201 changes due to fertilizers, 229 environmental factors, 201 –4 initial value, 201–2 liming effects, 209 –10, 211 management factors, 204–10 nutrient transformation, 230 plant N uptake, 195–7 plant residue returns, 197 –200 variation with depth, 190–2, 194, 197, 200, 204 soil quality, 89–138 see also soil fertility characteristics, 92 –6 contamination indicators, 112 –15 definition, 92 indexing systems, 124–8 microbial biomass indicators, 116–21 minimum data sets, 128 pH indicator, 121 –4 soil water, 3–4, 33–5, 202 –3 soil-induced acidification, 225 –6 speciation mechanisms, Gossypium, 155–8 starch quality (corn), 70– 2, 73–4, 78 substrate induced respiration (SIR), 97, 98, 100–1 subsurface layers, acidic, 187 –214 sulfate fertilizers, 228 –9 sulfides, 220 –1 sulfur, 220 liming effects, 243 –4, 252 –3 soil acidity effects, 232
transformation, 226 uptake and assimilation, 225 sulfur dioxide emissions, 222 T taxonomy, Gossypium, 145 –8, 150 tea bushes, 123– 4 tillage techniques, 208 toxic heavy metals, see heavy metals trace elements, 233 –4, 245– 6, 255–6 U Ultisols, 92, 95 see also acid soils urine excretion by stock, 200 –1 US agriculture, 46 –7, 50–1 V value-added trait analysis, 67 –8 van der Waals minimum, 22 W water, 189 –92 see also soil water wet milling efficiency (corn), 77 –9 wheat, 199 Y yields (corn), 47, 62 –5 Z zeta potential, 21 zinc, 233, 245, 256
ERRATUM Advances in Agronomy, Volume 75 In the References section, page 231, of the chapter “Quantitative Remote Sensing of Soil Properties” by E. Ben-Dor, the reference to the article by J. A. M. Dematteˆ and G. J. Garcia was printed incorrectly. The correct version is shown below: Dematteˆ, J. A. M., and Garcia, G. J. (1999). Alteration of soil properties through a weathering sequence as evaluated by spectral reflectance. Soil Sci. Soc. Am. J. 63, 327– 342.
The original (incorrect) citation was given as: Alexander, J., Dematte, M., and Garcia, G. J. (1999). Alteration of soil properties through a weathering sequence as evaluated by spectral reflectance. Soil Sci. Soc. Am. J. 63, 327 –342.