Advances in Agronomy, Volume 69

Agronomy D VA N C E S

VOLUME

I N

69

Advisory Board Martin Alexander

Ronald Phillips

Cornell University

University of Minnesota

Kenneth J. Frey

Larry P. Wilding

Iowa State University

Texas A&M University

Prepared in cooperation with the American Society of Agronomy Monographs Committee John Bartels Jerry M. Bigham Jerry L. Hatfield David M. Krell

Diane E. Stott, Chairman Linda S. Lee David Miller Matthew J. Morra John E. Rechcigl Donald C. Reicosky

Wayne F. Robarge Dennis E. Rolston Richard Shibles Jeffrey Volenec

Agronomy

DVANCES IN

VO L U M E

69

Edited by

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

San Diego

San Francisco

New York

Boston

London

Sydney

Tokyo

COPYRIGHT PAGE SUPPLIED BY AP

Contents Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

vii ix

THE MEASUREMENT AND INTERPRETATION OF SORPTION AND DESORPTION RATES FOR ORGANIC COMPOUNDS IN SOIL MEDIA Joseph J. Pignatello I. II. III. IV. V. VI.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Nature of Elementary Sorption Processes in Soils. . . . . . . . . . . . Slow Sorption and Desorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sorption Kinetic Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Experimental Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sorption Kinetics and Bioavailability . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2 3 16 27 45 56 65

ENVIRONMENTAL INDICATORS OF AGROECOSYSTEMS O. H. Smith, G. W. Petersen, and B. A. Needelman I. II. III. IV. V. VI. VII.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Agroecosystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Monitoring and Assessment Endpoints . . . . . . . . . . . . . . . . . . . . . . . . Environmental Indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Soil Organic Matter as a Candidate Environmental Indicator. . . . . . . Indicator Ranking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

76 76 77 78 85 90 91 92

GROWTH PROMOTION OF PLANTS INOCULATED WITH PHOSPHATE-SOLUBILIZING FUNGI M. A. Whitelaw I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Soil Phosphorus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. Phosphate-Solubilizing Soil Microorganisms. . . . . . . . . . . . . . . . . . . . IV. Liquid Medium Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

100 100 106 109

vi

CONTENTS

V. Plant Growth Promotion by Phosphate-Solubilizing Fungi . . . . . . . . VI. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

133 143 144

HYDROLOGICAL FACTORS FOR PHOSPHORUS TRANSFER FROM AGRICULTURAL SOILS P. M. Haygarth, A. L. Heathwaite, S. C. Jarvis, and T. R. Harrod I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II. Temporal Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . III. Spatial Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

154 155 162 173 173

CASSAVA, Manihot esculenta Crantz, GENETIC RESOURCES: THEIR COLLECTION, EVALUATION, AND MANIPULATION Nagib M. A. Nassar I. Wild Taxa of Cassava Manihot Species . . . . . . . . . . . . . . . . . . . . . . . . . II. Broadening the Genetic Base of Cassava, M. esculenta Crantz, and Development of Interspecific Hybridization . . . . . . . . . . . . . . . . . . . . III. Development and Selection for Apomixis in Cassava, M. esculenta Crantz. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. Production of Polyploid Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. Protein Contents in Cassava Cultivars and Its Hybrid with Wild Manihot Species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

180

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

231

198 210 215 225 227

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

T. R. HARROD (153), Soil Survey and Land Research Centre, Cranfield University, North Wyke, Okehampton, Devon EX20 2SB, United Kingdom P. M. HAYGARTH (153), Institute of Grassland and Environmental Research, North Wyke, Okehampton, Devon EX20 2SB, United Kingdom A. L. HEATHWAITE (153), Department of Geography, University of Sheffield, Sheffield S10 2TN, United Kingdom S. C. JARVIS (153), Institute of Grassland and Environmental Research, North Wyke, Okehampton, Devon EX20 2SB, United Kingdom NAGIB M. A. NASSAR (179), Departamento de Genética e Morfologia, Universidade de Brasília, Brasília 70919, Brazil B. A. NEEDELMAN (75), Department of Agronomy, Pennsylvania State University, University Park, Pennsylvania 16802 G. W. PETERSEN (75), Department of Agronomy, Pennsylvania State University, University Park, Pennsylvania 16802 JOSEPH J. PIGNATELLO (1), The Connecticut Agricultural Experiment Station, New Haven, Connecticut 06511 O. H. SMITH (75), Department of Agronomy, Pennsylvania State University, University Park, Pennsylvania 16802 M. A. WHITELAW (99), School of Wine and Food Sciences, Charles Sturt University, Wagga Wagga, New South Wales 2678, Australia

vii

This Page Intentionally Left Blank

Preface Volume 69 contains five excellent reviews dealing with crop and soil sciences. Chapter 1 is a comprehensive and timely review on the measurement and interpretation of sorption and desorption rates for organic compounds in soil media. Topics covered include the nature of elementary sorption processes in soil, hindered sorption and desorption processes, sorption kinetic models, experimental methods, and sorption kinetics and bioavailability. Chapter 2, by O. H. Smith and co-workers, is an excellent overview of environmental indicators of agroecosystems. Soil organic matter content is discussed in detail as a candidate environmental indicator. A ranking scheme is proposed for the use of multiple indicators in decision-making applications. Chapter 3, by M. A. Whitelaw, is an interesting treatise on plant growth as affected by phosphate-solubilizing soil microorganisms. The author provides a discussion on soil phosphorus, studies on P-solubilizing soil microorganisms, aspects of liquid medium studies, and plant growth promotion by phosphate-solubilizing fungi. Chapter 4, by P. M. Haygarth et al., is a fine review on hydrological factors affecting phosphorus (P) transfer from agricultural soils. The authors review current knowledge to define the spatial and temporal controls on P transfer from agricultural soils via the various hydrological pathways. Chapter 5, by N. M. A. Nassar, provides a thorough treatment of Cassava, Manihot esculenta Crantz, genetic resources. Topics that are discussed include wild taxa, the genetic base and development of interspecific hybrids, development and selection for apomixis, production of polyploid types, and protein content in cultivars. Many thanks to the authors for their first-rate contributions. Donald L. Sparks

ix

This Page Intentionally Left Blank

THE MEASUREMENT AND INTERPRETATION OF SORPTION AND DESORPTION RATES FOR ORGANIC COMPOUNDS IN SOIL MEDIA Joseph J. Pignatello The Connecticut Agricultural Experiment Station New Haven, Connecticut 06511

I. Introduction II. The Nature of Elementary Sorption Processes in Soils A. Intermolecular Interactions Available to Organic Molecules B. Properties of Soil Components and Mechanisms of Sorption C. Thermodynamic Driving Force for Sorption D. Rates of Elementary Processes III. Slow Sorption and Desorption A. Uptake and Release Profiles B. Retardation Mechanisms IV. Sorption Kinetic Models A. Models Based on Bond Energetics B. Driving Force Models C. Diffusion Models D. Stochastic Models V. Experimental Methods A. Batch Techniques B. Column Techniques C. Stirred-Flow Cell Technique D. Zero-Length Columns VI. Sorption Kinetics and Bioavailability A. Assimilation of Chemicals in Soil Systems B. Coupled Sorption–Biodegradation Kinetic Models References

Sorption controls the physical and biological availability of chemicals in soil. Most organic molecules undergo primarily weak physisorption interactions and the driving force for sorption is the hydrophobic effect. Sorption and desorption rates, therefore, are governed mainly by molecular diffusion through the fixed interstitial pores of particle aggregates and through the three-dimensional pseudophase of soil organic matter. Retardation in the fixed pore system is due to tortuosity, chromatographic adsorption to pore walls, and, in the smallest pores, steric hindrance. 1 Advances in Agronomy, Volume 69 Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved. 0065-2113/00 $30.00

2

JOSEPH J. PIGNATELLO Soil organic matter, which has the strongest affinity for most organic compounds, may exist in rubbery and glassy phases and retards sorption and desorption by its viscosity and by the presence of internal nanopores, which detain molecules and may sterically inhibit their migration. Soot carbon and/or ancient organic matter may be present in some soils but their roles are yet unclear. Desorption rates are correlated with the size and shape of the diffusant. Hysteresis is commonly observed but a satisfactory explanation for it has yet to emerge. Mathematical models based on bond energetics, driving force theory, diffusion, and stochastic analysis are discussed. These models have been used to describe batch experiments and have been coupled to advection–dispersion transport equations for use in flowing water systems. Diffusion models are the most realistic but also the most difficult to apply because diffusion is highly dependent on the geometry and composition of the sorbent. Soil heterogeneity impedes the mechanistic interpretation of rates. Particles span an extremely wide range of sizes. The appropriate diffusion length scale is often uncertain. The diffusion coefficient is expected to be concentration dependent in any diffusing medium in which sorption is nonlinear. Furthermore, the diffusant may alter the structure of soil organic matter. Bioavailability can be rate limited by desorption. Cells are believed to access only dissolved molecules, but organisms may affect sorption kinetics indirectly by steepening the concentration gradient or by altering soil properties through bioactivity. Coupled sorption–biodegradation models are necessary whenever nonequilibrium conditions prevail during exposure. Models coupling Monod or first-order biodegradation kinetics with “two-site,” driving-force, or diffusion models have been employed. Some have been used in conjunction with the advection–dispersion transport model. © 2000 Academic Press.

I. INTRODUCTION Sorption is fundamental to the fate of organic chemicals in soil environments. In order to assess the influence of sorption, it is important to understand the nature of the bonding forces between the sorbing molecule (sorbate) and the solid (sorbent), the thermodynamic driving forces responsible for establishing the position of equilibrium, and the rates of association and dissociation of sorbate with time. The primary focus of this chapter is on sorption kinetics. Sorption kinetics is an important field of investigation in soil and environmental science because nonequilibrium sorption conditions often apply as other fate processes, such as vaporor liquid-phase transport, uptake by organisms, and chemical reactions, are taking place. In fact, sorptive equilibrium may take as long as many months. Recent reviews (Alexander, 1995; Linz and Nakles, 1997; Pignatello and Xing, 1996) discussed the rate-limiting effects of sorption on mass transport and bioavailability and the ramifications thereof for the management and risk assessment of chemicals in soils and aquatic sediments. Soil particles are typically porous or have phases, such as soil organic matter (SOM), that are penetrable by the sorbate. Hence, sorption usually consists of at

SORPTION AND DESORPTION RATES

3

least three steps: (i) transport from the bulk fluid (vapor or liquid) to the vicinity of the external surface of the particle, (ii) transport through the pore structure or interstices of the particle, and (iii) formation of a “bond” at the “site” of sorption. This chapter will begin by discussing the properties of sorbate and sorbent and the nature of the sorption bond; in addition to providing a brief review of mechanisms, the purpose of this introductory material is to make the reader aware of the complexity of the sorption process, an essential prerequisite to understanding the kinetics of sorption. The chapter will then give an overview of the current state of our knowledge about the mechanisms that retard sorption and desorption. Next, mathematical approaches to describing sorption/desorption kinetics are discussed, followed by a discussion of the experimental techniques for measuring rates. The last section will address sorption kinetics in relation to bioavailability. The mathematical models discussed in this chapter will be presented only in their essential features to save space and spare the reader unnecessary mental toil; consequently, it is incumbent on an investigator to consult the original works before embarking on their use. It should be noted that a full understanding of the mechanisms that retard sorption has not been attained. As a result, there is plenty of opportunity for advancement in the field. It is still not generally possible to predict a priori the entire uptake or release curve for any given soil–chemical system. The goal of this chapter is to lay a foundation for understanding the causes of sorption/desorption rate limitations. In this chapter, we will deal with systems in a hydrated state.

II. THE NATURE OF ELEMENTARY SORPTION PROCESSES IN SOILS Defined broadly, the terms sorption and desorption refer to bulk mass-transfer phenomena in which molecules leave the fluid phase and become associated with an immobile phase and vice versa. The terms imply nothing about the nature of the interaction nor about the transport of the sorbate molecules once in the confines of the immobile phase. We speak of solution–solid and vapor–solid sorption. Sorption of organic compounds may be broadly divided into the following categories (Fig. 1): • Adsorption (A in Fig. 1) refers to association of molecules at the solid–fluid interface. The interface may exist on the external surface of the particle facing the bulk fluid or on the surface of a pore facing the fluid contained in that pore. • Absorption (B in Fig. 1) occurs when molecules penetrate the solid surface and intermingle with its three-dimensional molecular or atomic matrix. In natural soils, SOM is the only component that is penetrable in this manner. Polluted soils

4

JOSEPH J. PIGNATELLO

Figure 1 Different types of sorption available to organic molecules. A, Adsorption; B, absorption in SOM or NAPL phase; C, capillary condensation; D, dissolution in water film; E, adsorption to water film.

may contain additional absorptive phases in the form of nonaqueous phase liquids (NAPLs)—solvents, oils, tars, and so on. • Condensation (C in Fig. 1) refers to a phase change from the vapor or solution state to a nonaqueous liquid or solid state. Condensation may occur on any surface when the concentration is above the solubility or vapor pressure. However, it is facilitated in small pores (50 nm): As a pore width decreases there is a progression from monolayer adsorption to capillary condensation owing to the effect of surface tension, which reduces the vapor pressure below the value of the pure liquid in accordance with the Kelvin equation (Ruthven, 1984). Water competes effectively with organics for condensation in pores of minerals because such surfaces are ordinarily polar; however, recent studies of aquifer sediments suggest that capillary condensation of compounds such as benzene may occur even from aqueous solution and even at concentrations lower than their bulk water solubility (Corley et al., 1996).

SORPTION AND DESORPTION RATES

5

• Association with water films: Depending on the relative humidity, unsaturated soils contain liquid water in pores and as coatings of surfaces. When organic vapors contact unsaturated soils, dissolution in (D in Fig. 1) and adsorption on (E in Fig. 1) water films may occur (Kim et al., 1998; Ong and Lion, 1991; Petersen et al., 1995). Molecules in such states are technically sorbed because they are removed from the surrounding vapor state.

A. INTERMOLECULAR INTERACTIONS AVAILABLE TO ORGANIC MOLECULES Organic compounds can undergo chemisorption, physisorption, or ion pair formation (ion exchange) with natural particles. • Chemisorption involves significant atomic or molecular orbital overlap with the solid phase; that is, the formation of a covalent or coordination bond. Examples relevant to this chapter include “inner-sphere” coordination complexes between carboxylate, phenolate, amine, or sulfhydryl groups and metal ions; i.e., IMn+ – ZR, where IMn+ is a structural or adsorbed metal ion. Such bonds have both ionic and covalent character. Sorption accompanied by formation of a true covalent bond (such as a C–C bond) is seldom reversible and thus is not considered relevant to this chapter. • Physisorption involves weak intermolecular attractive forces between atoms and molecules, including “van der Waals,” hydrogen (H-) bonding, and charge transfer. Van der Waals force encompasses the following interactions (Castellan, 1971; Israelachvili, 1992): (i) dipole–dipole forces, resulting from mutual attraction between permanent dipoles; (ii) induced dipole–induced dipole (dispersion) forces, resulting from the synchronization of electronic motion in each molecule producing momentary dipole moments in each; (iii) Dipole-induced dipole, resulting from the attraction of a permanent dipole with the dipole it induces in its neighbor. Forces (i–iii) involve no appreciable molecular orbital overlap, are randomly oriented in space, and are only a few kilojoules per mole in energy. Force (ii) is available to all atoms and molecules. The total van der Waals energy is the sum of all individual interactions between the sorbate and the site, and it depends on the distance of approach, the sorbate size, and the polarizabilities and polarities of both sorbate and site. H-bonding (Schuster et al., 1976) involves interactions between acids and bases of the type –AH...:B–, where A and B are ordinarily N, O, or S atoms. Hbonding is a combination of the dipole–dipole force and a small degree of molecular orbital overlap. It is oriented in space (A–H–B angle  ⬃15) and ranges in strength from 10 to 25 kJ/mole.

6

JOSEPH J. PIGNATELLO

Charge-transfer interactions (often referred to as donor–acceptor interactions) may occur when an electron-poor acceptor (A) encounters an electron-rich donor (D) and forms a complex in which one resonance structure represents transfer of an electron (Foster, 1969): A  D S {A...D } A...D+}.

(1)

Charge-transfer complexes most relevant to soil systems are n r  and  r  types, where n refers to a nonbonding lone-pair electrons and  refers to an aromatic ring or other extended -conjugated system (Foster, 1969). Haderlein et al. (1996) proposed n r  charge-transfer complexation between permanent charges on clays (donor) and electron-deficient polynitroaromatic rings (acceptor).  r  charge-transfer bonds are possible between appropriate functional groups on sorbate and SOM (Müller-Wegener, 1987). • Ion-exchange force involves electrostatic attraction between an organic anion or cation and a charged group on the sorbent. It may be augmented by physisorption forces. For minerals, this type of sorption is best described as a concentration enhancement of the organic ion in the water near the surface, accompanied by depletion of the original (usually inorganic) ion. Ion exchange may also occur at charged sites in SOM, usually carboxylate or phenolate groups.

B. PROPERTIES OF SOIL COMPONENTS AND MECHANISMS OF SORPTION 1. Mineral Surfaces Two principal types of surface exist on natural minerals: 1. The hydroxylated surface consists of -OH groups protruding into solution from the topmost layer of metal ions (IMn+ –OH). It exists on all hydrous oxides of Si, Fe, and Al and on the edges of layer silicate clays. It has variable positive or negative charge density, depending on mineral, pH, and ionic strength. Regardless of charge, it is strongly hydrophilic; adsorption of water on this surface is more energetic than adsorption of nonpolar organic molecules (Curthoys et al., 1974), and it is believed that at ordinary humidities one or more layers of ordered water (“vicinal water”) are strongly under the influence of the surface. 2. The siloxane surface consists of oxygen atoms bridging underlying Si4+ ions. It exists on the faces of many layer silicate clays. It has permanent negative charge, depending on the degree of isomorphic substitution in the underlying lattice. The charged sites are closely associated with metal or organic cations and the surface in the vicinity of the charge is strongly hydrophilic. The neutral regions between charges are hydrophobic or only weakly hydrophilic (Chen, 1976; Jaynes and Boyd, 1991).

SORPTION AND DESORPTION RATES

7

Figure 2 Depiction of sorption. (a) Sorption to mineral surfaces: A1, solvent-separated physisorption; A2, physisorption with direct interaction with the surface; A3, chemisorption by coordination with underlying metal ion. (b) Sorption to SOM: B1, adsorption to the SOM-coated mineral surface; B2, adsorption to the extended organic surface; B3, absorption in the random network polymer phase.

Although not fully understood, several different modes of adsorption are believed occur on soil minerals (Fig. 2a). A1 in Fig. 2 refers to physisorption in which the sorbate is separated from the surface by solvent molecules (Goss, 1992). This occurs on hydroxylated surfaces for compounds that cannot displace adsorbed water. This type of adsorption might be best described as a concentration enhancement of the solute in the “vicinal water.” A2 refers to physisorption in which the sorbate is in direct contact with surface atoms. Direct contact occurs on neutral siloxane surfaces, as well as on hydroxylated surfaces, provided water is scarce or the compound’s H-bond ability is sufficiently great that it can displace tightly bound water. A3 refers to chemisorption through inner-sphere coordination with lattice or adsorbed metal ions. This mechanism requires appropriate coordinating functional groups on the molecule. D refers to pore condensation as discussed in reference to Fig. 1.

8

JOSEPH J. PIGNATELLO

2. Soil Organic Matter It is well established that sorption of hydrophobic compounds out of aqueous solution or at high relative humidity is dominated by the SOM fraction unless that fraction is very small (Schwarzenbach et al., 1993). For example, sorption of chlorinated benzenes and polycyclic aromatic hydrocarbons (PAHs) to nonporous inorganic oxides is so weak that it is expected to be insignificant when the fraction of soil organic carbon ( foc ) is ⬃0.0001 (Mader et al., 1997)! Situations in which the predominance of SOM does not necessarily hold include (i) very dry conditions, when capillary condensation or adsorption can be important, and (ii) when chemisorption is important. SOM consists of plant and microbial material in various stages of decomposition. Materials bearing little physical and chemical resemblance to their precursor biological polymers are called humic substances and make up the bulk of SOM (Hayes et al., 1989). Knowledge about humic substances is mainly inferred from studies of humic and fulvic acids, which are humic substances isolated from natural waters or extracted from soil with dilute base or polar solvents. Humic and fulvic acids are a refractory mixture of polyanionic macromolecules ranging from hundreds (Novotny et al., 1995) to hundreds of thousands of grams per mole (see Pellegrino and Piccolo (1999), however). Bearing in mind that each humic macromolecule may be unique, a hypothetical structure has been proposed on the basis of physical, spectroscopic, and fragmentation-identification studies (Schulten and Schnitzer, 1993) (Fig. 3a). It has both aliphatic and aromatic subunits and an abundance of oxygen functional groups. In solution, the macromolecules coil up in a random fashion and aggregate to form a spheroidal, water-swollen phase of entangled humic macromolecules (Fig. 3b). The density of the particle increases gradually from edge to center (Hayes and Himes, 1986; Swift, 1989). The unextractable SOM—typically more than half the total—is called humin. Humin is separated from minerals only by drastic treatment such as hydrofluoric acid digestion which dissolves the minerals (Preston et al., 1989). Humin may be separated into lipid-like and humic-like components (Rice and MacCarthy, 1990). Little is known about humin, even though it may have a greater affinity for organic compounds than whole SOM (Xing and Pignatello, 1997). The bulk of humin may consist simply of humic acid-like molecules of higher molecular weight and stronger affinity for mineral surfaces. Humin is more hydrophobic and more condensed than humic or fulvic acids. In the native state, SOM is usually bound to mineral particles on a scale ranging from a monolayer organic film to a discreet organic phase. The nature of SOM as a sorbent of organic compounds—obviously crucial to its role in sorption kinetics—is controversial. SOM has been modeled as a coating on mineral surfaces, an extended organic surface, or a random network polymer phase. These are depicted in Fig. 2b. As a coating, SOM is regarded to enhance the surface affinity for organic mol-

SORPTION AND DESORPTION RATES

9

ecules by making it more “hydrophobic,” similar to the effect of alkyl chains attached to the surface of silica gel used in reverse-phase liquid chromatography. On such a surface, the sorbate may be under the simultaneous influence of the mineral and the organic matter. Mayer (1999) provides evidence, however, that even in low-organic carbon (OC) aquifer sediments SOM exists in multilayer patches rather than as monolayers on the surface. The extended organic surface concept regards SOM to be an impenetrable adsorptive surface. The external surface area of SOM measured by N2 adsorption at 77 K using the B.E.T. equation is on the order of ⬃100 m2 /g (Chiou et al., 1993), which appears to be too low to account for the high affinity of SOM for organic compounds, implying that little impenetrable surface exists. The preponderance of evidence points to SOM behaving as a random network polymer phase that provides a three-dimensional hydrophobic environment for organic molecules. The “surface” of such a phase is expected to be diffuse rather than sharply defined due to more extensive solvation of the outer polar regions of the humic polymers that face the solvent (Hayes and Himes, 1986; Swift, 1989). If true, a long-lived surface-adsorbed state would be disfavored. Instead, the sorbate is expected to penetrate the phase and intermingle with the humic strands, much the same way in which small molecules interact with synthetic organic polymers (Rogers, 1965; Vieth, 1991; Frisch and Stern, 1983). The structure of lignin, the woody component of plant material and probably the main precursor of terrestrial humic substances, is also considered a “random network polymer” (Goring, 1989). According to the polymer phase concept, sorption is attributed to dissolution (absorption) of the hydrophobic solute in the liquid-like, organophilic phase in order to escape the polar environment of water (Chiou, 1989). Unlike a liquid, however, the sorption potential of SOM is not uniform (Pignatello, 1998, 1999; Xing et al., 1996; Xing and Pignatello, 1997; Young and Weber, 1995). Sorption isotherms tend to be nonlinear in the sense that sorption diminishes with increasing concentration. The isotherm can be fit to the Freundlich equation, qe KeCen,

(2)

where qe and Ce are the equilibrium sorbed and solution concentrations, Ke is the sorption coefficient, and n is a constant 1. Moreover, sorption in bisolute and multisolute systems is competitive. These behaviors indicate a more specific mechanism than ideal solid-phase dissolution and can be reconciled by considering SOM as a composite of “rubbery” and “glassy” polymers. Accordingly, the properties of SOM vary continuously from rubbery-like phases that have an expanded, flexible, and highly solvated structure to glassy-like phases that have a condensed, rigid, and less solvated structure (Pignatello, 1998, 1999; Xing et al., 1996; Xing and Pignatello, 1997). The glassy character has been suggested to increase with diagenetic alteration in the following natural progression: SOM r kerogen r coal and shale (Huang and Weber, 1997; Young and Weber, 1995). The nature of sorption is postulated to change along the rubbery–glassy con-

10

JOSEPH J. PIGNATELLO

Figure 3 Soil organic matter. (A) Hypothetical structure of a humic macromolecule (reprinted from Schulten and Schnitzer, 1993, with permission from Springer-Verlag). (B) Three-dimensional depiction of a natural organic matter colloid in aqueous solution. The colloid is an approximately spherical polymer mesh of entangled humic macromolecules that is swollen with water (water molecules not shown). The mass density increases toward the center. Some negative charges on the humic strands form ion pairs with metal cations, whereas others are balanced by counterions in solution. Cross-linking between strands is illustrated for the divalent cations Ca2+ and Mg2+. (Reprinted from Pignatello, 1998.)

tinuum in the same fashion as sorption of gases and organic molecules in polymers. In highly rubbery regions of SOM sorption occurs by solid-phase dissolution, whereas in glassy regions sorption occurs by a combined mechanism of solid-phase dissolution and site-specific, “hole-filling” processes. The holes are postulated to be nanometer-size pores made up of rigid humic segments, in which the guest molecules undergo an adsorption-like interaction with the pore walls. The sorption isotherm is thus given by the “dual-mode” equation (Eq. 3) (Pignatello, 1999; Xing et al., 1996; Xing and Pignatello, 1997), in which total sorption (qe ) is contributed by solid-phase dissolution (qD) and the sum of multiple site-selective processes (qL), each of which follows a Langmuir relationship:

Figure 3 Continued

12

JOSEPH J. PIGNATELLO

Figure 4 Rubbery–glassy polymer concept of soil organic matter. The perspective is intended to be three-dimensional. The rubbery and glassy phases both have dissolution domains in which sorption is linear and noncompetitive. The glassy phase, in addition, has pores of subnanometer dimension (“holes”) in which adsorption-like interactions occur with the walls, giving rise to nonlinearity and competitive sorption. The binding is analogous to host–guest inclusion complexes in chemistry. (Reprinted from Xing and Pignatello, 1997.)

qe = qD + qL = K D Ce +

n

b QC

∑ 1 +i bi Ce i =1

i

,

(3)

e

where KD is the (linear) dissolution domain coefficient made up of inseparable terms representing the rubbery phase and the dissolution domain of the glassy phases, Ce the equilibrium solution concentration, and bi and Qi are the Langmuir affinity and capacity constants for the ith unique site in the hole-filling or Langmuir domain. The dual-mode model is depicted in Fig. 4. Gas adsorption studies confirm the existence of internal nanoporosity in SOM samples which increases the total surface area by at least two orders of magnitude (Xing and Pignatello, 1997; de Jonge and Mittelmeijer-Hazeleger, 1996). The nanoporosity is correlated with the degree of nonlinearity in the isotherms and the degree of competition between compounds of like structure (Xing and Pignatello, 1997). Conditions that favor the rubbery state—increased temperature, the presence of cosolvents such as methanol, and high concentrations of cosolute—tend to make the isotherm more linear. The degree of nonlinearity follows the order expected on the basis of the glassy character of the material: humic acid  humin. As will be shown, there is increasing evidence that the mass transfer rates depend on the rubbery–glassy character of SOM.

SORPTION AND DESORPTION RATES

13

3. Carbonaceous Materials Other Than SOM Soils may contain forms of carbon not usually classified as SOM. These include ancient materials such as kerogen, coal, and shale, and “black carbon” (also known as “soot”), which refers to incompletely combusted organic material. Such materials are widely distributed in the environment and, because they are hydrophobic, are expected to have a high affinity for organic compounds (Kuhlbusch, 1998; McGroddy et al., 1996). The nature of these materials as sorbents of organic compounds is not well-known. Coal appears to have properties quite like glassy polymers—“internal microporosity” (Larson and Wernett, 1988) and demonstrable glass-to-rubber transition temperatures (above 300°C) (Lucht et al., 1987). Soots are expected to have some impenetrable hydrophobic surface. If this is true, sorption may occur by adsorption and condensation in fixed pores, as occurs in familiar inorganic materials. However, they may also have tar-like phases which behave more like absorption domains. PAHs, being products of incomplete combustion themselves, may become occluded in the interstices of soot particles during their formation in a way that makes them extremely unavailable (Gustafsson et al., 1997). Sorption of chemicals by NAPLs occurs by simple liquid-phase dissolution analogous to organic solvents such as hexane and octanol. The partitioning between the fluid phase and NAPLs is therefore governed by Raoult’s law (water– NAPL) or Henry’s law (vapor–NAPL); that is, the fluid-phase concentration is proportional to the mole fraction of contaminant in the NAPL times the solubility or vapor pressure, respectively, of a pure reference state (Schwarzenbach et al., 1993).

C. THERMODYNAMIC DRIVING FORCE FOR SORPTION Upon sorption from solution, an organic molecule exchanges one set of interactions with the solvent for another set of interactions with the sorbent. The molar free energy change at constant temperature and pressure encompasses free energy changes in sorbate–sorbent, sorbate–solvent, sorbent–solvent, and solvent– solvent interactions of all components involved in the sorption process, including displaced molecules such as water from the surface, and any reorganization occurring on the surface. For nonpolar and weakly polar compounds capable of interacting only by nonspecific physisorption mechanisms, sorption from water to mineral surfaces (Goss, 1997), as well as to SOM (Chiou, 1989), is driven principally by the hydrophobic effect. The hydrophobic effect results from the gain in free energy when a molecule possessing hydrophobic surface area is transferred out of the polar medium of water. The hydrophobic effect plays the same dominant role in aqueous solubility. Surface tension studies show that the hydrophobic

14

JOSEPH J. PIGNATELLO

effect is due almost entirely to the H-bonding cohesive energy of water (van Oss et al., 1988; van Oss and Good, 1988). It is thought that water molecules form an ordered cage around the hydrophobic portions of the solute, costing enthalpy, entropy, or both (Schwarzenbach et al., 1993; van Oss et al., 1988). H-bonding and dipolar interactions with the sorbent will increase the thermodynamic driving force for sorption only if such interactions with the sorbent are stronger than those with the solvent.

D. RATES OF ELEMENTARY PROCESSES It has been shown that sorption of organic compounds to soil particles usually involves the weak physisorption interactions. In solution, such interactions are practically instantaneous. For example, the lifetime of the H2O...NH2CH3 hydrogen bond in water is only 1.2 1011 s (Emerson et al., 1960). Van der Waals interactions are even shorter lived. The situation on the surface is more complex, however. Consider the elementary collision of a gas molecule with an unhindered plane surface having a number of “sites” of identical energy. The energy profile versus distance from the surface is illustrated in Fig. 5. As the adsorbate approaches the surface it descends into a potential energy well of depth Q. The instantaneous rate of adsorption is proportional to the pressure p and the concentration of vacant sites Sv. The instantaneous desorption rate is proportional to the concentration of occupied sites So. In the Arrenhius formulation, the rate expressions are Rate of sorption Aae(Ea*/RT )pSv ,

(4)

Rate of desorption Ade(Ed*/RT )So ,

(5)

Figure 5 Energy diagram for a physisorbing molecule approaching the surface.

SORPTION AND DESORPTION RATES

15

where Ea* and Ed* are the adsorption and desorption activation energies (kJ mole1), R is the universal gas constant, and Aa and Ad are the preexponential constants reflecting the “sticking probability.” Physisorption of a molecule from the vapor state to an unhindered surface is considered to be unactivated (Ea* near zero) (Adamson and Gast, 1997). Physisorption from solution may be slightly activated due to reorganization of solvent molecules around the sorbate and the surface. Desorption, however, is always activated because it requires “ascension” from the potential energy well. Assuming the principle of microscopic reversibility (that desorption follows the reverse pathway of sorption), Ed* Q  Ea*.

(6)

The Q is approximately equal to the isosteric heat of adsorption H (Ruthven, 1984). Values of H for vapor sorption on dry, nonporous silica gel (which represents a hydroxylated mineral surface) range from 36 to 63 kJ/mol for small nonpolar molecules, such as n-hexane, tetrachloromethane, and benzene, as well as polar molecules containing ether, cyano, nitro, ketone, and ester groups (Curthoys et al., 1974). Goss (1994) found that H for C1 –C10 compounds on moist surfaces of silica, quartz sand, and clays ranged from 28 to 50 kJ/mol. Adamson and Gast (1997) estimate the mean time of stay on the surface, , of a molecule at 25C, assuming Ea* 0 and a sticking probability of unity. For Q 37.6 kJ/mol, 106 s and for Q 83.7 kJ/mol, 102 s. Thus, we may expect that physisorption of typical compounds on a plane surface should occur in minutes or less. This is generally borne out experimentally (Adamson and Gast, 1997; Ruthven, 1984). The same conclusion of rapid equilibrium does not apply, however, when the elementary reaction involves forces stronger than physisorption. For example, in ligand exchange, Ea* may be appreciable due to such effects as orbital rehybridization and displacement of other ligands (OH, H2O, or organic ligand) from the inner coordination sphere of the metal ion (McBride, 1994). The Ed* will reflect the energy of the coordination bond, which is especially strong when the molecule coordinates through two or more functional groups (the “chelate effect”). There is little quantitative information available on the forward or reverse rates of ligandexchange reactions on soils or appropriate model sorbents. Also, the conclusion of rapid equilibrium does not apply to sorption in a sterically hindered environment such as the interstices of SOM. There, from the standpoint of the sorbate molecule as it tries to weave its way around the humic strands, each elementary jump from site to site is sterically hindered. Depending on the flexibility of the humic matrix, a cooperative movement in the position of humic macromolecules may be required in order to allow the sorbate molecule to pass. Thus, an elementary jump is expected to involve the making and breaking of numerous interactions simultaneously and therefore may be highly activated.

16

JOSEPH J. PIGNATELLO

III. SLOW SORPTION AND DESORPTION A. UPTAKE AND RELEASE PROFILES Evidence for slow (“nonequilibrium”) sorption phenomena has come from many sources—the extended tailing of breakthrough curves in soil column experiments, the formation of so-called “resistant” fractions in batch experiments, and the unexpected persistence of pesticides in the environment (Alexander, 1995; Pignatello, 1989). Since the late 1980s there have been many reports of chemical residues in samples collected from the field that appear to desorb extremely slowly, as well as reports of laboratory-spiked compounds that undergo sorption or de-

Figure 6 Sorption of phenanthrene in suspensions of Pahokee peat soil and derivatives. (a) Uptake by the peat soil showing slow approach to equilibrium. Apparent Koc is the ratio of sorbed to solution concentration referenced to organic carbon. (b) Normalized desorption of 2 g/g phenanthrene from the peat soil a in the presence of Tenax adsorbent beads after 3- to 100-day aging periods. (c) Same as b from peat humin (a is from White and Pignatello, unpublished; b and c are reprinted from White et al., 1999).

SORPTION AND DESORPTION RATES

17

sorption over weeks or months. This subject has been reviewed by Pignatello (1989) and Pignatello and Xing (1996). Figure 6a shows that phenanthrene, a three-ring PAH, required in excess of 110 days to reach equilibrium in a shaken suspension of a peat soil containing mostly organic matter (6.9% ash). Figure 6b shows the desorption of phenanthrene after various contact (“aging”) periods of 3–100 days (White et al., 1999). The desorptions were carried out in the presence of Tenax polymeric adsorbent, which rapidly sorbs phenanthrene as it leaves the soil, approximating conditions of zero concentration infinite bath and maximizing the driving force for desorption. One can see that the desorption rate slows with an increase in the aging period. It is worth noting that after only 3 days of sorption approximately 20% of phenanthrene strongly resists desorption over the subsquent 90 days in the presence of the infinite sink. Similar findings—that short-term contact can lead to formation of a strongly resistant fraction—have been reported by others (Kan et al., 1997, 1998). The following are observations pertaining to the resistant fraction: 1. Desorption is highly temperature dependent, being significantly enhanced by heating. For example, the apparent desorption activation enthalpy for aged 1,2-dibromoethane was 66 kJ/mol (Steinberg et al., 1987) and that of aged chlorobenzenes, polychlorinated biphenyls (PCBs), and PAHs ranged from 60 to 70 kJ/mol (Cornelissen et al., 1997b). 2. Desorption is accelerated by addition of cosolvents but only slightly by addition of surfactants (Deitsch and Smith, 1995). 3. Desorption is accelerated by breaking up particles (Steinberg et al., 1987; Ball and Roberts, 1991b). 4. Resistant fractions may be formed in soils containing no appreciable mineral matter (e.g., Fig. 6), in strictly inorganic porous particles (Farrell and Reinhard, 1994a,b; Werth and Reinhard, 1997), and perhaps even in colloidal-size particles (Maguire et al., 1995; Schlebaum et al., 1998).

B. RETARDATION MECHANISMS Slow kinetics has been exhibited by aliphatic and aromatic hydrocarbons, halogenated aliphatic and aromatic hydrocarbons, and agricultural chemicals. Generally, only physisorption interactions are open to them. Chemical and biological transformations, although quite possible, are irreversible in the sense that the byproducts cannot easily revert to starting compound and would not be identified as starting compound by the analyst using modern techniques. Therefore, the only reasonable explanation for slow kinetics for such compounds is mass transfer resistance—the resistance of the matrix to molecular diffusion. Diffusion is the tendency of molecules to migrate against a gradient in concen-

18

JOSEPH J. PIGNATELLO

tration (more correctly, a gradient in chemical potential) so as to achieve maximum entropy. Soil particles are characteristically porous and contain minerals such as SOM that can absorb small molecules within their interstices. These materials can provide resistance to diffusing molecules in many ways. Most suggested mechanisms for hindering the sorption process can be grouped into the following: pore diffusion (PD), and intraorganic matter diffusion (IOMD). Variations exist within each group, and in some cases there is some overlap. Although investigators have argued the merits of one compared to the other, it is likely that both operate, depending on soil properties. Resistant fractions can be generated in purely inorganic sorbents such as porous silica gel (Farrell and Reinhard, 1994b; Werth and Reinhard, 1997) and in pure organic materials such as low-ash peat soils (White et al., 1999). 1. Pore Diffusion PD attributes slow rates to hindered diffusion of molecules through the fixed intraparticle pore system. Fixed pores are more or less permanent and unaffected in shape by the presence of the diffusant. Porosity exists in cracks, lattice discontinuities, along grain boundaries, and in the interlayers of expandable clays. Pore sizes are classified by IUPAC according to their aperture (d ): Macropores: d  50 nm, Mesopores: 50  d  2 nm, Micropores: d  2 nm. In addition, there is a class of pores in the ⬃0.3- to 1-nm range referred to in the literature as “ultramicropores” or “nanopores.” For perspective, the C–C bond is ⬃0.15 nm long and CCl4 is ⬃0.5 nm in diameter. Researchers have different views on the nature of the pores and the root causes of hindered sorption. The pore surface may be organic or inorganic. In most PD models it is assumed that molecules instantaneously equilibrate locally between the pore liquid phase and the surface (“local equilibrium”). Diffusion in pores may be hindered with respect to diffusion in a bulk fluid by any or all of the following mechanisms: (i) tortuosity, a term encompassing elongation of diffusion paths relative to a straight line, variations in pore diameter, and the degree of pore connectivity as reflected by the presence of “dead-end” pores; (ii) sorption to pore walls analogous to a “chromatographic effect”; and (iii) steric interference from pore walls, especially in pores approaching the diffusant diameter. These will be discussed in more detail in Section IV,C. In addition, diffusion in small pores may be hindered by the viscous nature of water near hydrophilic surfaces where water molecules are strongly under the influence of the surface. Although the concept of pore diffusion is long known, Wu and Gschwend

SORPTION AND DESORPTION RATES

19

(1986) appear to be among the first to employ it to describe intraparticle diffusion of chemicals in soils and sediments. Ball and Roberts (1991b) used the PD model to describe sorption of trichloroethane (TCE) and tetrachlorobenzene in aquifer solids over long periods. They often obtained superior fits by including an instantaneously sorbing fraction of up to 30% of total. In these (Ball and Roberts, 1991b; Wu and Gschwend, 1986) and other studies (Kleineidam et al., 1999) the results were consistent with the nominal particle radius as the length scale over which diffusion occurs. By contrast, other studies (Carroll et al., 1994; Cornelissen et al., 1998b; Farrell and Reinhard, 1994b; Pignatello et al., 1993; Pignatello and Xing, 1996; Steinberg et al., 1987) found little or no dependence of diffusion rates on nominal soil particle radius, suggesting the appropriate diffusion length scale may be much smaller than the nominal particle radius, perhaps as small as 10 –100 nm. The diffusion length scale likely depends on the micromorphology of the soil particles in the sample. Pignatello (1990b) observed enhanced release of a portion of strongly resistant fractions of halogenated hydrocarbons by acidification of the suspension to pH  3. This suggested that some SOM, in particulate or coating form, had been shielded by mineral grains that were subsequently dispersed when the materials cementing them were acid dissolved. The results of Holmén and Gschwend (1997) on PAH transport in aquifer sand columns support this idea. They suggested that diffusion in porous oxide coatings on quartzite sand particles controls the rate of diffusion. The coatings, which consisted of fine-grained iron oxide and aluminosilicate clay particles, had porosities of 0.4 or 0.5, thicknesses up to ⬃200 m, and OC contents (0.7–1.6%) higher than the quartz substrate. Since the retardation of PAH transport was less than expected based on calculated Koc values, they inferred that only a fraction of SOM was accessed during a compound’s pass through the column. The flow velocities, however, were quite high–0.5–115 cm/ h for a 7-cm column or 1.7–400 column pore volumes per day. Farrell and Reinhard (1994b) and Werth and Reinhard (1997) desorbed TCE vapors from unsaturated silica gel columns or soils preequilibrated with TCE at fairly high relative pressures and 100% humidity. They observed biphasic kinetics (fast and slow phases). The small, highly resistant fraction of TCE that was formed was attributed to hindered diffusion in “hydrophobic micropores.” Corley et al. (1996) suggested that the resistant fraction of TCE and other volatile organic compounds (VOCs) might be associated with a neat VOC phase formed by capillary condensation in micropores or small mesopores during the sorption step. In their study of chlorinated benzenes and biphenyls in freshwater sediment (2.8–6.3% OC), Lick and coworkers (Borglin et al., 1996; Lick and Rapaka, 1996; Tye et al., 1996) proposed that sorption/desorption rates are controlled by diffusion in the pore network of flocs. Flocs result from the aggregation of sediment grains suspended in water. Their size and density is a function of sediment con-

20

JOSEPH J. PIGNATELLO

centration, fluid shear force, and water chemistry. Consistent with their mechanism, the effective diffusion coefficient depended on the floc size and porosity, sediment OC content, and (linear) partition coefficient of the sorbate. Another location where diffusion might be hindered is the interlayers of expandable clays. The interlayer gap is typically 1 nm—small enough to provide steric hindrance to diffusion or even size exclusion of some compounds. An important question that has not been satisfactorily answered regards availability of clay interlayers in natural soils to pesticides and other chemicals. The small amount of published information suggests that diffusion in the interlayer, when it is accessible, is relatively fast. Sawhney and Gent (1990) sorbed TCE and 1,2-dibromoethane vapors onto various expandable and nonexpandable clays under dry conditions. Desorption from the (expandable) smectite gave among the fastest rates, and X-ray analysis did not support penetration of the interlayer. In desorption of TCE from moist packed columns, Farrell and Reinhard (1994b) found that montmorillonite gave the smallest resistant fraction among many model and natural sorbents, but they, too, argued that interlayer penetration had not occurred. Huang et al. (1996) found that sorption of phenanthrene to bentonite was complete in a few hours, but no evidence of interlayer processes could be found. The remaining literature on the subject is confined to organoclays. Organoclays have quaternary ammonium ions (e.g., hexadecyltrimethyl ammonium) as exchangeable cations that are believed to provide an organophilic phase, or surface, with high affinity for hydrophobic compounds. Studies of organoclays [e.g., naphthalene and di-uron (Nzengung et al., 1997) and carbon tetrachloride and 1,2dichlorobenzene (Deitsch et al., 1998)] indicate that sorption equilibrium appears to be complete in hours and is much faster than sorption to SOM in the form of peat particles (Deitsch et al., 1998). Moreover, the solute–sorbent aging time did not significantly affect the rate of desorption (Deitsch et al., 1998). Questions remain, however, about how much sorption in these organoclays occurred in the interlayer versus on the edges (Nzengung et al., 1997). 2. Intraorganic Matter Diffusion a. General Considerations Since neutral organic compounds tend to associate predominantly with the SOM fraction, it is natural to consider whether SOM is the principal cause of hindered diffusion. SOM can hinder diffusion in at least two ways. First, even as a “rubbery” organic gel, SOM represents a highly viscous fluid that impedes molecular diffusion compared to water. Diffusion coefficients of small molecules in rubbery polymers compared to water are several orders of magnitude smaller and depend more strongly on the size and shape of the diffusant (Berens, 1989; Rogers, 1965). In the solid state, humic acid is believed to be a more rubbery form of organic matter than the SOM from which it was extracted (Xing and Pignatello, 1997). The

SORPTION AND DESORPTION RATES

21

Figure 7 Diffusion coefficient at 30C for gases and organic vapors in glassy () or rubbery () polyvinyl chloride. The rubbery state was obtained by adding phthalate ester plasticizers. (Redrawn from Fig. 9 of Berens, 1989, with permission.)

diffusion coefficients of toluene, n-hexane, and acetone is pressed humic acid disks range from 108 to 109 cm2 s1 (Chang et al., 1997), about the same as those in rubbery polymers at the same temperature, and may be compared to values of approximately 105 cm2 s1 in water. Second, glassy SOM offers a much greater impediment to diffusion than rubbery SOM because it is more rigid and condensed and it contains holes (nanopores) in which organic molecules can momentarily be detained (Pignatello, 1998; Pignatello and Xing, 1996; Xing and Pignatello, 1997). Figure 7 shows that the diffusion coefficient of gases and organic molecules in glassy polyvinyl chloride (PVC) is smaller than that in rubbery PVC for a given molecular diameter. Furthermore, they sharply diverge as the molecular size of the diffusant increases (Berens, 1989). Hole filling (and hole emptying) becomes an increasingly activated process as steric constraints at the hole increase. Figure 6c shows that desorption of phenanthrene from peat humin—the insoluble organic matter after humic acid is removed—is slower than that from the original peat SOM for a given aging period, reflecting the more glassy character of the humin compared to the native SOM (White et al., 1999). It has also been shown (White and Pignatello, 2000) that pyrene, a four-ring PAH, not only acts thermodynamically as a competing co-solute toward phenanthrene but also increases the rate of phenanthrene desorption, presumably by blocking nanopore sites ordinarily available to phenanthrene. This strongly suggests that the presence of nanopores impedes molecular diffusion inside SOM.

22

JOSEPH J. PIGNATELLO

Similar conclusions about the effect of SOM structure on diffusion rates have been reached by Weber and coworkers in their studies of hydrophobic compound sorption on soils and model materials (Weber and Huang, 1996). They proposed a three-domain model of soil. The domains fill up in the following order: Domain I: exposed inorganic surface Domain II: “amorphous” SOM (equivalent to rubbery) Domain III: “condensed” SOM (equivalent to glassy) Domain I, which is minor for hydrophobic compounds, is filled in minutes (Huang et al., 1996). The conclusion that domain III fills slowest is based on findings that the Freundlich exponent of phenanthrene (n) decreases with approach to equilibrium, especially in the first hours. The nonlinearity is assumed due to sorption in domain III. Similar changes in n with time were reported by Xing and Pignatello (1996) for 2,4-dichlorophenol, metolachlor, and 1,2-dichlorobenzene in two soils, including the 93% SOM peat soil. The decrease in linearity is due to the combined effects of increasing contribution from the glassy SOM with time and the intrinsic concentration dependence of diffusion in the glassy state (see Section IV,C,4). In the glassy state, diffusivity increases with sorbate concentration due to the following: (i) the decline in hole-filling sorption (see Eq. 3)—i.e., as the holes fill up, there is less impedance for subsequent molecules as they pass through. This is confirmed by the competition experiments between phenanthrene and pyrene mentioned above. (ii) At high enough concentrations the sorbate can “plasticize” the polymer—that is, bring about its conversion to a more rubbery state. b. Structure–Activity Relationships On the assumption that IOMD is the important limiting mechanism, many researchers have tried to relate the desorption rate parameter to molecular structure. Carroll et al. (1994) found that the effective diffusion coefficient (Deff) of PCBs in a sediment decreased with molecular size; about an order of magnitude decline in Deff occurred from monochlorinated to trichlorinated biphenyls. Brusseau and coworkers (Brusseau, 1993; Hu et al., 1995; Piatt and Brusseau, 1998) studied the transport of various compounds in packed soil columns. Through analysis of solute breakthrough curves they obtained a desorption mass transfer coefficient () for the noninstantaneous fraction (see discussion of the “two-site” model in sections IV,A and V,B). Since the residency time of the solutes in the columns was only a few minutes to a few hours, their results apply to short-timescale phenomena; applicability to longer timescale sorption requires caution. They found a linear log– log relationship between  and Kow: log  a log Kow  b,

(7)

where a and b are regression constants. This constitutes a linear free energy relationship (LFER) between sorption rate and sorption strength since log  is proportional to activation energy, E*d, and log Kow is proportional to log Ke, which in

SORPTION AND DESORPTION RATES

23

Figure 8 Linear free-energy relationship between the desorption mass transfer coefficient () and the first-order molecular connectivity index (1Xv) for PAHs, alkyl benzenes, chlorinated benzenes, and alkenes in two sandy aquifer samples (SB13-5 and SB-13-9) taken from a single bore hole at different depths. (Reprinted with permission from Piatt and Brusseau, 1998. Copyright 1998 American Chemical Society.)

turn is proportional to the thermodynamic free energy of sorption, Gsorp. The slope of Eq. (7) was negative, which means that the rate of desorption decreases with increasing affinity for the sorbent. Brusseau and coworkers interpreted the LFER in terms of a polymer diffusion concept. Thus, increasing molecular size results simultaneously in increasing hydrophobicity and decreasing mobility in the viscous organic phase. Such interpretation has also been given for diffusion of small- and medium-size molecules in polymers (Rogers, 1965). Brusseau (1993) and Piatt and Brusseau (1998) actually obtained slightly better LFERs between  and the molecular connectivity index X—a measure of topological size and degree of branching—than between  and Kow. Figure 8 presents such a correlation for hydrophobic compounds in two soils. They argued that diffusion through SOM is not just dependent on molecular polarity or hydrophobicity but also on size and shape. Such findings are consistent with diffusion limitations in an organic phase. They are not, however, inconsistent with diffusion limitations in fixed pore systems, as numerous studies of reference materials have shown (Kärger and Ruthven, 1992). 3. Effect of Soil Heterogeneity on Sorption Kinetics When dealing with a heterogeneous mixture of particles, the rate of sorptive uptake will be dominated by the faster-sorbing particles at short times and the slower-sorbing particles at long times. Pedit and Miller (1994, 1995) showed that better fits to the pore diffusion model could be obtained by incorporating different size classes into the model (see Section IV,C,6). The size classes not only have different diffusive path lengths but also may have different equilibrium sorption ca-

24

JOSEPH J. PIGNATELLO

pacities. Kleineidam et al. (1999) studied sorption to sands and gravels in southwest Germany and Switzerland. The samples, fragments of Triassic and Jurassic sedimentary rock, were separated according to size and lithographic type. They found that the rates of phenanthrene uptake depended on both particle size and properties. In general, dark-colored particles had the highest OC contents, lowest porosities, and highest sorption capacities while giving the slowest kinetics (e.g., 10% equilibrium in 500 days). The lighter-colored particles were just the opposite and showed the fastest kinetics (e.g., equilibrium in a few to 100 days). Most of the OC in these samples was ancient. 4. Hysteresis Hysteresis refers to the apparent asymmetry (nonsingularity) of the sorption/ desorption process. There is reference in the literature to asymmetry in the isotherm, where the curve defining the relationship between sorbed and fluidphase concentrations is different depending on whether it is determined in the forward (sorption) or the reverse (desorption) direction. There is also reference to nonsingularity in the rate parameters for sorption and desorption. Hysteresis has been observed in many soil–chemical systems but its causes have not been satisfactorily explained. Provided sorption is reversible and true thermodynamic equilibrium is attained, isotherms constructed from the sorptive and desorptive directions are expected to be superimposable. Figure 9 shows two examples of isotherm hysteresis by phenanthrene—in a riverine sediment (Fig. 9a; Kan et al., 1998) and in a shale material (Fig. 9b; Huang and Weber, 1997). In the former, a single sample was subjected to numerous desorption cycles, while in the latter, each sample was desorbed only once. As exemplified by Fig. 9a, the desorption curve often appears to intersect the ordinate at a nonzero value, indicating the presence of a “strongly resistant” fraction. Aside from method artifacts or chemical transformations (Rao and Davidson, 1980), there are several possible causes of isotherm hysteresis: 1. Formation of metastable states: Metastability plays an important role in the condensation/evaporation of gases in mesopores. The “hysteresis loop” commonly observed in gas adsorption isotherms is caused by formation of metastable films during uptake that abruptly coalesce to the condensed phase triggered by nucleation (Gregg and Sing, 1982). Hysteresis has also been observed in absorption of gases (e.g., CO2 and small hydrocarbons) by glassy, but not rubber, polymers (Kamiya et al., 1989, 1992). In this case the cause is believed to be slow volumestructural relaxation; that is, the microvoid volume which increases on sorption does not instantly relax to the original value on desorption. A mechanism involving metastable states in the context of sorption of dilute chemicals from soil solution, however, has not been articulated. 2. Insufficient time allowed for equilibrium: Nonattainment of equilibrium due

SORPTION AND DESORPTION RATES

25

Figure 9 Hysteretic isotherms of phenanthrene in two soils. Experiments were done in decant-reseal batch cycles with replacement of most of the fluid after each cycle. (a) Lula sediment. Adsorption: four cycles lasting 1–4 days each; desorption: 49 cycles lasting 1– 59 days each. (b) Norwood shale. Adsorption: 28 days; desorption, 14 days. [(a) Reprinted with permission from Kan et al., 1998. Copyright 1998 American Chemical Society. (b) Reprinted with permission from Huang and Weber, 1997. Copyright 1997 American Chemical Society.]

to rate-limited diffusion can lead to an underestimation of equilibrium sorbed concentration in the sorption direction and an overestimation in the desorption direction. A likely explanation for hysteresis in many cases, this is a vexing problem experimentally because true equilibrium can require very long times and may be concentration dependent. 3. Changes in the properties of the sorbent on sorption such that desorption takes place from a different molecular environment than sorption. For SOM it may be hypothesized that some sorbed molecules experience a conformational reararangement of the local humic matrix, resulting in encagement or at least an enhancement of the activation barrier for subsequent escape. An analogy has been

26

JOSEPH J. PIGNATELLO

made between sorbate-induced changes in the conformation of humic molecules and substrate-induced changes in the conformation of enzymes (Pignatello and Xing, 1996). Rearrangement has been observed in computational simulations of pollutant molecules [e.g., atrazine (Schulten, 1995) and pentachlorophenol (Schulten, 1996)] interacting with the hypothetical humic acid macromolecule shown in Fig. 3a. In order to explain isotherm hysteresis, Kan et al. (1998) proposed that total sorption includes reversible and irreversible components. The term irreversible, rather than implying permanent immobilization, is intended to mean that molecules leave a site by a different microscopic pathway than that by which they enter because of some kind of change of state taking place in the meantime (Adamson and Gast, 1997). Such behavior has been discussed in regard to adsorption of surfactants and polymers on oxides (Adamson and Gast, 1997, pp. 404–405) but without resolution of the cause. According to Kan et al. (1998), the “irreversible” compartment has a fixed maximum capacity for sorbate and fills in one or more steps in response to the solution-phase concentration. They proposed that the SOM matrix rearranges to trap the sorbate. Huang and Weber (1997) suggested that, in addition to nonattainment of equilibrium, hysteresis may be contributed by sorbate-induced expansion of condensed SOM to form pores that may “have no exits” once configurational changes in humic molecules occur. There are numerous examples of kinetic hysteresis, in which sorption appears to be faster than desorption (Connaughton et al., 1993; Farrell and Reinhard, 1994b; Harmon and Roberts, 1994; Pignatello et al., 1993). Harmon and Roberts (1994), for example, found the diffusion coefficient to be two to five times small-

Figure 10 Sorption and desorption rate curves for a hypothetical case. The cumulative mass gained or lost (M) relative to the mass gained or lost after infinite time (M) is shown for different Freundlich n values. The abscissa is the square root of dimensionless time. (Reprinted with permission from Lin et al., 1994. Copyright 1994 American Chemical Society.)

SORPTION AND DESORPTION RATES

27

er for desorption than sorption. “Thermodynamic” and “kinetic” hysteresis may have the same underlying cause; in studies of TCE and benzene vapor uptake by soil grains using an intragrain diffusion model, Lin et al. (1994) suggested that much of the diffusion asymmetry can be explained simply by nonlinearity of the isotherm. The results of the hypothetical case appear in Fig. 10. Note that the effect of nonlinearity is relatively minor unless the Freundlich exponent is less than about 0.75. Also, Farrell and Reinhard (1994b) found that the “slow fraction” of TCE remaining after N2 sparging was not well simulated by taking into account only equilibrium nonlinearity. A common assumption in many studies is that the rate parameter pertaining to sorption or desorption is single valued when in fact, because of the heterogeneous nature of soils, it is more likely to take on a range of values, depending on position along the uptake or release curve. Because most studies to date have focused on the behavior of the bulk of the chemical (first 80% sorbed or desorbed), much useful information has been missed.

IV. SORPTION KINETIC MODELS A. MODELS BASED ON BOND ENERGETICS The simple rate laws in Eqs. (4) and (5) seldom apply to real particles for two reasons. First, diffusion (mass transfer) is intrinsic to sorption kinetics because most sites are located in pores or within the SOM matrix and thus not directly accessible by molecules in the bulk fluid phase. Second, sites vary energetically because soils are heterogeneous. Nevertheless, kinetic models based on bond energetics, particularly those modified to account for soil heterogeneity, serve a purpose because, unlike diffusion models, they do not require knowledge about particle geometry. Only the essential features are presented for the models that follow: Readers are urged to consult the original papers for details about their application. The Langmuir kinetic model, reviewed by Adamson and Gast (1997), posits a collection of sites of uniform energy. Combining Eqs. (4) and (5) (since sorption and desorption events occur concurrently) and recognizing that the exponentials are constant at constant temperature,

[

]

[

]

* * dSo = A a e ( − Ea / RT ) pSv − A d e ( − Ed / RT ) So dt = ka′ p( ST − So ) − kd′ So

(8)

where total sorption ST Sv  So, and ka are the adsorption and desorption rate constants. Equation (5) may be put into a relevant soil–water frame of reference

28

JOSEPH J. PIGNATELLO

by replacing p with aqueous concentration C [M L3]; replacing So with sorbed concentration q [M M1]; and letting Q [M M1] be the capacity constant—the sorbed concentration when all sites are occupied. This gives dq ka  = C (Q − q ) − kd q dt 

(9)

where ka is in units of [L3M1T1] and kd is in units of [T1]. The Langmuir kinetic model is the basis for the familiar Langmuir isotherm since at equilibrium (dq/dt 0) Eq. (9) reduces to qe =

bQCe k , where b = a 1 + bCe kd 

(10)

Site nonuniformity has been dealt with customarily by including multiple sites. The two-site model (Hu and Brusseau, 1998) envisions an instantly reversible site S1, comprising a fraction f of the total sites, and a slower kinetic site S2: Ke C

S1

k2 k−2

S2

(11)

The equilibrium expressions are as follows. S1e f KeCen,

(12)

S2e (1  f )KeCen,

(13)

where Ke is determined on the basis of total sorbed concentration, q. The overall rate is the sum of the rates for each of the sites; ∂q ∂S1 ∂S = + 2 ∂t ∂t ∂t

(14)

Sorption at site 1 is governed by the equilibrium expression in Eq. (12), whereas site 2 follows a first-order reversible rate law in which the forward rate is proportional to S1 and the reverse rate is proportional to S2. ∂S2 = k2 S1 − k−2 S2 . ∂t

(15)

After differentiation and substitution to eliminate S1, and realizing that k2 /k2 (1  f )/f, the overall rate law becomes ∂q ∂C = n fK e C n −1 + k2 fK e C n − (1 − ∂t ∂t

(16)

SORPTION AND DESORPTION RATES

29

A Dutch group (Cornelissen et al., 1997a; 1998; 1999; Ten Hulscher et al., 1999) used a multicompartmental model (either two or three compartments) for desorption of polychlorinated benzenes and PAHs from sediments in the presence of Tenax TA polymeric beads as a third-phase adsorbent sink. The initial sorbed concentration q0 fiq0, where fi is the fraction in the ith compartment. Desorption in each compartment was regarded to occur in a first-order manner: dSi = − ki Si , (17) dt Si Si,0eki t.

(18)

Since the beads sorbed PAHs faster than does soil, the rate of resorption by the soil was assumed negligible. Thus, qt = ft e − kt t + fs e − ksi t + fvsi e − kvsi t q0

(19)

where fr, fsl, and fvsl refer to rapid, slow, and very slow fractions of initial chemical demarcated, somewhat arbitrarily, on the basis of discontinuities in the desorption curve. One could add as many different types of sites as one wanted. The “multireaction” model tested by Xue and Selim (1995) on alachlor sorption considers up to four sorption domains: an equilibrium domain Se obeying a Freundlich isotherm (n, Ke), a reversible kinetic domain S1, a “consecutive” irreversible kinetic domain S2 accessible only from S1, and a “concurrent” irreversible domain Sirr accessible only from solution.

B. DRIVING FORCE MODELS Several models are based on the idea that the rate is related to the degree that the system has reached equilibrium. The reversible model reviewed by Travis and Etnier (1981) assumes the rate is proportional to the difference between the equilibrium amount sorbed and the amount already sorbed: dq =  R ( qe − q ) =  R ( K e C n − q ) dt

(20)

where R is a rate parameter [T1]. Equation (20) is the reversible linear model or reversible nonlinear model, depending on the value of n. The reversible linear model is identical in form to the Langmuir model at very low concentration. The film resistance model (Eq. 21) assumes mass transfer resistance of molecules across a stagnant boundary layer (“film”) at the interface. The rate of sorp-

30

JOSEPH J. PIGNATELLO

tion is controlled by the difference between bulk solution concentration C and the concentration C* in equilibrium with the surface, dq =  FR (C − C*) dt

(21)

where  is the solids concentration [ML3] and  is the volumetric water content [L3L3]. One can see that if C* is related to q by the linear Freundlich isotherm then the film resistance model is of the same form as the reversible linear model. On the other hand, if C* is related to q by the Langmuir isotherm then it is of the same form as the Langmuir model. The second-order driving force model (Hendricks and Kuratti, 1982) regards the rate to be proportional to solution concentration times the difference between the equilibrium amount sorbed and the amount sorbed at time t: dq = C(qe − q ) dt

(22)

where  is now in units of [L3M1T1]. In sorption of a dye to Dowex 50 ionexchange resin or to activated carbon, Hendricks and Kuratti (1982) found this model to be superior to the reversible model, the Langmuir model, and two other models conceptually similar to the second-order and reversible models. The Fava–Eyring model (Fava and Eyring, 1956) is a nonlinear driving force model in which the rate is related to (t), defined as the distance from equilibrium divided by the initial distance from equilibrium; that is, (t) (q  qe)/(q0  qe), where q0 is the initial amount sorbed. The rate expression is given by Eq. (23), where a and b are constants. The hyperbolic sine term accounts for diminishing affinity for the sorbate with increasing loading: d = 2 sinh(b ) dt

(23)

C. DIFFUSION MODELS 1. General Considerations Soil particles are typically porous and contain highly viscous sorptive phases (i.e., SOM). Any mechanistic-based depiction of sorption would have to take diffusion of one kind or another into account. The form that the diffusion model takes is critically dependent on the geometry of the diffusing medium and the boundary conditions. To understand diffusion in heterogeneous systems such as soils we

SORPTION AND DESORPTION RATES

31

must consider studies of model sorbents. These studies are covered well by Kärger and Ruthven (1992) for fixed-pore sorbents and Frisch and Stern (1983) and Vieth (1991) for the organic solid state. The mathematics of diffusion is dealt with by Crank (1975). Transport diffusion is the nonequilibrium migration of molecules along the concentration gradient. Self-diffusion is scrambling of molecules due to their Brownian motions under equilibrium (no gradient) conditions and may be approximated by adding a tiny amount of radiolabeled tracer. Transport and self-diffusion are not necessarily equal. The fundamental diffusion equations are known as Fick’s first and second laws, which are given in Eqs. (24) and (25), respectively, for one-dimensional diffusion in the z direction and the general case: J = − D(c)

∂c ; ∂z

∂c ∂2c = D(c) 2 ; ∂t ∂z

J = − D(c) grad c

(24)

∂c = D(c) div(grad c) ∂t

(25)

where J(J) is the flux (E L2 T1], c is the total local volumetric concentration in the diffusing medium [M L3], and D(c) is the diffusion coefficient, or diffusivity [L2 T1]. Equation (26) gives Fick’s second law in radial coordinates for a vdimensional sample (v 1 for a slab, v 2 for a cylinder, and v 3 for a sphere), ∂c 1 ∂ ⎛ ( v −1) ∂c ⎞ = ( v −1) D(c) ⎟ ⎜r ∂t ∂r ⎝ ∂r ⎠ r

(26)

where r is the thickness (slab) or radius (cylinder, sphere). The equations apply as long as the sample is isotropic and homogeneous and not appreciably changed by the penetrant. The equations for the slab and cylinder assume no edge effects. In Eqs. (24)–(26), D is expressed as a function of concentration, although in many cases it may be constant at a given temperature; D is concentration dependent when 1. The diffusant alters the sorbent properties of the solid. For example, high loadings of an organic diffusant may cause an organic solid to soften and swell (Frisch and Stern, 1983; Lyon, 1995; Lyon and Rhodes, 1993; Vieth, 1991). This increases diffusivity because it makes the macromolecular chains more flexible, allowing the diffusant to pass more easily. 2. Sorption is nonlinear with concentration. Since the gradient in chemical potential is related to the logarithm of concentration, the transport diffusivity D is related to the self-diffusivity D (also called “corrected diffusivity”) (Kärger and Ruthven, 1992) by D =D

d ln p d ln c

(27)

32

JOSEPH J. PIGNATELLO

where p is the pressure of diffusant in the external fluid. It can be seen from Eq. (27) that D D when c is linear in p. If the isotherm is Langmuir-type, then D D (1  )1, where  is the fractional coverage, qeQ1. Sorption is always linear at infinite dilution; thus, D r D as c r 0. Diffusion is an activated process analogous to any elementary chemical reaction. In the Arrhenius formulation, D

DoeE*D /RT,

(28)

* is the diffusion activation energy and D is the preexponential constant. where ED o Analytical solutions to the diffusion equation are available for a variety of simple situations that might be encountered, including

• sorption/desorption in an infinite bath of constant or variable concentration; • sorption/desorption in a finite bath; and • “evaporation” at the surface to an infinite bath. The solutions are available for plane-sheet, cylinder, and sphere geometries in Crank (1975) and for cube geometry in Kärger and Ruthven (1992). They are too cumbersome to present here. Obviously, soil particles are not perfectly round; however, since the diffusion curves for the cylinder and cube cases are quite similar to that of the spherical case, except at long times, it is common practice to use the simpler spherical expression, whereupon the radius corresponds to that of a sphere having the same volume to external surface ratio. Analytical solutions are useful; however, the researcher must be aware that they come at the expense of many assumptions and simplifications. Some of the common ones include the following: 1. 2. 3. 4. 5.

Diffusion obeys Fick’s laws. The diffusion coefficient is concentration independent. The sample consists of particles of uniform size. The diffusant concentration is uniform throughout the particle at equilibrium. Local equilibrium exists in a finite element of the particle.

Assumption (1) may be invalid if the properties of the sorbent change in response to sorption, as could occur for SOM. Assumption (2) is not always valid but may be regarded to be valid over a narrow concentration range. Assumptions (3) and (4) are typically invalid for soils. Assumption (5) requires that, e.g., steric hindrance at the entrance/exit to sites is not rate limiting. 2. Diffusion in Fixed-Pore Systems Soil particles are aggregates of smaller grains that are cemented together with organic or inorganic matter (Greenland and Hayes, 1981). Figure 11 shows a rea-

SORPTION AND DESORPTION RATES

33

Figure 11 Soil particle aggregate with spherical-equivalent radius Ra made up of smaller grains of minerals of radius (rg ) and organic matter (rom). Illustrated is a micropore structure within grains and macropore/mesopore structure between grains.

sonable depiction of a soil aggregate having an equivalent spherical radius Ra. The aggregate consists of SOM particles, SOM coatings, mineral particles, and mineral particles coated with finer grains. It has a macropores/mesopores between the grains and a micropore network within individual grains. a. Macropores and Mesopores Diffusion is similar in macropores and mesopores in many respects and it is convenient to discuss the two sizes simultaneously. At relative humidities most relevant to the environmental (i.e., ⬃50%) the following conditions hold: (i) at least a monolayer of water is present on external mineral surfaces, (ii) micropores and mesopores are filled with water, and (iii) macropores are empty or only partially filled with water. As the humidity approaches 100%, the bulk Ke of a chemical is close to that observed under saturated conditions (Chiou, 1989). Diffusion of small molecules in macropores and mesopores is affected principally by tortuosity and sorption. For a particle of radius R, consider a volume element of that particle in which cp [M L3] is the local pore fluid concentration and

34

JOSEPH J. PIGNATELLO

sp [M L3] is the local sorbed concentration (i.e., c cp  (1  )sp, where  is the particle porosity). The governing equation in radial coordinates is 

∂cp ∂t

+ (1 − )

⎡ ∂ 2 cp 2 ∂cp ⎤ = Dp ⎢ 2 + ⎥ ∂t R ∂R ⎦⎥ ⎢⎣ ∂R

∂sp

(29)

and Dp is the local pore diffusivity which includes diffusion along the pore surface and diffusion in the pore fluid. Equation (29) assumes Dp to be concentration independent. By further assuming instantaneous local equilibrium, Eq. (29) becomes ⎡ ∂ 2 cp 2 ∂cp ⎤ = Deff ⎢ 2 + ⎥ ∂t R ∂R ⎦⎥ ⎢⎣ ∂R

∂cp

(30)

If the isotherm is linear —sp K cp, where K [dimensionless] is the local sorption coefficient—the effective diffusivity is given by Deff =

Dp  + (1 − )nK c

(31)

Pore diffusivity taking into account surface diffusion can be expressed as ⎛ 1 − ⎞ Dp = Dpf + ⎜ ⎟ K Dps ⎝  ⎠

(32)

where Dpf is the pore fluid diffusivity and Dps is the pore surface diffusivity. Surface diffusion may be important in pores with a large surface to volume ratio. Surface diffusivities are generally determined by the difference between observed pore diffusivity and the estimated value of pore fluid diffusivity. Surface diffusivity tends to increase with diffusant concentration and temperature. Surface diffusivities have been determined for gases in activated carbons and mesoporous glasses (Kapoor et al., 1989; Kärger and Ruthven, 1992). The contribution of surface diffusion in water-filled soil pores, however, is not well established. Pore fluid diffusivity is reduced with respect to bulk fluid diffusivity by a tortuosity factor (1) that reflects deviation from straight-line paths and pore interconnectedness and by a steric parameter  (1) that reflects steric hindrance by the pore walls: Dpf Db / .

(33)

Theoretical models predict to be proportional to 1 (Currie, 1960; Kärger and Ruthven, 1992; Wakao and Smith, 1962). Currie (1960) studied H2 diffusion in

SORPTION AND DESORPTION RATES

35

beds of glass beads, sand, carborundum, several soils, and sodium chloride packed at various bulk densities. The beds had porosities between 0.18 and 0.65 consisting primarily of macropores and mesopores. Currie found the exponent of  to vary with the material and with . Columns of high porosity gave an exponent of  close to 1, but this exponent became less negative with decreasing porosity to values as high as 0.38. For lower porosities and smaller pore sizes, tortuosity appears to be greater than predicted on the basis of  1. For example, for intraparticle diffusion of chlorinated hydrocarbons in aquifer sediments, was 102 or 103 times greater than predicted (Ball and Roberts, 1991b; Grathwohl and Reinhard, 1993). This can be attributed in part to incomplete interconnectedness of pores and in part to steric hindrance of diffusion. Steric effects begin to show when the ratio () of the minimum critical molecular diameter to the pore diameter reaches ⬃0.1 and become severe as  approaches unity. Most molecules of interest are ⬃1.5 nm in their smallest dimension. Hence, steric effects are important in micropores and smaller mesopores. Many theoretical models have been presented for steric hindrance that fit specific data reasonably well (Kärger and Ruthven, 1992; Lee et al., 1991). Satterfield et al. (1973) found  2.0 (0.09    0.5) for nonsorbing sorbates in microporous (3.2 nm diameter) silica–alumina beads. For sorbing compounds, the effects of tortuosity and steric hindrance are difficult to separate. The diffusivities of Cd2+ ion vs the larger SeOe2 ion in aluminas were consistent with a greater steric effect in micropores than in mesopores (Papelis et al., 1995). Combining Eqs. (31–33), we get Deff =

 −1 Dpf + (1 − ) K Dps  + (1 − ) K

(34)

To simplify modeling it has been customary to (i) neglect surface diffusion; (ii) combine tortuosity and steric hindrance into a single “effectve tortuosity” parameter (/  e ); and (iii) relate the local sorption coefficient to the bulk equilibrium sorption cofficient to the bulk equilibrium sorption coefficient (K  Ke , where  [M L3] is the particle density inclusive of intraparticle porosity). b. Micropores In the narrow confines of a micropore the diffusant is always under the influence of the surface; therefore, it is meaningless to distinguish between fluid and sorbed concentrations. The following is the governing equation in a spherical particle of radius r : ⎡ ∂2 s ∂s 2 ∂s ⎤ = D ⎢ 2 + ⎥, ∂t r ∂r ⎦ ⎣ ∂r

(35)

36

JOSEPH J. PIGNATELLO

Figure 12 (a) Schematic of biporous diffusion model: microporous spherical grains inside a mesoporous or macroporous spherical aggregate. (b) Coupled pore–intraorganic diffusion model: (i) completely mixed aggregate of uniform mineral microparticles and uniform spherical SOM microparticles or (ii) a porous mineral particle in which SOM microparticles are uniformly distributed throughout the internal surface. (Reprinted from Yiacoumi and Tien, 1994, with permission from the American Geophysical Union.)

where s(r) is the local concentration and D is the effective micropore diffusivity. Equation (35) is a restatement of Eq. (26) for a sphere and concentration-independent D . c. Dual-Resistance Models One may consider the case in which diffusion takes place in a biporous particle such as the one shown in Fig. 12a, which depicts a macroporous (or mesoporous)

SORPTION AND DESORPTION RATES

37

aggregate (Ra) made up of individual microporous grains (r ). This model combines Eqs. (29) and (35) and has the following boundary conditions: s(r, 0) = cp ( R, 0) = 0

(36a)

∂cp ∂s (0, t ) = (0, t ) = 0 ∂r ∂R

(36b)

s(r , t ) = Kcp ( Ra , t )

(36c)

3 sp ( R, t ) = s ( R, t ) = 3 r

r

∫ s ( r )r

2

dr

(36d)

0

(36e) where C(t) [ML3] is the external solution concentration. An analytical solution of the biporous dual-resistance model is found in Ruckenstein et al. (1971) and Lee (1978) for linear sorption under two types of wellmixed batch conditions: (i) a step change in surface concentration, in which the external fluid concentration is kept constant, and (ii) variable surface concentration, in which diffusion occurs from a finite medium. The mathematics are similar for two other conceptual systems: particle-scale independent macropore and micropore networks (not realistic for natural particles) and a particle with in-series macropore and micropore networks. Arocha et al. (1996) describe a numeric solution to cases in which sorption in the macroporous and microporous systems is nonlinear and obeys different Freundlich equations; they applied this nonlinear biporous dual-resistance model to toluene vapor sorption to dry soil crumbs and Na–montmorillonite. The diffusivity of toluene in the micropore zone was approximately 1012 cm2 s1, which is consistent with values reported for zeolites (Kärger and Ruthven, 1992). 3. Diffusion in Organic Matter Rubbery SOM, in which diffusant molecules are considered to be dissolved, is analogous to a highly viscous fluid. Equation (35) applies if the particle is spherical. However, as discussed previously, SOM may have both a dissolution domain and a Langmuir domain in which adsorption-like interactions occur, with an overall isotherm given by the dual-mode model [Eq. (3); for polymers, it is commonly assumed that n 1]. In this case, diffusion is more complex. The model of Vieth and Sladek (1965) assumes that diffusion occurs only in the dissolution domain (DD) and that molecules at Langmuir sites are totally immobile (DL 0). For diffusion through a planar sheet of polymer,

38

JOSEPH J. PIGNATELLO

J = Deff

∂s ∂s = − DD D ∂z ∂z

(37)

∂( sD + sL ) ∂2 s (38) = DD 2 dt ∂z Analogous equations can be written for other geometries. If we assume local equilibrium exists between sD and sL , then sL =

(bsL0 / K D )sD 1 + ( b / K D ) sD

(39)

Therefore, we can eliminate sL in Eq (38). After derivatizing we find that ⎡ ⎤ K′ Deff = ⎢1 + 2⎥ (1 + sD ) ⎦ ⎣

−1

(40)

where K bsoL /KD, and  b/KD. Note that Deff is a function of the local dissolution domain concentration and thus changes as sorption or desorption progresses. The model of Paul and Koros (1976) considers dual but independent diffusivities. This means that molecular jumps are allowed within the dissolution domain and between sites in the Langmuir domain, but cross-jumps are ignored. For diffusion through a planar sheet of polymer, J = Deff

∂s ∂s ∂s = − DD D − DL L ∂z ∂z ∂z

(41)

where s is the total local volumetric concentration in the polymer. Assuming local equilibrium and performing the mathematics in the analogous manner, it can be shown that

Deff

K ′( DL / DD ) ⎤ ⎡ ⎢1 + (1 + s )2 ⎥ D ⎥ = DD ⎢ K′ ⎥ ⎢1+ ⎢⎣ (1 + sD )2 ⎥⎦

(42)

The models of Barrer (1984) and Fredrickson and Helfand (1985) allow crossjumping between the Henry and Langmuir domains. The expression for effective diffusivity then becomes DDD + Deff =

⎡ ⎤ ( DDL + DLL ) K ′ sD K ′ sD K ′ 2 + D 1 − + DL ⎢ 2 0 3 0⎥ (1 + sD ) (1 + sD )sL (1 + sD ) sL ⎦ ⎣ K′ 1+ (1 + sD )2

(43)

SORPTION AND DESORPTION RATES

39

where DDD and DLL are the intradomain diffusivities, and DDL is the cross-domain diffusivity. For SOM, it must be assumed that there are many different kinds of sites. This makes the solution very complicated, indeed. Horas and Nieto (1994) derived an expression for Deff for the general case in which the energy is distributed: Deff = f

∞

∞

∫ ∫

−∞ −∞

l ij2

∂ Si vij ( Si0 − S j ) exp( − Eij / k T ) dEi dE j ∂S

[

]

(44)

where f is the fraction of jumps in the direction of the macroscopic flux, Si is the concentration of occupied sites of type i, Sio is the total number of sites of type i, ᐍij is the average distance between sites i and j, vij is the vibrational frequency of the molecule at site i, Eij is the potential energy barrier between sites i and j, T is the temperature (oK), and k is the Boltzman constant (1.38044 1016 erg/deg). The derivation of Eq. (44) requires many assumptions, some of which are debatable when applied to soils: (i) Only one molecule can occupy each site; (ii) the values of ᐍij , f, and vij are constant for a given system; (iii) the approximate jump frequency of molecules from site i to site j is a function of vij , the probability of finding a vacancy at j, and Eij (this follows from statistical mechanics and transition-state theory); and (iv) Eij is assumed to be equal to the thermodynamic depth of the potential energy well of site i (see Fig. 5), which is equivalent to saying that the energy level of the transition state is the same throughout the sorbent. Equation (44) has been integrated for the case in which the energy distribution is Gaussian. The solution is not presented here due to space limitations. Horas and Nieto have shown that it simplifies to Eq. (42) or Eq. (43) when the corresponding assumptions intrinsic to those models are employed. 4. Combined Pore Diffusion/Organic Matter Diffusion Model Yiacoumi and coworkers (Yiacoumi and Rao, 1996; Yiacoumi and Tien, 1994, 1995) developed a model that includes both pore and intraorganic matter processes simultaneously. It is mathematically equivalent for either of the situations depicted in Fig. 12b in which the particle aggregate of radius Ra is (i) a completely mixed aggregate of uniform mineral microparticles and uniform spherical SOM microparticles (radius rom) or (ii) a porous mineral particle in which SOM microparticles are uniformly distributed throughout the internal surface. The model takes into account pore diffusion (Dp) composed of pore fluid (Dpf) and pore surface (Dps) terms; adsorption on the mineral surface, assuming linearity and local equilibrium (sp Kmf cp); pore surface diffusion (Dps); sorption in SOM, assuming SOM is a homogeneous Henry’s law partition medium (som Komcp); and radial diffusion in SOM (Dom). The model also assumes that mass transfer across the bulk solution–macroparticle interface, as well as across the pore liquid–SOM interface, is never rate limiting.

40

JOSEPH J. PIGNATELLO

Figure 13 Sorption predicted by the combined pore–intraorganic matter diffusion model of Yiacoumi and coworkers. (Top) Approach to equilibrium in completely mixed batch experiment. (Bottom) Column breakthrough curve. (Bulk solution concentrations: cb, time ; cb0, time zero; ce, equilibrium; cin, input eluent concentration. is dimensionless time.). The  refers to the ratio of diffusion time scales in pores vs OM (Eq. 43). Small  reflects dominance of intraorganic matter diffusion resistance. (Reprinted from Yiacoumi and Tien, 1994, with permission from the American Geophysical Union.)

The governing equation is

[ + (1 − )Kmf ]

∂cp ∂t

+

⎛ ∂cp2 ∂N 2 ∂cp ⎞ = Dpf + (1 −  ) K mf Dps ⎜ 2 + (45) ∂t R ∂R ⎟⎠ ⎝ ∂R

[

]

Equation (45) is identical in form to Eq. (29) except that it includes a term for uptake of solute in pore liquid by SOM:

SORPTION AND DESORPTION RATES

∂c 3f ∂N ⎛ ∂c ⎞ = fom om = om Dom ⎜ om ⎟ ⎝ ∂r ⎠ r = r ∂t ∂t rom om

41 (46)

where fom is the fraction of organic matter and c¯om is the average sorbed concentration in SOM. The solution to Eqs. (45) and (46) for typical initial and boundary conditions was found by applying the method of Laplace transforms. The Laplace transform and the numerical techniques for inverting it are described in the original papers by Yiacoumi and coworkers cited previously. The solution was then used to provide aqueous solute concentration as a function of time in batch experiments and aqueous solute concentration as a function of both time and position in columns. On the basis of calculations, Yiacoumi and Tien (1994) examined the influence of pore diffusion resistance vs intraorganic matter diffusion resistance. The ratio of diffusion timescales  in pores vs OM is =

Dom / rom 2 . ( Dp /  )/ Ra 2

(47)

In batch systems (see Fig. 13, top), they found that , the dimensionless time for equilibrium, increased as intraorganic matter resistance increased in importance (i.e., as  decreased). In column experiments (Fig. 13, bottom), the breakthrough curves became more spread out (less like piston flow) as the intraorganic matter resistance increased in importance. The mentioned effects were more pronounced at low fomKom (i.e., at low SOM concentrations or for compounds with low affinity for SOM). They also applied their model to data existing in the literature on short-term sorption desorption studies. When the model was applied to the data of Nkedi-Kizza et al. (1989) on transport of atrazine and diuron in columns of Eustis soil, Yiacoumi and Tien concluded that diffusion resistance in SOM rather than the pore system was rate controlling. When applied to batch desorption of tetrachlorobenzene from Charles River and North River sediments by Wu and Gschwend (1986), the results were inconclusive in this respect. 5. Mixtures of Particle Sizes Particle size distributions in soil span many orders of magnitude. For a distribution of sphere sizes whose probability density function is (a), where a is Ra or r , the average concentration is obtained by mass balance as ∞

⎛ q − q0 ⎞ ⎛ Mt ⎞ = =⎜ ⎜ ⎟ ⎟ ⎝ M∞ ⎠ mix ⎝ q0 − q∞ ⎠ mix

⎡ Mt ⎤ 3 ⎥ ( a) a da ⎣ ∞⎦

∫ ( a ) ⎢ M 0

∞

∫ (a)a 0

3

da

(48)

42

JOSEPH J. PIGNATELLO

where the quantity [Mt /M](a) is the functional expression for the fractional mass of chemical taken up or released for each discrete particle size a. For narrow particle size fractions, Eq. (48) may be approximated as ⎛ Mt ⎞ =1− ⎜ ⎟ ⎝ M∞ ⎠ mix

∞

⎡M ⎤

∑ Xi ∑ ⎢ M t ⎥ ( a ) i

n =1 ⎣

∞

⎦

(49)

where Xi is the mass fraction of particles having geometric average radius a R¯ai or r¯ i. If experiments are done on pooled size fractions, the Sauter mean radius is recommended (Ball and Roberts, 1991b; Pedit and Miller, 1995): ⎡ n ⎤⎡ n F ⎤ as = ⎢∑ Fi ⎥ ⎢∑ i ⎥ ⎣i =1 ⎦ ⎣i =1 ai ⎦

−1

(50)

where Fi and ai are the mass fraction and geometric mean radius of each of the n subfractions pooled. Particle size can have an enormous effect on the diffusion curves (Fig. 14). The approach to equilibrium in a finite bath at fixed D is shown for uniform particles whose radii span a range of 100 arbitrary units. The small, medium, and large particles reach 95% equilibrium at approximately 0.14, 14, and 1400 units of time,

Figure 14 Effect of spherical particle radius on uptake or release of a solute in a mixed finite bath. Mt /Minf is mass of solute taken up or released divided by mass taken up or released at equilibrium. Radius (left to right: 1, 10, and 100) and time are in arbitrary units. Equations in Chapter 6 of Crank (1975) were used. Final mass taken up was 50% of total present.

SORPTION AND DESORPTION RATES

43

respectively. Multiple-particle class models better represented long-term sorption rates than did single-particle class models (Pedit and Miller, 1995). Some researchers have used a two-compartment model, with one instantaneous and the other diffusion limited (Ball and Roberts, 1991b; Lorden et al., 1998; Pignatello et al., 1993) (see Section V,C,2,c); in this case, the instantaneous compartment may partially account for the smaller particles if the nominal particle radius is indeed the appropriate length scale for diffusion.

D. STOCHASTIC MODELS Researchers are well aware of the high degree of spatial variability of soils and the difficulty of applying models designed for homogeneous systems. Several investigators have therefore attempted to model sorption kinetics stochastically. In such models it is assumed that sorption rate can be described with an array of firstorder rate constants k that are continuously distributed according to a probability density function (PDF). A useful PDF is the gamma function (Chen and Wagenet, 1995; Connaughton et al., 1993; Gustafson and Holden, 1990; Pedit and Miller, 1994, 1995), which expresses the frequency distribution of k as f (k ) =

 k  −1e −k ; ()

() =

∞

∫x

 −1 − x

e dx

(51)

0

where  and  are the shape and scale parameters, respectively, of the PDF curve, and () is a mathematical statistical function. The mean of k is 1 and the variance of k is ¹⁄₂ 1. The shape of the PDF ranges from Gaussian at  ⬃10 to positively skewed (i.e., high frequency of small k values) at  ⬃1. The fraction of initial mass remaining after time t in any compartment having rate constant k is simply ekt. Thus, the mass remaining after t for all compartments is Mt = Mi

∞

∫ 0

⎛  ⎞ f ( k )e − kt dt = ⎜ ⎟ ⎝  + t⎠



(52)

The values of  and  are obtained by regression. The time t1y necessary for the concentration to reach the fraction y of the initial concentration is t1y [y1/  1]. The gamma model has been applied to sorption/desorption of naphthalene in a freshly equilibrated soil and a field soil long contaminated with coal tar (Connaughton et al., 1993), spiked lindane in a soil (Pedit and Miller, 1994), diuron in an aquifer sand (Pedit and Miller, 1995), and TCE during transport of its vapors in an aquifer soil column (Lorden et al., 1998). Interestingly, Gustafson and Holden

44

JOSEPH J. PIGNATELLO

Figure 15 TCE vapor elution curves from an aquifer sand at 90% relative humidity and under “low concentration, low flow” (LC–LF) conditions (Lorden et al., 1998) showing the superiority of the gamma stochastic model (GS) over the two-site first-order (TSFO) model and the two-site spherical diffusion (TSSD) model. The latter two assumed one of the sites to be instantaneous with a sorption distribution coefficient KD determined experimentally; when KD was included as a third fitting parameter, both TSFO and TSSD fit better but still not as good as the GS model. (Reprinted with permission from Lorden et al., 1998. Copyright 1998 American Chemical Society.)

(1990) successfully applied the gamma PDF to field and laboratory “dissipation” data of 45 different pesticide–soil systems, where dissipation means all fate processes including sorption. Pedit and Miller (1994, 1995) investigated a log-normal PDF, which is given by f (k ) =

⎡ 1 ⎛ ln k − ⎞ 2 ⎤ 1 exp ⎢− ⎝ ⎥  ⎠ ⎦ (2  )1 / 2 k ⎣ 2

(53)

where the arithmetic mean of k is k¯ exp(  2 /2) and the variance of k is V k¯ 2exp(2  1). The log-normal PDF simulated lindane uptake by soil more accurately than did the gamma PDF. Not surprisingly, stochastic models often outperform other models, given the same number of adjustable parameters. They outperform the single-site first-order model (Pedit and Miller, 1994), the two-site first-order model (Connaughton et al.,

SORPTION AND DESORPTION RATES

45

1993; Lorden et al., 1998), and the two-site diffusion model (one instantaneous and one diffusion controlled) (Lorden et al., 1998; Pedit and Miller, 1994). Figure 15 shows the superiority of the gamma model in describing the tailing of TCE elution curves observed by Lorden et al. (1998). The value of a stochastic approach depends on if and how well it can predict behavior under different conditions or different soil–chemical systems than the one in which the model parameters were obtained. Lorden et al. (1998) demonstrated predictive capability of the gamma function for elution of TCE vapors at different flow rates and TCE pressures. The ability of stochastic models to translate to different soil–chemical systems has not been established, however. Also, these models do not offer much mechanistic insight.

V. EXPERIMENTAL METHODS In a review such as this, it is worth discussing the methodology used in sorption/desorption rate studies. The sorption time frame may be transient or steady state. In the transient time frame one observes a changing concentration gradient, whereas in the steady-state time frame one establishes a stable concentration gradient and measures the mass of chemical entering or leaving the system. The transient time frame is the one most often used for soils. Popular techniques for measuring transient processes include batch, column, zero-length column (ZLC), and stirred-flow cell (SFC). Batch techniques measure phenomena on the intraparticle scale since the external surfaces are exposed to a mixed fluid of uniform concentration. Column techniques measure phenomena on intra- and interparticle scales concurrently. The column’s effluent concentration profile depends on both dispersion (longitudinal mixing) and sorption/desorption processes. Column techniques avoid abrasion of particles and can be more realistic. ZLC and SFC are hybrid techniques that attempt to combine the advantages of each.

A. BATCH TECHNIQUES In solution–solid systems, agitation (stirring or shaking) is usually required to achieve uniform conditions, whereas in vapor–solid systems bulk gas diffusion is rapid enough to maintain uniformity at all but the fastest uptake rates, provided the sorbent bed is sufficiently thin. Sorption may be carried out under infinite or finite source conditions, which require different approaches. Under infinite source conditions, the fluid-phase concentration is not reduced by sorption and uptake has to be measured by analysis of the solid. This approach is commonly used in vapor–

46

JOSEPH J. PIGNATELLO

solid studies in which uptake is followed gravimetrically. When the source is finite, the bulk fluid-phase concentration changes, providing a convenient way to follow uptake, assuming no chemical transformations occur. 1. Measurement Techniques Batch solution–solid experiments are commonly performed by mixing the components in sealed flasks and, at predetermined times, measuring solution concentration after first separating the phases. In estimating Ke, the optimal degree of sorption is about 55% (McDonald and Evangelou, 1997); the same degree of final sorption would apply to an uptake kinetic experiment if the objective is to quantify changes in Ke with time. Although this methodology appears simple to manage, the researcher must be aware of the following potential complications: 1. “Bottle losses”: Adsorption on glass walls can be important if it is more than a few percent of sorption to the soil. It is seldom an issue except for highly hydrophobic compounds and may be corrected by running controls. Diffusion through polymer materials is greater for small molecules and occurs regardless of whether the chemical comes into contact with the material via the solution or vapor phase (the chemical potentials in the two phases are equal). Little or no loss through Teflon cap liners was reported for lindane during 84 days (Miller and Pedit, 1992) or for phenanthrene over several weeks (White et al., 1999). Polymer vessels and soft polymer stoppers or cap liners (e.g., “butyl” rubber or “phenolic”) should be avoided. Polymer materials can be avoided altogether by using flamesealed glass ampules (Ball and Roberts, 1991a) or screw-cap vials with foil stretched tightly over the lip of the vial before the cap with its liner is screwed on (Huang et al., 1997; Xia and Ball, 1999). Piercing the septum liner with a needle to remove sample will expose hydrophobic septum material (usually silicone rubber) to the flask contents, resulting in noticeable extraction even within hours; thus, most researchers use the “bottle point” method in which replicates are sacrificed at time points. 2. Phase distribution artifacts, such as headspace partitioning of volatile compounds and sorption to colloidal particles not removed by centrifugation or filtration: The former can be assessed by considering Henry’s law. The latter is potentially significant for highly hydrophobic compounds (log Kow  ⬃4) or compounds that otherwise interact strongly with colloids (Schwarzenbach et al., 1993). Assay of the liquid phase may not distinguish between colloid-bound and truly dissolved fractions. Moreover, one cannot assume the concentration of nonsettling colloids will remain constant throughout an experiment. The problem of sorption to colloids can be avoided by retaining the soil in dialysis tubing (Allen-King et al., 1995), provided a correction can be made for sorption to the tubing. 3. Artifacts caused by agitation: Shear forces generated by vigorous agitation may change the particle size distribution and expose additional surface area, thus

SORPTION AND DESORPTION RATES

47

affecting rates. Stir bars grind particles (Wu and Gschwend, 1986). The minimum degree of agitation necessary to achieve rapid mixing on the timescale of sorption is recommended; slow end-over-end tumbling seems to be the best technique. 4. Method of analyte introduction: Strongly sorbing compounds tend to sorb to the first surfaces encountered and may not readily redistribute to other particles; therefore, attention must be paid to the way in which initial contact between particle and chemical is achieved. Homogeneous particle coverage requires fast dispersion of particles into a large volume of a solution. When spiking bulk soils in preparation for desorption studies the method of Karickhoff et al. (1979) is recommended: First, the compound is deposited on the walls of the flask by evaporating a solution in a volatile carrier (e.g., CH2Cl2) and then the soil suspension is added, followed by agitating for 24–48 h. 5. Biodegradation: Sterilization may be accomplished by heat, irradiation, or chemical treatment. Wolf et al. (1989) compared sterilization techniques based on plate counts of bacteria and fungi. The following were completely effective on three soils: double or triple autoclaving, propylene oxide (48-h exposure), 60Co irradiation (0.05 MGy), and HgCl2 (500 g/g). The remaining treatments were less effective and followed the order: once-autoclaving  oven-drying NaN3 ⬃ CHCl3 ⬃ microwave treatment  control ⬃ antibiotics. Autoclaving is generally avoided because of concerns it can affect soil chemistry, particularly SOM structure. Propylene oxide may react chemically with SOM, potentially affecting sorption of the compound of interest. [The same holds for formaldehyde, which was not tested by Wolf et al. (1989) but is widely known to be effective.] 60Co -irradiation equipment is not widely available. Mercuric chloride had the least effect of all treatments on soil CEC, pH, and extractable metal ions. Xing et al. (1996) showed that HgCl2 had no effect on the 48-h Freundlich sorption parameters of metolachlor. Mercury presents a disposal problem, however. Sodium azide (NaN3) is a popular bacteriostat that binds to cytochromes, inhibiting terminal electron transport. However, its efficacy cannot be taken for granted since it may not inhibit enzymatic transformations that require little energy, and because bioactivity may be revived if dilution of the NaN3 occurs in subsequent steps [e.g., as it did in the protocol of Wolf et al. (1989)]. To separate the solid and liquid phases, centrifugation is not suitable for sampling frequencies more often than about one per 10 minutes. Filtration through a microporous filter (e.g., 0.2 m) can increase sampling frequency to perhaps one per minute. Filtration has been employed in experiments at temperatures other than ambient in order to minimize temperature effects during sample handling (Xing and Pignatello, 1997). Wu and Gschwend (1986) used an air-stripping apparatus which constantly recirculated air through the soil suspension and then through a detector. This technique, which is suitable for sorption occurring over a few seconds to 48 h, avoids the colloid problem because activities are measured in the gas phase and related to solution-phase concentration via Henry’s law. Similar appa-

48

JOSEPH J. PIGNATELLO

ratuses are described by Brusseau et al. (1990) and Benzing et al. (1996). For semivolatile compounds Karickhoff and Morris (1985) designed a gas purge cell in which air was passed through glass frits in contact with the suspension and the vapors were collected in external traps containing Tenax, a polymeric hydrophobic resin. Harmon and Roberts (1994) describe a purge apparatus that fits around a glass ampule after the flame-sealed top is broken off. The apparatus of Benzing et al. (1996) can operate in both recirculate and purge modes. A popular way to monitor desorption is by use of the successive-dilution technique, in which a known volume of the supernatant is replaced with “clean” water. The mathematical model that one uses needs to take into account the stepped boundary conditions that this technique entails. Pignatello and coworkers (Pignatello, 1990a,b; White et al., 1999) and Cornellison et al. (1997a,b, 1998a) used an in situ trap of Tenax polymeric adsorbent beads. Tenax has a high affinity for nonpolar compounds, even C1 and C2 halogenated hydrocarbons (Pignatello, 1990b). The beads are readily separated from the soil after centrifugation because they float. Soil particles have little tendency to adhere to the beads and are easily rinsed off. Carroll et al. (1994) used XAD-4 resin as an in situ trap; however, K2CO3 had to be added prior to sampling to increase the solution density in order to make the beads float. The purge gas or in situ polymer trap rapidly lowers the solution concentration to near zero. Hence, diffusion kinetics can be modeled using equations that describe “diffusion with surface evaporation” (Crank, 1975). These equations take forms that depend on the extent of diffusion and the sorbent shape. If the particle is a sphere of radius Ra and initially at uniform concentration, the surface boundary condition uses the linear driving force assumption −D

∂c = if (cs − co ) ∂R

(54)

where cs is the concentration just within the sphere, co is the concentration required to maintain equilibrium with the solution, and if is the interfacial mass transfer coefficient [L T1]. Provided that enough time has passed so that the rate of purging from the solution is equal to the rate of desorption, the governing equation for the fractional amount of substance leaving the particle at any time is Mt =1− M∞

∞

∑

(

6 L2 exp −2n Dt / a 2

2 n =1  n

{

2n

}

+ L( L − 1)

),

(55a)

where L a/D and the n values are the roots of n cot n + L − 1 = 0 .

(55b)

SORPTION AND DESORPTION RATES

49

In their studies of tetrachloroethane (PCE) desorption from aquifer sediments, Harmon and Roberts (1994) purged the suspension only intermittantly during desorption; therefore, they used a model combining boundary conditions for diffusion with surface evaporation and diffusion to a solution of finite volume. O’Dell et al. (1992) describe a static batch reactor in which nearly saturated soil containing a herbicide (imazethapyr) was placed in syringe barrels. At predetermined times the soil solution was displaced with saturated CaSO4 when the syringe was placed under vacuum. By including a tracer in the CaSO4, they verified that the recovered soil solution was uncontaminated with the displacing solution. Direct analysis of the solid is sometimes carried out. Sorbed concentration is conveniently measured by solvent extraction after correcting for solute in the water removed along with the solid. However, resistant sorbed fractions may not be completely recovered unless extraction is fairly vigorous. Sawhney et al. (1988), Pignatello (1990a), and Huang and Pignatello (1990) found that complete solvent extraction of even weakly sorbing compounds required several hours at 75C. Similarly, Ball et al. (1997) found that a 16-h methanol extraction at 70C was required to remove volatile organic contaminants in field aquifer sediments. A watermiscible solvent (e.g., acetone, methanol, or acetonitrile) will more effectively penetrate SOM and intraparticle pores than an immiscible solvent. Harmon and Roberts (1994) found that extraction of [14C]PCE from aquifer sediments by nonmiscible scintillation cocktail required tens of days. Tognotti et al. (1991) describe a single-particle technique in which the particle is suspended in a vapor-laden gas stream within an electrodynamic thermogravimetric analyzer (EDTGA). Uptake or release of toluene or carbon tetrachloride was measured by the voltage required to keep the particle stationary in the EDTGA as a function of time. To determine the effect of temperature, a CO2 laser was used to heat the particle.

B. COLUMN TECHNIQUES 1. General Considerations Glass columns with perforated Teflon end plates and Teflon adjustable plunger are commercially available. Stainless-steel preparatory-scale liquid chromatography columns have also been widely used. Column packing techniques apparently have not undergone systematic investigation. Typically, columns are packed by layering dry material a little at a time in an attempt to create a uniform bed. The investigator should be aware, however, that dry soil has a tendency to segregate by particle size, making it difficult to scoop up a representative sample from the container from which the soil is taken. Moistening the soil (e.g., to 5 or 10% wa-

50

JOSEPH J. PIGNATELLO

ter by weight), if appropriate for the experiment, helps alleviate this problem. Pignatello et al. (1993) filled the column with air-dry soil, allowing about one-fourth of the volume to be free of soil; the soil was then homogenized by manually rocking and rotating the column axially for several minutes. After packing, the column is then slowly saturated from the bottom with eluent. Most investigators use dilute (e.g., 0.01 M) CaCl2 as the eluent to help prevent dispersion of clays and mobilization of colloids. The column is often “conditioned” by passing through many pore volumes of eluent prior to injecting the solute solution. Eluent is pumped through the column with a constant-volume displacement pump such as the kind used in liquid chromatographic systems. A three-way valve is installed before the column to allow switching between reservoirs containing eluent with or without solute. The effluent may be collected with a fraction collector for later analysis or passed through a UV-visible, scintillation counter, fluorescence, or other type of flowthrough detector. For vapor-elution studies, Lorden et al. (1998) describe an apparatus that diverts flow to a gas chromatograph through a programmable automatic switching valve. Injection of the solute may occur according to a step function or a pulse function. For a step function, the concentration of solute is increased abruptly from an initial value (usually zero) to a second value, and the solute “breakthrough curve” at the terminus is observed until the concentration reaches the stepped value. Desorption may be observed at some later time by switching the inflow to clean eluent. For pulse injection, the solute is introduced over a finite period and the effluent concentration is monitored at the terminus. Many reviews are available that discuss contaminant transport in soil columns (Brusseau and Rao, 1989; Parker and van Genuchten, 1984; van Genuchten and Wagenet, 1989). Solute transport when no decomposition occurs is described by the advection–dispersion (A–D) equation with sorption: 

∂C ∂q ∂2C ∂C + = Dh 2 − v ∂t ∂t ∂x ∂x

(56)

where  is bulk porosity [L3 L3],  is bulk solids density [M3 L3], Dh is the hydrodynamic dispersion coefficient [L2 T1], v is the linear average flow velocity [L T1], and x is longitudinal distance [T1]. The term of focus, !q/!t, represents the rate of change in average sorbed concentration as seen by the flow regime. In addition to the complicating effects of dispersion, another limitation of column studies is that the degree of sorption that occurs depends on the residency time of the solute in the column and hence the flow rate. That is, slow sorption states may not be fully accessed by solute molecules in a fast-moving fluid. For example, Brusseau (1992), using the two-site model, showed more than an order of magnitude decrease in the sorption rate parameter for several compounds with a decrease in flow rate from ⬃45–90 to ⬃5 cm/h (13 to 0.7 pore volumes per hour).

51

SORPTION AND DESORPTION RATES

Likewise, Lorden et al. (1998) showed a decrease in the characteristic time for sorption of TCE vapors with increasing flow rate. It is difficult to achieve flow rates of less than one pore volume per day; however, batch studies show that a substantial fraction of sorption may require weeks or months to complete. Obviously, in the above-cited studies the equilibrium and kinetic fractions of the two-site formalization are accommodating to a shift in timescale of elution. One must also be aware that significant amounts of injected chemical may reach slow sites but not desorb on the timescale of the experiment. For all practical purposes, such amounts are “irreversibly” sorbed and thus missed by the model. For example, Spurlock et al. (1995) carried out delayed elution experiments with monuron and fenuron, relatively polar compounds with high water solubilities, in a soil (3.4% OC). They found that the elution curves were affected little by the length of the prior contact period (8, 80, or 240 days). However, appreciable amounts (10–30%) remained sorbed after the effluent concentration declined to C/Co  0.01. This residual sorbed herbicide increased with contact time, became increasingly greater than predicted on the basis of the equilibrium law, and was uniform along the length of the column. 2. Transport Models with Sorption Kinetic Term a. Two-Region Model In this model (van Genuchten and Wierenga, 1976, 1977; van Genuchten et al., 1977), the pore space is divided into mobile water (m) and immobile water (im) regions, with advection and dispersion occurring only in the mobile region. The A–D equation is thus m

∂Cm ∂S ∂Cim ∂S ∂ 2 Cm ∂Cm + (1 − f ) im =  m Dh + f m +  im 2 − m v ∂t ∂t ∂t ∂t ∂x ∂x

(57)

where f is the fraction of sorption occurring in the mobile region. Suppose sorpn and tion is locally instantaneous and obeys the Freundlich equation Sm f KeCm n Sim (1  f )KeCim. Upon differentiating Sm and Sim with respect to t and substituting into Eq. (57), one obtains

(m + fnKcCim n −1 ) ∂C∂tm + (im + (1 − f )nKc Cim n −1 ) ∂C∂tim

= (58)

The relationship between Cm and Cim is established by assuming that the driving force for mass transfer between the regions is simply proportional to the difference in their respective liquid-phase concentrations; thus,

52

JOSEPH J. PIGNATELLO

(im + (1 − f )Kc nCim n −1 ) ∂C∂tim

= (Cm − Cim )

(59)

where  is a mass transfer coefficient. An analytical solution to Eqs. (58) and (59) for the linear sorption case is available (van Genuchten and Wierenga, 1976). For the nonlinear case the solution must be obtained numerically. If Ke is known, the fitting parameters are , f, and im. A reasonable assumption is that f and im are equal (Nkedi-Kizza et al., 1984) or im may be obtained by fitting the model to the elution curve of a nonsorbing tracer (Spurlock et al., 1995). b. First-Order Models A widely used model is the two-site (equilibrium and kinetic) one, described in Section IV,A (Eqs. 11–15) (Cameron and Klute, 1977; Nkedi-Kizza et al., 1984; van Genuchten and Wagenet, 1989). The appropriate expression for !q/!t is Eq. (16). The A–D equation becomes ∂2C   ⎛ ⎞ ∂C 1 + f nK e C n −1 +  f K e C n − (1 − f )q = Dh 2 − ⎝ ⎠ ∂t   ∂x

(

)

(60)

Ordinarily, Ke is determined in separate batch experiments or calculated by moments analysis (Valocchi, 1985), and Dh is determined by eluting a nonsorbing tracer, such as 3H2O. Thus, Eq. (60) has two fitting parameters, k2 and f. An analytical solution is available for the linear case only (van Genuchten and Wagenet, 1989). c. Diffusion Models Diffusion concepts have been used directly or in conjunction with other models. Pignatello et al. (1993) and Lorden et al. (1998) used a two-compartment model having a rapid equilibrium compartment (S1, f ) and a second slow compartment (SD, 1  f ) governed by radial diffusion so that q S1  SD: C

Ke

D S1 ⇔ SD

(61)

The model regards the diffusion domain as being a collection of idealized equal spheres (radius, a) in which diffusion obeys a Fickian diffusion law Eq. (62a) (analogous to Eq. 26) with boundary conditions as given in Eq. (62b): ∂s D ∂ ⎛ 2 ∂s ⎞ = 2 ⎜r ⎟ ∂t r ∂r ⎝ ∂r ⎠

(62a)

∂s(r ) =0 ∂r r = 0

(62b)

and s(r = a, t ) = (1 − f ) K e C

SORPTION AND DESORPTION RATES

53

The volume-averaged sorbed concentration in the diffusion domain is given by a

SD =

3 s(r, t )r 2 dr a 3 ∫0

(63)

where  is the particle density (g/cm3). Curve fitting requires numerical techniques. A two-region diffusion concept was proposed by Nkedi-Kizza et al. (1982) for inorganic solutes and later used by Young and Ball (1994) to simulate breakthrough of PCE in aquifer sediments. In this conceptualization, the “immobile zone” is the intraparticle porosity () and its associated sorption capacity, and the “mobile zone” is the interparticle porosity and its associated sorption capacity. The same two-region A–D equation (Eq. 57) applies, except that im is replaced by , and Cim is replaced by the volume-averaged concentration inside the particles C¯im (spherical, Ra): Cim =

3 Ra3

Ra

∫0

Cim ( R, t ) R 2 dR

(64)

The term !C¯im(R,t)/!t is expressed in terms of intraparticle radial diffusion laws (Eq. 30; cp C¯im) and the series of equations solved numerically. This model assumes that particles are uniform in size, porosity, and chemistry; that Ke is the same in the mobile and immobile regions; and that Cm cim at the particle/mobile–water interface. The  was determined (Ball et al., 1990) by mercury porisimetry which measures intraparticle porosity down to pore widths on the order of ⬃10 nm (mesopore to macropore range). Piatt and Brusseau (1998) analyzed elution curves using the two-site model and then related the S2 site first-order desorption rate constant k2 (see Section IV,A, Eq. 11) to an organic matter diffusivity Dom, a particle shape factor ", and a diffusion length scale, l: k2 "Dom /l 2 (1  f ).

(65)

A solution for solute transport through a bed of uniform porous pellets for a step function change in solute concentration in the influent is provided by Yiacoumi and Rao (1996) and Yiacoumi and Tien (1994). Diffusion occurs within the pellet pore system and SOM microparticles contained therein, as described in Section IV,C,5. The solute uptake per pellet is given by t ⎛ ∂cp ⎞ M = ∫ 4 Ra2 Dp ⎜ ⎟ dt ⎝ ∂R ⎠ R 0

(66)

a

where the (!cp /!R)Ra term is from Eq. (45). Upon considering solute uptake on a per-gram sorbent basis, the rate of sorption is given by

54

JOSEPH J. PIGNATELLO t ⎫⎪ ∂ ⎧⎪ 3(1 − ) Dp ⎛ ∂cp ⎞ ∂q ∂ ⎡ 1− ⎤ = M dt = ⎬ ⎨ ⎜ ⎟ Ra ∫o ⎝ ∂R ⎠ R ⎪ ∂t ∂t ⎢  4 R3 ⎥ ∂t ⎪  a ⎥ ⎢ ⎭ ⎩ a ⎦ ⎣ 3

(67)

where  is the particle density (g/cm3). Equation (67) is then substituted into the A–D equation (Eq. 56). The combined equations were solved by Yiacoumi et al. by the method of Laplace transforms. This gives a complex expression (not reproduced here) relating the observed bulk solute concentration in the Laplace domain C˜ as a function of the following: initial bulk solute concentration Co (if other than zero), the input concentration Cin, diffusivities (Dpf, Dps, and Dom), partition coefficients (Kmf and Kom ), pellet and SOM microparticle radii (Ra and rom), bulk moduli (, , and ), flow velocity (v), dispersion coefficient (Dh), and distance along the column. Inversion of the Laplace transform was performed by a Fourier series approximation algorithm (Yiacoumi and Tien, 1992).

C. STIRRED-FLOW CELL TECHNIQUE The SFC is known in engineering as a “continuous-flow stirred tank reactor.” Eluent is passed through a chamber of volume VR[L3] containing a well-mixed suspension of soil and out to a fraction collector, adsorbent cartridge, or flowthrough detector (Deitsch and Smith, 1995; Eick et al., 1990; Miller et al., 1989; Seyfried et al., 1989; Sparks, 1989; Zhang and Sparks, 1993). Frits are positioned at either end of the cell to hold in particles. As a hybrid of batch and column techniques, the SFC eliminates hydrodynamic dispersion but has the sampling advantages of a column. The method is capable of measuring sorption or desorption with “half-lives” as short as ⬃1 min or desorption with characteristic times of a few tens of hours. Uptake rates are measured by pumping influent of solute concentration Cin [M L3] at flow rate u [L3 T1] through the suspension and measuring the effluent concentration Ceff [M L3]. Release rates are measured likewise by pumping clean liquid through the suspension of known initial sorbed and aqueous concentrations, q0 [M M1] and C0 [M L3]. Provided adequate mixing occurs in the cell, the mass balance equation is 

∂q ∂C u + = (Cim − Ceff ) ∂t ∂t VR

(68)

where q and C represent the instantaneous sorbed and aqueous concentrations in the cell,  is the bulk solids density [M L3], and  is the volumetric liquid content [L3 L3]. Under well-mixed conditions C Ceff. The Ceff is made up of a component cs

SORPTION AND DESORPTION RATES

55

representing uptake or release from the sorbent plus a component C* representing the liquid phase of the chamber that would be observed in the case of no sorption: Ceff = cs + C*

(69)

⎛ −ut ⎞ C* = (C0 − Cin ) exp⎜ ⎟ + Cin ⎝ VR ⎠

(70)

The value of C* is obtained by solving Eq. (68) for the case of no sorption. Verification of Eq. (70), which signifies thorough mixing conditions in the cell, can be performed using a nonsorbing solute such as 3H2O. After subtracting C* from Ceff (it can be seen that C* approaches a constant value at long times), Eq. (68) in terms of cs becomes ∂q  ∂cs u = − cs − ∂t  ∂t VR

(71)

The first term on the right in Eq. (71) represents efflux from the chamber due to the sorption process, whereas the second represents solution-phase chemical inside the cell due to the sorption process. If an in-line detector is used and concentrations are measured at intervals of t, then the finite difference approximation of Eq. (71), correct to terms of order t2, is q u( c j − 1 + 2 c j + c j + 1 )  ( c j + 1 − c j − 1 ) =− − t 4VR  2t

(72)

where the superscript j represents the jth measurement. If samples are collected, reactor efflux is known exactly from c j and, therefore, the analogous approximation is q uc j  ( c j +1 − c j −1 ) =− − t 4VR  2t

(73)

The cumulative sorbed concentration is given by q q0  t(q/t).

(74)

For measurements with natural soils, the SCF technique suffers from many problems. It is not very well suited for long-term uptake experiments because the differences between Cin and Ceff quickly become too small. Some cells have builtin stirring mechanisms: Magnetic stir bars grind particles and propeller seals have a finite life because of continual abrasion from particles. Deitsch and Smith (1995)

56

JOSEPH J. PIGNATELLO

simply placed the cell on a rotary shaker to effect mixing. There is always a tradeoff between loss of particles from the cell and clogging of the frits. Small particles can be generated from large particles over time owing to particle–particle and particle–wall collisions.

D. ZERO-LENGTH COLUMNS ZLCs (Eic and Ruthven, 1988; Ruthven and Eic, 1988) are columns that are short enough (or the flow rate is fast enough) that dispersion is insignificant and thus the response is sensitive only to sorption processes. The method is most suitable for desorption experiments. Basically, a short column of preequilibrated material is swept with a stream of gas or liquid at a rate fast enough to maintain “zero” concentration at the external particle surfaces, and desorbed chemical in the effluent is monitored with a sensitive detector or is trapped. This method is useful because it retains the advantages of columns in terms of relevance and ease of measurement while eliminating the effects of longitudinal dispersion. Its advantage over batch experiments is that it eliminates the need for agitation. For a bed of spherical particles of radius a the total amount of diffusing substance leaving the column compared with the initial amount is given by Eq. (55) (see Section V,A,1) provided the ratio of the rate parameter for external mass transfer to the diffusivity is large. The observation that verifies this condition is that the rate of efflux from the column is independent of flow velocity. The derivative of Eq. (55) with respect to time is the desorption rate. The ZLC technique has been employed in studies of TCE desorption from columns of soil and silica sorbents (Farrell and Reinhard, 1994b; Grathwohl and Reinhard, 1993).

VI. SORPTION KINETICS AND BIOAVAILABILITY A. ASSIMILATION OF CHEMICALS IN SOIL SYSTEMS It is well-known that bioavailability of chemicals (uptake, toxic effect, etc.) is lower in a soil–water mixture compared to water alone. Bioavailability is reduced for both thermodynamic and kinetic reasons: thermodynamic because a fraction of the chemical is partitioned to the soil and is not available there; kinetic because desorption can be rate limiting to uptake by the organism from the fluid phase. Although the terms “available” and “sequestered” are often used in the literature in regard to fractions of a sorbed chemical in soil, realistically there is a continuum from instantly available to completely unavailable. Furthermore, the term se-

SORPTION AND DESORPTION RATES

57

questered has meaning only in the context of a given receptor, chemical, soil environment, mode, and duration of uptake. Clearly, large multicell organisms assimilate chemicals only through the fluid phase (liquid or vapor) and not directly from the particle surface or interior, although they may be able to indirectly affect the flux of chemical from the particle. For single-cell organisms the situation is less clear. Cells may attach to surfaces by molecular forces or via extracellular exudates. Whether attached cells are able to abstract sorbed organic molecules directly from the surface is inconclusive but the preponderance of evidence is in the negative, at least for soil particles (Crocker et al., 1995; Shelton and Doherty, 1997). Experimental results are supported by logic: (i) Most sorption sites lie within SOM interstices, which are physically inaccessible to cells; (ii) most of the surface area of a particle is contained in mesopores and micropores (i.e., 50 nm), where even the smallest cells cannot fit; and (iii) if we assume rapid local equilibrium sorption at the solution–solid interface in the vicinity of the cell (Ke), the chemical potential, and therefore the activity, of substrate is the same for dissolved and adsorbed forms. The following rate expressions apply, Rate of uptake from solution kwa*fw

(75)

Rate of uptake from surface ksa* (1  fw)

(76)

where kw and ks are the rate constants for uptake from water and surface, a* is the activity of the chemical in solution or on the surface (in equivalent units), and fw is the fraction of cell surface area, , exposed to the water. A surface abstraction mechanism can enhance bioavailability only if ks(1  fw)  kw fw, which is doubtful because molecules on the surface are likely to be less mobile than molecules in solution. An organism may affect the flux of chemical from soil particles indirectly in various ways. First, it can do so by steepening the concentration gradient across the particle–fluid interface as a result of uptake from the fluid. This will accelerate desorption and may explain why some bacteria seem able to access sequestered fractions (Guerin and Boyd, 1992; Schwartz and Scow, 1999). Second, it can do so by causing changes in soil properties through biological activity in ways that affect Ke. Such change may result through direct action on the particle or through

58

JOSEPH J. PIGNATELLO

effects on the surrounding medium. For example, dermal contact may involve transfer of skin or hair oils to the particle that can facilitate uptake of hydrophobic chemicals. Ingestion may expose particles abruptly to biosurfactants and radically different pH regimes. Weston and Mayer (1998) found that stomach fluids increase bioavailability of PAHs in soil. Solution pH affects soil minerology and SOM structure; acidification of a soil to below pH ⬃2 released sequestered fractions of halogenated hydrocarbons possibly by dissolving metal oxide cements (Pignatello, 1990b). Soil ingested by birds may be pulverized in the gizzard resulting in shorter diffusion path lengths. Grinding in a ball mill has been shown to release resistant fractions (Ball and Roberts, 1991b; Pignatello, 1990a; Steinberg et al., 1987). Plant exudates may increase desorption by a surfactant effect or by a competitive sorption effect; natural aromatic acids that are produced by living and decomposing plants were shown to increase desorption of chlorinated aromatic hydrocarbons and phenols by competitive displacement (Xing and Pignatello, 1998). Recent studies show that competitive solutes increase sorption and desorption rates of the principal solute (J. White and J. Pignatello, submitted for publication).

B. COUPLED SORPTION –BIODEGRADATION KINETIC MODELS 1. General Considerations When biological uptake is relatively slow, or when the receptor moves rapidly through the contaminated medium, the solution concentration is not altered appreciably and bioavailability may be controlled simply by the existing solution concentration. The equilibrium partition model being considered by the U.S. Environmental Protection Agency for setting sediment quality criteria (Ankley et al., 1996; Di Toro et al., 1991) is based on the assumption that bioavailability, or biological effect, can be predicted knowing the equivalent pore water concentration. The pore water concentration is calculated from the total concentration present in the solids (determined by exhaustive extraction) and the Koc determined experimentally or calculated from established Kow- or solubility-based LFERs (Schwarzenbach et al., 1993). Although the database of Koc values and LFERs is extensive, the values are primarily based on short equilibration times (48 h). Their relevance to aged-contaminated systems, therefore, is highly questionable. In many cases, the apparent Koc in historically contaminated samples has been as much as two orders of magnitude greater than values obtained in freshly spiked samples (Pignatello and Xing, 1996). Ronday (1997) found that, although the toxicity of pesticides to the springtail (Folsomia candida) correlated well with the pore water concentration, the toxicity decreased over time and did not correlate well with short-term Koc values. When nonequilibrium conditions prevail during exposure it is necessary to consider mass transfer rate laws describing the flux of chemical through the par-

SORPTION AND DESORPTION RATES

59

Figure 16 The fraction or initial rate of phenanthrene desorbed (in the presence of Tenax infinite sink) or mineralized to CO2 by two bacteria. The coincidence indicates that phenanthrene metabolism is rate limited by desorption. The soils contained 1.4% OC (silt loam) and 44.5% OC (peat) (––) Desorption, (–䉭••) strain R biodeg, (– –••) strain P5-2 biodeg. (Reprinted with permission from Environmental Toxicology and Chemistry, 1999. Correlation between the biological and physical availabilities of phenanthrene in soils and soil humin in aging experiments, by J. C. White, M. Hunter, K. Nam, J. J. Pignatello, and M. Alexander, 18, 1720–1727. Copyright Society of Environmental Toxicology and Chemistry (SETAC), Pensacola, FL, 1999.)

ticle, across the particle–bulk fluid interface, and across the fluid–biomembrane interface. An accurate bioavailability model will require linkage of uptake/depuration kinetics with sorption/desorption kinetics. Some coupled models will be discussed in this section. The discussion is restricted to the most widely researched systems—those involving degradation of chemicals by microorganisms. In the models we assume that the substrate is available only through the aqueous phase. There are many studies showing that biodegradation is rate limited by desorption of the substrate (White et al., 1999; Bosma et al., 1997; Rijnaarts et al., 1990). For example, Fig. 16 shows the results of experiments (White et al., 1999) in which phenanthrene was allowed to “age” in contact with soil for various times before either adding bacterial degraders or carrying out desorption in the presence of the infinite sink, Tenax. Normalized plots of initial desorption rate or initial biodegradation rate, each as a function of aging time, coincide. Likewise, normalized plots of the amount desorbed or mineralized vs aging time coincide. This indicates that the degraders metabolize phenanthrene molecules as they desorb.

60

JOSEPH J. PIGNATELLO

Substrate is supplied to the solution by desorption and consumed via the solution phase by biodegradation: 

∂C ∂q ⎛ ∂C ⎞ = − − ⎜ ⎟ ⎝ ∂t ⎠ bd ∂t ∂t

(77)

Biodegradation of substrate when no other nutrient limitations exist is governed by Monod kinetics:

max C X (t ) ⎛ ∂C ⎞ − ⎜ ⎟ = ⋅ ⎝ ∂t ⎠ bd Km + C YS

(78)

where C is the solution concentration experienced at cell surfaces [M L3], max is the maximum growth rate [T1], Km [M L3] is the “half-saturation constant” (the substrate concentration at which the rate is 50% of maximum), X the cell mass concentration [M L3], and YS [M M1] is the specific bioconversion factor for growth on the substrate (i.e., biomass produced per mass substrate consumed). The values of max, Km, and Ys are obtained in separate soil-free growth experiments, assuming the surface has no influence. When C is well below Km substrate utilization is simplified to an expression that is first order in C:

X (t ) ⎛ ∂C ⎞ C − ⎜ ⎟ = max ⎝ ∂t ⎠ bd K m Ys

(79)

The cell mass concentration is a function of the rates of growth and decay, including death and inactivation by the soil. Cells may grow on the chemical and on utilizable natural organic matter (UOM). The general expression for the change in cell mass is ∂X max CX ⎛ ∂[UOM] ⎞ = + YUOM ⎜ ⎟ − X ⎝ Km + C ∂t ∂t ⎠

(80)

where YUOM [M M1] is the corresponding bioconversion factor for growth on UOM, and  [T1] is a first-order decay coefficient. If degradation by indigenous organisms is being considered, natural growth and decay may be assumed to have reached a steady state and the last two terms on the right of Eq. (80) cancel out. If, however, degradation by cultured organisms is being considered, the last two terms—especially the decay term—may be significant. Growth on UOM and decay processes in soils are complex and poorly understood. The values of YUOM, ![UOM]/!t, and  are thus difficult to acquire, especially since accurate assays for active cell population in the presence of soil particles are lacking. [Somewhat better than order-of-magnitude estimates of active degrader population may be made by a 14C-most-probable-number technique

SORPTION AND DESORPTION RATES

61

if radiolabeled compound is available (Lehmicke et al., 1979).] If cells can be accurately monitored, it may be possible to establish an empirical growth and decay curve in the absence of substrate and input it into the model. 2. Biodegradation Coupled with First-Order-Type Sorption Models In their study of 2,4-dichloroacetic acid (2,4-D) degradation by a 2,4-D degrading Alcaligenes species in unsaturated soils, Shelton and Doherty (1997) employed a four-compartment model: the biomass (X), the solution (C), sorbed available (A), and sorbed unavailable (U) compartments:

Mass transfer between the compartments obeyed the following Monod and simple first-order expressions:

max C dX X = dt Km + C

(81a)

max C X dC = − k1C + k−1SA − dt K m + C YS

(81b)

dA = − k1C − ( k−1 + k2 ) SA + k−1SU dt

(81c)

dU = k2 SA − k−1SU dt

(81d)

They assumed that the pesticide is the primary growth substrate and limiting nutrient, that there was no interference from indigenous organisms, and (apparently) that there was no natural decay or growth. The sorption rate constants (k1, k1) and (k2, k2) were obtained in independent experiments performed over 3- and 48h periods, respectively. Hence, the model is specific to the time frame of the experiment. 3. Biodegradation Coupled with Linear Driving Force Sorption Model Bosma et al. (1997) studied the biodegradation of -hexachlorocyclohexane residues in field-contaminated soil. They assumed that the rate of desorption was

62

JOSEPH J. PIGNATELLO

proportional to the concentration gradient of solute between liquids in distant pores, where cells cannot enter, and the sites of bacterial colonies (0.8 to 3- m macropores):

−

 ∂q = (Cd − Cv )  ∂t

(82)

where Cd is the distal and Cc is the vicinal concentration with respect to the cell surface, and  [T1] is a desorption rate parameter. Under steady-state conditions Eqs. (78) and (82) are equal, and through further manipulation it is possible to obtain the so-called Best equation: v = vmax

Cd + K m + vmax  −1 2 vmax  −1

⎧⎪ ⎨1 − ⎪⎩

⎡ ⎤ 4Cd vmax  −1 ⎢1 − −1 2 ⎥ ( ) C + K + v  max d m ⎣ ⎦

1/ 2 ⎫

⎪ ⎬ ⎪⎭

(83)

where v is the specific degradation velocity [M T1]; vmax ( maxX/YS ) is the maximum degradation velocity [M T1], and the other variables are as defined previously. The Best number (Bn), 1), Bn /(vmaxKm

(84)

is the index of mass transfer to biodegradation; the reaction is rate limited by biodegradation when Bn  1 and rate limited by mass transfer when Bn  1. The Bn for -hexachlorocyclohexane in soil slurry was 0.016 –0.03, indicating mass transfer limitation. 4. Biodegradation Kinetics Coupled with Radial Diffusion Laws These models (Rijnaarts et al., 1990; Scow and Alexander, 1992; Scow and Hutson, 1992) employ a simple radial pore diffusion law such as the one in Eq. (30)

63

SORPTION AND DESORPTION RATES

in order to calculate an effective diffusivity using analytical (Crank, 1975) or numerical solutions. It is normally assumed that the substrate concentration at the cell surface is near zero. Rijnaarts et al. (1990) used the measured mean particle diameter (122–182 m) to obtain Deff, however, this value of Deff resulted in poor fits in the coupled model. Running the coupled model instead with a fitting parameter  representing the average “intraparticle diffusion distance” resulted in good fit when  14–18 m. Rijnaarts et al. hypothesized that the bacteria were able to slightly penetrate the particle. However, it is more likely that the length scale over which diffusion occurs is simply smaller than the actual particle radius. 5. Biodegradation Coupled with Transport With biodegradation the A–D equation for solute transport (Eq. 56; see Section IV,A) becomes 

∂C ∂q ∂2C ∂C ∂ + = Dh 2 − v − ∂t ∂t ∂x ∂t ∂x

(∑ i Ci ) ,

(85)

biodegradation term

where i refers to each solution compartment if more than one is applicable. Models published to date have assumed first-order biodegradation kinetics; that is, that substrate concentration falls in the low-concentration region of the Monod curve and the degrader population is at steady state. The two-region (mobile–immobile) A–D model (Eq. 58) incorporating degradation is thus

(m + fnKe Cim n −1 ) ∂C∂tm + (im + (1 − f )nKe Cim n −1 ) ∂C∂tim − m v

=  m Dh

∂Cm −  m m Cm −  im im Cim , ∂x

∂ 2 Cm ∂x 2 (86)

where the first-order biodegradation rate constants [T1] in the mobile ( m) and immobile ( im) regions may be different due to nutrient availability, different populations, or other factors. Likewise, the two-site (equilibrium–kinetic) A–D model of Eq. (60) is given by  ∂2C ∂C ⎛ ⎞ ∂C  1 + fnK e C n −1 + k−2 ( Ke C n − q ) = Dh 2 − v − C , ⎝ ⎠  ∂t  ∂x ∂x

(87)

where is the first-order biodegradation rate constant [T1]. An analytical solution to Eq. (86) or Eq. (87) is readily obtained in the linear sorption case (van Genuchten and Wagenet, 1989).

64

JOSEPH J. PIGNATELLO

Figure 17 Effect of dimensionless desorption rate parameter (*) and biodegradation rate parameter ( *) on the number of pore volumes needed to decrease the initial contaminant mass by a factor of 103 in an aquifer. Here, * L/v, * L/v,* Ke /, and D* Dh /vL. DII is the Damkohler number II, the ratio of degradation rate to mass transfer rate. (Reprinted from Fry and Istok, 1994, with permission from the American Geophysical Union.)

Researchers have used these models experimentally with some success (Angley et al., 1992; Gamerdinger et al., 1990; Hu and Brusseau, 1998). Degradation delays the breakthrough of the solute (Angley et al., 1992; Hu and Brusseau, 1998) and, of course, decreases the amount recovered. Angley et al. studied alkylben-

SORPTION AND DESORPTION RATES

65

zene transport in columns of nonsterile aquifer material, taking sorption to be linear. Predicted elution curves using the nonequilibrium two-site model were superior to the corresponding equilibrium single-site model ( f 1). The ’s were highly dependent on flow velocity, however, increasing by a factor of three to eight with increasing flow velocity from 5.76 to 65.8 pore volumes per day. Even at the slowest flow velocity, the ’s were an order of magnitude greater than those in nonagitated batch microcosm studies. This result underscored the “pseudo, or nonconstant nature of [ ]” and rendered extrapolation to the field “problematic” (Angley et al., 1992). The relationship between desorption and bioavailability in an aquifer remediation scenario was examined theoretical by Fry and Istok (1994). They assumed first-order biodegradation and the existence of a single sorption domain having a linear isotherm and first-order desorption rate coefficient. Figure 17 shows the number of pore volumes needed to decrease the initial contaminant mass in the aquifer by three orders of magnitude as a function of the dimensionless desorption coefficient, *, and the dimensionless biodegradation rate coefficient *. When degradation is rate limiting (* is large relative to *), increasing the degradation rate decreases the number of pore volumes needed to remediate the aquifer. However, when desorption is rate limiting (* is small relative to *), increasing the degradation rate is predicted to be futile.

REFERENCES Adamson, A. W., and Gast, A. P. (1997). “Physical Chemistry of Surfaces,” Sixth Edition/Ed. Wiley, New York. Alexander, M. (1995). How toxic are toxic chemicals in soil? Environ. Sci. Technol. 29, 2713 –2717. Allen-King, R. M., Groenevelt, H., and Mackay, D. M. (1995). Analytical method for the sorption of hydrophobic organic pollutants in clay-rich materials. Environ. Sci. Technol. 29, 148 –153. Angley, J. T., Brusseau, M. L., Miller, W. L., and Delfino, J. J. (1992). Nonequilibrium sorption and aerobic biodegradation of dissolved alkylbenzenes during transport in aquifer material: Column experiments and evaluation of a coupled-process model. Environ. Sci. Technol. 26, 1404 –1410. Ankley, G. T., Berry, W. J., Di Toro, D. M., Hansen, D. J., Hoke, R. A., Mount, D. R., Reiley, M. C., Swartz, R. C., and Zarba, C. S. (1996). Use of equilibrium partitioning to establish sediment quality criteria for nonionic chemicals: A reply to Iannuzzi et al. Environ. Toxicol. Chem. 15, 1019 – 1024. Arocha, M. A., Jackman, A. P., and McCoy, B. J. (1996). Adsorption kinetics of toluene on soil agglomerates: Soil as a bioporous sorbent. Environ. Sci. Technol. 30, 1500 –1507. Ball, W. P., and Roberts, P. V. (1991a). Long-term sorption of halogenated organic chemicals by aquifer material. 1. Equilibrium. Environ. Sci. Technol. 25, 1223 –1235. Ball, W. P., and Roberts, P. V. (1991b). Long-term sorption of halogenated organic chemicals by aquifer material. 2. Intraparticle diffusion. Environ. Sci. Technol. 25, 1237–1249. Ball, W. P., Buehler, C., Harmon, T. C., Mackay, D. M., and Roberts, P. V. (1990). Characterization of a sandy aquifer material at the grain scale. J. Contam. Hydrol. 5, 253 –295. Ball, W. P., Xia, G., Durfee, D. P., Wilson, R. D., Brown, M. J., and Mackay, D. M. (1997). Hot-

66

JOSEPH J. PIGNATELLO

methanol extraction for the analysis of volatile organic chemicals in subsurface core samples from Dover Air Force Base, Delaware. Ground Water Monitoring Remediation 17, 104 –121. Barrer, R. M. (1984). J. Membrane Sci. 14, 25 – 85. Benzing, T. R., Rao, P. S. C., and Byd, S. A. (1996). Construction and performance of a gas purge system. J. Environ. Quality 25, 1436 –1441. Berens, A. (1989). Transport of organic vapors and liquids in poly(vinyl chloride). Makromol. Chem. Macromol. Symp. 29, 95 –108. Borglin, S., Wilke, A., Jepsen, R., and Lick, W. (1996). Parameters affecting the desorption of hydrophobic organic chemicals from suspended sediments. Environ. Toxicol. Chem. 15, 2254–2262. Bosma, T. N. P., Middeldorp, P. J. M., Schraa, G., and Zehnder, A. J. B. (1997). Mass transfer limitation of biotransformation: Quantifying bioavailability. Environ. Sci. Technol. 31, 248 –252. Brusseau, M. L. (1992). Nonequilibrium transport of organic chemicals: The impact of pore-water velocity. J. Contam. Hydrol. 9, 353 –368. Brusseau, M. L. (1993). Using QSAR to evaluate phenomenlogical models for sorption of organic compounds by soil. Environ. Toxicol. Chem. 12, 1835 –1846. Brusseau, M. L., and Rao, P. S. C. (1989). Sorption nonideality during organic contaminant transport in porous media. Crit. Rev. Environ. Control 19, 33 – 99. Brusseau, M. L., Jessup, R. E., and Rao, P. S. C. (1990). Sorption kinetics of organic chemicals: Evaluation of gas-purge and miscible-displacement techniques. Environ. Sci. Technol. 24, 727–735. Cameron, D. R., and Klute, A. (1977). Convective–dispersive solute transport with a combined equilibrium and kinetic adsorption model. Water Resour. Res. 13, 183 –188. Carroll, K. M., Harkness, M. R., Bracco, A. A., and R. R. B. (1994). Application of a permeant/polymer diffusional model to the desorption of polychlorinated biphenyls from Hudson River sediments. Environ. Sci. Technol. 28, 253 –258. Castellan, G. W. (1971). “Physical Chemistry,” 2nd ed. Addison-Wesley, Reading, MA. Chang, M., Wu, S., and Chen, C. (1997). Diffusion of volatile organic compounds in pressed humic acid disks. Environ. Sci. Technol. 31, 2307–2312. Chen, N. Y. (1976). Hydrophobic properties of zeolites. J. Phys. Chem. 80, 60 – 64. Chen, W., and Wagenet, R. J. (1995). Solute transport in porous media with sorption-site heterogeneity. Environ. Sci. Technol. 29, 2725 –2734. Chiou, C. T. (1989). Theoretical considerations of the partition uptake of nonionic organic compounds by soil organic matter. In “Reactions and Movement of Organic Chemicals in Soil” (B. L. Sawhney and K. Brown, Eds.), pp. 1–29. Soil Science Society of America, Madison, WI. [Special publication] Chiou, C. T., Rutherford, D. W., and Manes, M. (1993). Sorption of N2 and EGME vapors on some soils, clays, and mineral oxides and determination of sample surface areas by use of sorption data. Environ. Sci. Technol. 27, 1587–1594. Connaughton, D. F., Stedinger, J. R., Lion, L. W., and Schuler, M. L. (1993). Description of time-varying desorption kinetics: Release of napthalene from contaminated soils. Environ. Sci. Technol. 27, 2397–2403. Corley, T. L., Farrell, J., Hong, B., and Cponklin, M. H. (1996). VOC accumulation and pore filling in unsaturated porous media. Environ. Sci. Technol. 30, 2884 –2891. Cornelissen, G., van Noort, P. C. M., and Govers, H. A. J. (1997a). Desorption kinetics of chlorobenzenes, polycyclic aromatic hydrocarbons, and polychlorinated biphenyls: Sediment extraction with Tenax and effects of contact time and solute hydrophobicity. Environ. Toxicol. Chem. 16, 1351–1357. Cornelissen, G., van Noort, P. C. M., Parsons, J. R., and Govers, H. A. J. (1997b). The temperature dependence of slow adsorption and desorption kinetics of organic compounds in sediments. Environ. Sci. Technol. 31, 454 – 460. Cornelissen, G., Rigterink, H., Ferdinandy, M. M. A., and Van Noort, P. C. M. (1998a). Rapidly de-

SORPTION AND DESORPTION RATES

67

sorbing fractions of PAHs in contaminated sediments as a predictor of the extent of bioremediation. Environ. Sci. Technol. 32, 966 – 970. Cornelissen, G., van Zuilen, H., and van Noort, P. C. M. (1999). Particle size dependence of slow desorption of in situ PAHs from sediments. Chemosphere 37. Crank, J. (1975). “The Mathematics of Diffusion,” 2nd ed. Clarendon, Oxford, UK. Crocker, F. H., Guerin, W. F., and Boyd, S. A. (1995). Bioavailability of naphthalene sorbed to cationic surfactant-modified smectite clay. Environ. Sci. Technol. 29, 2953 –2958. Currie, J. A. (1960). Gaseous diffusion in porous media. Part 2. Dry granular materials. Br. J. Appl. Phys. 11, 318–324. Curthoys, G., Davydow, V. Y., Kiseliv, A. V., Kiselev, S. A., and Kuznetsov, B. V. (1974). Hydrogen bonding in adsorption on silica. J. Colloid Interface Sci. 48, 58 –72. Deitsch, J. J., and Smith, J. A. (1995). Effect of Triton X-100 on the rate of trichloroethene desorption from soil to water. Environ. Sci. Technol. 29, 1069 –1080. Deitsch, J. J., Smith, J. A., Arnold, M. B., and Bolus, J. (1998). Sorption and desorption rates of carbon tetrachloride and 1,2-dichlorobenzene to three organobentonites and a natural peat soil. Environ. Sci. Technol. 32, 3169 – 3177. de Jonge, H., and Mittelmeijer-Hazeleger, M. C. (1996). Adsorption of CO2 and N2 on soil organic matter: Nature of porosity, surface area, and diffusion mechanisms. Environ. Sci. Technol. 30, 408– 413. Di Toro, D. M., Zarba, C. S., Hansen, D. J., Berry, W. J., Swartz, R., Cowan, C. E., Pavlou, S. P., Allen, H. E., and Paquin, P. R. (1991). Technical basis for establishing sediment quality criteria for nonionic organic chemicals using equilibrium partitioning. Environ. Toxicol. Chem. 10, 1541–1583. Eic, M., and Ruthven, D. M. (1988). A new experimental technique for measurement of intracrystalline diffusivity. Zeolites 8, 40 – 45. Eick, M. J., Bar-Tal, A., Sparks, D. L., and Feigenbaum, S. (1990). Analyses of adsorption kinetics using a stirred-flow chamber: II. Potassium–calcium exchange on clay minerals. Soil Sci. Soc. Am. J. 54, 1278–1282. Emerson, M. T., Grunwald, E., Kaplan, M. L., and Kromhout, R. A. (1960). Proton transfer studies by nuclear magnetic resonance. III. The mean life of the amine-water hydrogen bond in aqueous solution. J. Am. Chem. Soc. 82, 6307– 6314. Farrell, J., and Reinhard, M. (1994a). Desorption of halogenated organics from model solids, sediments, and soil under unsaturated conditions. 1. Isotherms. Environ. Sci. Technol. 28, 53 – 62. Farrell, J., and Reinhard, M. (1994b). Desorption of halogenated organics from model solids, sediments, and soil under unsaturated conditions. 2. Kinetics. Environ. Sci. Technol. 28, 63 –72. Fava, A., and Eyring, H. (1956). Equilibrium and kinetics of detergent adsorption—A generalized equilibration theory. Langmuir 60, 890 – 898. Foster, R. (1969). “Organic Charge-Transfer Complexes.” Academic Press, London. Fredrickson, G. H., and Helfand, E. (1985). Dual-mode transport of penetrants in glassy polymers. Macromolecules 18, 2201–2207. Frisch, H. L., and Stern, S. A. (1983). Diffusion of small molecules in polymers. CRC Crit. Rev. Solid State Mater. Sci. 11, 123 –186. Fry, V. A., and Istok, J. D. (1994). Effects of rate-limited desorption on the feasibility of in situ bioremediation. Water Resour. Res. 30, 2413 –2422. Gamerdinger, A. P., Wagenet, R. J., and van Genuchten, M. T. (1990). Application of two-site/two region models for studying simultaneous nonequilibrium transport and degradation of pesticides. Soil Sci. Soc. Am. J. 54, 957– 963. Goring, D. A. I. (1989). The lignin paradigm. In “Lignin Properties and Materials. ACS Symposium Series 397” (W. G. Glasser and S. Sarkanen, Eds.). American Chemical Society, Washington, DC. Goss, K. (1992). Effects of temperature and relative humidity on the sorption of organic vapors on quartz on sand. Environ. Sci. Technol. 26, 2287–2294.

68

JOSEPH J. PIGNATELLO

Goss, K. (1994). Adsorption of organic vapors on polar mineral surfaces and on a bulk water surface: Development of an empirical predictive model. Environ. Sci. Technol. 28, 640 – 645. Goss, K.-U. (1997). Conceptual model for the adsorption of organic compounds from the gas phase to liquid and solid surfaces. Environ. Sci. Technol. 31, 3600 – 3605. Grathwohl, P., and Reinhard, M. (1993). Desorption of trichloroethylene in aquifer material: Rate limitation at the grain scale. Environ. Sci. Technol. 27, 2360 –2366. Greenland, D. J., and Hayes, M. H. B. (1981). Soil processes. In “The Chemistry of Soil Processes” (D. J. Greenland and M. H. B. Hayes, Eds.), pp. 1– 35. Wiley, Chichester, UK. Gregg, S. J., and Sing, K. S. W. (1982). “Adsorption, Surface Area and Porosity,” 2nd ed. Academic Press, London. Guerin, W. F., and Boyd, S. A. (1992). Differential bioavailability of soil-sorbed naphthalene to two bacterial species. Appl. Environ. Microbiol. 58, 1142–1152. Gustafson, D. I., and Holden, L. R. (1990). Nonlinear pesticide dissipation in soil: A new model based on spatial variability. Environ. Sci. Technol. 24, 1032–1038. Gustafsson, O., Haghseta, F., Chan, C., MacFarlane, J., and Gschwend, P. M. (1997). Quantification of the dilute sedimentary soot phase: Implications for PAH speciation and bioavailability. Environ. Sci. Technol. 31, 203 –209. Haderlein, S. B., Weissmahr, K. W., and Schwarzenbach, R. P. (1996). Specific adsorption of nitroaromatic explosives and pesticides to clay minerals. Environ. Sci. Technol. 30, 612– 622. Harmon, T. C., and Roberts, P. V. (1994). Comparison of intraparticle sorption and desorption rates for a halogenated alkene in a sandy aquifer material. Environ. Sci. Technol. 28, 1650 –1660. Hayes, M. H. B., and Himes, F. L. (1986). Nature and properties of humus–mineral complexes. In “Interaction of Soil Minerals with Natural Organics and Microbes” (P. M. Huang and M. Schnitzer, Eds.), pp. 103–157. Soil Science Society of America, Madison, WI. Hayes, M. H. B., MacCarthy, P., Malcolm, R. L., and Swift, R. S. (Eds.) (1989). “Humic Substances II. In Search of Structure.” Wiley, Chichester, UK. Hendricks, D. W., and Kuratti, L. G. (1982). Derivation of an empirical sorption rate equation by analysis of experimental data. Water Res. 16, 829 – 837. Holmén, B. A., and Gschwend, P. M. (1997). Estimating sorption rates of hydrophobic organic compounds in iron oxide- and aluminosilicate clay-coated aquifer sands. Environ. Sci. Technol. 31, 105–113. Horas, J. A., and Nieto, F. (1994). A generalization of dual mode transport theory for glassy polymers. J. Polymer Sci. B Polymer Phys. 32, 1889 –1898. Hu, M. Q., and Brusseau, M. L. (1998). Coupled effects of nonlinear, rate-limited sorption and biodegradation on transport of 2,4-dichlorophenoxyacetic acid in soil. Environ. Toxicol. Chem. 17, 1673–1680. Hu, Q., Wang, X., and Brusseau, M. L. (1995). Quantitative structure–activity relationships for evaluating the influence of sorbate structure on sorption of organic compounds by soil. Environ. Toxicol. Chem. 14, 1133 –1140. Huang, L. Q., and Pignatello, J. J. (1990). Improved extraction of atrazine and metolachlor in field soil samples. J. Assoc. Official Anal. Chemists 73, 443 – 446. Huang, W., and Weber, W. J., Jr. (1997). A distributed reactivity model for sorption by soils and sediments. 10. Relationships between desorption, hysteresis, and the chemical characteristics of organic domains. Environ. Sci. Technol. 31, 2562–2569. Huang, W., Schlautman, M. A., and Weber, W. J., Jr. (1996). A distributed reactivity model for sorption by soils and sediments. 5. The influence of near-surface characteristics in mineral domains. Environ. Sci. Technol. 30, 2993 – 3000. Huang, W., Young, T., Schlautman, M. A., Yu, H., and Weber, W. J., Jr. (1997). A distributed reactivity model for sorption by soils and sediments. 9. General isotherm nonlinearity and applicability of the dual reactive domain model. Environ. Sci. Technol. 31, 1703 –1710. Israelachvili, J. N. (1992). “Intermolecular and Surface Forces,” 2nd ed. Academic Press, London.

SORPTION AND DESORPTION RATES

69

Jaynes, W. F., and Boyd, S. A. (1991). Hydrophobicity of siloxane surfaces in smectites as revealed by aromatic hydrocarbon adsorption from water. Clays Clay Miner. 39, 428 – 436. Kamiya, Y., Mizoguchi, K., Hirose, T., and Naito, Y. (1989). Sorption and dilation in poly(ethyl methacrylate)–carbon dioxide system. J. Polymer Sci. B. 27, 879 – 892. Kamiya, Y., Bourbon, D., Mizoguchi, K., and Naito, Y. (1992). Sorption, dilation, and isothermal glass transition of poly(ethyl methacrylate)–organic gas systems. Polymer J. 24, 443 – 449. Kan, A. T., Fu, G., Hunter, M. A., and Tomson, M. B. (1997). Irreversible adsorption of naphthalene and tetrachlorobiphenyl to lula and surrogate sediments. Environ. Sci. Technol. 31, 2176 –2186. Kan, A. T., Fu, G., Hunter, M., Chen, W., Ward, C. H., and Tomson, M. B. (1998). Irreversible sorption of neutral hydrocarbons to sediments: Experimental observations and model predictions. Environ. Sci. Technol. 32, 892– 902. Kapoor, A., Yang, R. T., and Wong, C. (1989). Surface diffusion. Catal. Rev. Sci. Eng. 31, 129 –214. Kärger, J., and Ruthven, D. M. (1992). “Diffusion in Zeolites and Other Microporous Solids.” Wiley, New York. Karickhoff, S. W., and Morris, K. R. (1985). Sorption dynamics of hydrophobic pollutants in sediment suspensions. Environ. Toxicol. Chem. 4, 469 – 479. Karickhoff, S. W., Brown, D. S., and Scott, T. A. (1979). Sorption of hydrophobic pollutants on natural sediments. Water Res. 13, 241–248. Kim, H., Annable, M. D., and Rao, P. S. C. (1998). Influence of air–water interfacial adsorption and gas-phase partitioning on the transport of organic chemicals in unsaturated porous media. Environ. Sci. Technol. 32, 1253 –1259. Kleineidam, S., Rügner, H., and Grathwohl, P. (1999). The impact of grain scale heterogeneity on slow sorption kinetics. Environ. Toxicol. Chem. 18, 1673 –1678. Kuhlbusch, T. A. J. (1998). Black carbon and the carbon cycle. Science 280, 1903 –1904. Larson, J. W., and Wernett, P. (1988). Pore structure of Illinois No. 6 coal. Energy Fuels 2, 719 –720. Lee, L. K. (1978). The kinetics of sorption in a biporous adsorbent particle. Am. Inst. Chem. Eng. J. 24, 531–533. Lee, S. Y., Seader, J. D., Tsai, C. H., and Massoth, F. E. (1991). Restrictive diffusion under catalytic hydroprocessing conditions. Ind. Eng. Chem. Res. 30, 29 – 38. Lehmicke, L. G., Williams, R. T., and Crawford, R. L. (1979). 14-C Most-probable-number method for enumeration of active heterotrophic microorganisms in natural waters. Appl. Environ. Microbiol. 38, 644–649. Lick, W., and Rapaka, V. (1996). A quantitative analysis of the dynamics of the sorption of hydrophobic organic chemicals to suspended sediments. Environ. Toxicol. Chem. 15, 1038 –1048. Lin, T., Little, J. C., and Nazaroff, W. W. (1994). Transport and sorption of volatile organic compounds and water vapor within dry soil grains. Environ. Sci. Technol. 28, 322– 330. Linz, D. G., and Nakles, D. V. (Eds.) (1997). “Environmentally Acceptable Endpoints in Soil: RiskBased Approach to Contaminated Site Management Based on Availability of Chemicals in Soil.” American Academy of Environmental Engineers, Annapolis, MD. Lorden, S. W., Chen, W., and Lion, L. W. (1998). Experiments and modeling of the transport of trichloroethene vapor in unsaturated aquifer material. Environ. Sci. Technol. 32, 2009 –2017. Lucht, L. M., Larson, J. M., and Peppas, N. A. (1987). Macromolecular structure of coals. 9. Molecular structure and glass transition temperature. Energy Fuels 1, 56–58. Lyon, W. G. (1995). Swelling of peats in liquid methyl, tetramethylene and propyl sulfoxides and in liquid propyl sulfone. Environ. Toxicol. Chem. 14, 229 –236. Lyon, W. G., and Rhodes, D. E. (1993). Molecular size exclusion by soil organic materials estimated from their swelling in organic solvents. Environ. Toxicol. Chem. 12, 1405 –1412. Mader, B. T., Goss, K.-U., and Eisenreich, S. J. (1997). Sorption of nonionic, hydrophobic organic chemicals to mineral surfaces. Environ. Sci. Technol. 31, 1079 –1086. Maguire, R. J., Batchelor, S. P., and Sullivan, C. A. (1995). Reduction of the efficiency of extraction of lipophilic chemicals from water by dissolved organic matter. Environ. Toxicol. Chem. 14, 389–393.

70

JOSEPH J. PIGNATELLO

Mayer, L. (1999). Extent of coverage of mineral surfaces by organic matter in marine sediments. Geochim. Cosmochim. Acta, 63, 207–215. McBride, M. B. (1994). “Environmental Chemistry of Soils,” Oxford Univ. Press, New York. McDonald, L. M., Jr., and Evangelou, V. P. (1997). Optimal solid-to-solution ratios for organic chemical sorption experiments. Soil Sci. Soc. Am. J. 61, 1655 –1659. McGroddy, S. E., Farrington, J. W., and Gschwend, P. M. (1996). Comparison of the in situ and desorption sediment–water partitioning of polycyclic aromatic hydrocarbons and polychlorinated biphenyls. Environ. Sci. Technol. 30, 172–177. Miller, C. T., and Pedit, J. A. (1992). Use of a reactive surface-diffusion model to describe apparent sorption–desorption hysteresis and abiotic degradation of lindane in a subsurface material. Environ. Sci. Technol. 26, 1417–1427. Miller, D. M., Miller, W. P., and Sumner, M. E. (1989). A continuous-flow stirred reaction cell for studying adsorption in suspensions. Soil Sci. Soc. Am. J. 53, 1407–1411. Müller-Wegener, U. (1987). Electron donor acceptor complexes between organic nitrogen heterocycles and humic acid. Sci. Total Environ. 62, 297– 304. Nkedi-Kizza, P., Rao, P. S. C., Jessup, R. E., and Davidson, J. M. (1982). Ion exchange and diffusive mass transfer during miscible displacement through an aggregated oxisol. Soil Sci. Soc. Am. J. 46, 471–476. Nkedi-Kizza, P., Biggar, J. W., Selim, H. M., van Genuchten, M. T., Wierenga, P. J., Davidson, J. M., and Nielsen, D. R. (1984). Ion exchange and diffusive mass transfer during miscible displacement through an aggregated oxisol. Soil Sci. Soc. Am. J. 46, 471– 476. Nkedi-Kizza, P., Brusseau, M. L., Rao, P. S. C., and Hornsby, A. G. (1989). Nonequilibrium sorption during displacement of hydrophobic organic chemicals and 45Ca through soil columns with aqueous and mixed solvents. Environ. Sci. Technol. 23, 814 – 820. Novotny, F. J., Rice, J. A., and Well, D. A. (1995). Characterization of fulvic acid by laser-desorption mass spectrometry. Environ. Sci. Technol. 29, 2464 –2466. Nzengung, V. A., Nkedi-Kizza, P., Jessup, R. E., and Voudrias, E. A. (1997). Organic cosolvent effects on sorption kinetics of hydrophobic chemicals by organoclays. Environ. Sci. Technol. 31, 1470–1475. O’Dell, J. D., Wolt, J. D., and Jardine, P. M. (1992). Transport of imazethapyr in undisturbed soil columns. Soil Sci. Soc. Am. J. 56, 1711–1715. Ong, S., and Lion, L. (1991). Mechanisms for trichloroethylene vapor sorption onto soil minerals. J. Environ. Quality 20, 180–188. Papelis, C., Roberts, P. V., and Leckie, J. O. (1995). Modeling the rate of cadmium and selenite adsorption on micro- and mesoporous transition aluminas. Environ. Sci. Technol. 29, 1099 –1108. Parker, J. C., and van Genuchten, M. T. (1984). Determining transport parameters from laboratory and field tracer experiments. bull. Virginia Agric. Exp. Station 84 – 3. Paul, D. R., and Koros, W. J. (1976). Effect of partially immobilizing sorption on permeability and the diffusion time lag. J. Polymer Sci. Polymer Phys. 14, 675 – 685. Pedit, J. A., and Miller, C. T. (1994). Heterogeneous sorption processes in subsurface systems. 1. Model formulations and applications. Environ. Sci. Technol. 28, 2094 –2104. Pedit, J. A., and Miller, C. T. (1995). Heterogeneous sorption processes in subsurfaces systems. 2. Diffusion modeling approaches. Environ. Sci. Technol. 29, 1766 –1772. Pellegrino, C., and Piccolo, A. (1999). Conformational arrangement of dissolved humic substances: Influence of solution composition on association of humic molecules. Environ. Sci. Technol. 33, 1682–1690. Petersen, L. W., Moldrup, P., El-Farhan, Y. H., Jacobsen, O. H., Yamaguchi, T., and Rolston, D. E. (1995). The effect of moisture and soil texture on the adsorption of organic vapors. J. Environ. Quality 24, 752–759. Piatt, J. J., and Brusseau, M. L. (1998). Rate-limited sorption of hydrophobic organic compounds by soils with well-characterized organic matter. Environ. Sci. Technol. 32, 1604 –1608. Pignatello, J. J. (1989). Sorption dynamics of organic compounds in soils and sediments. In “Reactions

SORPTION AND DESORPTION RATES

71

and Movement of Organic Chemicals in Soils,” pp. 45 – 80. Soil Science Society of America, Madison, WI. [Special publication] Pignatello, J. J. (1990a). Slowly reversible sorption of aliphatic halocarbons in soils. I. Formation of residual fractions. Environ. Toxicol. Chem. 9, 1107–1115. Pignatello, J. J. (1990b). Slowly reversible sorption of aliphatic halocarbons in soils. II. Mechanistic aspects. Environ. Toxicol. Chem. 9, 1117–1126. Pignatello, J. J. (1998). Soil organic matter as a nanoporous sorbent of organic pollutants. Adv. Colloid Interface Sci. 7677, 445 – 467. Pignatello, J. J. (1999). A revised physical concept of natural organic matter as a sorbent of organic compounds. In “Kinetics and Mechanisms of Reactions at the Mineral/Water Interface” (D. L. Sparks and T. J. Grundl, Eds.). American Chemical Society, Washington, DC. Pignatello, J. J., and Xing, B. (1996). Mechanisms of slow sorption of organic chemicals to natural particles. Environ. Sci. Technol. 30, 1–11. Pignatello, J. J., Ferrandino, F. J., and Huang, L. Q. (1993). Elution of aged and freshly added herbicides from a soil. Environ. Sci. Technol. 27, 1663 –1671. Preston, C. M., Schnitzer, M., and Ripmeester, J. A. (1989). A spectroscopic and chemical investigation on the de-ashing of a humin. Soil Sci. Soc. Am. J. 53, 1442–1447. Rao, P. S. C., and Davidson, J. M. (1980). Estimation of pesticide retention and transformation parameters required in nonpoint source pollution models. In “Environmental Impact of Nonpoint Source Pollution” (M. R. Overcash and J. M. Davidson, Eds.), pp. 23 – 67. Ann Arbor Science, Ann Arbor, MI. Rice, J. A., and MacCarthy, P. (1990). A model of humin. Environ. Sci. Technol. 24, 1875 –1877. Rijnaarts, H. H. M., Bachmann, A., Jumelet, J. C., and Zehnder, A. J. B. (1990). Effect of desorption and intraparticle mass transfer on the aerobic biomineralization of -hexachlorocyclohexane in a contaminated calcareous soil. Environ. Sci. Technol. 24, 1349 –1354. Rogers, C. E. (1965). Solubility and diffusivity. In “Physics and Chemistry of the Organic Solid State” (D. Fox, M. M. Labes, and A. Weissberger, Eds.), Vol. 2, pp. 510 – 601. Interscience, New York. Ronday, R., van Kammen-Polman, A. M. M., and Dekker, A. (1997). Persistence and toxicological effects of pesticides in topsoil: Use of the equilibrium partitioning theory. Environ. Toxicol. Chem. 16, 601–607. Ruckenstein, E., Vaidyanathau, A. S., and Youngquist, G. R. (1971). Sorption by solids with bidisperse pore structures. Chem. Eng. Sci. 26, 1305 –1318. Ruthven, D. M. (1984). “Principles of Adsorption and Adsorption Processes.” Wiley, New York. Ruthven, D. M., and Eic, M. (1988). Diffusion of linear paraffins and cyclohexane in NaX and 5A zeolite crystals. Zeolites 8, 472– 479. Satterfield, C. N., Coltion, C. K., and Pitcher, W. H., Jr. (1973). Restricted diffusion in liquids within fine pores. Am. Inst. Chem. Eng. J. 19, 628 – 635. Sawhney, B. L., and Gent, M. P. N. (1990). Hydrophobicity of clay surfaces: Sorption of 1,2-dibromoethane and trichloroethene. Clays Clay Miner. 38, 14 –20. Sawhney, B. L., Pignatello, J. J., and Steinberg, S. M. (1988). Determination of 1,2-dibromoethane (EDB) in field soils: Implications for volatile organic compounds. J. Environ. Quality 17, 149–152. Schlebaum, W., Badora, A., Schraa, G., and van Riemsdijk, W. H. (1998). Interactions between a hydrophobic organic chemical and natural organic matter: Equilibrium and kinetic studies. Environ. Sci. Technol. 32, 2273 –2277. Schulten, H.-R. (1995). The three-dimensional structure of humic substances and soil organic matter studied by computational analytical chemistry. Fresenius J. Environ. Anal. Chem. 351, 62–73. Schulten, H.-R. (1996). Three-dimensional, molecular structures of humic acids and their interactions with water and dissolved contaminants. Int. J. Anal. Chem. 64, 147–162. Schulten, H.-R., and Schnitzer, M. (1993). A state of the art structural concept for humic substances. Naturwissenschaften 80, 29 – 30. Schuster, P., Zundel, G., and Sandorfy, C. (1976). “The Hydrogen Bond.” North-Holland, Amsterdam.

72

JOSEPH J. PIGNATELLO

Schwartz, E., and Scow, K. M. (1999). Using biodegradation kinetics to measure the availability of aged phenanthrene to a bacteria inoculated into soil. Environ. Toxicol. Chem., in press. Schwarzenbach, R. P., Gschwend, P. M., and Imboden, D. M. (1993). “Environmental Organic Chemistry.” Wiley, New York. Scow, K. M., and Alexander, M. (1992). Effect of diffusion on the kinetics of biodegradation: Experimental results with synthetic aggregates. Soil Sci. Soc. Am. J. 56, 128 –134. Scow, K. M., and Hutson, J. (1992). Effect of diffusion and sorption on the kinetics of biodegradation: Theoretical considerations. Soil Sci. Soc. Am. J. 56, 119 –127. Seyfried, M. S., Sparks, D. L., Bar-Tal, A., and Feigenbaum, S. (1989). Kinetics of calcium–magnesium exchange on soil using a stirred-flow reaction chamber. Soil Sci. Soc. Am. J. 53, 406 – 410. Shelton, D. R., and Doherty, M. A. (1997). A model describing pesticide bioavailability and biodegradation in soil. Soil Sci. Soc. Am. J. 61, 1078 –1084. Sparks, D. L. (1989). “Kinetics of Soil Chemical Processes.” Academic Press, San Diego. Spurlock, F. C., Huang, K., and van Genuchten, M. T. (1995). Isotherm nonlinearity and nonequilibrium sorption effects on transport of fenuron and monuron in soil columns. Environ. Sci. Technol. 29, 1000–1007. Steinberg, S. M., Pignatello, J. J., and Sawhney, B. L. (1987). Persistence of 1,2-dibromoethane in soils: Entrapment in intraparticle micropores. Environ. Sci. Technol. 21, 1201–1208. Swift, R. S. (1989). Molecular weight, shape, and size of humic substances by ultracentrifugation. In “Humic Substances II. In Search of Structure” (M. H. B. Hayes, P. MacCarthy, R. L. Malcolm, and R. S. Swift, Eds.). Wiley, Chichester, UK. Ten Hulscher, T. E. M., Vrind, B. A., Van den Heuvel, J., Van der Velde, L. E., Van Noort, P. C. M., Beurskens, J. E. M., and Govers, H. A. J. (1999). Triphasic desorption of highly resistant chlorobenzenes, polychlorinated biphenyls, and polycyclic aromatic hydrocarbons in field contaminated sediment. Environ. Sci. Technol. 33, 126–132. Tognotti, L. M., Flytzani-Stephanopoulos, M., Sarofim, A. F., Kopsinis, H., and Stoukides, M. (1991). Study of adsorption–desorption of contaminants on single soil particles using the electrodynamic thermogravimetric analyzer. Environ. Sci. Technol. 25, 104 –109. Travis, C. C., and Etnier, E. L. (1981). A survey of sorption relationships for reactive solutes in soil. J. Environ. Quality 10, 8 –17. Tye, R., Jepsen, R., and Lick, W. (1996). Effects of colloids, flocculation, particle size, and organic matter on the adsorption of hexachlorobenzene to sediments. Environ. Toxicol. Chem. 15, 643 – 651. Valocchi, A. J. (1985). Validity of the local equilibrium assumption for modeling sorbing solute transport through homogenous soils. Water Resour. Res. 21, 808 – 820. van Genuchten, M. T., and Wagenet, R. J. (1989). Two-site/two-region models for pesticide transport and degradation: Theoretical development and analytical solutions. Soil Sci. Soc. Am. J. 53, 1303 – 1310. van Genuchten, M. T., and Wierenga, P. J. (1976). Mass transfer studies in sorbing porous media I. Analytical solutions. Soil Sci. Soc. Am. J. 40, 473 – 480. van Genuchten, M. T., and Wierenga, P. J. (1977). Mass transfer studies in sorbing porous media: II. Experimental evaluation with tritium (3H2O). Soil Sci. Soc. Am. J. 41, 272–278. van Genuchten, M. T., Wierenga, P. J., and O’Connor, G. A. (1977). Mass transfer studies in sorbing porous media: III. Experimental evaluation with 2,4,5-T. Soil Sci. Soc. Am. J. 41, 278 –285. van Oss, C. J., and Good, R. J. (1988). On the mechanism of hydrophobic interactions. J. Dispersion Sci. Technol. 9, 355 – 362. van Oss, C., Chaudhury, M. K., and Good, R. J. (1988). Interfacial Lifshitz–van der Waals and polar interactions in macroscopic systems. Chem. Rev. 88, 927– 941. Vieth, W. R. (1991). “Diffusion in and through Polymers.” Hanser, Munich. Vieth, W. R., and Sladek, K. J. (1965). A model for diffusion in glassy polymer. J. Colloid Sci. 20, 1014–1033.

SORPTION AND DESORPTION RATES

73

Wakao, N., and Smith, J. M. (1962). Diffusion in catalyst pellets. Chem. Eng. Sci. 17, 825 – 834. Weber, W. J., Jr., and Huang, W. (1996). A distributed reactivity model for sorption by soils and sediments. 4. Intraparticle heterogeneity and phase-distribution relationships under nonequilibrium conditions. Environ. Sci. Technol. 30, 881– 888. Werth, C. J., and Reinhard, M. (1997). Effects of temperature on trichloroethylene desorption from silica gel and natural sediments. 2. Kinetics. Environ. Sci. Technol. 31, 697–703. Weston, D. P., and Mayer, L. M. (1998). In vitro digestive fluid extraction as a measure of the bioavailability of sediment-associated polycyclic aromatic hydrocarbons: Sources of variation and implications for partitioning models. Environ. Toxicol. Chem. 17, 820 – 829. White, J., and Pignatello, J. (2000). Influence of bisolute competition on the kinetics of sorption and desorption of polycyclic aromatic hydrocarbons in soil. Environ. Sci. Technol., in press. White, J. C., Hunter, M., Nam, K., Pignatello, J. J., and Alexander, M. (1999). Correlation between the biological and physical availabilities of phenanthrene in soils and soil humin in aging experiments. Environ. Toxicol. Chem. 18, 1720 – 1727. Wolf, D. C., Dao, T. H., Scott, H. D., and Lavy, T. L. (1989). Influence of sterilization methods on selected soil microbiological, physical, and chemical properties. J. Environ. Quality 18, 39 – 44. Wu, S., and Gschwend, P. M. (1986). Sorption kinetics of hydrophobic organic compounds to natural sediments and soils. Environ. Sci. Technol. 20, 717–725. Xia, G., and Ball, W. P. (1999). Adsorption-partitioning uptake of nine low-polarity organic chemicals on a natural sorbent. Environ. Sci. Technol. 33, 262–269. Xing, B., and Pignatello, J. J. (1996). Time-dependent isotherm shape of organic compounds in soil organic matter: Implications for sorption mechanism. Environ. Toxicol. Chem. 15, 1282–1288. Xing, B., and Pignatello, J. J. (1997). Dual-mode sorption of low-polarity compounds in glassy poly(vinyl chloride) and soil organic matter. Environ. Sci. Technol. 31, 792–799. Xing, B., and Pignatello, J. J. (1998). Competitive sorption between 1,3-dichlorobenzene or 2,4dichlorophenol and natural aromatic acids in soil organic matter. Environ. Sci. Technol. 32, 614 – 619. Xing, B., Gigliotti, B., and Pignatello, J. J. (1996). Competitive sorption between atrazine and other organic compounds in soils and model sorbents. Environ. Sci. Technol. 30, 2432–2440. Xue, S. K., and Selim, H. M. (1995). Modeling adsorption–desorption kinetics of alachlor in a typic Fragiudalf. J. Environ. Quality 24, 896 – 903. Yiacoumi, S., and Rao, A. V. (1996). Organic solute uptake from aqueous solutions by soil: A new diffusion model. Water Resour. Res. 32, 431– 440. Yiacoumi, S., and Tien, C. (1994). A model of organic solute uptake from aqueous solutions by soils. Water Resour. Res. 30, 571– 580. Yiacoumi, S., and Tien, C. (1995). Uptake of organic compounds from aqueous solutions by soils—A comparison of two Laplace transform inversion techniques. Computers Chem. Eng. 19, 1041– 1050. Young, D. F., and Ball, W. P. (1994). A priori stimulation of tetrachloroethene transport through aquifer material using an intraparticle diffusion model. Environ. Prog. 13, 9 –20. Young, T. M., and Weber, W. J., Jr. (1995). A distributed reactivity model for sorption by soils and sediments. 3. Effects of diagenetic processes on sorption energetics. Environ. Sci. Technol. 29, 92–97. Zhang, P.-C., and Sparks, D. L. (1993). Kinetics of phenol and aniline adsorption and desorption on an organo-clay. Soil Sci. Soc. Am. J. 57, 340 – 345.

This Page Intentionally Left Blank

ENVIRONMENTAL INDICATORS OF AGROECOSYSTEMS O. H. Smith,1 G. W. Petersen, and B. A. Needelman Department of Agronomy Pennsylvania State University University Park, Pennsylvania 16802

I. II. III. IV.

Introduction Agroecosystems Monitoring and Assessment Endpoints Environmental Indicators A. Biological Indicators B. Physical Indicators C. Chemical Indicators D. Landscape Indicators V. Soil Organic Matter as a Candidate Environmental Indicator A. The Functions of SOM in Soils B. SOM and Atmospheric Carbon C. Factors Controlling the SOM Content of Soils D. Absolute and Relative Measures of SOM E. The Measurement and Expression of SOM Quantities VI. Indicator Ranking VII. Conclusions and Recommendations References

Conventional production agricultural practices are partly responsible for intensifying the degradation of productive lands throughout the world. In monitoring the impacts of these practices, a variety of biological, physical, chemical, landscape, and economic measures are being used as indicators of environmental change. This chapter is largely a review of both common and uncommonly used environmental indicators of agricultural systems. Soil organic matter content is discussed in detail as a candidate environmental indicator. A ranking scheme is proposed for the use of multiple indicators in decision-making applications. © 2000 Academic Press.

1

Current address: VantagePoint Network LLC., 2057 Vermont Drive, Fort Collins, CO 80525. 75 Advances in Agronomy, Volume 69 Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved. 0065-2113/00 $30.00

76

O. H. SMITH ET AL.

I. INTRODUCTION Increased use of synthetic chemicals and intensification of management have increased agricultural production but are often accompanied by deleterious environmental effects. At a local level, these negative effects may include increases in soil erosion and reductions in biodiversity. Regionally, incidents of groundwater contamination and eutrophication of rivers and lakes are becoming more common, and rates of pesticide resistance in insect populations and plant pathogens are increasing (Matson et al., 1997). Regional challenges also include the loss of highvalue farmland for commercial, industrial, and residential land uses (U.S. Department of Agriculture [USDA], 1997). Indicators of such changes require some a priori knowledge of the scale of change in space and time and which indicators are most appropriate for identifying critical impacts. The problem scope and its spatial and temporal scales of resolution need to be defined, ideally prior to sampling, and realistic scales of observation and management need to be selected (Thomann, 1994). In general, less is known about indicators for terrestrial ecosystems than for aquatic systems (U.S. Geological Survey Biological Resource Division [USGSBRD], 1995). There are many interesting leads, but overall we are still far from adequately characterizing major environmental indicators (EIs).

II. AGROECOSYSTEMS An agroecosystem is an ecological system that is intensively managed for the purposes of producing food, feed, and fiber. The functional components of an agroecosystem relate to fluxes of energy, nutrients, matter, and biological species (Altieri, 1995) within and across physical boundaries that are not spatially continuous (Lanyon, 1995). Cropland, pasture, and range comprise more than 55% of the total land area of the contiguous United States (USDA, 1997). Agroecosystems are typically studied at field, whole farm, or regional scales (Food and Agriculture Organization [FAO], 1994) and vary with physical, biological, socioeconomic, and cultural influences, although methods are being developed in precision agriculture for study and management at the subfield scale. Major resource classes in agroecosystems are (i) annually harvested herbaceous crops, (ii) perennial fruit and nut crops, and (iii) pasture or range (U.S. Environmental Protection Agency/ Agroecosystems Resource Group [USEPA/ARG], 1993). In developed countries, agroecosystems are generally managed for high production and are characterized by intensive nutrient and pesticide inputs, fast growth and harvest cycles (hence, disturbance dominated), and low crop and animal genetic diversity (Odum, 1984; Matson et al., 1997). Agroecosystems usually contain parcels of unmanaged or lightly managed areas, such as woodlots, fencerows, and riparian areas, that can

ENVIRONMENTAL INDICATORS OF AGROECOSYSTEMS

77

act both as refuges for beneficial predators of insect pests (Letourneau, 1997) and as reservoirs of insect pests, weed seeds, and sources of plant pathogens. These areas harbor alternate hosts to fungal pathogens of crops such as cedar apple rust and crown rust of oats (Schumann, 1991). Natural areas bordering agricultural fields can act as similar refuges or reservoirs. Recent analyses have estimated national and global economic benefits from ecosystem services of soil formation, nitrogen fixation, decomposition, pest biocontrol, pollination, and others (Costanza and Daly, 1992; Pimentel et al., 1997). Intensive agricultural management practices cause damage or loss of ecosystem services, in the form of changes in nutrient cycling, primary productivity, species diversity, species dominance, and population fluctuations (Odum, 1984; Rapport et al., 1985; Duelli, 1997; Matson et al., 1997; Vitousek, 1997), in exchange for economic productivity. Humans play a major role in determining the nutrient, energy, water, and biological species inputs and outputs in agroecosystems. Therefore, to fully characterize an agroecosystem or identify possible alternative management options for a given agroecosystem, indicators of economic, cultural, and social forces that determine human activities need to be included in an agroecosystem analysis (Aber and Melillo, 1991; Altieri, 1995). Agricultural practices that are ecologically sustainable may not be profitable, thereby being economically unsustainable. Measuring crop productivity or animal production alone also is not a sufficient indicator of agroecosystem status because practices that achieve high yields may not be ecologically or socioeconomically sustainable.

III. MONITORING AND ASSESSMENT ENDPOINTS Previous and continuing federal efforts to monitor environmental conditions in the United States include the National Agricultural Statistical Service (NASS; USDA, 1997), USGS’s Biomonitoring of Environmental Status and Trends program (National Research Council [NRC], 1995), and EPA’s Environmental Mapping and Assessment Program (EMAP; USEPA, 1997). The EMAP–ARG (USEPAARG, 1993) is an interagency partnership between the USEPA and the USDA Agricultural Research Service (ARS), the USDA NASS, and the USDA Natural Resources Conservation Service. The ARG produced a seminal paper (Meyer et al., 1992) which defined the following three assessment endpoints that summarize issues of concern to society and can guide environmental scientists to appropriately interpret indicators: • Sustainability: The capacity of agroecosystems to maintain commodity production through time without threatening ecosystem structure and function. • Contamination of natural resources: The degradation of the quality of air, soil,

78

O. H. SMITH ET AL.

water, or associated biota with by-products of agricultural practices, such as fertilizers, pesticides, pathogens, and sediments. • Quality of agricultural landscapes: The modification of land-use patterns that comprise a matrix of landscape elements. Most significant is the ability of landscapes to support noncrop vegetation and wildlife populations. The pressure–state–response framework (PSR; Pieri, 1995) is a different guide used to conceptualize indicators of sustainability and land degradation that are defined in groups as resource pressure indicators (pressure from human activities), state indicators (resulting states of land quality and changes in their state, such as from agricultural to urban land use), and response indicators (societal responses). The PSR uses land quality indicators that differ by agroecological zones. Indicators include intensity and diversity of land use, soil erosion, sediment loads in surface water, soil fertility, price of farmlands, soil conservation practices, evenness in income distribution, access to markets and services, and potential versus actual farm yields (Pieri, 1995).

IV. ENVIRONMENTAL INDICATORS Meyer et al. (1992) associated various indicators with assessment endpoints, shown in Table I. All the candidate indicators discussed in this report are effectively “state variables” that are integrative of environmental effects and therefore cannot be considered statistically independent variables in subsequent analyses. Of the 21 indicators identified, Meyer et al. (1992) selected the following 6 as the most important, recommending that these should be evaluated by appropriate research: • Crop productivity: Crop yields are maintained by management inputs such as seed, fertilizers, irrigation water, and pesticides. Crop productivity, therefore, is a measure of efficiency, using crop-yield equivalents while accounting for the inputs used to sustain them. • Soil productivity: Soil productivity is characterized by the capacity of soil to supply and maintain fertility, with soil fertility being an integrated measure of nutrient-holding capacity, microbial activity to maintain nutrient delivery and soil structure, the extent of contamination, and erosion rate. • Quantity and quality of irrigation water: Impacts to irrigated agroecosystems from water availability and quality as well as impacts of dissolved salts, sediments, fertilizers, and pesticides from agroecosystems on water quality. • Abundance and diversity of beneficial insects: An indicator of the abundance and diversity of insect pollinators as well as parasites and predators of insect pests of crops.

ENVIRONMENTAL INDICATORS OF AGROECOSYSTEMS

79

Table I Association between Agroecosystem Assessment Endpoints and Potential Indicatorsa

Indicator

Sustainability

Crop productivity Soil productivity Nutrient-holding capacity Erosion Contaminants Microbial component Land use Landscape descriptors Wildlife populations Beneficial insect density Pest density Status of biomonitor species Irrigation water quantity Irrigation water quality Agricultural chemical use Non-point source loading Foliar symptoms Livestock production Socioeconomic factors Genetic diversity

X X X X X X X

Contamination of natural resources

Quality of agricultural landscape

X

X X X

X X X

X X X X X X X X X X

X X X X

a From “Ecological Indicators” (1992), pp. 629– 658. Indicators of the ecological status of agroecosystems, J. R. Meyer, C. L. Campbell, T. J. Moser, G. R. Hess, J. O. Rawlings, S. Peck, and W. W. Heck; Table 35.2; with kind permission from Kluwer Academic Publishers.

• Agricultural chemical use: Crop productivity, non-point source pollution, wildlife, beneficial species, and microbial processes are affected by amounts, frequency, and types of agricultural chemicals used. • Genetic diversity: Continued genetic improvement of crops and livestock relies on the existence of wild-type species. The use of monocultures in agriculture threatens the continuation of genetically diverse populations and invites potential epidemics.

A. BIOLOGICAL INDICATORS Measurement of species abundance, diversity, and extent are also useful as agroecosystem EIs. Soil microbial status, invertebrate assemblages or invertebrate exposures to toxins, pesticide resistance in insect pests, crop diversity, and waterborne pathogens are discussed as important indicators.

80

O. H. SMITH ET AL.

1. Soil Microbial Status Soils are possibly the most critical nonrenewable resource in terrestrial ecosystems. Depletion of the soil resource can have catastrophic impacts on ecosystem structure and function. Measurable effects may include reductions in net primary productivity and species biodiversity following land-use conversion of forest ecosystems to pasture or cropland. The disturbance effect may be attributable, in part, to changes in soil quality and biota. It has been suggested that, because of their rapid turnover times and genetic information transfer, soil microbial communities are especially responsive to ecosystem disturbance (Edwards et al., 1996; Elmholt, 1996; Scow, 1997). The difficulty in testing this hypothesis lies in problems of identifying and culturing soil organisms without biasing the assessment toward those organisms that simply grow well on the media chosen. Therefore, if soil microbial diversity and microbially mediated processes are to be included as EIs, selection should emphasize measurements of microbial diversity that minimizes bias and important specific ecosystem processes, such as nitrification, that are mediated by a select group of microorganisms and that are particularly sensitive to stress, such as nitrification or CO2 fixation by algae. Based on these considerations, two soil microbial indicators may be considered: (i) some index of microbial community diversity. Several have been suggested (Garland and Mills, 1991; Zak et al., 1994), but none is widely accepted so additional evaluations are required; and (ii) assessment of changes in ecosystem function by measuring nitrifying potential of soil (Schmidt and Belser, 1994) and photosynthetic activity by surface-dwelling cyanobacteria and algae. The former approach is based on the fact that an important soil process whereby ammonium is oxidized to nitrate is mediated by a relatively small group of bacteria that are quite sensitive to environmental conditions. The latter approach relies on a process that is sensitive to other types of ecosystem stress. 2. Earthworms Earthworms are important decomposers that make plant residues and microbial biomass available to other organisms. Earthworm activity contributes to soil structure and water infiltration (Altieri, 1995). Earthworms have also been used as indicators of soil sustainability due to their sensitivity to pesticides (Paoletti et al., 1991; Edwards et al., 1996; Bohlen et al., 1997). 3. Honeybees Honeybees, which collect pollens from up to 7 km2, have been used to assess regional pollutant levels (Bromenshenk, 1988). Recently, exposure to organophos-

ENVIRONMENTAL INDICATORS OF AGROECOSYSTEMS

81

phate and carbamate insecticides was determined using acetylcholinesterase activity of honeybees (Stefanidou et al., 1996). 4. Other Biological Indicators Other species used as EIs in agricultural systems include carabid beetles (Kromp, 1990), oribatid mites (Franchini and Rockett, 1996; Lebrun and van Straalen, 1995), lichens (Loppi, 1996), invertebrate assemblages (Paoletti et al., 1991; Samways et al., 1996), protozoa (Foissner, 1997), avian communities (Brooks et al., 1998), and plants used to monitor pest populations (Berlinger et al., 1996). Further development and standardization of biological indicators is greatly needed. 5. Pesticide Resistance in Insect Pests Pesticide resistance is another useful metric of agroecosystem sustainability. Pesticide resistance in arthropods, plant pathogens, and weed species is a major obstacle to sustaining current levels of agricultural production (Georghiou, 1986; Hoy, 1990). Ever-increasing resistance effectively forces farmers to make more frequent pesticide applications and to use new pesticides, increasing chemical loadings to fields, groundwater, and surface waters. Although crop rotations and integrated pest management tactics are locall effective, current practices result in the loss of susceptible genotypes on a regional basis. Because of these losses, regional coordination of agricultural practices to slow resistance development is required. Changes in pesticide resistance can be interpreted by rightward shifts in dose–mortality curves (ffrench-Constant and Roush, 1990). New transgenic crops such as Bt corn (Roush, 1997) express insecticidal proteins in the crop to which insect pests are expected to develop quick resistance (Alstad and Andow, 1995; Roush, 1997; Tabashnik, 1997; Fox, 1998). 6. Crop Diversity Crop diversity at the field, regional, and global scale is an important indicator of agroecosystem sustainability and stability. At the field and regional scale, crop diversity is an indicator of the vulnerability of a cropping system to pest outbreaks (NRC, 1972). If an agroecosystem is dominated by a few crop species representing a few cultivars (genetic composition), a new pest could spread rapidly across a large region. Such an event occurred in 1970 and 1971 when almost all of the corn harvest in Illinois and Indiana was lost due to the fungus causing the southern corn leaf blight (Helminthosporium maydis) and in the mid- and late 1980s when an epidemic of blackleg (Leptosphaeria maculans) in rapeseed caused immense yield losses ( Juska et al., 1997). Crop species diversity and the diversity of genotypes of a given species at the global scale is an important indicator of agro-

82

O. H. SMITH ET AL.

ecosystem sustainability. Loss of crop and related wild species, as well as their natural habitats, indicates the loss of genetic resources important to the future of crop breeding programs (Gliessman, 1998). 7. Fecal Pathogens Surface water runoff from agricultural fields where livestock are present usually contains fecal pathogens. Fecal pathogens and parasitic protozoa can also be used as biological indicators of water quality resulting from agricultural practices. However, more research is needed to identify appropriate indicators of pathogens and their biology in order to determine cost-effective water quality protection programs. For instance, fecal coliforms and fecal streptococci are not pathogens, but they are indicators of water-borne pathogens that are difficult to monitor (Doran et al., 1981). Confirmed human pathogens such as Cryptosporidium are a result of runoff from dairy operations (Atwill, 1996). Particularly sensitive areas include those underlain by limestone bedrock, where common sinkholes can allow surface water to contaminate groundwater aquifers. Challenges exist with the use of bioindicators: If a particular bioindicator species exists in one area, it may be absent from another region. Related species used as indicators may have different susceptibilities than those of the primary species. Soil microorganisms, lichens, earthworms, beetles, bees, and birds all have merits as indicators of agroenvironmental condition, each with varying ranges of applicability. These ranges need to be experimentally determined. Changes in pest resistance to pesticides will also be valuable indicators of impacts due to the widespread adoption of transgenic crops such as Bt corn and Bt cotton.

B. PHYSICAL INDICATORS Physical EIs in agroecosystems were discussed in part in Table I in relation to indicators of crop and soil productivity. Special mention, however, needs to be made about soil degradation. Soil degradation, which reduces both profitability and sustainability, is caused by water and wind erosion, excess salts, leaching of nutrients, buildup of toxins, and changes in porosity, permeability, bulk density, structural stability, and rates of organic matter decomposition (FAO, 1990). Soil loss alone is reported to occur 13–40 times faster than soil renewal rates, mainly from erosion on tilled and bare lands (Pimentel and Kounang, 1998). Soil degradation is currently being assessed using remote sensing in conjunction with field observations, soil databases, simulation models, and geographic information systems (GIS; FAO, 1990; Petersen et al., 1997; Nizeyimana and Petersen, 1997). Landscape indicators, discussed later, can also be useful for coarse-scale assessments of agroecosystem change.

ENVIRONMENTAL INDICATORS OF AGROECOSYSTEMS

83

An important database of landscape-scale natural resource trends is the National Resource Inventory (NRI). The NRI is a comprehensive sampling of land cover, land use, soil erosion, prime farmland, wetlands, and other characteristics on nonfederal rural land in the United States (USDA, 1997). Every 5 years since 1982, the USDA has published summaries of natural resource trends based on 800,000 sites in the NRI database. Several important discoveries were made using these data: Although the rates of soil erosion and agricultural wetland loss have decreased, between 1982 and 1992, 6 million acres of prime farmland were converted to nonagricultural uses. Future NRI surveys should continue to elucidate important natural resource trends.

C. CHEMICAL INDICATORS Chemical processes that occur in natural and managed ecosystems affect the bioavailability of nutrients and toxins and can therefore impose significant constraints on ecosystem status. An essential or a toxic element is bioavailable if it is in a chemical form that plants can absorb readily and if, once absorbed, it affects the life cycle of the plant (Barber, 1984). Therefore, key to assessing bioavailability is the knowledge of the form, speciation, or sequestration of a nutrient or potentially toxic element in the ecosystem matrix. Soil chemical and biochemical reactions such as complexation, sorption, ion exchange, mineral solubility, and organic matter transformation set the initial constraints on the availability of essential elements for plant uptake (Sposito, 1989). In addition to adequate supply, plant growth is affected by the balance of nutrients in the soil. Nutrient imbalances do not usually induce deficiency symptoms unless the supply of a given nutrient reaches very low levels, but they do cause plants to grow more slowly (Clarkson and Hanson, 1980). Conversely, the balance of nutrient and toxic elements can induce significant stress in ecosystems in which chemical species are taken up competitively from the soil. In such cases, molar ratios can provide a valuable EI for stress “thresholds” (Cronan and Grigal, 1995). For example, the calcium to aluminum (Ca/Al) molar ratio of the soil solution has been used successfully to identify approximate thresholds beyond which the risk of forest damage from Al stress and nutrient imbalances increases. It has also been used as an indicator to assess forest ecosystem changes over time in response to acidic deposition, forest harvesting, and other soil acidifying processes (Cronan and Grigal, 1995). Selective extraction procedures have been developed to quantify available nutrient species in soils (Van der Zee et al., 1987; Shuman et al., 1988; Bundy and Meisinger, 1994). However, the problem with routine application of operationally defined extractions to quantify chemical species is that the methods are not sufficiently specific for target compounds and the actual species extracted are rarely known with precision. Recent advances in molecular spectroscopy provide a set

84

O. H. SMITH ET AL.

of tools that may be used to assess chemical speciation in complex matrices such as soils (Bertsch and Hunter, 1998). Spectroscopic approaches should be exploited in future work to corroborate the results of selective extractions and to assess the viability of both approaches to evaluate element speciation and bioavailability. Excess salinity, extremes in pH, and the status of nitrogen and phosphorous are also important chemical indicators in agricultural systems. Crop harvest involve the intentional removal of nutrients from agricultural lands. Unintentional nutrient removal occurs through soil erosion, nutrient runoff or leaching, and volatilization, depleting nutrient stocks that are necessary for sustained yields (Magdoff et al., 1997). To offset nutrient depletion, synthetic fertilizers and/or manure are added to the soil, which results in temporary excesses in N and P that are subject to runoff and leaching, causing eutrophication of rivers and lakes. Nutrient flow pattern, or nutrient dislocation, has demonstrable consequences at the field, farm, region, and global scales (Magdoff et al., 1997).

D. LANDSCAPE INDICATORS Landscape variables have been used extensively in efforts to link landscape patterns with ecological processes (Turner, 1989; Forman, 1995; Gustafson and Gardner, 1996; Ives et al., 1998). The theoretical underpinnings of pattern/process interactions are related to issues of hierarchy and scale (Allen and Hoekstra, 1992; Riera et al., 1998). Landscape descriptors, or metrics, reflect levels of landscape disturbance, which have been used to test hypotheses relating to the effects of landscape structure on such diverse entities as forest communities (Franklin and Forman, 1987), beetles in rangelands (Wiens and Milne, 1989) and in croplands (Colunga-Garcia et al., 1997), crop pest parasitoids (Marino and Landis, 1996), bobcats (Litvaitis, 1993), prairie grasses (Wu and Levin, 1994), and avian guilds (Herkert, 1994; Miller et al., 1997). Combinations of GIS and knowledge-based systems have been employed to interpret relations within and among landscape data themes for decision making on agricultural and natural resources (Coulson et al., 1990; Petersen et al., 1995, 1997). Spatial patterns of land cover types are characterized by a variety of landscape metrics. Riitters et al. (1995) determined that 55 discrete metrices could be reduced by factor analysis to the following metrics, exhibiting relative statistical independence: 1. Average patch perimeter/area ratio (called fractal dimension) 2. Contagion 3. Average patch area normalized to the area of a square with the same perimeter (linear vs planar patch shapes) 4. Patch perimeter/area scaling 5. Number of attribute classes

ENVIRONMENTAL INDICATORS OF AGROECOSYSTEMS

85

The USEPA (1994) ranked landscape metrics with regard to their suitability in landscape characterization studies. The list, shown in Table II, shows that contagion, fractal dimension, and dominance were ready for field test and implementation. Equations describing each metric are beyond the scope of this chapter; however, readers are referred to Gardner and O’Neill (1990) for further details. Contagion is a measure of clumping, or aggregations of patches (Gardner and O’Neill, 1990). Fractal dimension, more accurately described as average patch perimeter/area ratio, is a measure of patch edge complexity (Wiens and Milne, 1989). Dominance is the proportion to which one or few land cover types occur predominantly in the landscape (Gardner and O’Neill, 1990). Major improvements to contagion and fractal dimension that are dramatically less sensitive to spatial resolution and tesellation, termed the patch per unit area metric and square-pixel metric, respectively, have been developed by Frohn (1997).

V. SOIL ORGANIC MATTER AS A CANDIDATE ENVIRONMENTAL INDICATOR No indicator can fully characterize agroecosystem status at multiple scales. However, soil organic matter (SOM) content correlates with many aspects of agroecosystem productivity, sustainability, and environmental integrity. Stevenson’s (1994, p. 1) definition of SOM as “the whole of the organic material in soils, in-

Table II Ranking of Landscape Stability and Resilience Indicators by the EPA (1994)a Indicator

Status

Contagion Fractal dimension Dominance Lacunarity Diffusion rates Percolation backbone Percolation thresholds Miles of road Recovery time Landcover transition matrix

C C C A A B B B A A

aAbbreviations used: A, requiring further conceptual development; B, requiring testing for feasilbility/sensitivity; C, ready for field test and implementation.

86

O. H. SMITH ET AL.

cluding litter, light fraction, microbial biomass, water-soluble organics, and stabilized organic matter (humus)” will be used in this text. In general, SOM content is positively correlated with positive soil status (Reeves, 1997; Bauer and Black, 1994). The methodology for SOM measurement is established and in widespread use (Nelson and Sommers, 1996). With proper techniques, the measurement of SOM is very precise. In comparison to many soil measures, SOM content has small spatial and short-term temporal variability. There is a large body of basic and applied research focused on SOM character and the relationship between SOM and agroecosystem properties. The SOM capacity of a soil is dependent on other soil properties such as soil texture. Therefore, changes in SOM content must be interpreted in the context of site-specific variables. SOM is not a comprehensive indicator. For example, SOM content is not an accurate indicator of excess salinity and poor drainage. In addition, many larger scale parameters, such as land conversion and rural economic prosperity, cannot be quantified with the SOM indicator.

A. THE FUNCTIONS OF SOM IN SOILS SOM plays a mostly beneficial role in determining the biological, physical, and chemical qualities of a soil (Stevenson, 1994, 1986). SOM is a nutrient and energy source for soil organisms and a nutrient source for plants. It improves soil structure, strengthens soil aggregates, increases water retention, chelates metals, buffers the soil pH, interacts with xenobiotics, and retains cations and anions in the soil system. The positive relationship between SOM content and crop yield has been observed in many agroecosystems (Reeves, 1997). Bauer and Black (1994) correlated an increase of 15.6 kg ha1 of wheat grain yield to a 1 Mg ha1 increase of SOM in the northern Great Plains. The influences of increased SOM on soil physical properties tends to reduce runoff and erodibility. This may be a result of increased infiltration and percolation rates, water-holding capacity, aggregate strength, and crusting resistance. The relationships described previously between SOM and soil properties are not necessarily causative. For example, SOM is correlated with the clay content of a soil, making it difficult to separate the effects of SOM and clay on soil properties.

B. SOM AND ATMOSPHERIC CARBON The carbon component of SOM is a dynamic pool in the global carbon cycle and forms the largest terrestrial carbon pool (Paul and Clark, 1996). Kern and Johnson (1993) sought to estimate the potential effect of conservation tillage on

ENVIRONMENTAL INDICATORS OF AGROECOSYSTEMS

87

U.S. terrestrial carbon pools. They reviewed 17 studies that compared SOM contents in conventional tillage and no-tillage systems. They estimated that the increase in SOM carbon that would result from the adoption of no-till agriculture on 76% of the planted cropland in the United States would be equivalent to 0.7–1.1% of the total projected U.S. fossil fuel carbon emissions for the next 30 years. The authors assumed that the effects of tillage were not dependent on climatic and soil variables.

C. FACTORS CONTROLLING THE SOM CONTENT OF SOILS 1. The Factors of Soil Formation The five factors of soil formation—climate, organisms, relief, parent material, and time—largely determine the SOM content of soils ( Jenny, 1941; Stevenson, 1994). In general, the degree of impact of the factors of soil formation on SOM decreases in the order climate  vegetation  topography parent material  age for loamy soils in the United States. Climate influences SOM content primarily through temperature and precipitation. Temperature is negatively correlated to SOM content, probably due to increased microbial activity at higher temperatures (Stevenson, 1986). In general, precipitation stimulates plant growth, thereby increasing organic matter inputs into soils (Stevenson, 1994). Soil moisture changes may decrease or increase SOM decomposition rates, depending on the balance between water and oxygen availability. An example of the effect of vegetation on SOM is the greater quantity of SOM found in grassland (Mollisols) soils than in forest (Alfisols) soils. The mechanisms of this effect may include increased biomass production, nitrification inhibition, decreased aeration, and a more extensive rhizosphere in grasslands (Stevenson, 1986). Topography influences SOM contents through impacts on microclimate, drainage, and erosion. Anaerobic conditions in poorly drained soils cause SOM accumulation due to slow decomposition rates and incomplete catabolism. Erosion tends to remove SOM-enriched soil (Rasmussen and Collins, 1991). The impact of parent material is primarily through its influence on soil texture (Stevenson, 1986). The positive relationship between SOM content and fine soil texture is well established. Soil age is most important in young soils, in which SOM accumulation rates typically exceed rates of decay. 2. Management The tillage of natural lands leads to a loss of SOM (Jenny, 1941; Paustian et al., 1997). This may be a result of erosion, increased oxygen availability, reduced biomass production, and the disruption of aggregates. Erosion removes the surface

88

O. H. SMITH ET AL.

layer, which is frequently the soil with the greatest SOM concentration. In addition, erosion preferentially removes fine soil particles, which are SOM enriched. Rasmussen and Collins (1991) argue that soil erosion is often overlooked in SOM field experiments because soil erosion rates are rarely measured and field experiments are usually located on fertile, level to gently sloping land. In general, management practices that are considered beneficial to the health of the agroecosystem, such as cover crops, conservation tillage, manure and residue inputs, and erosion prevention practices, also increase SOM contents. Many agronomic practices that increase yields will also increase biomass production, thereby increasing organic matter inputs. However, there are many exceptions to these generalizations. The impacts of management practices on SOM are a result of a complex and interacting set of factors. Most of the research concerning the impacts of management practices on SOM contents has been conducted using long-term field plot studies. Although much has been learned in these studies (Paustian et al., 1997), additional research is needed to quantify the interactions between and among the factors of soil formation and management practices. For example, Campbell et al. (1999) investigated the impacts of texture in an 11-year comparison of no-till and conventional tillage practices at three sites in the Brown soil zone in semiarid southwestern Saskatchewan. They observed that the no-till systems contained more SOM than conventionally tilled systems only in finer-textured soils.

D. ABSOLUTE AND RELATIVE MEASURES OF SOM SOM content may be quantified in absolute or relative terms. Absolute measures are simply the amount of SOM in a system. Relative measures estimate the change in SOM content over time. The use of absolute measures would require the estimation of SOM baseline conditions, most likely based on the factors of soil formation and past management practices. Although this is an area of major research, the relationship between the factors of soil formation, management practices, and SOM contents is not sufficiently understood to make widespread, accurate baseline estimates. In addition, the data set that would be required is not available. For these reasons, relative measures of SOM content should be used. With sufficient sampling intensity, a time step of 5 years should be appropriate for most applications. The impacts of management on SOM content are generally detectable after several years but may take decades. Relative measures need to be understood in the context of soil variability. For example, a 10% increase in SOM content in a clay soil and in a sandy soil should be interpreted differently. These interpretations are a function of the particular use of the SOM measure. For example, for issues of atmospheric carbon, net changes should be considered. The topic requires further research.

ENVIRONMENTAL INDICATORS OF AGROECOSYSTEMS

89

E. THE MEASUREMENT AND EXPRESSION OF SOM QUANTITIES There are three standard methods to express SOM quantities: gravimetric, volumetric, and equivalent mass. The gravimetric and volumetric methods may not adequately account for soil depth and density (Hammer et al., 1995). The simplest way to express the quantity of SOM in soil is as the mass of carbon or nitrogen per mass of dry soil (mass basis). This method characterizes the average mass SOM per mass soil in the sampling depth. Expression of SOM on a mass basis cannot account for the total quantity of soil present (Doran and Parkin, 1994). For example, this method may not accurately estimate SOM losses due to erosion. In principle, the solution is to express SOM quantities as a mass carbon or nitrogen per unit area of soil—in effect integrating the mass per unit volume over sufficient depth to include all the organic matter. However, it is generally impractical to sample to a depth that would include all the organic matter in the system. A more practical solution is to express SOM quantities per unit volume of soil (volumetric basis). This method is most effective when most of the SOM of the site is contained in the sampling depth, which is often not the case. This method may overestimate SOM in high bulk density soils. For example, consider a study that compares a low and a high bulk density soil of otherwise identical composition. Both soils will have identical gravimetric SOM contents, but the high bulk density soil will have a greater volumetric SOM content. This is particularly problematic when treatments, such as tillage, affect bulk density. One solution to this problem is to express the quantity of SOM on an equivalent mass basis (Ellert and Bettany, 1995). This approach adds a step to the volumetric method of calculation. Volumetric measurements are corrected by adjusting the sampling depth to give an equivalent mass. This is done by sampling a smaller depth in heavier soils or a greater depth in lighter soils. In the example of a low and a high bulk density soil, a portion of the high bulk density soil would be removed mathematically, restoring the equivalence in SOM content estimation. Soil density, particularly that of the surface soil, is variable over short periods of time. The equivalent mass basis is not sensitive to soil density changes. Surface residues are rarely included in the quantification of SOM (Paustian et al., 1997). This underestiimates SOM sequestration in high-residue systems such as conservation tillage and forests. However, in soils with large SOM contents, residues account for a small percentage of the total SOM storage of the system. Carbon storage in residue is more important to consider in lower SOM content soils. Typically, SOM concentrations are stratified in soils with the greatest concentration near the surface. The degree of stratification is dependent on soil properties and management. For example, tillage homogenizes SOM in the plow layer. At the other extreme, forest soils contain a surface layer composed primarily of organic materials. A vertically stratified sampling system should be used in order

90

O. H. SMITH ET AL.

to estimate SOM distribution. The vertical distribution of SOM is important in understanding SOM function and the causes of soil degradation. In general, sampling depths should be based on soil horizons as opposed to standard depths. However, standard depths may be used to divide soil horizons (e.g., 0–2 cm, 2–6 cm, and 6 cm to the bottom of the A horizon).

VI. INDICATOR RANKING Although SOM as a candidate EI was discussed in particular detail, the caveats for measuring, determining, and using indicator values apply to any indicator variable that is examined. Furthermore, using a combination of indicators presents special challenges in determining the weight and ranking assigned to each indicator. Development of indicator rankings should include criteria of indicator appropriateness based on biological relevance, repeatability of measure, integration of effects, existing databases, cost-effectiveness, and other criteria. When does a state variable become an EI? In an interactive group decision-making process that involves multiple and conflicting criteria, many experts, and many potential outcomes, the Delphi method (Erffmeyer et al., 1986) can be useful in gaining consensus. Characteristics of the Delphi technique include anonymity, controlled feedback, and statistical group response (Khorramshahgol and Moustakis, 1988). Once the evaluation criteria are weighted (Table III), each of the indicators can be ranked (Table IV) according to the evaluation criteria by circulating brief, successive questionnaires among experts in the field, yielding a total numeric score for each indicator. Table V contains the final indicator rankings, lumped by score ranges into numeric categories, with 1 being the highest rank. Applying EIs for regulatory or policy purposes necessitates the use of multiple criteria decision-making tools such as the Delphi process. Table III Example Criteria, with Associated Example Weights, That May Be Used to Rank Any Indicator Criteria

Average weight (sums to 100)

Biological relevance Repeatability Integrated measurement Cost-effectiveness Existing database Infrastructural support Etc.

35 20 20 10 10 5 —

91

ENVIRONMENTAL INDICATORS OF AGROECOSYSTEMS Table IV

Example Criteria for One Indicator and Its Associated Ranking and Weighted Rank Scores Criteria relating to crop productivity indicator Cost-effectiveness Biological relevance Repeatability Integrated measurement Existing database Infrastructural support Etc. Total for crop productivity

Rank (0–9)

Criteria weights

3 9 3 4 4 2 —

10 35 20 20 10 5 —

Indicator weighted rank 30 315 60 80 40 10 — 535

VII. CONCLUSIONS AND RECOMMENDATIONS Concerns regarding issues of ecology and sustainability of agricultural systems as well as detrimental environmental impacts from agriculture have resulted in the development of indicators to estimate environmental trends and conditions. To appropriately describe and evaluate the sustainability of an agroecosystem, a wide

Table V Matrix of Candidate Agroecosystem Indicators Based on Potential Rankings Indicator

Rank

Crop productivity Soil productivity Nutrient-holding capacity Erosion Contaminants Microbial status Irrigation water quantity and quality Beneficial insect abundance and diversity Agricultural chemical use Genetic diversity Status of biomonitor species Landscape descriptors Pest-resistance status Socioeconomic factors Etc.

1 1 3 2 3 1 2 2 1 2 1 3 1 3 —

92

O. H. SMITH ET AL.

range of biological, physical, chemical, and economic indicators need to be assessed. The following recommendations can be made with regard to the use of EIs: • Determine which indicators are the most sensitive to change. • Develop standards for validating bioindicators as EIs. • Validate the use of landscape metrics with remote-sensing and GIS products for monitoring landscape change. • Use a multiscale approach to indicator validation, with studies targeted to local, landscape, and regional levels. • Provide funds for programs designed to improve terrestrial assessment methods and long-term monitoring of soil biological processes, soil organic matter, crop diversity, movement of pests and pathogens, ecological impacts of propagating transgenic crops, changes in resistance to pesticides of insects, weeds, and plant pathogens. • Conduct field studies to explore linkages between ecological processes and spatial/temporal patterns in agroecosystems. • Institute policies to preserve crop and wild crop species diversity, which is critical for maintaining a genetic resource base for future plant breeding programs. • Amend the USDA NASS questionnaire to include bioindicator information and other data relevant to USEPA EMAP assessments.

ACKNOWLEDGMENTS We thank John Chorover, Shelby Fleischer, Heather Karsten, and Andy Rogowski for their constructive criticisms and contributions.

REFERENCES Aber, J. D., and Melillo, J. M. (1991). “Terrestrial Ecosystems.” Saunders, Philadelphia. Allen, T. F. H., and Hoekstra, T. W. (1992). “Toward a Unified Ecology.” Columbia Univ. Press, New York. Alstad, D. N., and Andow, D. A. (1995). Managing the evolution of insect resistance to transgenic plants. Science 268, 1894 –1896. Altieri, M. A. (1995). “Agroecology: The Science of Sustainable Agriculture,” 2nd ed. Westview, Boulder, CO. Atwill, E. R. (1996). Assessing the link between rangeland cattle and water-borne Cryptosporidium parvum infection in humans. Rangelands 18(2), 48 – 51. Barber, S. A. (1984). “Soil Nutrient Bioavailability.” Wiley, New York. Bauer, A., and Black, A. L. (1994). Quantification of the effect of soil organic matter content on soil productivity. Soil Sci. Soc. Am. J. 58, 185 –193. Berlinger, M. J., Dijk, B. L., Dahan, R., and Lebiush-Mordechi, S. (1996). Indicator plants for monitoring pest population growth. Ann. Entomol. Soc. Am. 89(5), 611– 622. Bertsch, P. M., and Hunter, D. B. (1998). Elucidating fundamental mechanisms in soil and environ-

ENVIRONMENTAL INDICATORS OF AGROECOSYSTEMS

93

mental chemistry: The role of advanced analytical and spectroscopic methods. In “The Future of Soil Chemistry” (P. M. Huang, Ed.). SSSA, Madison, WI. Bohlen, P. J., Parmelee, R. W., McCartney, D. A., and Edwards, C. A. (1997). Earthworm effects on carbon and nitrogen dynamics of surface litter in corn agroecosystems. Ecol. Appl. 7(4), 1341– 1349. Bromenshenk, J. J. (1988). Regional monitoring of pollutants with honeybees. In “Progress in Environmental Specimen Banking” (S. A. Wise, R. Zeisler, and G. M. Goldstein, Eds.), Special Publ. No. 740. U.S. Department of Commerce, National Bureau of Standards, Washington, DC. Brooks, R. P., O’Connell, T. J., Wardrop, D. H., and Jackson, L. E. (1998). Toward a regional index of biological integrity: The example of forested riparian ecosystems. Environ. Monitoring Assessment 51(1), 131–143. Bundy, L. G., and Meisinger, J. J. (1994). Nitrogen availability indices. P. 951– 984, In “Methods of Soil Analysis, Part 2—Microbiological and Biochemical Properties” (R. W. Weaver et al., Eds.), SSSA Book Series No. 5. SSSA, Madison, WI. Campbell, C. A., Biederbeck, V. O., McConkey, B. G., Curtin, D., and Zentner, R. P. (1999). Soil quality: Effect of tillage and fallow frequency. Soil organic matter quality as influenced by tillage and fallow frequency in a silt loam in Southern Saskatchewan. Soil Biol. Biochem. 31, 1–7. Clarkson, D. T., and Hanson, J. B. (1980). The mineral nutrition of higher plants. Annu. Rev. Plant Phys. 31, 239–298. Colunga-Garcia, M., Gage, S. H., and Landis, D. A. (1997). Response of an assemblage of Coccinellidae (Coleoptera) to a diverse agricultural landscape. Environ. Entomol. 26, 797– 804. Costanza, R., and Daly, H. (1992). Natural capital and sustainable development. Conservation Biol. 1, 37–45. Coulson, R. N., Lovelady, C. N., Flamm, R. O., Spradling, S. L., and Saunders, M. C. (1990). Intelligent geographic information systems for natural resource management. In “Quantitative Methods in Landscape Ecology: The Analysis and Interpretation of Landscape Heterogeneity.” (M. G. Turner and R. H. Gardner, Eds.), pp. 153 –172. Springer-Verlag, New York. Cronan, C. S., and Grigal, D. F. (1995). Use of calcium/aluminum ratios as indicators of stress in forest ecosystems. J. Environ. Quality 24, 209 –226. Doran, J. W., and Parkin, T. B. (1994). Defining and assessing soil quality. p. 3 –21. In “Defining Soil Quality for a Sustainable Environment” ( J. W. Doran, D. C. Coleman, D. F. Bezdicek, and B. A. Stewart, Eds.), SSSA Special Publ. No. 35. SSSA, Madison, WI. Doran, J. W., Slepers, J. S., and Swanson, N. P. (1981). Chemical and bacteriological quality of pasture runoff. J Soil Water Conservation 36(3), 166 –171. Duelli, P. (1997). Biodiversity evaluation in agricultural landscapes: An approach at two different scales. Agric. Ecosystems Environ. 62, 81– 91. Edwards, C. A., Subler, S., Chen, S. K., and Bogomolov, D. M. (1996). Essential criteria for selecting bioindicator species, processes, or systems to assess the environmental impact of chemicals on soil ecosystems. In “Bioindicator Systems for Soil Pollution” (N. M. van Straalenand and D. A. Krivolutskii, Eds.), pp. 67– 84. Kluwer, Dordrecht. Ellert, B. H., and Bettany, J. R. (1995). Calculation of organic matter and nutrients stored in soils under contrasting management regimes. Can. J. Soil Sci. 75, 529 – 538. Elmholt, S. (1996). Microbial activity, fungal abundance, and distribution of penicillium and fusarium as bioindicators of a temporal development of organically cultivated soils. Biol. Agric. Hort. 13(2), 123. Erffmeyer, R. C., Erffmeyer, E. S., and Lane, I. M. (1986). The delphi technique: An empirical evaluation of the optimal number of rounds. Groups Org. Stud. 11, 120 –128. ffrench-Constant, R. H., and Roush, R. T. (1990). Resistance detection and documentation: The relative roles of pesticidal and biochemical assays. In “Pesticide Resistance in Arthropods” (R. T. Roush and B. E. Tabashnik, Eds.). Chapman & Hall, New York.

94

O. H. SMITH ET AL.

Foissner, W. (1997). Protozoa as bioindicators in agroecosystems, with emphasis on farming practices, biocides, and biodiversity. Agric. Ecosystems Environ. 62, 93 –103. Food and Agriculture Organization (FAO) (1990). Use of high-resolution satellite data and geographic information system for soil erosion mapping, FAO RSC Series No. 56. FAO, Rome. Food and Agriculture Organization (FAO) (1994). Agro-Ecological land resources assessment for agricultural development planning: A case study of Kenya. Making land use choices for district planning, World Soil Resource Reports 71/9. FAO, Rome. Forman, R. T. (1995). “Land Mosaics: The Ecology of Landscapes and Regions.” Cambridge Univ. Press, Cambridge, UK. Fox, T. (1998). EPA seeks refuge from Bt resistance. Nature Biotechnol. 15(5), 409. Franchini, P., and Rockett, C. L. (1996). Oribatid mites as “indicator” species for estimating the environmental impact of conventional and conservation tillage practices. Pedobiologia 40(3), 217– 225. Franklin, J. F., and Forman, R. T. T. (1987). Creating landscape patterns by forest cutting: Ecological consequences and principles. Landscape Ecol. 1, 5 –18. Frohn, R. C. (1997). “Remote Sensing for Landscape Ecology: New Metric Indicators for Monitoring, Modeling, and Assessment of Ecosystems.” CRC Press, Boca Raton, FL. Gardner, R. H., and O’Neill, R. V. (1990). Pattern, process, and predictability: The use of neutral models for landscape analysis. In “Quantitative Methods in Landscape Ecology: The Analysis and Interpretation of Landscape Heterogeneity” (M. G. Turner and R. H. Gardner, Eds.), pp. 289– 307. Springer-Verlag, New York. Garland, J. I., and Mills, A. I. (1991). Classifications and characterization of heterotrophic microbial communities on the basis of community level sole-carbon source utilization. Appl. Environ. Microbiol. 57, 2351–2359. Georghiou, G. P. (1986). “Pesticide Resistance: Strategies and Tactics for Management.” National Academy Press, Washington, DC. Gliessman, S. R. (1998). “Agroecology. Ecological Processes in Sustainable Agriculture.” Ann Arbor Press, Ann Arbor, MI. Gustafson, E. J., and Gardner, R. H. (1996). The effect of landscape heterogeneity on the probability of patch colonization. Ecology 77, 94 –107. Hammer, R. D., Henderson, G. S., Udawatta, R., and Brandt, D. K. (1995). Soil organic carbon in the Missouri Forest–Prairie Ecotone. In “Carbon Forms and Functions in Forest Soils” (W. W. McFee and J. W. Kelly, Eds.), pp. 201–231. SSSA, Madison, WI. Herkert, J. R. (1994). The effects of habitat fragmentation on midwestern grassland bird communities. Ecol. Appl. 4(3), 461– 471. Hoy, M. A. (1990). Pesticide resistance in arthropod natural enemies: Variability and selection responses. In “Pesticide Resistance in Arthropods” (R. T. Roush and B. E. Tabashnik, Eds.). Chapman & Hall, New York. Ives, A. R., Turner, M. G., and Pearson, S. M. (1998). Local explanations of landscape patterns: Can analytical approaches approximate simulation models of spatial processes? Ecosystems 1, 35 – 51. Jenny, H. (1941). “Factors of Soil Formation.” McGraw-Hill, New York. Juska, A., Busch, L., and Tanaka, K. (1997). The Blackleg epidemic in Canadian rapeseed as a “normal agricultural accident.” Ecol. Appl. 7(4), 1350 –1356. Kern, J. S., and Johnson, M. G. (1993). Conservation tillage impacts on national soil and atmospheric carbon levels. Soil Sci. Soc. Am. J. 57, 200 –210. Khorramshahgol, R., and Moustakis, V. S. (1988). Delphic hierarchy process (DHP): A methodology for priority setting derived from the Delphi method and analytic hierarchy process. Euro. J. Operations Res. 37, 347– 354. Kromp, B. (1990). Carabid beetles (Coleoptera, Carabidae) as bioindicators in biological and conventional farming in Austrian potato fields. Biol. fertil. Soils 9(2), 182.

ENVIRONMENTAL INDICATORS OF AGROECOSYSTEMS

95

Lanyon, L. E. (1995). Does nitrogen cycle? Changes in the spatial dynamics of nitrogen with industrial nitrogen fixation. J. Prod. Agric. 8, 70 –78. Lebrun, P., and van Straalen, N. M. (1995). Oribatid mites: Prospects for the use in ecotoxicology. Exp. Appl. Acarol. 19(7), 361– 379. Letourneau, D. K. (1997). Plant–arthropod interaction in agroecosystems. In “Ecology in Agriculture” (L. E. Jackson, Ed.), pp. 239 –290. Academic Press, San Diego. Litvaitis, J. A. (1993). Response of early successional vertebrates to historic changes in land use. Conservation Biol. 7(4), 866 – 872. Loppi, S. (1996). Lichens as bioindicators of geothermal air pollution in central Italy. Bryologist 99, 41. Magdoff, F., Lanyon, L., and Liebhardt, B. (1997). Nutrient cycling, transformations, and flows: Implications for a more sustainable agriculture. Adv. Agron. 60, 1–73. Marino, P. C., and Landis, D. A. (1996). Effect of landscape structure on parasitoid diversity and parasitism in agroecosystems. Ecol. Appl. 6, 276 –284. Matson, P. A., Parton, W. J., Power, A. G., and Swift, M. J. (1997). Agricultural intensification and ecosystem properties. Science 277, 504 – 509. Meyer, J. R., Campbell, C. L., Moser, T. J., Hess, G. R., Rawlings, J. O., Peck, S., and Heck, W. W. (1992). Indicators of the ecological status of agroecosystems. In “Ecological Indicators” (D. H. McKenzie, D. E. Hyatt, and V. J. McDonald, Eds.), pp. 629 – 658. Elsevier, Amsterdam. Miller, J. N., Brooks, R. P., and Croonquist, M. J. (1997). Effects of landscape patterns on biotic communities. Landscape Ecol. 12, 137–153. National Research Council (1972). “Committee on Genetic Vulnerability of Major Crops. Genetic Vulnerability of Major Crops.” National Academy of Science, Washington, DC. National Research Council (1989). “Alternative Agriculture.” National Academy Press, Washington, DC. National Research Council (1995, February). “A Review of the Biomonitorning of Environmental Status and Trends (BEST) Program: The Draft Detailed Plan. Committee to Review the Department of the Interior’s Biomonitoring of Environmental Status and Trends Program.” National Academy Press, Washington, DC. Nelson, D. W., and Sommers, L. E. (1996). Total carbon, organic carbon, and organic matter. In “Methods of Soil Analysis, Part 3. Chemical Methods” (D. L. Sparks et al., Eds.), SSSA Book Ser. No. 5, pp. 961–1010. ASA/SSSA, Madison, WI. Nizeyimana, E., and Petersen, G. W. (1997). Remote sensing applications to soil degradation assessments. In “Methods for Assessment of Soil Degradation” (R. Lal, W. H. Blum, C. Valentine, and B. A. Stewart, Eds.), pp. 393– 403. CRC Press, Washington, DC. Odum, E. P. (1984). Properties of agroecosystems. In “Agricultural Ecosystems: Unifying Concepts” (R. Lowrance, B. R. Stinner, and G. J. House, Eds.), pp. 3–12. Wiley–Interscience, New York. Paoletti, M. G., Favretto, M. R., Stinner, B. R., Purrington, F. F., and Bater, J. E. (1991). Invertebrates as bioindicators of soil use. Agric. Ecosystems Environ. 34, 341– 362. Paul, E. A., and Clark, F. E. (1996). “Soil Microbiology and Biochemistry,” 2nd ed. Academic Press, San Diego. Paustian, K., Collins, H. P., and Paul, E. A. (1997). Management controls on soil carbon. In “Soil Organic Matter in Temperate Agroecosystems” (E. A. Paul, E. T. Elliot, K. Paustian, and C. V. Cole, Eds.), pp. 15–49. CRC Press, Boca Raton, FL. Petersen, G. W., Bell, J. C., McSweeney, K., Nielsen, G. A., and Robert, P. C. (1995). Geographic information systems in agronomy. Adv. Agron. 55, 67–111. Petersen, G. W., Nizeyimana, E., and Evans, B. M. (1997). Applications of geographic information systems in soil degradation assessments. In “Methods for Assessment of Soil Degradation” (R. Lal, W. H. Blum, C. Valentine, and B. A. Stewart, Eds.), pp. 377– 391. CRC Press, Washington, DC. Pieri, C., Dumanski, J., Hamblin, A., and Young, A. (1995). “Land Quality Indicators,” World Bank Discussion Paper No. 315. World Bank, Washington, DC.

96

O. H. SMITH ET AL.

Pimentel, D., Wilson, C., McCullum, C., Huang, R., Dwen, P., Flack, F., Tran, Q., Saltman, T., and Cliff, B. (1997). Economic and environmental benefits of biodiversity. BioScience 47, 747–757. Pimentel, D. C., and Kounang, N. (1998). Ecology of soil erosion in ecosystems. Ecosystems 1, 416 – 426. Rapport, D. J., Regier, H. A., and Hutchinson, T. C. (1985). Ecosystem behavior under stress. Am. Nat. 125, 617–640. Rasmussen, P. E., and Collins, H. P. (1991). Long-term impacts of tillage, fertilizer, and crop residue on soil organic matter in temperate semiarid regions. Adv. Agron. 45, 93 –134. Reeves, D. W. (1997). The role of soil organic matter in maintaining soil quality in continuous cropping systems. Soil Tillage Res. 43, 131–167. Riera, J. L., Magnuson, J. J., Vande Castle, J. R., and MacKenzie, M. D. (1998). Analysis of large-scale spatial heterogeneity in vegetation indices among North American landscapes. Ecosystems 1, 268–282. Riitters, K. H., O’Neill, R. W., Hunsaker, C. T., Wickham, J. D., Yankee, D. H., Timmons, S. P., Jones, K. B., and Jackson, B. L. (1995). A factor analysis of landscape pattern and structure metrics. Landscape Ecol. 10, 23 – 40. Roush, R. T. (1997). Bt-transgenic crops: Just another pretty insecticide or a chance for a new start in resistance management? Pesticide Sci. 51, 328 – 334. Samways, M. J., Caldwell, P. M., and Osborn, R. (1996). Ground-living invertebrate assemblages in native, planted and invasive vegetation in South Africa. Agric. Ecosystems Environ. 59, 19 – 32. Schmidt, E. L., and Belser, L. W. (1994). Autotrophic nitrifying bacteria. In “Methods of Soil Analysis, Part 2—Microbiological and Biochemical Properties” (R. W. Weaver et al., Eds.), SSSA Book Ser. No. 5, pp. 159 –178. SSSA, Madison, WI. Schumann, G. L. (1991). “Plant Diseases and Their Social Impact.” American Phytopathological Society, St. Paul, MN. Scow, K. M. (1997). Soil microbial communities and carbon flow in agroecosystems. In “Ecology in Agriculture” (L. E. Jackson, Ed.), pp. 367– 403. Academic Press, New York. Shuman, L. M., Rayner, P. L., Day, J. L., and Cordonnier, M. J. (1988). Comparison of four phosphorus extraction methods on three acid southeastern soils. Commun. Soil Sci. Plant Anal. 19, 579–595. Sposito, G. (1989). “The Chemistry of Soils.” Oxford Univ. Press, New York. Stefanidou, M., Koutselinis, A., Pappas, F., and Methenitou, G. (1996). Bee head acetylcholinesterase as an indicator of exposure to organophosphate and carbamate insecticides. Vet. Hum. Toxicol. 38(6), 420–422. Stevenson, F. J. (1986). “Cycles of Soil: Carbon, Nitrogen, Phosphorous, Sulfur, Micronutrients.” Wiley, New York. Stevenson, F. J. (1994). “Humus Chemistry: Genesis, Composition, Reactions.” Wiley, New York. Tabashnik, B. E. (1997). Seeking the root of insect resistance to transgenic plants. Proc. Natl. Acad. Sci. 94, 3488–3490. Thomann, R. V. (1994). The significance of resource scale in water quality and ecosystem modeling and decision making. In “Towards a Sustainable Coastal Watershed: The Chesapeake Experiment,” conference proceedings, June 1–3, pp. 20–27. Chesapeake Research Consortium, Edgewater, MD. Turner, M. G. (1989). Landscape ecology: The effect of pattern on process. Annu. Rev. Ecol. Systematics 20, 171–197. U.S. Department of Agriculture (USDA) (1997). Agricultural resources and environmental indicators, 1996–97, Agricultural Handbook No. 712. USDA Economic Research Service, Natural Resources and Environment Division, Washington, DC. U.S. Environmental Protection Agency (1994). Landscape monitoring and assessment research plan, EPA No. 620/R-94/009. Office of Research and Development, Washington, DC. U.S. Environmental Protection Agency (1997). EMAP research strategy [Online]. http://www.epa.gov. [1998, February 15]

ENVIRONMENTAL INDICATORS OF AGROECOSYSTEMS

97

U.S. Environmental Protection Agency–Agroecosystems Resource Group (1993). EMAP—Agroecosystems program background and conceptual framework [Online]. http://www.darwin.arrc. ncsu.edu, September 7. [1998, February 15] U.S. Geological Survey/BRD (1995). Summary of a workshop for the selection of biomonitoring methods [Online]. http://orion.mesc.nbs.gov. [1998, February 15] Van der Zee, S.E.A.T.M., Fokkink, L. G. J., and van Riemsdijk, W. H. (1987). A new technique for assessment of reversibly adsorbed phosphate. Soil Sci. Soc. Am. J. 51, 599 – 604. Vitousek, P. M., Mooney, H. A., Lubchenco, J., and Melillo, J. M. (1997). Human domination of earth’s ecosystems. Science 277, 494 – 499. Wiens, J. A., and Milne, B. T. (1989). Scaling of landscapes in landscape ecology, or, landscape ecology from a beetle’s perspective. Landscape Ecol. 3(2), 87–96. Wu, J., and Levin, S. A. (1994). A spatial patch dynamic modeling approach to pattern and process in an annual grassland. Ecol. Monogr. 64(4), 447– 464. Zak, J. C., Willig, M. R., Moorhead, D. L., and Wildman, H. G. (1994). Functional diversity of microbial communities: A quantitative approach. Soil Biol. Biochem. 26, 1101–1108.

This Page Intentionally Left Blank

GROWTH PROMOTION OF PLANTS INOCULATED WITH PHOSPHATE-SOLUBILIZING FUNGI M. A. Whitelaw School of Wine and Food Sciences Charles Sturt University Wagga Wagga, NSW 2678, Australia

I. Introduction II. Soil Phosphorus A. Phosphate Sorption B. Precipitation of Phosphate Compounds C. The Soil Phosphorus Cycle D. The Relationship of Soil pH to Phosphate Solubility E. Plant Uptake of Sparingly Soluble Phosphate III. Phosphate-Solubilizing Soil Microorganisms A. Solubilization of Phosphates by Free-Living Fungi B. Precipitated Phosphate Agar Studies IV. Liquid Medium Studies A. Solubility of Phosphate Compounds B. Production of Organic Acids in Liquid Medium Studies C. Chelation of Cations by Organic Acids D. Titratable Acidity E. pH F. N Source G. Fluctuations in Soluble Phosphate Levels over Time V. Plant Growth Promotion by Phosphate-Solubilizing Fungi VI. Conclusion References

Phosphorus is an important plant nutrient which is in short supply in many agricultural soils. Because much of the soluble phosphate (P) applied to soils as fertilizer is “fixed” by the soil and rendered less available to plants, the long-term application of P fertilizers has resulted in an accumulation of total soil P, most of which is poorly soluble. Many soil fungi, predominantly of the genera Aspergillus and Penicillium, have been shown to possess the ability to solubilize sparingly soluble phosphates in vitro by secreting inorganic or organic acids. Growth promotion and

99 Advances in Agronomy, Volume 69 Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved. 0065-2113/00 $30.

100

M. A. WHITELAW increased uptake of P by plants inoculated with P-solubilizing fungi have also been reported by many investigators. © 2000 Academic Press.

I. INTRODUCTION Many soils, such as those found in much of Australia, are low in phosphate (P) readily available for plant growth (i.e., “available P”). Historically, deficiency of soil P has been one of the most important chemical factors restricting plant growth in the acid soils of the southern wheat belt of New South Wales, Australia (Colwell, 1963). Because low P status tends to limit the production of arable crops, the application of soluble P fertilizers such as superphosphate has been widely used, especially for cereal crops. However, much of the soluble P applied as fertilizer may react with the soil and be “fixed” or converted into one of the many sparingly soluble forms which are less available for uptake by plants than rapidly exchangeable forms of P (Stevenson, 1986). It has long been known that soil microorganisms play an important role in the cycling of many soil nutrients including phosphate. Bacteria, yeasts, actinomycetes, mycorrhizal fungi, and free-living fungi have been reported to cause increases in plant-available P in the soil. Many in vitro studies have demonstrated the presence of soil microorganisms capable of transforming soil P to forms available to plants. Many investigators have also reported increases in yield and P uptake due to inoculation of plants by P-solubilizing fungi. The use of vesicular mycorrhizal (VAM) fungi to inoculate plants is limited by the inability of scientific researchers to grow the fungi in pure culture and produce large amounts of inoculum (Kucey et al., 1989). This review attempts to summarize information from microbial P solubilization in vitro studies and from glasshouse and field studies involving free-living P-solubilizing fungi.

II. SOIL PHOSPHORUS Much of the soluble P applied to soil as fertilizer may react with the soil and be converted to one of many sparingly soluble forms of P. The major P fixation reactions are shown in Fig. 1. Both biological and chemical processes are involved in P fixation but the chemical processes of P sorption and precipitation are more important in the retention of fertilizer P (Stevenson, 1986). McLaughlin et al. (1988) reported that wheat plants grown on a solonized brown soil (pH 8.3) used only 12% of the fertilizer P applied with the crop. Of the remainder, 71% was either precipitated by aluminium, ferric, or calcium compounds or adsorbed by hydrous oxides of aluminium and ferric or by aluminosilicate (clay) materials. Four percent was immobilized by soil microbes and 13% was incorporated into soil organic matter.

PHOSPHATE-SOLUBILIZING FUNGI

101

Figure 1 Phosphate fixation reactions in soil. [Adapted from Sauchelli (1951) in “Cycles of Soil, Carbon, Nitrogen, Phosphorus, Sulfur, Micronutrients,” F. J. Stevenson. Copyright 1986 John Wiley & Sons. Reprinted by permission of John Wiley & Sons.]

A. PHOSPHATE SORPTION “P sorption” is a term used to describe two types of chemical reactions which remove P from the soil solution: adsorption and absorption. Adsorption is the initial rapid process in which a layer of H 2PO 4 forms on the surface of solid aluminium or ferric hydrous oxides, whereas absorption is the slow diffusive penetration into the bulk of the adsorbing solid which occurs after adsorption (Barrow, 1987). Because of similarities in sorption isotherms, several researchers have concluded that the sorption mechanisms for aluminosilicates (clays) are similar to those of the hydrous oxides (Sample et al., 1980). The adsorption of anions by the edge faces of the clay, kaolinite, is believed to be similar to the adsorption by hydrous oxides (Hingston et al., 1972). Because montmorillonite and kaolinite clays retain similar amounts of P, their P sorption mechanisms are also considered to be similar (Wild, 1950). Olsen and Watanabe (1957) found that acidic soils fixed twice the amount of added P per unit surface area than neutral or calcareous soils. They also reported that the P fixed was held with five times more bonding energy in acidic soils compared to calcareous soils. Much of the P retained by acidic soils is specifically

102

M. A. WHITELAW

adsorbed by iron and aluminium hydrous oxides and aluminosilicate minerals. Leaver and Russell (1957) demonstrated the importance of hydrated iron and aluminium oxides in P retention in soils. They found that treating soils with reagents which formed insoluble or un-ionized compounds with Fe3+ and Al3+ reduced P retention. Oxides and hydrous oxides of iron and aluminium exist as discrete compounds in soil, coatings on other soil particles, and amorphous aluminium-hydroxy compounds between layers of expandable aluminium silicates (Sample et al., 1980). Soil is a dynamic system with surface-reactive amorphous aluminium and iron hydrous oxides being continually added to the system through the weathering process (Hsu, 1965). The surfaces of these oxides and hydrous oxides have a high affinity for P and the adsorption reactions onto these are rapid (Van der Zee et al., 1987). The P adsorption capacity of neutral or high pH calcareous soils is influenced mainly by the amount of exchangeable calcium ions and by the CaCO3 content (Kamprath and Watson, 1980). However, when Holford and Mattingly (1975) studied P adsorption by 24 calcareous soils, they found strongly bonded adsorption surfaces to be closely related to the content of iron hydrous oxides, indicating that even in calcareous soils hydrous oxides are important in P adsorption.

B. PRECIPITATION OF PHOSPHATE COMPOUNDS Because of the high surface area of most soils and the rapidity of adsorption reactions, it has been assumed that P fixation in soils is controlled mainly by adsorption. However, precipitation reactions could be an important mechanism for P fixation in some soils and Holford (1983) suggested that precipitation by aluminium was a major cause of P fixation in several acidic soils (pH  4.5). Precipitation was found to be greatest in a sandy soil which had the lowest quantity of major P adsorbents such as hydrous oxides. In nonsandy soils, precipitation was a relatively minor but still significant P-fixation mechanism. The type of precipitate formed from dissolved P depends largely on the soil pH. In acidic soils, P can be precipitated by the Fe3+ and Al3+ released when the aluminium and iron hydrous oxides are decomposed by acidic, highly concentrated P solutions, whereas in alkaline soils dissolved P can be precipitated by Ca2+. The initial products of the reaction of superphosphate in alkaline soils are mainly calcium monohydrogen phosphate dihydrate (CaHPO4#2H2O) and calcium monohydrogen phosphate (CaHPO4), whereas the initial products in acidic soils are amorphous ferric phosphate and amorphous aluminium phosphate. In acidic soils, aluminium phosphates are precipitated in preference to iron phosphates unless the soils contain a large quantity of amorphous ferric hydroxide (Russell, 1980). Initially, the precipitates are amorphous and P is moderately available to plants (Rajan, 1976). Hsu (1982) observed that the reaction products may remain in amor-

PHOSPHATE-SOLUBILIZING FUNGI

103

phous forms for at least 66 months. The precipitates are then thought to be slowly converted to less soluble forms such as calcium orthophosphate [Ca3(PO4)2], hydroxyapatite [(Ca5(PO4)3(OH)], and fluoroapatite [Ca10(PO4)6F2] in alkaline soils or crystalline ferric phosphate [FePO4#2H2O(cr) or strengite] and crystalline aluminium phosphate [AlPO4#2H2O(cr) or variscite] in acidic soils (Lindsay et al., 1962; Rajan, 1976; Sample et al., 1980).

C. THE SOIL PHOSPHORUS CYCLE The soil P cycle is a dynamic one involving soil, plants, and microorganisms. In natural ecosystems there is virtually a closed cycle, with the P consumed by plants being returned to the soil in plant or animal residues which are broken down by soil microorganisms. However, soil under conventional cultivation contains lower levels of organic matter and P is regularly removed in the harvest (Stevenson, 1986). In order to produce high crop yields, farmers must apply larger quantities of P fertilizer than is taken up by their crops. In many cases they may be advised to apply up to four times the P requirements of the crop to overcome the problem of the conversion of P to unavailable forms (Goldstein, 1986). This situation inevitably leads to an accumulation of sparingly soluble inorganic P in the soil, with the soil acting as a “sink” for P (Cosgrove, 1977). P concentrations in soil solutions are small, ranging from 0.01 to 0.3 mg P liter1 (Ozanne, 1980). Although the soil solution P is only a small proportion of the total soil P content, this is where plants derive most of their immediate requirements (Bolan, 1991). In the absence of significant levels of organic P, the solution P is in equilibrium with a quantity of relatively labile inorganic P such that in any one soil the ratio of labile inorganic P to solution P is constant (Stewart and Sharpley, 1987). The soil solution is only adequate if the labile inorganic P is solubilized at least as quickly as the roots can extract if from the soil solution (Russell, 1980). When plants take up P from the soil solution, the depleted solution P pool is immediately replenished from labile and moderately labile inorganic P forms. If these pools are depleted, nonlabile P sources determine the soluble P concentration in soil (Stewart and Sharpley, 1987). The solution inorganic P equilibrium was described by Larsen (1967). When the soil solution is depleted, there is rapid movement from labile soil P to soil solution P. However, when labile soil P is depleted, the movement from nonlabile soil P to labile soil P is slow. Soil solution P

B

labile soil P (solid phase)

B

nonlabile soil P (solid phase)

The overall supply of P to plants is represented in Fig. 2 (Bolan, 1991). Labile soil P, consisting mainly of P adsorbed by iron and aluminium oxides and alumi-

104

M. A. WHITELAW

Figure 2 Schematic representation of the supply of phosphate to plant roots in soil systems. (1) Adsorption–desorption, (2) solid state diffusion, (3) precipitation–dissolution, (4) immobilization– mineralization, (5) diffusion in solution, (6) movement into roots. [Reproduced from Plant Soil 134 (1991), 189–207. A critical review on the role of mycorrhizal fungi in the uptake of phosphorus by plants. N. S. Bolan. © 1991, with kind permission from Kluwer Academic Publishers.]

nosilicate (clay) minerals, is able to replenish depleted P in the soil solution in a short period of time. Nonlabile inorganic soil P (i.e., P from ferric, aluminium, and calcium–phosphate precipitates and organic P and “occluded” P) is slower to replenish soil solution P (Murrmann and Peech, 1969). Soil is deficient in P when the rate of supply to plant roots is less than the rate required for optimum growth. One can increase the rate of P supply by supplying a low level of freshly applied fertilizer but with time the rate decreases again and more P fertilizer is needed (Costin and Williams, 1983).

D. THE RELATIONSHIP OF SOIL PH TO PHOSPHATE SOLUBILITY Controversy exists regarding the effect of pH change on P solubility in soil. Depending on the soil and experimental conditions, sorption has been found to be decreased, increased, or not affected by increasing soil pH. This topic has been reviewed by White (1983), who points out that most studies of acidic soils in which aluminosilicates (i.e., clays) are the dominant P adsorbents show that P solubility either decreases or exhibits no significant change with increasing pH. The chemical basis for this common observation is thought to be the hydrolysis of exchangeable aluminium at high pH and the subsequent reaction with soluble P in the soil solution to form an insoluble aluminium–phosphate compound (Haynes, 1982). Changes in P desorption with changes in soil pH have also been attributed to other causes. Soil pH affects the charge on the P species in solution and on the sur-

PHOSPHATE-SOLUBILIZING FUNGI

105

faces of the adsorbing particles. As the soil pH rises the P adsorbing surfaces become increasingly negatively charged. This makes the reaction with negatively charged phosphate ions more difficult and therefore tends to decrease P adsorption as pH rises. However, raising the pH also changes the proportions of P species in solution. In the normal range of soil pH, P is present in solution mainly as 2 2 H2PO 4 and HPO4 but the affiinity of reactive soil surfaces for HPO4 is much  2 greater than that for H2PO4 . As the concentration of HPO4 increases with pH, this effect tends to offset the increase in the negative charge of the adsorbing particles and therefore increases P adsorption as pH rises (Barrow, 1990). Whether a decrease in soil pH will increase or decrease P solubilization sometimes depends on which of the effects is dominant. The effects of pH on P sorption also differ depending on the amount of P which had previously been added to the soil. Barrow (1984) found that increases in pH up to 5.5 decreased sorption in unfertilized soils but increased sorption in fertilized soils. However, in both fertilized and unfertilized soils, soil P desorption was largest at low pH. Barrow offered two explanations for the increased P desorption at low pH. First, acidity may have decomposed some of the P-retaining material in soils. Second, desorption involved diffusion of P from the interior of the particles back to the surface and this diffusion probably becomes more rapid as the pH decreases and the surface charge becomes less negative. Decreases in soil pH can also enhance the release of P from precipitated calcium phosphates in neutral to slightly acidic soils (White, 1983).

E. PLANT UPTAKE OF SPARINGLY SOLUBLE PHOSPHATE There are three main mechanisms by which plants may be able to extract P from normally insoluble sources in the soil. First, acid production (of plant or microbial origin) may cause a lowering of rhizosphere pH. Second, organic acids capable of chelating metal ions may compete with P for adsorption sites in the soil. Third, organic acids may form soluble complexes with metal ions associated with insoluble P (Ca2+, Al3+, and Fe3+) thus releasing P (Kepert et al., 1979). Hedley et al. (1982) demonstrated that the rhizosphere pH of rape (a plant which is known to be efficient in absorbing P from P-deficient soils) grown in thin layers of 32P-labeled soil decreased from 6.1 to 4.1 in 28 days, leading to the dissolution of acid-soluble forms of inorganic P which were not exchangeable with 32P. After 41 days, an extra 3.2 mmol P kg1 soil was released due to the decrease in soil pH. The acidification was reported to be mainly due to H+ release from the roots during periods when cation uptake by the plant exceeded that of anions and not to organic acid production. Trolldenier (1992) also provided evidence for the solubilization of calcium phosphates due to acidification of the rhizosphere by a variety of plant species grown in agar. Hinsinger and Gilkes (1995) reported rootinduced dissolution of rock phosphate by two species of lupin and concluded that

106

M. A. WHITELAW

dissolution was probably due to proton excretion, as evidenced by a decrease in the rhizosphere pH of about two pH units. Plants which have the dense proliferation of rootlets arising from their lateral roots known as “proteoid roots” are notable for their ability to grow on soil low in available P (Grierson and Attiwill, 1989). Grierson and Attiwill reported that Banksia proteoid roots were able to solubilize amorphous ferric phosphate. Grierson (1992) later demonstrated that large amounts of organic acids, of which 50% was citrate, 18% malate, and 17% aconitic, were excreted by these proteoid roots. Gardner et al. (1982) reported that both ferric and aluminium phosphates were dissolved in agar films around the proteoid roots of lupin plants and suggested that chelation of the Fe3+ and Al3+ by a substance excreted by the roots was responsible. Gardner et al. (1983) later demonstrated that large amounts of H+ and organic acids including citric acid were excreted by these proteoid roots. They suggested that the excreted citrate reacts with ferric phosphate in the soil, allowing the eventual release of Fe3+ from ferric phosphate and enabling the absorption of P by the roots. Armstrong et al. (1993) found that the availability to maize of various sparingly soluble P sources compared to KH2PO4 (100%) was 53% for amorphous aluminium phosphate, 39% for amorphous ferric phosphate, 3% for AlPO4(cr), and 2% for FePO4(cr). These results demonstrate the uptake of sparingly soluble sources of mineral P by some plants, although the mechanisms for solubilization is not known. Sainz and Arines (1988), in a sequential P extraction study of three acidic soils, reported that applied P was mainly retained in the inorganic P fractions extracted with NH4F (i.e., P bound to Al3+) and NaOH [i.e., P bound to Fe(III)] and that the P bound to Al3+ was the fraction mainly utilized by red clover plants which were infected by native VAM fungi.

III. PHOSPHATE-SOLUBILIZING SOIL MICROORGANISMS The existence of soil microbes capable of transforming soil P to forms available to the plant has been recorded by many investigators. Many bacterial, fungal, yeast, and actinomycete species capable of solubilizing sparingly soluble P in pure culture have been isolated from soil and rhizosphere samples (Kucey et al., 1989). Gerretsen (1948) provided evidence that microorganisms in the rhizosphere can dissolve sparingly soluble inorganic P. It has since been observed by many investigators that a high proportion of P-solubilizing microorganisms are concentrated in the rhizosphere of plants (Sperber, 1957; Raghu and MacRae, 1966; Khan and Bhatnagar, 1977; Whipps and Lynch, 1986; Martinez Cruz et al., 1990). Although Katznelson and Bose (1959) did not find selective stimulation of P-solubilizing

PHOSPHATE-SOLUBILIZING FUNGI

107

microorganisms in the rhizosphere, they did find that rhizosphere bacteria had greater metabolic activity and suggested that they might contribute significantly to the phosphate economy of the plant. Microorganisms in the rhizosphere obtain their nutrition from root exudates (low-molecular-weight compounds such as sugars, organic acids, and amino acids which leak from root cells), plant mucigel (an actively secreted substance), and root lysates (compounds released by autolysis of root cortical cells, older epidermal cells, and root hairs) (Martin, 1977; Rovira et al., 1979). Whipps (1984) showed that in wheat plants up to 33–40% of the total carbon fixed as photosynthate could be excreted into the rhizosphere. Rhizosphere microorganisms are normal heterotrophs but they live in an environment with high levels of nutrients, such as carbon and nitrogen, and tend to adapt rapidly to improvements of nutrient supply (Tinker, 1984).

A. SOLUBILIZATION OF PHOSPHATES BY FREE-LIVING FUNGI Sperber (1958a) reported that many P-solubilizing bacteria lost the ability to solubilize P on serial subculturing. However, Kucey (1983) found that fungi, in contrast to bacteria, retained their P-solubilizing activity even after serial subculturing and could be kept actively solubilizing P for many years. P-solubilizing fungi also showed greater P-solubilizing activity both on precipitated phosphate agar and in liquid media than did bacteria in studies by Sperber (1958b), Kucey (1983), Venkateswarlu et al. (1984), Darmwal et al. (1989), Chabot et al. (1993), and Nahas (1996). Scanning electron microscopy by Chabot et al. (1993) showed that in liquid culture the hyphae of fungi were attached to P mineral particles, whereas bacteria were not. Fungi in soil are able to traverse distances more easily than bacteria and thus may be more important to P solubilization in soils (Kucey, 1983). Psolubilizing fungi in the rhizosphere or bulk soil were often found to be predominantly Penicillium and Aspergillus spp. (Katznelson et al., 1962; Ramos et al., 1968; Bardiya and Gaur, 1974; Khan and Bhatnagar, 1977; Banik and Dey, 1981a; Kucey, 1983; Molla et al., 1984; Thomas et al., 1985; Darmwal et al., 1989; Martinez Cruz et al., 1990; Narsian et al., 1994).

B. PRECIPITATED PHOSPHATE AGAR STUDIES Solubilization of various forms of precipitated calcium phosphate in unbuffered solid agar medium plates has been used widely as the initial criterion for the isolation of P-solubilizing microorganisms (Sackett et al., 1908; Pikovskaia, 1948; Gerretsen, 1948; Sperber, 1957; Katznelson and Bose, 1959). Microorganisms grown on precipitated clacium phosphate agar produce clear zones around their

108

M. A. WHITELAW

colonies if they are capable of solubilizing calcium phosphate minerals. Other forms of sparingly soluble P, such as ferric or aluminium phosphate (Rose, 1957; Sperber, 1957; Louw and Webley, 1959; Barthakur, 1978; Banik and Dey, 1983; Martinez Cruz et al., 1990; Toro et al., 1996), have also been included in precipitated phosphate agar for the isolation of P-solubilizing microorganisms. Rock phosphate suspended in unbuffered solid agar plates has also been used for the isolation of microorganisms capable of solubilizing rock phosphate (Ahmad and Jha, 1968; Bardiya and Gaur, 1974; Khan and Bhatnagar, 1977; Mba, 1996; Singh et al., 1984; Toro et al., 1996). The precipitated phosphate agar techniques are useful for isolating and selecting microorganisms for further investigations but have limited sensitivity. Factors such as the rate of diffusion of excreted organic acids and the colony growth rate affect the size of the clear zone. Cunningham and Kuiack (1992) reported that both the nutrient composition (e.g., carbon source and nitrogen source) and the buffering capacity of the medium influenced the diameter of the zone of calcium phosphate solubilization. The day on which the clear zone is measured was also important: Some bacterial isolates were reported to show a clear zone early in growth but later failed to keep the clear zone beyond their colony edge (Kucey, 1983). However, Kucey (1983) found that fungal isolates which expressed P-solubilizing ability early in colony growth continued to maintain a clear zone even when the colonies were growing rapidly and that there was a significant correlation (r 0.70) between the ability of 23 fungal isolates to produce a clear zone on precipitated phosphate agar and their ability to solubilize rock phosphate in a liquid medium. Some investigators reported low correlation between the size of the clear zone in precipitated phosphate agar studies and the more quantitative data from P solubilization in liquid media. For example, some isolates with little or no clear zone on solid agar exhibited high efficiency for dissolving insoluble phosphates in a liquid medium showing that the plate technique was insufficient to detect all Psolubilizing microorganisms (Rose, 1957; Louw and Webley, 1959; Das, 1963; Ahmad and Jha, 1968; Gupta et al., 1994). Conversely, some isolates showed large clear zones on agar but low P solubilization in a liquid medium (Ahmad and Jha, 1968). Gupta et al. (1994) modified the precipitated phosphate agar technique by including bromophenol blue, an acid indicator. Cunningham and Kuiack (1992) made a similar modification to the precipitated phosphate agar technique by including “alizarin red S” as the acid indicator. Gupta et al. (1994) recorded that many fungi which were unable to produce a clear zone on calcium phosphate agar produced yellow acidic zones on agar which contained bromophenol blue. These fungi were able to solubilize Ca3(PO4)2 in a liquid medium with acid production and the authors found a significant correlation between the yellow acidic zone diameter on agar and the amount of P solubilized in the liquid medium (r 0.69).

PHOSPHATE-SOLUBILIZING FUNGI

109

IV. LIQUID MEDIUM STUDIES Many researchers have quantitatively investigated the ability of soil fungi to solubilize sparingly soluble inorganic phosphates in unbuffered pure liquid medium cultures (Table I). A wide range of media have been used to allow the study of nutritional effects, and levels of solubilized phosphate and metabolites such as organic acids have also been measured. Fungal strains differed widely in their abilities to solubilize P. P solubilization was often higher when the initial insoluble P levels were high, as long as the volume of liquid medium was adequate for dissolution. The amounts of insoluble phosphate P initially included in the medium varied from 87 mg P liter1 (Illmer and Schinner, 1992) to 4460 mg P liter1 (Cunningham and Kuiack, 1992). Low levels of P solubilization by Aspergillus candidus and Aspergillus fumigatus were reported by Banik and Dey (1982) despite high initial insoluble P levels. In this case, the dissolution of P may have been limited by the low volume (15 ml) of liquid medium used (Table I). The initial pH of the liquid medium probably affected the amount of P solubilized in many liquid medium studies. Ahmad and Jha (1968) found that P solubilization was highest when the initial pH was in the optimum pH range for the growth of the isolates (e.g., a Penicillium isolate was most active when the initial pH was 4.0). P solubilization was also affected by whether or not the medium was shaken. Oxygen absorption rates in cultures are markedly increased by shaking (Corman et al., 1957). Cunningham and Kuiack (1992) reported that the production of citric acid by the P-solubilizing fungus Penicillium bilaii was promoted by shaking, and Ahmad and Jha (1968) found that P solubilization by a Penicillium isolate was greater when the culture was shaken (Table I). The presence of alkaline substances in liquid media was detrimental to P solubilization in some cases. Sperber (1958b) found that the presence of calcium carbonate greatly reduced solubilization of apatite by lactic acid, and Ahmad and Jha (1968) found that the presence of calcium carbonate greatly reduced microbial solubilization of rock phosphate. Banik and Dey (1982) included calcium carbonate in their culture medium for the microbial solubilization of aluminium and ferric phosphates in an attempt to avoid the partial hydrolysis of AlPO4 and FePO4 during autoclaving. Because the calcium carbonate would have neutralized any acid produced, the solubilization of the aluminium and ferric phosphates was negligible (Table I).

A. SOLUBILITY OF PHOSPHATE COMPOUNDS Another factor affecting the amount of P solubilized in liquid medium studies is the solubility of the P mineral. The solubility of sparingly soluble P compounds

Table I Phosphate Solubilization by Fungi

Microorganism

Phosphate solubilized (mg P liter1 culture)a

Organic acids produced

Phosphate source

Titratable acidity (mmol H liter1) and change in pH

1000e (14 days)

2000 mg P liter1 CaHPO4b

Oxalate, citrate, aspartate (PC)

ND

1000e (14 days)

Oxalate, citrate (PC)

ND

Oxalate (PC)

ND

Oxalate and H2S production (PC) Oxalate (PC)

ND

Oxalate (PC)

ND

Glyoxylate, citrate (PC)

ND

460e (max. at 3 days)

2000 mg P liter1 Ca3(PO4)2c (volume unknown 2000 mg P liter1 FePO4d (volume unknown) 2000 mg P liter1 FePO4d 2000 mg P liter1 CaHPO4b (volume unknown) 2000 mg P liter1 Ca3(PO4)2c 2000 mg P liter1 FePO4d (volume unknown) 460 mg P liter1 CaHPO4b in 50 ml

TA 55

340e (max. at 12 days)

460 mg P liter1 CaHPO4b in 50 ml

Citrate, glycollate, succinate, gluconate (PC) Lactate, glycollate (PC)

599 (time unknown)

1000 mg P liter1 Ca3(PO4)2 in 50 ml

ND

ND

Cephalosporium sp.

55

ND

ND

Alternaria sp.

62

1000 mg P liter1 Ca3(PO4)2 in 50 ml 1000 mg P liter1 Ca3(PO4)2 in 50 ml

ND

ND

Aspergillus niger (isolated from New Zealand, pH unknown)

100e (14 days)

110

Aspergillus terreus (isolated from soil) Sclerotium rolfsii (isolated from soil)

1000e (14 days) 1000e (14 days)

1000e (14 days) 400e (14 days)

A. niger (isolated from soil or rhizoshere, Australia Penicillium sp. (isolated from soil or rhizosphere) Penicillium sp. (from surface of legume root nodule, India)

N source and sugar concentration in medium (% w/v)

Reference

0.2% asparagine, 1% glucose (temperature unknown, static)

Rose (1957)

0.05% yeast extract, 1% glucose (25°C, shaken)

Sperber (1958b)

0.01% N (from NH4), 0.05% yeast extract, 1% glucose (temperature unknown)

Subba-Rao and Bajpai (1965)

ND

TA 25

111

Penicillium lilacinum (isolated from legume root nodules, India) A. niger (isolated from legume root nodules, India) Aspergillus sp. (isolated from legume root nodules, India) Aspergillus flavus (isolated from legume root nodules, India) A. terreus (isolated from legume root nodules, India) Penicillium sp. (Siwan) (isolated from soil, Bihar, India) Penicillium sp. (isolated from soil, Delhi and Ludhiana, India) Aspergillus sp. (isolated from soil, Delhi and Ludhiana, India) Fusarium sp. (isolated from soil, Delhi and Ludhiana, India) A. niger

78 (21 days)

200 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 6.7 to 4.7

50 (21 days)

200 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 6.7 to 1.8

35 (21 days)

200 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 6.7 to 5.5

35 (21 days)

200 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 6.7 to 6.8

18 (21 days)

200 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 6.7 to 8.6

260e (14 days)

1000 mg P liter1 (volume unknown)

ND

ND

103 (21 days)

200 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 6.6 to 6.4

99 (21 days)

200 mg P liter1 Ca3(PO4)2c in 50 ml

ND

pH 6.6 to 4.0

85 (21 days)

200 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 6.6 to 5.1

175 (20 days)

1000 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 7.0 to 4.1

Aspergillus ustus

157 (20 days)

ND

pH 7.0 to 6.0

Fusarium solani

159 (20 days)

ND

pH 7.0 to 4.7

Mortierella nana

156 (20 days)

ND

pH 7.0 to 5.1

S. rolfsii

139 (20 days)

ND

pH 7.0 to 3.2

A. niger

162 (20 days)

1000 mg P liter1 Ca3(PO4)2 in 50 ml 1000 mg P liter1 Ca3(PO4)2 in 50 ml 1000 mg P liter1 Ca3(PO4)2 in 50 ml 1000 mg P liter1 Ca3(PO4)2 in 50 ml 200 mg P liter1 fluoroapatite in 50 ml

ND

pH 7.0 to 5.6

0.1% N (from NO3 ), 5% sucrose (28°C, static)

Chhonkar and Subba-Rao (1967)

0.5% peptone, 1% sucrose (room temperature, shaken) 0.1% N (from NO3 ), 5% sucrose (28°C, static)

Ahmad and Jha (1968)

0.1% N (from NO3 ), 3% glucose (25°C static)

Sethi and Subba-Rao (1968)

Agnihotri (1970) (all pure culture of species commonly found in forest beds, Canada)

continues

Table I—Continued

Microorganism

Phosphate solubilized (mg P liter1 culture)a

Organic acids produced

Phosphate source 1

Titratable acidity (mmol H liter1) and change in pH

112

200 mg P liter fluoroapatite in 50 ml 200 mg P liter1 fluoroapatite in 50 ml 200 mg P liter1 fluoroapatite in 50 ml 200 mg P liter1 fluoroapatite in 50 ml 200 mg P liter1 hydroxyapatite in 50 ml 200 mg P liter1 hydroxyapatite in 50 ml 874 mg P liter1 Ca3(PO4)2 in 100 ml

ND

pH 7.0 to 5.9

ND

pH 7.0 to 5.9

ND

pH 7.0 to 4.0

ND

pH 7.0 to 6.2

ND

pH 7.0 to 4.6

ND

pH 7.0 to 3.1

ND

pH 7.0 to 4.4

874 mg P liter1 Ca3(PO4)2 in 100 ml 874 mg P liter1 Ca3(PO4)2 in 100 ml 874 mg P liter1 Ca3(PO4)2 in 100 ml 874 mg P liter1 Ca3(PO4)2 in 100 ml

ND

pH 7.0 to 4.7

ND

pH 6.5 to 2.7

ND

pH 6.5 to 6.1

ND

pH 6.5 to 6.4

57e (7 days)

874 mg P liter1 Ca3(PO4)2 in 100 ml

ND

pH 6.5 to 7.1

116e (21 days)

874 mg P liter1 Mussoorie RP in 100 ml

ND

pH 6.5 to 3.4

Cylindrocladium sp.

156 (20 days)

Penicillium sp.

131 (20 days)

S. rolfsii

155 (20 days)

Verticillium alboatrum

132 (20 days)

A. niger

119 (20 days)

S. rolfsii

46 (20 days)

Aspergillus awamori

750e (7 days)

Penicillium digitatum

579e (7 days)

Aspergillus carbonarum (isolated from RP, India) A. flavus (isolated from RP, India) Aspergillus wentii (isolated from Delhi alluvial soil, India) Aspergillus fumigatus f (isolated from Delhi alluvial soil, India) A. awamori (isolated from rhizosphere of soybean)

143e

(7 days)

89e

(7 days)

82e

(7 days)

N source and sugar concentration in medium (% w/v)

Reference

0.01% NH4, 0.05% yeast extract, 1% glucose (temperature unknown)

Gaur (1972)

0.01% N (from NH4) 0.05% yeast extract, 1% glucose (30°C, static)

Gaur et al. (1973)

0.01% N (from NH4), 0.05% yeast extract, 1% glucose (30°C static)

Bardiya and Gaur (1974)

P. digitatum (isolated from rhizosphere of soybean) Curvularia sp. (1F) (isolated from rhizosphere of cowpea) Trichoderma sp. (26F) (isolated from rhizosphere of black gram) A. niger (isolated from rhizosphere soil, Sindri, India. Soil available P: 6 ppm, pH 6.4)

77e (21 days)

48e (21 days)

45e (21 days)

298e (7 days)

133e (7 days)

113

A. niger van Tiegh (isolated from rice rhizosphere, Assam, India)

46 (15 days)

67 (15 days) A. niger

35 (15 days) 65 (15 days)

Trichoderma hamatum

33 (15 days) 67 (15 days)

Mucor flavus

28 (15 days) 51 (15 days)

F. solani

42 (15 days)

Fusarium oxysporum

20 (15 days)

F. oxysporum

39 (15 days)

Penicillium simplicissimum

37 (15 days)

874 mg P liter1 Mussoorie RP in 100 ml 874 mg P liter1 Mussoorie RP in 100 ml 874 mg P liter1 Mussoorie RP in 100 ml 1000 mg P liter1 Ca3(PO4)2 in 200 ml

ND

pH 6.5 to 4.0

ND

pH 6.5 to 5.3

ND

pH 6.5 to 5.4

ND

pH 7.0 to 3.7

669 mg P liter1 Jhabua RP in 20 ml 100 mg P liter1 Fe3(PO4)2 in 50 ml

ND

pH 7.0 to 3.8

ND

pH 4.8 to 3.2

100 mg P liter1 FePO4 in 50 ml 100 mg P liter1 A1PO4 in 50 ml 100 mg P liter1 Ca3(PO4)2 in 50 ml 100 mg P liter1 A1PO4 in 50 ml 100 mg P liter1 Ca3(PO4)2 in 50 ml 100 mg P liter1 A1PO4 in 50 ml 100 mg P liter1 Ca3(PO4)2 in 50 ml 100 mg P liter1 Ca3(PO4)2 in 50 ml 100 mg P liter1 FePO4 in 50 ml 100 mg P liter1 Ca3(PO4)2 in 50 ml 100 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 5.2 to 2.8

ND

pH 6.0 to 3.7

ND

pH 6.6 to 4.9

ND

pH 6.0 to 3.6

ND

pH 6.6 to 4.7

ND

pH 6.0 to 3.3

ND

pH 6.6 to 4.7

ND

pH 6.6 to 4.8

ND

pH 5.2 to 3.0

ND

ph 6.6 to 5.6

ND

ph 6.6 to 3.4

0.2% asparagine, 1% glucose (28°C, static)

Khan and Bhatnagar (1977)

0.01% N (from NH4), 1% glucose (33°C, shaken once every 2 days)

Barthakur (1978)

continues

Table I—Continued

Microorganism A. terreus

Phosphate solubilized (mg P liter1 culture)a 16 (15 days) 39 (15 days) 15 (15 days)

A. awamori

642 (21 days)

625 (21 days)

92 (21 days)

114

A. niger

534 (21 days)

533 (21 days)

95 (21 days)

P. digitatum

399 (21 days)

345 (21 days)

68 (21 days)

A. niger (Spain)

150 (max. at 7 days)

Organic acids produced

Phosphate source

Titratable acidity (mmol H liter1) and change in pH

100 mg P liter1 AlPO4 in 50 ml 100 mg P liter1 Ca3(PO4)2 in 50 ml 100 mg P liter1 FePO4 in 50 ml 874 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 6.0 to 3.6

ND

pH 6.6 to 5.3

ND

pH 5.2 to 3.0

ND

874 mg P liter1 hydroxyapatite in 50 ml 874 mg P liter1 Mussoorie RP in 50 ml 874 mg P liter1 Ca3(PO4)2 in 50 ml

ND

874 mg P liter1 hydroxyapatite in 50 ml 874 mg P liter1 Mussoorie RP in 50 ml 874 mg P liter1 Ca3(PO4)2 in 50 ml

ND

874 mg P liter1 hydroxyapatite in 50 ml 874 mg P liter1 Mussoorie RP in 50 ml Ca3(PO4)2 (amount unknown) in 250 ml

ND

pH 7.0 to between 2.9 and 3.6 pH 7.0 to between 2.9 and 3.6 pH 7.0 to between 2.9 and 3.6 pH 7.0 to between 3.3 and 3.8 pH 7.0 to between 3.3 and 3.8 pH 7.0 to between 3.3 and 3.8 pH 7.0 to between 3.3 and 3.8 pH 7.0 to between 3.3 and 3.8 pH 7.0 to between 3.3 and 3.8 pH from 4.5 to 3.2

ND

ND

ND

ND

ND

ND

N source and sugar concentration in medium (% w/v)

Reference

0.01% N (from NH4), 1% glucose (33°C, shaken once every 2 days) 0.01% N (from NH4), 0.05% yeast extract, 1% glucose (30°C, static)

Arora and Gaur (1979)

0.05% N (from NO3 ), 0.5% yeast extract, 3% sucrose

Ortuno et al. (1979)

A. niger van Tieghem (isolated from soil, West Bengal, India. Soil available P: 7 ppm, pH 5.4) Aspergillus candidus (isolated from alluvial soil, West Bengal, India. Soil available P: 7 ppm, pH 7.4)

4 (10 days)

Ca3(PO4)2 (amount unknown) in 15 ml

Unidentified organic acid (PC)

ND

0.01% N (from NH4), 0.05% yeast extract, 1% sucrose (37°C, static)

Banik and Dey (1981a)

20 (10 days)

1000 mg P liter1 Ca3(PO4)2 in 15 ml

Oxalate (1.2 mM ), tartate (0.7 mM ) (PC)

ND

0.01% N (from NH4), 0.05% yeast extract, 1% sucrose (37°C, static)

Banik and Dey (1982)

2 (10 days)

1000 mg P liter1 AlPO4 in 15 ml 1000 mg P liter1 FePO4 in 15 ml 1000 mg P liter1 Mussoorie RP in 15 ml 1000 mg P liter1 Ca3(PO4)2 in 15 ml

ND

ND

ND

ND

ND

ND

Above with 5 mM CaCO3 Above with 5 mM CaCO3 Above without 5 mM CaCO3

Oxalate (4.5 mM ), tartate (1.2 mM ), citrate (1.1 mM ) (PC) ND

ND

Above with 5 mM CaCO3

ND

ND

ND

ND

ND

Above with 5 mM CaCO3 Above with 5 mM CaCO3 Above without 5 mM CaCO3

ND

pH 5.2 to 2.7

3 (10 days) 2 (10 days)

115

A. fumigatus f (isolated from alluvial soil, West Bengal, India. Soil available P: 7 ppm, pH 7.4)

19 (10 days)

1 (10 days) 1 (10 days) 2 (10 days)

441 (15 days)

1000 mg P liter1 AlPO4 in 15 ml 1000 mg P liter1 AlPO4 in 15 ml 1000 mg P liter1 Mussoorie RP in 15 ml 1000 mg P liter1 Ca3(PO4)2 in 100 ml

A. awamori (isolated from soil or compost, India) P. digitatum

395 (15 days)

1000 mg P liter1 Ca3(PO4)2 in 100 ml

ND

pH 5.2 to 2.7

Paecilomyces fusisporus

388 (15 days)

ND

pH 5.2 to 3.3

Papulaspora mytiline

284 (15 days)

ND

pH 5.2 to 3.5

Masoniella grisea

191 (15 days)

1000 mg P liter1 Ca3(PO4)2 in 100 ml 1000 mg P liter1 Ca3(PO4)2 in 100 ml 1000 mg P liter1 Ca3(PO4)2 in 100 ml

ND

pH 5.2 to 4.0

0.1% N (from NO3 ), 3% glucose (static)

Singh et al. (1982)

continues

Table I—Continued

Microorganism

116

Penicillium sp. (isolated from Fluvaquent soil, West Bengal, India; pH unknown) Fungus “93” (isolated from soil, Southern Alberta, Canada) Fungus “98” (isolated from soil, Southern Alberta, Canada) Fungus “96” (isolated from soil, Southern Alberta, Canada) Aspergillus sp. (isolated from soil and rye grass or wheat rhizosphere, Bangladesh) Penicillium sp. A. awamori

A. niger (isolated from soil, Western Rajasthan, India; pH unknown) Pencillium pinophillum (isolated from soil, Western Rajasthan, India; pH unknown) A. niger (isolated from soil, Western Rajasthan, India; pH unknown)

Phosphate solubilized (mg P liter1 culture)a

Organic acids produced

Phosphate source 1

Titratable acidity (mmol H liter1) and change in pH

N source and sugar concentration in medium (% w/v) NH4),

3 (10 days)

1000 mg P liter Ca3(PO4)2 in 15 ml

Oxalate (trace) (PC)

125e (8 days)

412 mg P liter1 Idaho RP in 50 ml

ND

ND

121e (8 days)

412 mg P liter1 Idaho RP in 50 ml

ND

ND

110e (8 days)

412 mg P liter1 Idaho RP in 50 ml

ND

ND

524e (max. at 11 days)

Unknown amount of CaHPO4b in 100 ml

ND

pH 7.4 to 6.5

513e (max. at 11 days) 4 (14 days)

Unknown amount of CaHPO4b in 100 ml 1000 mg P liter1 Mussoorie RP in 50 ml

ND

pH 7.4 to 6.7

ND

pH 8.0 to 6.9

666e (max. at 15 days)

1000 mg P liter1 Ca3(PO4)2 in 50 ml

Lactate, glycollate, citrate, succinate (PC)

pH 6.5 to 2.3

440e (max. at 9 days)

1000 mg P liter1 Ca3(PO4)2 in 50 ml

Lactate, glycollate, citrate, succinate (PC)

pH 6.5 to 3.7

375e (max. at 12 days)

875 mg P liter1 RP in 50 ml

Lactate, glycollate, citrate, succinate (PC)

ND

Reference

0.01% N (from 0.05% yeast extract, 1% sucrose (37°C, static)

Banik and Dey (1983)

0.005% N (from NO3 ) and 0.01% N (from NH4), 1% glucose (20°C, shaken)

Kucey (1983)

0.05% yeast extract, 1% sucrose (30°C, static)

Molla et al. (1984)

0.01% N (from NH4), 0.05% yeast extract, 1% glucose (30°C, shaken 10 h day1) 0.01% N (from NH4), 0.05% yeast extract, 1% glucose (28°C, static)

Singh et al. (1984)

As above except 2% glucose

Venkateswarlu et al. (1984)

117

Cylindrocarpon obtusisporum (isolated from lateritic forest soil, Maharashtra, India; pH unknown) Spegazzinia tessarthra (isolated from lateritic forest soil, Maharashtra, India; pH unknown) Beltraniella humicola (isolated from lateritic forest soil, Maharashtra, India; pH unknown) Scopulariopsis brumptii

209e (7 days)

994 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 6.5 to 6.0

209e (7 days)

994 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 6.5 to 5.4

189e (7 days)

994 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 6.5 to 6.0

156e (7 days)

ND

pH 6.5 to 5.4

Phoma exigua

139e (7 days)

ND

pH 6.5 to 5.1

Curvularia lunata

120e (7 days)

ND

pH 6.5 to 5.5

Myrothecium roridum

120e (7 days)

ND

pH 6.5 to 5.5

Humicola fuscoatra

117e (7 days)

ND

pH 6.5 to 4.7

Robillarda sessilis

117e (7 days)

994 mg P liter1 Ca3(PO4)2 in 50 ml 994 mg P liter1 Ca3(PO4)2 in 50 ml 994 mg P liter1 Ca3(PO4)2 in 50 ml 994 mg P liter1 Ca3(PO4)2 in 50 ml 994 mg P liter1 Ca3(PO4)2 in 50 ml 994 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 6.5 to 5.0

Gliomastix murorum

107e (7 days)

ND

pH 6.5 to 5.0

Syncephalastrum racemosum

97e (7 days)

ND

pH 6.5 to 5.7

Aspergillus sp. (isolated from soil, coconut plantations, Kerala, India. Soil available P: 7–11 mg P kg1, pH 5.5–6.3) Penicillium sp. (24) (isolated from soil, coconut plantations, Kerala, India. Soil available P: 7–11 mg P kg1, pH 5.5–6.3) Penicillium bilaii

148e (15 days)

994 mg P liter1 Ca3(PO4)2 in 50 ml 994 mg P liter1 Ca3(PO4)2 in 50 ml 200 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 7.0 to 2.5; TA 50

138e (15 days)

200 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 7.0 to 2.9; TA 46

298 (max. within 12 days)

206 mg P liter1 Idaho RP in 100 ml

ND

pH 7.2 to 3.7

0.01% N (from NH4), 0.05% yeast extract, 1% glucose (27°C, shaken)

Surange (1985)

0.01% N (from NH4), 0.05% yeast extract, 1% glucose (30°C, static)

Thomas et al. (1985)

0.01% N (from NH4) and 0.01% N (from NO3 ), 1% glucose (24°C, shaken)

Asea et al. (1988)

continues

Table I—Continued

Microorganism Penicillium fuscum P. bilaii P. fuscum (Sopp) Biorge Sensu. A. niger van Tieghem

Phosphate solubilized (mg P liter1 culture)a

118

Penicillium sp. (isolated from garden soil)

N source and sugar concentration in medium (% w/v)

Reference

206 mg P liter1 Idaho RP in 100 ml 206 mg P liter1 Idaho RP in 100 ml 206 mg P liter1 Idaho RP in 100 ml 86 mg P liter1 fluoroapatite in 30 ml

ND

pH 7.2 to 4.1

ND

pH 7.2 to 4.0

ND

pH 7.2 to 6.2

ND

pH 7.0 to 3.9; TA 17

60 (11 days)

86 mg P liter1 fluoroapatite in 30 ml 86 mg P liter1 fluoroapatite in 30 ml 437 mg P liter1 Ca3(PO4)2 (volume unknown) 437 mg P liter1 Mussoorie RP (volume unknown) 500 mg P liter1 FePO4 in 50 ml

ND

ND

pH 7.0 to 2.5; TA 40 pH 7.0 to 5.3; TA 7 pH 4.9 to 3.2

ND

pH 4.8 to 3.8

ND

pH 7.0 to 2.4

0.01% N (from NH4), 0.05% yeast extract, 1% glucose (30°C, static)

Martinez Cruz et al. (1990)

1640 mg P liter1 Araxá fluoroapatite RP in 30 ml 204 mg P liter1 hydroxyapatite in 100 ml spent culture

ND

pH 4.2 to 3.2; TA 59

Nahas et al. (1990)

Oxalate (15.9 mmol liter1) and itaconate (HPLC)

pH 7.0 before inoculation to approx. 2.5 after incubation without hydroxyapatite; to 5.4 after equilibration with hydroxyapatite

N from vinasse 4.5° Brix (% glucose and fructose mixture) (30°C, static) 0.01% N (from NH4), 1% glucose (28°C, shaken)

108 (max. at 7 days) 89 (max. at 10 days)

A. niger (isolated from soil. Soil available P: 18–37 ppm, pH 6.2 or 7.1; or from sugarcane rhizosphere, Havana, Cuba) A. niger van Tieghem (isolated from soil)

Phosphate source

Titratable acidity (mmol H liter1) and change in pH

203 (max. within 12 days) 46 (max. within 12 days) 7 (max. within 12 days) 76 (11 days)

8 (11 days) A. niger (isolated from compost material, Faizabad, India)

Organic acids produced

455e (14 days)

401e (13 days)

41e (max. at 4.5 hours)

ND

Asea et al. (1988) As above except 0.01% N from NO3 only As above except 0.01% N from N03 only 0.02% N (from NH4) and 0.02% N (from NO3 ), 1% glucose 13.3 mM sodium citrate (30°C, static) As above except 0.02% N from NH4 only As above except 0.01% N from NO3 0.01% N (from NH4), 0.05% yeast extract, 1% glucose (28°C, static)

Cerezine et al. (1988)

Darmwal et al. (1989)

Parks et al. (1990)

P. bilaii

119 Penicillium sp. (isolated from forest soils, Austria. Soil available P: 0.9–1.66 ppm, pH 4.2–4.6) Aspergillus sp. (KAR021) (isolated from soils, Indonesia) Penicillium sp. (KAR023) (isolated from soils, Indonesia) A. awamori (isolated from soil, Bhavnagar, India)

36e (max. at 4.5 hours)

180 mg P liter1 Ca3(PO4)2 in 100 ml spent culture

Oxalate (15.9 mmol liter1) and itaconate (HPLC)

⬇8e (max. at 10 min.)

⬇40 mg P liter1 iron ore concentrate in 100 ml spent culture

Oxalate and itaconate (HPLC)

1686 (1 h)

4560 mg P liter1 CaHPO4 in 100 ml “fresh” (72-h) culture

Nil citrate (TLC)

2412 (1 h)

4560 mg P liter1 CaHPO4 in 100 ml “spent” (72-h) culture

Citrate and oxalate (amount unknown) (TLC)

65 (max. at 8 days)

44 mg P liter1 hydroxyapatite and 43 mg P liter1 CaHPO4 (volume unknown)

Gluconate (0.6 mM) and traces of lactate and citrate (isotachophoresis)

162 (22 days)

1000 mg P liter1 AlPO4 in 25 ml

ND

pH 4.7 to 2.7

64 (22 days)

1000 mg P liter1 AlPO4 in 25 ml

ND

pH 4.7 to 2.7

154e (max. at 12 days)

500 mg P liter1 Ca3(PO4)2 in 100 ml

ND

pH 7.0 to 4.4

pH 7.0 before inoculation to 2.5 after incubation without Ca3(PO4)2; to 5.4 after equilibration with Ca3(PO4)2 pH 7.0 before inoculation to 2.8 after incubation in absence of iron ore; to 5.7 after equilibration with iron ore pH 4.5 to 5.9 after equilibration with CaHPO4 pH 5.0 to 5.6 after equilibration with CaHPO4 pH 7.0 to ⬇3.7

0.05% N (from NH4), 2% sucrose (20°C, shaken)

Cunningham and Kuiack (1992)

0.03% N (from NO3 ), 2% sucrose (20°C, shaken)

0.01% N (from NH4) and 0.01% N (from NO3 ), 0.2% glucose, 0.2% sucrose; spent culture, diluted 1:2 (30°C, shaken) 0.01% N (from NH4), 0.05% yeast extract, 1% glucose (25°C, shaken)

0.01% N (from NH4), 0.05% yeast extract, 1% glucose (28°C, static)

Illmer and Schinner (1992)

Goenadi and Saraswati (1993)

Narsian et al. (1993)

continues

Table I—Continued

Microorganism

Phosphate solubilized (mg P liter1 culture)a e

Organic acids produced

Phosphate source liter1

Titratable acidity (mmol H liter1) and change in pH

500 mg P CaHPO4 in 100 ml 500 mg P liter1 Fe3(PO4)2 in 100 ml 500 mg P liter1 AlPO4 in 100 ml 1000 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 7.0 to 2.2

ND

pH 7.0 to 3.4

ND

pH 7.0 to 2.4

Aspergillus tamarii

218 (max. at 9 days) 175e (max. at 15 days) 190e (max. at 15 days) 222e (7 days)

Oxalate (PC)

pH 5.9 to 4.8

Aspergillus clavatus

192e (7 days)

ND

pH 5.9 to 3.6

Penicillium nigricans

191e

(7 days)

ND

pH 5.9 to 4.9

Aspergillus foetidus

205e

(7 days)

Oxalate (PC)

pH 5.9 to 5.0

A. awamori

190e

(7 days)

Oxalate (PC)

pH 5.9 to 4.8

Aspergillus aculeatus

144e

(7 days)

ND

pH 5.9 to 4.4

Aspergillus terricola

142e

(7 days)

Oxalate (PC)

pH 5.9 to 2.5

A. terreus

139e

(7 days)

ND

pH 5.9 to 4.3

A. fumigatus

135e

(7 days)

ND

pH 5.9 to 3.1

A. candidus

95e

(7 days)

ND

pH 5.9 to 4.2

Aspergillus amstelodemi

86e

(7 days)

Oxalate (PC)

pH 5.9 to 3.2

Aspergillus fischeri

72e

(7 days)

ND

pH 5.9 to 4.6

A. aculeatus (isolated from rhizosphere of Gram, Bhavnagar, India)

471e (max.

1000 mg P liter1 Ca3(PO4)2 in 50 ml 1000 mg P liter1 Ca3(PO4)2 in 50 ml 1000 mg P liter1 Ca3(PO4)2 in 50 ml 1000 mg P liter1 Ca3(PO4)2 in 50 ml 1000 mg P liter1 Ca3(PO4)2 in 50 ml 1000 mg P liter1 Ca3(PO4)2 in 50 ml 1000 mg P liter1 Ca3(PO4)2 in 50 ml 1000 mg P liter1 Ca3(PO4)2 in 50 ml 1000 mg P liter1 Ca3(PO4)2 in 50 ml 1000 mg P liter1 Ca3(PO4)2 in 50 ml 1000 mg P liter1 Ca3(PO4)2 in 50 ml 500 mg P liter1 Ca3(PO4)2 in 100 ml 500 mg P liter1 CaHPO4 in 100 ml

ND

pH 7.0 to 3.6

ND

pH 7.0 to 3.2

A. awamori

120

at 4 days) e 273 (max. at 5 days)

N source and sugar concentration in medium (% w/v)

Reference Narsian et al. (1993)

0.01% N (from NH4), 0.05% yeast extract, 1% glucose (30°C, static)

Gupta et al. (1994)

0.01% N (from NH4), 0.05% yeast extract, 1% sucrose (30°C, static, shaken every 12 h)

Narsian et al. (1994)

A. aculeatus A. aculeatus A. niger (isolated from rhizosphere soil, Bhavnagar, India)

121

Aspergillus japonicus (isolated from low-grade RP, India)

Aspergillus foetidus (isolated from low-grade RP, India) Penicillium aurantiogriseum (isolated from forest soil, Austria) A. niger

A. niger (isolated from forest soil, Austria) A. aculeatus (isolated from rhizosphere of Gram (Cicer ariatenum), India)

144e (max. at 11 days) 159e (max. at 13 days) 101e (max. at 15 days) 449e (max. at 5 days)

500 mg P liter1 Fe3(PO4)2 in 100 ml 500 mg P liter1 AlPO4 in 100 ml 219 mg P liter1 Sonari RP (India) in 100 ml 500 mg P liter1 Ca3(PO4)2 in 100 ml

ND

pH 7.0 to 2.8

ND

pH 7.0 to 2.0

ND

pH 7.0 to 3.0

ND

pH 7.0 to 3.5

231e (max. at 12 days) 184e (max. at 15 days) 130e (max. at 13 days) 64e (max. at 15 days) 180 (7 days)

500 mg P liter1 CaHPO4 in 100 ml 500 mg P liter1 Fe3(PO4)2 in 100 ml 500 mg P liter1 AlPO4 in 100 ml 219 mg P liter1 Sonari RP (India) in 100 ml 437 mg P liter1 Parulia RP in 50 ml

ND

pH 7.0 to 2.0

ND

pH 7.0 to 3.4

ND

pH 7.0 to 2.4

ND

pH 7.0 to 2.8

Oxalate (PC)

pH 8.0 to 3.6

211 (7 days)

437 mg P liter1 Jhabua RP in 50 ml 437 mg P liter1 Sagar RP in 50 ml

Oxalate (PC)

pH 8.0 to 3.8

Oxalate (PC)

pH 8.0 to 3.3

154 mg P liter1 AlPO4(cr) (volume unknown) 154 mg P liter1 AlPO4(cr) (volume unknown) 154 mg P liter1 AlPO4(cr) (volume unknown) 500 mg P liter1 Ca3(PO4)2 in 50 ml

Nil organic acid (HPLC)

pH 5.9 to 6.0

Citrate (18 M ), oxalate (5 M), gluconate (0.5 M) (HPLC) Citrate (18 M ), oxalate (5 M), gluconate (0.5

M ) (HPLC) ND

pH 7.0 to 2.4

pH 7.0 to 4.0

500 mg P liter1 Ca3(PO4)2 in 50 ml

ND

pH 7.0 to 4.4

214 (7 days)

4e (max. time unknown) 55e (unknown incubation time) 100e (max. at 16 days but still increasing) 470e (max. at 2 days) 209e (2 days)

pH 6.9 to 1.8

0.01% N (from NH4), 0.05% yeast extract, 1% glucose (37°C, static)

Singal et al. (1994)

0.02% N (from NH4), 0.8% glucose, 0.8% sucrose As above except 0.01% N from NH4

Illmer and Schinner (1995a)

0.01% N (from NH4) 0.8% glucose, 0.8% sucrose 0.01% N (from NH4), 0.05% yeast extract, 1% glucose (28°C, static) As above except 0.01 % N from NO3

Illmer et al. (1995) Narsian et al. (1995)

continues

Table I—Continued

Microorganism

Phosphate solubilized (mg P liter1 culture)a

Organic acids produced

Phosphate source

Titratable acidity (mmol H liter1) and change in pH

122

A niger (NB2)

292 (10 days)

384 mg P liter1 RP fluoroapatite from Morocco in 50 ml

Citrate (22 mM ) (colorimetric methods)

TA 72 (10 days); pH 6.5– 7.0 to 3.0 (20 days)

A. niger (isolated from soil, Sao Paulo State, Brazil)

219e (9 days)

540 mg P liter1 CaHPO4 # 2H2O

ND

TA 10; pH 7.0 to 5.4

47e (9 days)

520 mg P liter1 Patos de Minas (Brazil) RP 540 mg P liter1 CaHPO4 # 2H2O

ND

TA 2; pH 7.0 to 6.2 TA 13; pH 7.0 to 5.1

Penicillium solitum (isolated from soil, Sao Paulo State, Brazil)

302e (9 days)

52e (9 days) Eupenicillium shearii (isolated from soil, Sao Paulo State, Brazil)

285e (9 days)

111e (9 days) Aspergillus ochraceus (isolated from soil, Sao Paulo State, Brazil)

200e (9 days)

13e (9 days) Penicillium implicatum (isolated from soil, Sao Paulo State, Brazil)

20e (9 days)

144e (9 days)

520 mg P liter1 Patos de Minas (Brazil) RP 540 mg P liter1 CaHPO4 # 2H2O 520 mg P liter1 Patos de Minas (Brazil) RP 540 mg P liter1 CaHPO4 # 2H2O 520 mg P liter1 Patos de Minas (Brazil) RP 540 mg P liter1 CaHPO4 # 2H2O 520 mg P liter1 Patos de Minas (Brazil) RP

ND

ND ND

ND ND

ND ND

ND

TA 2; pH 7.0 to 5.1 TA 15; pH 7.0 to 5.2 TA 4; pH 7.0 to 5.6 TA 15; pH 7.0 to 5.5 TA 8; pH 7.0 to 4.0 TA 22; pH 7.0 to 5.0 TA 4; pH 7.0 to 6.0

N source and sugar concentration in medium (% w/v) 0.05% N (from NO3 ), 0.5% yeast extract, 3% sucrose, 10% sugar beet waste (30°C, shaken for 5 min day1) Unknown

Reference Vassilev et al. (1995)

Nahas (1996)

Penicillium minioluteum (isolated from soil, Sao Paulo State, Brazil)

Penicillium viridicatum (isolated from soil, Sao Paulo State, Brazil)

290e (9 days)

540 mg P liter1 CaHPO4 # 2H2O

ND

TA 16; pH 7.0 to 5.3

117e (9 days)

520 mg P liter1 Patos de Minas (Brazil) RP 540 mg P liter1 CaHPO4 # 2H2O

ND

TA 3; pH 7.0 to 6.3 TA 13; pH 7.0 to 5.0

160e (9 days)

106e (9 days) Penicillium purpurogenum (isolated from soil, Sao Paulo State, Brazil)

319e (9 days)

142e (9 days) Penicillium variabile P16 (Spain)

123

Penicillium radicum (isolated from wheat rhizosphere, Australia)

279 (5th batch, 2 days) 347 (3rd batch, 2 days) 475 (max. at 6 days) 360 (max. at 14 days) 207 (max. at 14 days) 213 (max. after 31 days) 176 (max. at 5 days)

520 mg P liter1 Patos de Minas (Brazil) RP 540 mg P liter1 CaHPO4 # 2H2O 520 mg P liter1 Patos de Minas (Brazil) RP 1790 mg P liter1 sedimentary RP in 50 ml

ND

ND ND

ND Gluconate (86 mM; 5th batch) (HPLC and enzymatic methods)

TA 2; pH 7.0 to 6.5 TA 12; pH 7.0 to 5.5 TA 4; pH 7.0 to 5.9 ND

1000 mg P liter1 CaHPO4 in 50 ml

Gluconate (17 mM) (HPLC)

pH 6.4 to 3.4; TA 29

1000 mg P liter1 Ca3(PO4)2 in 50 ml 1000 mg P liter1 amorphous aluminium phosphateg in 50 ml 1000 mg P liter1 Ca3(PO4)2 in 50 ml 1000 mg P liter1 CaHPO4 in 50 ml

Gluconate (20 mM) (HPLC)

pH 6.1 to 4.1; TA 23 pH 3.7 to 2.7; TA 24

Gluconate (9 mM) (HPLC)

Gluconate (31 mM) (HPLC) Gluconate (23 mM) (HPLC)

pH 6.1 to 4.5; TA 20 pH 6.6 to 4.6; TA 12

0.05% N (from NO3 ), 8% glucose. The fungus was immobilized on 0.3 cm3 polyurethane sponge cubes (28°C, shaken) 0.01% N (from NH4), 3% glucose (25°C, shaken)

0.01% N (from NO3 ), 3% glucose (25°C, shaken)

Note. Abbreviations used: HPLC, high-performance liquid chromatography; ND, not determined; PC, paper chromatography; TLC, thin-layer chromatography; RP, rock phosphate. at which determination was made is given. When solubilized P was monitored over a period of time, only the maximum (max.) solubilized P is given. in situ by method of Gerretson (1948): K2HPO4  CaCl2 r CaHPO4  2KCl. cPrecipitated in situ by method of Rose (1957): 2K HPO  3CaCl r Ca (PO )  6KCl. 3 4 2 3 4 2 dPrecipitated in situ by method of Rose (1957): K HPO  FeCl r FePO  2KCl  HCl. 2 4 3 4 eNo indication of statistical significance was given by these authors. f Human pathogen. gPrepared by method of Deming and Cate (1963). a Time

bPrecipitated

Vassilev et al. (1996)

Whitelaw et al. (1999)

124

M. A. WHITELAW

in water varies widely. A brief comparison of solubility equilibrium constants (logKsp) for a range of common crystalline P compounds in pure water at 25C is shown in Table II. The approximate water solubilities (i.e., the maximum amount which can be solubilized in water) are shown. The pH of the solvent plays a major role in the dissolution of P compounds. Poorly soluble P compounds, such as variscite and strengite, can only be effectively dissolved in water if the pH is amended. In aqueous solution, AlPO4 has a minimum solubility at a pH between 6.0 and 6.5 and FePO4 has a minimum solubility at a pH between 5.0 and 6.0. As the pH decreases further, the solubility of these P compounds tends to increase. Most of the calcium phosphates have minimum solubilities above pH 8 in water and their solubility increases as the pH decreases below pH 8. The solubility of some common calcium, ferric, and aluminium phosphates is represented in Fig. 3 (Stumm and Morgan, 1995). The solubility of P compounds also depends on the degree of crystallinity and particle size. Amorphous compounds do not have a well-organized crystal structure and are almost always more soluble than the corresponding crystalline solid.

Table II Solubility of Common Crystalline Phosphate Compunds

Phosphate compound

Equilibrium reaction for the dissolution of the phosphate compound

Ca(H2PO4)2 (monocalcium phosphate or “triple superphosphate”) CaHPO4 (calcium monohydrogen phosphate) -Ca3(PO4)2 (-calcium orthophosphate Ca5(PO4)3OH (hydroxyapatite)

Ca(H2PO4)2(s) B Ca2  2H2PO 4

CaHPO4(s) B Ca2  HPO42

Log Ksp 1.14

Snoeyink and Jenkins (1980)

6.6

Stumm and Morgan (1995) Snoeyink and Jenkins (1980) Snoeyink and Jenkins (1980) Stumm and Morgan (1995) Stumm and Morgan (1995)

-Ca3(PO4)2(s) B 3Ca2  2PO43

24.0

Ca5(PO4)3OH(s) B 5Ca2  3PO43  OH

55.9

AlPO4 # 2H2O (variscite)

AlPO4 # 2H2O(s) B Al3  PO43  2H2O

21.0

FePO4 # 2H2O (strengite)

FePO4 # 2H2O(s) B Fe3  PO43  2H2O

26.0

aAylward

and Findlay (1994). insoluble.”

b“Practically

Reference for log Ksp

Solubility (in water, 25°C)a (g/100 g) 18

0.14

0.02

b

b

b

PHOSPHATE-SOLUBILIZING FUNGI

125

Figure 3 Solubility as a function of pH for selected discrete phosphate compounds in water. (Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters,” 3rd ed., W. Stumm and J. J. Morgan. Copyright 1995 John Wiley & Sons. Reprinted by permission of John Wiley & Sons, Inc.)

Amorphous solids are not stable and will transform slowly to the stable crystalline solid phase (Snoeyink and Jenkins, 1980). The initial reaction product of the reaction of superphosphate in acidic soils (amorphous aluminium phosphate) is more soluble than crystalline AlPO4#2H2O (Lindsay, 1979), and this is probably due to the lower particle size and thus higher surface area of the amorphous compound. The surface area of amorphous aluminium phosphate is 10.5 m2 g1 (Juo and Ellis, 1968), whereas the surface area of crystalline AlPO4#2H2O is 1.54 m2 g1 (Taylor and Gurney, 1964). The equilibrium constant (logKsp) for the dissolution of AlPO4#2H2O, AlPO4#2H2O B Al3+  H2PO4  20H is higher for the more soluble amorphous aluminium phosphate (logKsp 28.1; Veith and Sposito, 1977) than for crystalline AlPO4#2H2O (logKsp 30.5; Bache, 1963). Many investigators have selected CaHPO4 or CaHPO4#2H2O as the “insoluble” source of P in agar and liquid medium studies or in pot and field studies of plant growth promotion by P-solubilizing microorganisms. These calcium phosphate minerals, however, are poor models for soil P because they are not insoluble. CaHPO4 is a moderately effective fertilizer in acidic soils in higher rainfall areas. Although it is far less soluble than Ca(H2PO4)2#H2O in water, it is soluble enough in a soil environment to allow plant growth. The consequence of using relatively soluble forms of P when investigating plant growth promotion by P-sol-

126

M. A. WHITELAW

ubilizing microorganisms is that the ability of the P-solubilizing microorganisms to dissolve soil P is potentially underestimated.

B. PRODUCTION OF ORGANIC ACIDS IN LIQUID MEDIUM STUDIES Laboratory studies, reviewed by Stevenson (1967), Kucey et al. (1989), and BarYosef (1991), have shown that the microbial solubilization of soil phosphates in liquid medium studies has often been due to the excretion of organic acids. In many studies the presence of organic acids in liquid culture filtrates was determined by paper chromatography or thin-layer chromatography, but more modern techniques such as high-performance liquid chromatography, isotachophoresis, and enzymatic methods have been used by others to allow more accurate identification of unknown organic acids (Rose, 1957; Sperber, 1958b; Banik and Dey, 1982, 1983; Venkateswarlu et al., 1984; Parks et al., 1990; Cunningham and Kuiack, 1992; Illmer and Schinner, 1992, 1995a; Gupta et al., 1994; Singal et al., 1994; Illmer et al., 1995; Vassilev et al., 1995, 1996; Whitelaw et al., 1999) (Table I). Production of oxalic acid as a major organic acid was reported by many investigators from Aspergillus amstelodemi, A. awamori, A. candidus, A. foetidus, A. fumigatus, A. japonicus, A. niger, A. tamarii, A. terreus, A. terricola, an unknown species of Penicillium, and Sclerotium rolfsii (Rose, 1957; Banik and Dey, 1982, 1983; Parks et al., 1990; Gupta et al., 1994; Singal et al., 1994; Illmer and Schinner, 1995a). Cunningham and Kuiack (1992) also reported that oxalate was a major organic acid anion produced by P. bilaii when the N source was nitrate, although it was not detected when the N source was ammonium (Table I). The production of citric acid as a major organic acid by A. niger was reported by Rose (1957), Sperber (1958b), Venkateswarlu et al., (1984), Illmer and Schinner (1995a), and Vassilev et al. (1996). Cunningham and Kuiack (1992) also reported that citrate was a major organic acid anion produced by P. bilaii when the N source was nitrate, although, as with oxalic acid, citric acid was not detected when the N source was ammonium (Table I). The production of lactic acid has been reported from an isolate of Penicillium and from A. niger and Penicillium pinophillum, whereas tartaric acid production was reported from A. candidus and A. fumigatus (Sperber, 1958b; Banik and Dey, 1982; Venkateswarlu et al., 1984). Gluconate was reported to be the principal organic acid anion formed during P solubilization by Penicillium aurantiogriseum (Illmer and Schinner, 1992), P. variabile (Vassilev et al., 1996), and P. radicum (Whitelaw et al., 1999), with smaller quantities being produced by A. niger (Sperber, 1958b; Illmer and Schinner, 1995a) (Table I). Petruccioli et al. (1994) demonstrated the production of large quantities of glucose oxidase and gluconic acid by P. variabile immobilized on polyurethane sponge. Phosphate limitation has been reported to favor gluconic

PHOSPHATE-SOLUBILIZING FUNGI

127

acid synthesis in A. niger (Berry, 1975). De La Torre et al. (1993), in a study of biochemical weathering of stone by P. frequentans, reported that gluconic acid produced by the fungus released a large amount of Ca2+ from limestone and appreciable amounts of Fe3+, Ca2+, and Al3+ from sandstone and granite. Robert and Razzaghe-Karimi (1975) and Eckhardt (1980) also observed that gluconic acid was able to induce significant cation release from biotite.

C. CHELATION OF CATIONS BY ORGANIC ACIDS The solubilization of soil P by plant organic acids and the mechanisms by which fungal organic acids solubilize phosphates in liquid media are reported to be due to either the lowering of pH or the chelation of the cation bound to P. Chelation of cations has been shown to be an important mechanism for P solubilization when the organic acid structure is favorable. Chelation involves the formation of two or more coordinate bonds between a molecule (the “ligand”) and a metal ion, thereby creating a ring structure complex. Chelation by an organic acid ligand occurs through oxygen containing hydroxyl and carboxyl groups and takes place only when either five-membered rings or the less stable six-membered rings can be formed (Albert and Serjeant, 1984). Many investigators have recorded that organic acids were able to solubilize more P than was solubilized by inorganic acid at the same pH with the difference presumed to be primarily due to chelation. Johnston (1959) found that HCl at pH 1.0 was able to solubilize less ferric and aluminum phosphate than many organic acids, including citric, oxalic, malic, and lactic acids, at higher pH. Kim et al. (1997) reported that HCl was able to solubilize less P from hydroxyapatite than citric acid and oxalic acid at the same pH. Cunningham and Kuiack (1992) observed that the presence of citrate at pH 4.5 increased the abiotic solubilization of CaHPO4 by 837 mg P liter1 compared to inorganic acid alone at the same pH, indicating that chelation by citric acid was probably involved. Artificial acidification of culture media with HCl also solubilized less P from insoluble calcium phosphates than that solubilized by unknown species of both Penicillium and Pseudomonas (Illmer and Schinner, 1992) and Enterobacter agglomerans (Kim et al., 1997), suggesting that chelation by organic acids produced by the microorganisms may have been involved. Whitelaw et al. (1999) observed that either gluconic acid alone or P. radicum inoculation alone was able to solubilize more amorphous aluminium phosphate than HCl at the same pH. Johnston (1954) also reported that at a similar pH, gluconic acid was able to solubilize more P from Ca3(PO4)2 than could be solubilized by HCl alone, indicating that chelation was probably occurring. Many studies have investigated the ability of organic acid anions to release P from soil or the P-adsorbing components of soil. Citrate was able to release P from

Table III Stability Constants for Al-Organic Acid Complexes Organic acid

Structure

Log KAl

Reference

9.6

Bar-Yosef (1991)

7.3

Bar-Yosef (1991)

5.3

Bar-Yosef (1991)

5.4

Bolan et al. (1994)

2.4

Bolan et al. (1994)

2.0

Motekaitis and Martell (1984)

1.5

Bar-Yosef (1991)

1.4

Bar-Yosef (1991)

H 冟 H%C%COOH 冟 HOOC%C%OH 冟 H%C%COOH 冟 H

Citric

Tricarboxylic acid, one -hydroxyl group HO

O C 冟 C

O

Oxalic

OH

Dicarboxylic acid, no hydroxyl groups H 冟 HOOC%C%OH 冟 HO%C%COOH 冟 H

Tartaric

Dicarboxylic acid, two -hydroxyl group H 冟 HOOC%C%OH 冟 H%C%COOH 冟 H

Malic

Dicarboxylic acid, one -hydroxyl group OH 冟 H%C%COOH 冟 H%C%H 冟 H

Lactic

Monocarboxylic acid, one -hydroxyl group COOH 冟 H%C%OH 冟 HO%C%H 冟 H%C%OH 冟 H%C%OH 冟 CH2OH

Gluconic

Monocarboxylic acid, one -hydroxyl group HO

O

C 冟 H%C%H 冟 H

Acetic

Monocarboxylic acid, no hydroxyl groups HO

O C 冟 H

Formic

Monocarboxylic acid, no hydroxyl groups

PHOSPHATE-SOLUBILIZING FUNGI

129

kaolin, goethite, ferric phosphate, and aluminum phosphate (Swenson et al., 1949); citrate, tartrate, oxalate, and malate were able to desorb P from kaolinite and gibbsite (Nagarajah et al., 1970); citrate and tartrate reduced P sorption by soil (Earl et al., 1979); and citrate, oxalate, and malate were able to desorb P from acidic soils over a range of pH (Lopez-Hernandez et al., 1979). Nagarajah et al. (1970) suggested that the decrease in P adsorption caused by the organic acid anions was due to competition with P for adsorption sites which would be determined by the stability of the ferric or aluminium–organic anion complex. The ability to desorb P generally decreased with a decrease in the stability constants for the Fe(III)– or Al–organic acid complex (logKA1 or logKFe) in the following order: citrate  oxalate  malonate  tartrate  acetate. Other studies reported that logKA1 was a good indicator of the ability of organic acids to release P from soils (Fox et al., 1990) and rock phosphate (Kpomblekou-A and Tabatabai, 1994). Bolan et al. (1994) reported that the addition of organic acids caused the dissolution of soil components such as ferric and aluminium oxides and thereby decreased P adsorption by soil. They suggested that dissolution of ferric and aluminium oxides was caused by the complexation of these metal ions with the organic acids. The decreases in P adsorption increased with an increase in logKA1. Organic acids also increased the dissolution of rock phosphate, with P solubilization increasing with increasing logKA1 values. (LogKA1 values for a selection of organic acids commonly produced by soil microorganisms are included in Table III). The extent to which an organic acid is capable of chelating metal cations is greatly influenced by its molecular structure. Struthers and Sieling (1950) demonstrated that the ability of organic acids to prevent the precipitation of P by Fe3+ and Al3+ (and by inference the ability to chelate Al3+ and Fe3+) increased progressively as the number of hydroxyl groups was increased in an otherwise unchanged molecule. Johnston (1956) and Johnston and Miller (1959) described the importance of hydroxyl and carboxyl structures to the P-solubilizing ability of carboxylic acids. Ca2+ was chelated most strongly by tricarboxylic acids such as citric acid, less strongly by dicarboxylic acids such as malic and tartaric acids, and to a smaller extent by some monobasic acids such as lactic acid, which had hydroxyl groups adjacent to carboxylic groups (i.e., -hydroxy acid structures). Among the dicarboxylic acids, those with hydroxyl groups formed the strongest complexes and hydroxy compounds were more effective than compounds which had hydroxyl groups located on the second carbon atom from the carboxylic groups (i.e., -hydroxy compounds). The ability to solubilize calcium phosphates was attributed both to the increased acid strength of -hydroxy acids in comparison to unsubstituted acids and to the enhancement of Ca2+ chelation by organic acids with such structures. Bolan et al. (1994) also found that the ability of organic acids to prevent P adsorption in soil decreased in the following order: tricarboxylic acid  dicarboxylic acid  monocarboxylic acid. Hue et al. (1986) found that the Al3+ detoxifying capacities of organic acids (and

130

M. A. WHITELAW

by inference Al3+ chelating ability) were correlated with the relative positions of hydroxyl and carboxylic groups on their main carbon chain. Many effective chelators of Al3+ had -hydroxy acid structures, which favored the formation of stable five-bond ring structures with Al3+. Kpomblekou-A and Tabatabai (1994) also observed that aliphatic acids with hydroxyl and carboxyl groups in positions suitable for the formation of complexes with metal cations were more effective than other aliphatic or aromatic acids in releasing P from rock phosphate. On the other hand, Moghimi and Tate (1978) demonstrated that the ability of ketogluconic acid to solubilize hydroxyapatite was due to its ability to lower the pH and not to the chelation of calcium ions. The stability constant for the Ca2+ – -ketogluconic acid complex (i.e., logKCa ) was found to be negligible. -Ketogluconic acid is produced by many P-solubilizing bacteria and is one of the strongest monobasic carboxylic acids, having a low acid dissociation constant (pKa) of 2.66. Illmer and Schinner (1995a) suggested that solubilization of AlPO4 by A. niger and Penicillium simplicissimum in a liquid medium was probably caused by acidolysis (i.e., the lowering of pH) followed by complex formation by citrate, whereas calcium phosphates such as CaHPO4#2H2O and hydroxyapatite were solubilized by acidolysis alone. This suggestion is partly supported by the fact that the stability constant for the Al3+ –citrate complex (logKA1 9.6; Bar-Yosef, 1991) is higher than the constant for the Ca2+ –citrate complex (logKCa 3.5; Martell and Smith, 1977).

D. TITRATABLE ACIDITY Titratable acidity (TA), determined by titrating the culture filtrate with NaOH solutions using phenolphthalein indicator, was included in many P-solubilization studies as a measure of the concentration of total dissociated plus undissociated H+ (Table I). Peaks in TA and organic acid concentration often coincided with peaks in soluble P, suggesting correlations between these variables. High correlations between solubilized P and TA were reported by Thomas et al. (1985), Nahas (1996), Vassilev et al. (1996), and Whitelaw et al. (1999), with r values of 0.57, 0.83, 0.81, and 0.91, respectively, indicating that acid production was the key Psolubilization mechanism in many cases.

E. pH In most cases, acidification (i.e., pH decrease) was a major P-solubilizing mechanism. High P solubilization was often associated with a low pH in the final cul-

PHOSPHATE-SOLUBILIZING FUNGI

131

ture solution and, conversely, low P solubilization was often associated with a high pH of the culture solution pH (Table I). Some liquid medium studies compared the changes in medium pH and P solubilization for many different soil fungi. Significant negative correlations for linear regression equations linking media pH and solubilization of calcium phosphate minerals by a range of different soil microorganisms were found by Venkateswarlu et al. (1984), Thomas et al. (1985), and Illmer and Schinner (1992), with correlation coefficients (r) of 0.93, 0.75, and 0.49, respectively. A negative correlation between pH and P released by different soil fungi was not found by some investigators (Sethi and Subba-Rao, 1968; Gaur et al., 1973; Bardiya and Gaur, 1974; Kucey, 1983; Surange, 1985; Salih et al., 1989; Nahas, 1996). Sometimes the culture filtrate pH was relatively high and yet medium to high P solubilization occurred. Cylindrocarpon obtusisporum and Beltraniella humicola solubilized 209 and 189 mg P liter1, respectively, after the pH decreased from 6.5 to only 6.0 (Surange, 1985), and A. fumigatus increased the culture filtrate pH from 6.5 to 7.1 while solubilizing 57 mg P liter1 (Gaur et al., 1973). In the study by Bardiya and Gaur (1974), two fungal isolates caused the culture pH to decrease to almost the same level (viz. A. awamori strain 18F and an unidentified fungal isolate “21F” reached pH 3.4 and 3.3, respectively), whereas the amounts of P solubilized from rock phosphate were quite different (viz. 116 and 48 mg P liter1, respectively), indicating that fungal P solubilization probably depends not only on the ultimate pH of the culture solution but also on other factors such as the type of organic acids excreted (Table I). Other liquid medium studies monitored changes in the pH and the amount of P solubilized over time by a particular soil fungus. Ortuno et al. (1979) reported that throughout the incubation period, the pH and the solubilization of calcium phosphate minerals by A. niger were negatively correlated (r 0.99, p  0.001). Cerezine et al. (1988) also found a negative correlation between pH and the solubilization of fluoroapatite by A. niger (r 0.45, p  0.01). Whitelaw et al. (1999) found a positive correlation between the activity of H+ and P solubilized from CaHPO4, Ca3(PO4)2, and amorphous aluminum phosphate. In contrast, Narsian et al. (1995) reported that over an incubation period of 7 days, pH and solubilization of Ca3(PO4)2 by A. aculeatus were not correlated. However, Ca3(PO4)2 solubilization by A. aculeatus reached a maximum, coinciding with a minimum in pH, on the second day of incubation. After this time, the amount of P solubilized decreased, fluctuating in a manner which did not appear to be related to pH. The decrease in the soluble P concentration after the initial peak was probably due to factors other than pH. The tendency for the relationship between pH and P solubilization to become less significant after the initial peak in solubilized P was also seen in P solubilization by P. aurantiogriseum (Illmer and Schinner, 1992), a Penicillium isolate (Goenadi and Saraswati, 1993), and P. radicum (Whitelaw et al., 1999).

132

M. A. WHITELAW

F. N SOURCE The N source affected P solubilization in many liquid medium studies. Assimilation of ammonium often favored higher acid production (i.e., lower pH and higher TA) and greater solubilization of P in comparison to nitrate. Higher acid production and P solubilization from ammonium compared to nitrate assimilation has also been reported for the solubilization of fluoroapatite by A. niger (Cerezine et al., 1988); rock phosphate by P. bilaii (Asea et al., 1988); Ca3(PO4)2 by A. aculeatus (Narsian et al., 1995); and CaHPO4, Ca3(PO4)2, and colloidal aluminium phosphate by P. radicum (Whitelaw et al., 1999) (Table I). Illmer et al. (1995) also found that solubilization of AlPO4 by P. aurantiogriseum and P. simplicissimum was more effective when small amounts of ammonium were included in the medium compared to higher amounts of nitrate. However, Cunningham and Kuiack (1992) reported greater P solubilization of CaHPO4 with nitrate than with ammonium by a fresh P. bilaii culture (Table I). Acid production in the form of H+ release in response to the assimilation of cations such as ammonium is a well-known fungal phenomenon (Taha et al., 1969; Banik and Dey, 1983; Kucey, 1983; Roos and Luckner, 1984; Asea et al., 1988). The uptake of ammonium by fungi in a liquid medium commonly leads to a rapid decrease in the pH of the medium (Cochrane, 1958). Roos and Luckner (1984) found that an equal number of moles of H+ were produced by Penicillium cyclopium per gram of mycelial growth from the same number of moles of ammonium in a sucrose medium because of ammonium assimilation. In contrast, the initial stages of nitrate assimilation produce an increase rather than a decrease in pH. The reduction of nitrate to ammonia results in the internal release of OH or consumption of H+ (Haynes, 1990). Nitrate-fed plant cells can cope with excess internal OH production by excreting OH or producing organic acid anions (Haynes, 1990). The relatively high pH of culture filtrates in nitrate media found by many investigators may be consistent with the excretion of either OH or organic acid anions in response to nitrate consumption.

G. FLUCTUATIONS IN SOLUBLE PHOSPHATE LEVELS OVER TIME In many liquid medium studies, soluble P concentrations, TA, and organic acid concentrations increased and decreased several times during the incubation period. Solubilization of P by A. niger (Cerezine et al., 1988; Nahas et al., 1990; Vassilev et al., 1995), P. bilaii (Cunningham and Kuiack, 1992, P. aurantiogriseum (Illmer and Schinner, 1992), A. aculeatus (Narsian et al., 1995), and P. radicum (Whitelaw et al., 1999) followed a pattern whereby the amount of P solubilized reached a peak after a number of days and then decreased. This indicates the importance of mon-

PHOSPHATE-SOLUBILIZING FUNGI

133

itoring microbial P solubilization at regular intervals over the entire incubation period. Studies in which the soluble P concentration is determined at specific times and not throughout the whole incubation period may not indicate the correct maximum amount of P solubilization. Illmer and Schinner (1992) suggested that the rise and fall in P concentration observed when calcium phosphates were solubilized by P. aurantiogriseum might be caused by the formation and secondary solubilization of organic P complexes which might be used as energy or nutrient sources are required by the fungus. A pattern of increases and decreases in P concentration was also found by Illmer and Schinner (1995a) when a mixture of CaHPO4#2H2O and hydroxyapatite was solubilized by a Pseudomonas sp. Because the decreases in soluble P occurred at the same time as the increases in the mass of precipitated calcium phosphate, it was suggested that hydroxyapatite was crystallized out of the solution once a threshold value in soluble P concentration was reached. Parks et al. (1990) also reported that P solubilized from iron ore increased to a maximum within 5 min after contact with spent medium containing Penicillium sp. fungal metabolites but then decreased during the next 2 h. It was suggested that this was due either to reprecipitation or to adsorption to ore particles or the vessel walls. Vassilev et al. (1995) suggested that the decrease in both TA and P concentration observed during solubilization of rock phosphate by A. niger on sugar beet waste was most likely a result of citric acid utilization by the fungus under conditions of nutrient depletion which also caused sporulation of the fungal culture. This explanation is in agreement with Cochrane (1958), who stated that with few exceptions acids which accumulate in fungal cultures are later utilized for energy and growth.

V. PLANT GROWTH PROMOTION BY PHOSPHATE-SOLUBILIZING FUNGI The ability of P-solubilizing fungi to solubilize soil P compounds in vitro indicates that inoculation by such fungi could enhance the utilization of sparingly soluble soil phosphate by plants. Although it is unlikely that organic acids produced in the rhizosphere would remain untouched for long enough to affect bulk P release from soil, it is possible that the short-term reduction of the rhizosphere pH and the complexation of cations could produce an effective P-solubilizing microenvironment, resulting in increased P uptake by the plant roots (Kucey et al., 1989). Much of the solubilized P would be quickly reprecipitated or readsorbed by the soil particles, but plant roots might be able to take up some of the released P before this occurs. P-solubilizing rhizosphere fungi would be supplied with nutrients from the plant root (Martin, 1977; Rovira et al., 1979; Whipps, 1984; Rovira, 1991).

Table IV Plant Growth Promotion and Enhancement of Soil P Availability by Phosphate-Solubilizing Fungi Microorganism

Soil type or growth mediuma

Aspergillus niger

Hydroponics glasshouse experiment: nutrient solution, sand

Aspergillus awamori

Greenhouse experiment: alluvial soil (alkaline) from Delhi, India. Soil available P: 9 mg kg1, pH unknown

Penicillium digitatum A. awamori (isolated from soil, Tamil Nadu, India)

Greenhouse experiment: red soil

134

A. awamori with Pseudomonas striata

Field trial: IARI, New Delhi, India

A. niger van Tieghen (isolated from Lateritic soil, India)

Flask culture experiment: typic Ochragual soil, Kapgari, West Bengal, India. Soil available P: 7 mg P kg1, pH 5.4 Greenhouse experiment: sandy loam, India. Soil available P: 3 mg P kg1, pH 5.6

A. niger [isolated from rhizosphere of cowpea (Vignia unguiculata (L.) alp., India] A. niger with Bacillus mobilis (N fixing bacterium, isolated from rhizosphere of cotton, India) A. niger with Bacillus mobilis and Glomus fasciculatus

Plant American elm (Ulmus americana) harvested at 3 months Wheat (Triticum aestivum Kalyan sona) harvested at maturity Groundnut (Arachis hypogaea L.) harvested at 30 days

Wheat (Sonalika and HD2122) harvested at maturity No plant

Onion (Allium cepa var. chikkaballapur red.)

P source

Yield increase (dry weight) (%)

Increase in P (and N) uptake

RP

Mussoorie RP (219 kg P ha1)

84 (grain)

83% (grain P)

Gaur (1972)

80 (grain) ND

98% (grain P) 30%b (total plant P)

Vidhyasekaran et al. (1973)

ND

383%b (total plant P)

16 (straw) 11 (grain)

ND

Gaur et al. (1980)

ND

2 mg P kg1 (soil available P)

Banik and Dey (1981b)

24 (aboveground) NS (root) 62 (bulb fresh weight)

24% (shoot P) 51% more VAM spores in soil

Manjunath et al. (1981)

25 (aboveground) 16 (roots) 88 (bulb fresh weight)

25% (shoot P) 44% (shoot N) 54% more VAM spores in soil

57 (shoot) 82 (roots) 155 (bulb fresh weight)

48% (shoot P) 71% (shoot N) 84% (root P) 125% more VAM sproes in soil

Nil P fertilizer

Farmyard manure (10 tons ha1) 44 kg P ha1 Mussoorie RP and paddy straw Nil P fertilizer

Nil P fertilizer

(aboveground)

900%b,c

Reference

600b,c

(aboveground P)

Rosendahl (1942)

Aspergillus fumigatusd [isolated from a Gangetic alluvial, (Fluvaquent) soil, West Bengal, India; pH 7.4] Aspergillus candidus (isolated from a Gangetic alluvial soil, West Bengal, India; pH 7.4) A. awamori

Greenhouse experiment: partially sterilized Gangetic alluvial soil (Fluvaquent) silty clay, West Bengal, India. Soil available P: 7 mg P kg1, pH 7.4

Penicillium bilaii (isolated from soil, Canada)

Greenhouse experiment: brown Chernozemic loamy sand from Alberta, Canda. Soil sterilized and allowed to reequilibrate for 6 weeks. Soil available P: 2 mg kg1, pH 7.2

Field trial

135 Penicillium bilaii Field trial: orthic brown Chernozemic clay loam, Lethbridge, Canada. Soil available P: 2 mg kg 1, pH unknown

P. bilaii plus mixed culture of VAM fungi

P. bilaii

Greenhouse experiment: brown Cherozemic loamy sand, Alberta, Canada. Soil sterilized to kill native VAM fungi. Soil available P: 2 mg kg1, pH 7.2 Greenhouse experiment: Orthic Brown Chernozem from Canada. Soil available P: 3 mg kg1, pH 8.0

No plant

RP (30 kg P ha1) and farmyard manure (40 kgN ha1)

ND

3 kg P ha1 (soil available P)

No plant

RP (30 kg P ha1) and farmyard manure (40 kgN ha1)

ND

NS (soil available P)

Soybean harvested at maturity Field beans (Phaseolus vulgaris) harvested at maturity

Nil P fertilizer

NS (grain)

44% (grain P)

Gaur (1985)

Idaho RP (45 mg P kg1)

10.3 (aboveground)

NS (aboveground P)

Kucey (1987)

Field beans (P. vulgaris) harvested at maturity Wheat (Chester) harvested at maturity Wheat (Chester) harvested at maturity Wheat (Chester) harvested at maturity

Nil P fertilizer

53 (aboveground)

31% (aboveground P)

Idaho RP (45 mg P kg1 soil) Nil P fertilizer

21 (aboveground)

9% decrease (aboveground P)

5 (aboveground)

3% (aboveground P)

Straw and Idaho RP (20 kg P ha1)

25 (grain) 9 (straw)

31% (aboveground P)

Wheat (Chester) harvested at maturity Wheat (Chester) harvested at maturity

Nil P fertilizer

27 (grain) 17 (straw) 39 (aboveground)

34% (aboveground P)

Wheat (Neepawa) harvested at the early heading stage

Idaho RP (20 mg P kg1 soil)

28% (aboveground)

25% (aboveground P)

Straw and Idaho RP (45 mg P kg1 soil)

Banik and Dey (1982)

46% (aboveground P)

Asea et al. (1988)

continues

Table IV—Continued Microorganism

Soil type or growth mediuma

P. bilaii

Greenhouse experiment: dark brown Chernozemic clay loam soil, Lethbridge, Alberta, Canada. Soil available P: 4 mg kg1, pH 7.7

136

Field trial: orthic dark brown Chernozemic clay loam soil, Lethbridge, Alberta, Canada. Soil available P: 4 mg kg1, pH 7.7

P. bilaii

Greenhouse experiment: brown Chernozem soil (Loamy sand), Canada. Soil available P: 2 mg kg1, pH 7.2

P. bilaii

Penicillium sp. (isolated from rhizosphere of tomato, eggplant, or cucumber, Baghdad, Iraq)

Greenhouse experiment: calcareous soil (typic Torrifluvent) from Iraq. Soil available P: 4 mg kg1, pH8.2

Plant

P source

Yield increase (dry weight) (%)

Increase in P (and N) uptake

Wheat (Neepawa) harvested at the early heading stage Wheat (Chester) harvested at 8 weeks

Nil P fertilizer

35% (aboveground)

27% (aboveground P)

Idaho RP (700 mg P kg1 soil)

93 (aboveground)

Wheat (Chester) harvested at 8 weeks

Nil P fertilizer

86 (aboveground)

Wheat (Chester) harvested at maturity

RP (20 kg P ha1)

11 (grain) 4 (straw)

47% (aboveground P) 2 mg kg1 (soil available P) 214% more P-solubilizing fungi in the rhizosphere 73% (aboveground P) 2 mg kg1 (soil available p) 283% more P-solubilizing fungi in rhizosphere 8% (aboveground P)

Wheat (Chester) harvested at maturity Canola (Brassica napus L. “Westar”) harvested at maturity

Nil P fertilizer

6 (straw) 10 (grain) NS (straw or pods)

36% (aboveground P)

MAP (20 mg P kg1) nil P fertilizer

NS (straw or pods)

19% (aboveground P)

NS (straw or pods)

62% (aboveground P)

RP from Akashatmine, Iraq (45 mg P kg1)

11 (aboveground)

17% (aboveground P) 2 mg kg1 (soil available P)

TSP (45 mg P kg1)

16 (aboveground)

8% (aboveground P) 9 mg kg1 (soil available P)

Wheat (Chester) harvested at 8 weeks Wheat (Chester) harvested at 8 weeks Sorghum (Sorghum bicolor Moench) harvested at 70 days

Wheat (Chester) harvested at 8 weeks

Florida RP (20 mg P kg1)

Reference

Kucey (1988)

15% (aboveground P) Kucey and Leggett (1989)

Salih et al. (1989)

Aspergillus foetidus (isolated from rhizosphere of tomato, eggplant, or cucumber, Baghdad, Iraq)

A. awamori

P. bilaii

137 P. bilaii and Rhizobium leguminosarum P. bilaii

P. bilaii

Greenhouse experiment: calcareous soil (typic Torrifluvent) from Iraq. Soil available P: 4 mg kg1, pH 8.2

Field trial: alluvial loamy soil (Pura Farm). Soil available P: 6 kg ha1, pH 7.5

Field trial: moderate alkaline loamy soil (Chakeri farm). Soil available P: 12 kg P ha1, pH 8.6 Pot trial: Shellbrook orthic gray-black Chernozemic fine sandy loam from Porcupine Plain, Saskatchewan, Canada. Soil available P: 4 mg kg1, pH 7.3

Field trials: 37 locations, 3 Canadian provinces, over 3 years. Low to medium available P soils (0–22 kg P ha1).

Greenhouse experiment: brown Chernozemic soil (loamy sand) from Saskatchewan, Canada. Soil available P: 13.5 mg kg1, pH 7.6

Sorghum (Moench) harvested at 70 days

RP from Akashat mine, Iraq (45 mg P kg1)

13 (aboveground)

19% (aboveground P) 1 mg kg1 (soil available P)

Sorghum (Moench) harvested at 70 days

TSP (45 mg P kg1)

4 (aboveground)

Rice (Saket-4) harvested at maturity

RP (13 kg P ha1) and TSP (13 kg P ha1) RP (26 kg P ha1)

12 (grain)

4% (aboveground P)  5 mg kg1 (soil available P) ND

RP (26 kg P ha1)

NS (grain)

110% more rhizobium nodules ND

RP (26 kg P ha1)

14 (grain)

ND

Peas (Pisum sativum) harvested at 43 days

Nil P fertilizer

22 (roots plus aboveground)

NS (roots plus aboveground P)

Peas harvested at 43 days

TSP (44 kg P ha1)

NS (aboveground)

NS (aboveground P)

17% decrease cf. Rhizobium alone (roots plus aboveground) 2 (grain)

ND

Chickpea (T3) harvested at maturity Wheat (HD-1553) harvested at maturity Wheat (HD-1553) harvested at maturity

23 (rain)

Wheat harvested at maturity

TSP (4.4 kg P ha1)

Wheat harvested at maturity Pea (Trapper) harvested as first flower bud opened at 41–51 days

Nil P fertilizer

2 (grain)

ND

Nil P fertilizer

48 (aboveground)

39% (aboveground P) 55% (aboveground N)

Tiwari et al. (1989)

Downey and van Kessel (1990)

Gleddie et al. (1991)

Gleddie (1993)

continues

Table IV—Continued Microorganism

Soil type or growth mediuma

P. bilaii

Field trials: four P-fertilizer responsive locations, 1989 and 1991. Thin black and black soil zones, Western Canada. Available soil P: “low to medium levels”

A. awamori

138

Greenhouse experiment: Patharchatta sandy loam (typic Hapludoll), India. Soil available P: 27 mg kg1, pH 6.2

Plant

Field trial: moderately alkaline soil (loam), Chakeri, Kanpur (U.P.),

Increase in P (and N) uptake

TSP (10 mg P kg1)

12 (aboveground)

18% (aboveground P) NS (aboveground N)

Nil P fertilizer

NS (aboveground)

NS (aboveground P) 7% (aboveground N)

Pea (Trapper) harvested at 8 weeks after emergence Pea (Trapper) harvested at 8 weeks after emergence Soybean (Glycine max, L. merr.) (Bragg) harvested at maturity

TSP (4.4 kg P ha1)

16 (aboveground)

18% (aboveground P) 19% (aboveground N)

TSP (8.7 kg P ha1)

NS (aboveground)

7% (aboveground) NS (aboveground N)

Nil P fertilizer

34 (straw) 44 (grain)

Mussoorie RP (86 mg P kg1)

10 (straw) 5 (grain)

Nil P fertilizer

24 (grain) compared to inoculation by Bradyrhizobium sp. alone

Wheat (HD 1553) harvested at maturity

Nil P fertilizer

NS (grain) 12 (straw)

58% (grain P) 8 mg P kg1 (soil available P) 40% (N uptake) 100% (nodule dry weight) 5900% more P-solubilizing microbes in soil 13% (grain P) 3 mg P kg1 (soil available P) 11% (N uptake) NS (nodule dry weight) 271% more P-solubilizing microbes in soil 35% (grain P) NS (soil available P) 21% (grain N) NS (nodule dry weight) ND

Wheat (HD1553) harvested at maturity Wheat (HD1553) harvested at maturity

Mussoorie RP (26 kg P ha1) Nil P fertilizer

NS (grain or straw)

ND

8 (grain) 7 (straw)

ND

A. awamori and Bradyrhizobium sp.

Field trial: alluvial loamy soil at Pura, India. Soil available P: 6.3 kg ha1, pH 7.5

Yield increase (dry weight) (%)

Pea (Trapper) harvested as first flower bud opened at 41–51 days Pea (Trapper) harvested at 8 weeks after emergence

Soybean (G. max, L. merr.) (Bragg) harvested at maturity

A. awamori

P source

Reference

Singh and Singh (1993)

Tiwari et al. (1993)

India. Soil available P: 12 kg ha1, pH 8.6

Aspergillus sp. (KAR0210)

Aspergillus sp. (KAR0210) Penicillium glaucum (HE4) (isolated from sunflower rhizoshpere, Bangalore, India) Penicillium aurantiogriseum (isolated from forest soil, Austria)

139 Paecilomyces fussiporus Aspergillus sp.

A. awamori

Greenhouse experiment: autoclaved Sang Hyang Damar Ultisol soil. Soil available P: 0.5 mg P kg1

Greenhouse experiment: Rajamandala Ultisols soil. Soil available P: 37 kg ha1, pH 3.9 Greenhouse experiment: red soil GKVK farm, Bangalore, India. 1% farmyard manure. Soil available P: 7 mg kg1, pH unknown Greenhouse experiment (no plant): soil from Austrian alps. Soil available P: 0.6 or 0.7 g P dm1, pH 4.2 or 4.6. Soil amended with 0.2% glucose, 0.2% sucrose, 0.002% N (from NH  4) Field trial: clayey soil, Junagadh, India. Soil available P: 31 kg P ha1, pH 7.9 Field trial: sandy loam Bangalore, India. Soil available P: 14 kg P ha1, pH 6.2

Greenhouse experiment: soil, India. Soil available P: 1.2 kg P ha1, pH 7.1

Wheat (HD1553) harvested at maturity No plant (15 days)

Mussoorie RP (26 kg P ha1) Nil P fertilizer

NS (grain or straw)

ND

ND

14 mg P kg1 (soil available P)

No plant (30 days)

RP (300 mg P kg1) RP (300 mg P kg1)

ND

38 mg P kg1 (soil available P) 24 mg P kg1 (soil available P)

Sunflower (Helianthus annuus L.) (BSH-1)

SSP (82 mg P kg1 soil)

NS (plant height, 60 days) NS (leaf area, 45 days) 35 (seed yield at harvest)

NS (shoot P at harvest) 47% more P-solubilizing fungi in the rhizosphere

Gururaj and Mallikarjunaiah (1995)

No plant

Nil P fertilizer

ND

2 ng P ml1 soil solution (soil available P) (No significant difference when soil not amended with glucose, sucrose, or N)

Illmer and Schinner (1995b)

Groundnut (Arachis hypogaea)

Nil P fertilizer

NS (pod)

Mehta et al. (1996)

Soybean (Hardee) harvested at maturity

RP (17.5 kg P ha1)

7 (seed) 7 (oil)

19% (P in kernel) 24% (P in haulm) 8% (nodule dry weight) NS (protein)

RP (35 kg P ha1) RP (17.5 kg P ha1) and SSP (17.5 kg P ha1) Nil P fertilizer

5 (seed) 8 (oil) 12 (seed) 12 (oil)

NS (protein)

34 (aboveground)

36% (aboveground P) 206% (nodule dry weight)

SSP (20 kg P ha1)

14 (aboveground)

24% (aboveground P) 22% (nodule dry weight)

No plant (14 days)

Gram (JG 315) (legume) harvested at flowerinitiation stage Gram (JG 315) (legume) harvested at flowerinitiation stage

ND

Goenadi (1995)

Goenadi et al. (1995)

Thimmegowda and Devakumar (1996)

7% (protein)

Vaishya et al. (1996)

continues

Table IV—Continued Microorganism

Soil type or growth mediuma

A. awamori

Penicillium radicum (isolated from wheat rhizosphere, Australia)

Greenhouse experiment: red earth soil. Soil available P: 17 mg kg1 pH 4.6

140

Field trial: red earth soil, Wagga Wagga, NSW, Australia. Soil available P: 16 mg kg1, pH 4.9

Plant

P source

Yield increase (dry weight) (%)

Increase in P (and N) uptake

Gram (JG 315) (legume) harvested at flowerinitiation stage Gram (JG 315) (legume) harvested at flowerinitiation stage Wheat (Dollarbird) harvested at maturity at 20 weeks

SSP (40 kg P ha1)

13 (aboveground)

52% (aboveground P) 22% (nodule dry weight)

RP (17 kg P ha1)

NS (above ground)

38% (aboveground P) 20% (nodule dry weight)

Nil P fertilizer

26 (grain) 14 (aboveground)

23% (grain protein) NS (grain P)

Wheat (Dollarbird) harvested at maturity at 20 weeks Wheat (Dollarbird) harvested at maturity at 20 weeks Wheat (Dollarbird) harvested at maturity at 28 weeks Wheat (Dollarbird) harvested at maturity at 28 weeks Wheat (Dollarbird) harvested at maturity at 28 weeks Wheat (Dollarbird) harvested at maturity at 28 weeks

KH2PO4 (5 kg P ha1)

10 (grain)

NS (grain P)

KH2PO4 (15 kg P ha1)

15 (grain)

20% (grain P) 15% (grain protein)

Nil P fertilizer

NS

NS

SSP (5 kg P ha1)

18 (grain)

18% (grain protein)

SSP (15 kg P ha1)

25 (grain)

33% (grain protein)

SSP (20 kg P ha1)

16 (grain)

NS

Reference

Whitelaw et al. (1997)

Note: Abbreviations used: MAP, monoammonium phosphate; ND, not determined; NS, not statistically different; RP, rock phosphate; SSP, single superphosphate [mixture of Ca(H2PO4)2 and CaSO4 produced by the action of H2SO4 on RP]; TSP, triple superphosphate [Ca(H2PO4)2 “monocalcium phosphate” produced by action of H3PO4 on RP]; VAM, vesicular arbuscular mycorrhizal fungi. a“Soil-available P” determined by extraction with NaHCO unless otherwise indicated. 3 bStatistical significance not given. cCompared to sterile plants. dHuman pathogen.

PHOSPHATE-SOLUBILIZING FUNGI

141

Growth promotion and increased P uptake by plants inoculated with P-solubilizing fungi have been reported by many investigators (Table IV). Many of the studies reported in Table IV have investigated the ability of P-solubilizing fungi to promote P uptake and plant growth in soil under greenhouse conditions. Under these conditions, rooting volumes are usually restricted so that if microbial P solubilization does take place, the plant response may be higher than that in field trials (Kucey et al., 1989). In an early sand and nutrient solution greenhouse study, inoculation of American elm with A. niger was able to increase the yield and P uptake by 600 and 900%, respectively, but in this study inoculated plants were compared to plants in a sterile medium (Rosendahl, 1942). Illmer and Schinner (1995b) point out that nearly all rhizosphere microorganisms, not only those which solubilize P, increase the nutrient supply of plants. This means it is advisable to use nonsterile soil to ensure that uninoculated control plants have adequate nutrition. In soil greenhouse trials, yield and/or P uptake has been increased by inoculation of wheat, onion, sorghum, soybean, and gram with P-solubilizing fungi (Gaur, 1972; Manjunath et al., 1981; Salih et al., 1989; Singh and Singh, 1993; Vaishya et al., 1996; Whitelaw et al., 1999). Yield alone was also increased by inoculation of sunflower and soybean (Gururaj and Mallikarjunaiah, 1995; Thimmegowda and Devakumar, 1996). Penicillium bilaii (ATCC strain No. 20851) is commercially available under the trade name Provide and has consistently increased grain yield and P uptake by wheat grown on neutral or alkaline soils under greenhouse conditions. Penicillium bilaii has also increased the yield and P uptake by field beans and peas and the uptake of P by canola (Kucey, 1987, 1988; Asea et al., 1988; Kucey and Leggett, 1989; Downey and Van Kessel, 1990; Gleddie, 1993) (Table IV). Growth promotion of plants by P-solubilizing fungi under field conditions has also been reported by many investigators. P uptake was increased in groundnut inoculated with Paecilomyces fussiporus (Mehta et al., 1996), whereas both wheat yields and P uptake were increased by inoculation with P. bilaii (Kucey, 1987, 1988). Wheat yield has been increased by inoculation with P. bilaii, A. awamori, and P. radicum (Gleddie et al., 1991; Tiwari et al., 1989, 1993; Whitelaw et al., 1999) and yields of pea, rice, chickpea, and soybean have been increased by inoculation with P-solubilizing fungi (Gleddie et al., 1991; Gleddie, 1993; Tiwari et al., 1989; Thimmegowda and Devakumar, 1996) (Table IV). Some studies of P-solubilizing fungi included the effect of P-solubilizing bacteria in a mixed inoculum. A mixed inoculum of A. awamori with Pseudomonas striata increased the yield of wheat under field conditions (Gaur et al., 1980). VAM fungi, which are known to enhance the ability of the host plant to absorb P, have also been included in mixed inocula. The individual growth-promoting effects of P-solubilizing fungi, bacteria, and VAM fungi have been reported to be additive. An inoculum consisting of a mixture of A. niger, Bacillus mobilis, and the VAM

142

M. A. WHITELAW

fungus Glomus fasciculatus increased the P content of shoots and roots and the bulb weight of onion grown on unsterilized soil to a greater extent than the individual microbial components (Manjunath et al., 1981). Wheat plants inoculated with both P. bilaii and VAM fungi and grown on soil which was sterilized to kill native VAM fungi received greater benefit from rock phosphate addition than plants receiving only one of the microorganisms (Kucey, 1987) (Table IV). Many studies have been made on the effect of P-solubilizing fungi on legumes. Fungi such as unknown species of Penicillium, Cephalosporium, and Alternaria and Penicillium lilacinum, A. niger, A. flavus, and A. terreus were associated with legume root nodules and were capable of solubilizing Ca3(PO4)2 (Subba-Rao and Bajpai, 1965; Chhonkar and Subba-Rao, 1967) (Table I). Increased soil P availability is known to increase nodulation in legumes (Jones et al., 1977) and in some studies inoculation of legumes with P-solubilizing fungi increased nodulation and N uptake (Gleddie, 1993; Tiwari et al., 1989; Singh and Singh, 1993; Mehta et al., 1996; Vaishya et al., 1996) (Table IV). The effect of inoculation of unplanted soil on the NaHCO3-extractable P or “soil available P” levels was investigated in several greenhouse studies. Increases in soil available P in unplanted soil were reported by Banik and Dey (1981b) with A. niger, Goenadi et al. (1995) with an unidentified Aspergillus sp., and Illmer and Schinner (1995b) with P. aurantiogriseum. Soil available P was also determined in some plant growth studies. Increases in the levels of plant available P found in the soil after wheat, sorghum, and soybean experiments were reported by Kucey (1988), Salih et al., (1989), and Singh and Singh (1993) (Table IV). Inoculation of plants with P-solubilizing fungi was sometimes reported to encourage the proliferation of other P-solubilizing fungi in the rhizosphere. Kucey (1988) found that after 8 weeks of growth, the total number of P-solubilizing fungal isolates from the rhizosphere of wheat inoculated with P. bilaii was 283% higher than in uninoculated plants. Gururaj and Mallikarjunaiah (1995) also reported an increase (47%) in the number of P-solubilizing fungi in the rhizosphere of sunflower inoculated with Penicillium glaucum at harvest. The relative benefits of inoculation with P-solubilizing fungi have been observed to decrease as the soil P availability increases. Downey and Van Kessel (1990) and Gleddie (1993) found that the response to inoculation of pea with P. bilaii under greenhouse conditions depended on whether fertilizer P had been added to the soil. When soluble P fertilizer was added, the response to inoculation was lower than when no P was added. Whitelaw et al. (1999) also found that the response to inoculation of wheat with P. radicum under greenhouse conditions was lower when fertilizer P was added. Gleddie et al. (1991) reported results of 55 separate field trials of wheat inoculated with P. bilaii and found that in soils which had high available P levels, neither P fertilization nor inoculation induced a yield response. Salih et al. (1989) reported higher responses in yield and P uptake to inoculation of sorghum with an unidentified Penicillium sp. or with A. foetidus in

PHOSPHATE-SOLUBILIZING FUNGI

143

soil treated with rock phosphate in comparison with soil treated with triple superphosphate. They surmised that this was probably due to the presence of adequate quantities of readily available P in the soil treated with triple superphosphate, whereas available P in the soil treated with rock phosphate appeared to be the limiting factor for plant growth. Many greenhouse or field experiments have investigated plant growth promotion by P-solubilizing microorganisms using only a single level of P application. Abbott and Robson (1984), while discussing VAM fungi which are capable of increasing P availability to plants, state that there are advantages in studying a complete P response curve (i.e., a full range of applied P with a low and at least two high P application levels resulting in a plateau defining maximum yield response to P). The study of the complete response curve is important to demonstrate that the effect of the inoculum can be eliminated by applying P fertilizer, thus implying that the P fertilizer has replaced the advantage gained by the P-solubilizing effect of the inoculum (Abbott and Robson, 1984). Where less data are available (e.g., less than a full response curve), confirmation of a negative interaction between inoculation and the P rate is evidence for a P-solubilizing effect but it is not possible to separate alleviation of P deficiency from a combination of alleviation of P deficiency and other plant growth-promotion mechanisms. Three classes of response to the inoculation and P application treatments need to be considered when interpreting the data in the previously discussed experiments. First, if the effect of inoculation on plant growth is caused by the alleviation of P deficiency as a result of fungal-induced solubilization of soil inorganic phosphates, then application of P fertilizer at a rate sufficient to eliminate the deficiency should eliminate the response to inoculation (i.e., a negative interaction between inoculation and P application rate would be expected). Second, if the effects of inoculation are other than the alleviation of P deficiency, then the response should not be eliminated by P fertilizer. Therefore, a negative interaction between inoculation and P application rate would not be expected and there would be a positive yield response to inoculation no matter how high the P application rate. Third, if both P-solubilizing and other mechanisms are present, then a negative interaction between inoculation and P application rate could be expected but a difference in yield would still be expected at P rates high enough to eliminate P-deficiency effects.

VI. CONCLUSION Phosphorus is an important plant nutrient which is in short supply in many agricultural soils. Because a large percentage of phosphatic fertilizer is fixed by soil and thus rendered less available to plants, the long-term application of P fertiliz-

144

M. A. WHITELAW

ers has resulted in an accumulation of total soil P, most of which is poorly soluble. Many soil fungi, predominantly of the genera Aspergillus and Penicillium, have been shown to possess the ability to solubilize sparingly soluble phosphates in vitro by secreting inorganic or organic acids. The microorganisms and soils reported here are very diverse and it is difficult to generalize about the success of plant growth promotion by P-solubilizing soil fungi. The relationship between soil pH and phosphate solubility is not a simple one, and controversy exists over the effect of increasing or decreasing pH on P solubility in soil. However, in unbuffered liquid media studies mentioned previously, P solubilization appeared to often be associated with a lowering of pH. Some of the fungi tested appeared to solubilize P in soils and in unbuffered liquid media. Field and greenhouse trials have demonstrated that inoculation of plants with some P-solubilizing fungi increased the concentration of “available” P in the soil and enhanced the yield and P uptake by the plant. Continued work in this area could yield plant fungal inoculants capable of more consistent performance over a range of soil and climatic conditions.

ACKNOWLEDGMENTS I thank Associate Professor Terence Harden of the School of Wine and Food Sciences, Charles Sturt University, Wagga Wagga, and Dr. Mark Conyers of N.S.W. Agriculture, Wagga, Wagga, N.S.W. Australia, for critically reviewing the manuscript.

REFERENCES Abbott, L. K., and Robson, A. D. (1984). The effect of VA Mycorrhizae on plant growth. In “VA Mycorrhiza” (C. L. Powell and D. J. Bagyaraj, Eds.) pp. 113 –130. CRC Press, Boca Raton, FL. Agnihotri, V. P. (1970). Solubilization of insoluble phosphates by some soil fungi isolated from nursery seed beds. Can. J. Microbiol. 16, 877– 880. Ahmad, N., and Jha, K. K. (1968). Solubilization of rock phosphate by micro-organisms isolated from Bihar soils. J. Gen. Appl. Microbiol. 14, 89 – 95. Albert, A., and Serjeant, E. P. (1984). Chelation and the stability constants of metal complexes. In “The Determination of Ionization Constants—A Laboratory Manual” (A. Albert and E. P. Serjeant, Eds.), 3rd ed. Chapman & Hall, London. Armstrong, R. D., Helyar, K. R., and Prangnell, R. (1993). Direct assessment of mineral phosphorus availability to tropical crops using 32P labelled compounds. Plant Soil 50, 279 –287. Arora, D., and Gaur, C. (1979). Microbial solubilization of different inorganic phosphates. Indian J. Exp. Biol. 17, 1258 –1261. Asea, P. E. A., Kucey, R. M. N., and Stewart, J. W. B. (1988). Inorganic phosphate solubilisation by 2 Penicillium species in solution culture and soil. Soil Biol. Biochem. 20, 459 – 464. Aylward, G., and Findlay, T. (1994). “SI Chemical Data,” 3rd ed. Wiley, Brisbane. Bache, B. W. (1963). Aluminium and iron phosphate studies relating to soils. I. Solution and hydrolysis of variscite and strengite. J. Soil Sci. 14, 113 –123. Banik, S., and Dey, B. K. (1981a). Phosphate-solubilizing microorganisms of a lateritic soil. (I). Solubilization of inorganic phosphates and production of organic acids by microorganisms, isolated

PHOSPHATE-SOLUBILIZING FUNGI

145

in sucrose calcium phosphate agar plates. Zentralblatt Bakteriol. Parasitenkunde Infektionskarankheiten, Hygiene. 2. Naturwiss. Miikrobiol. Landwirtsch 136, 478 – 486. Banik, S., and Dey, B. K. (1981b). Phosphate-solubilizing microorganisms of a lateritic soil. (II). Effect of some tricalcium phosphate-solubilizing microorganisms on available phosphorus content of the soil. Zentralblatt Bakteriol. Parasitenkunde Infektionskarankheiten, Hygiene. 2. Naturwiss. Miikrobiol. Landwirtsch 136, 487– 492. Banik, S., and Dey, B. K. (1982). Available phosphate content of an alluvial soil as influenced by inoculation of some isolated phosphate-solubilizing micro-organisms. Plant Soil 69, 353 – 364. Banik, S., and Dey, B. K. (1983). Alluvial soil microorganisms capable of utilizing insoluble aluminium phosphate as a source of phosphorus. Zentralblatt Mikrobiol. 138, 437– 442. Bardiya, M. C., and Gaur, A. C. (1974). Isolation and screening of microorganisms dissolving low-grade rock phosphate. Folia Microbiolo. (Praha. Acad. Scientiarium Bohemoslovaca) 19, 386–389. Barrow, N. J. (1984). Modelling the effects of pH on phosphate sorption by soils. J. Soil Sci. 35, 283 – 297. Barrow, N. J. (1987). “Reactions with Variable-Charge Soils.” Nijhoff, Dordrecht. Barrow, N. J. (1990). Chemical processes and management systems. In “Phosphorus Requirements for Sustainable Agriculture in Asia and Oceania,” pp. 199 –209. International Rice Research Institute, Manilla. Barthakur, H. P. (1978). Solubilization of relatively insoluble phosphate by some fungi isolated from the rhizosphere of rice. Indian J. Agric. Sci. 48, 762–766. Bar-Yosef, B. (1991). Root excretions and their environmental effects: Influence on availability of phosphorus. In “Plant Roots: The Hidden Half ” (Y. Waisel, A. Eshel, and U. Kafkafi, Eds.), pp. 529–557. Dekker, New York. Berry, D. R. (1975). The environmental control of the physiology of filamentous fungi. In “The Filamentous Fungi, Vol. 1, Industrial Mycology” (J. E. Smith and D. R. Berry, Eds.), pp. 16–32. Edward Arnold, UK. Bolan, N. S. (1991). A critical review on the role of mycorrhizal fungi in the uptake of phosphorus by plants. Plant Soil 134, 189 –207. Bolan, N. S., Naidu, R., Mahimairaja, S., and Baskaran, S. (1994). Influence of low-molecular-weight organic acids on the solubilization of phosphates. Biol. Fertil. Soils 18, 311– 319. Cerezine, P. C., Nahas, E., and Banzatto, D. A. (1988). Soluble phosphate accumulation by Aspergillus niger from fluoroapatite. Appl. Microbiol. Biotechnol. 29, 501– 505. Chabot, R., Cescas, M. P., and Antoun, H. (1993). Microbiological solubilization of inorganic P-fraction normally encountered in soils. Phosphorus Sulfur Silicon 77, 329. Chhonkar, P. K., and Subba-Rao, N. S. (1967). Phosphate solubilization by fungi associated with legume root nodules. Can. J. Microbiol. 13, 749 –753. Cochrane, V. W. (1958). “Physiology of Fungi.” Wiley, New York. Colwell, J. D. (1963). The estimation of the P requirements of wheat in southern NSW by soil analysis. Austr. J. Exp. Agric. Anim. Husbandry 3, 190 –197. Corman, J., Tsuchiya, H. M., Koepsell, H. J., Benedict, R. G., Kelley, S. E., Feger, V. H., Dworschack, R. G., and Jackson, R. W. (1957). Oxygen absorption rates in laboratory and pilot plant equipment. Appl. Microbiol. 5, 313 – 318. Cosgrove, D. J. (1977). Microbial transformations in the phosphorus cycle. In “Advances in Microbial Ecology” (M. Alexander, Ed.). Plenum, New York. Costin, A. B., and Williams, C. H. (1983). “Phosphorus in Australia.” Centre for Resource and Environmental Studies, A.N.U., Canberra, Australia. Cunningham, J. E., and Kuiack, C. (1992). Production of citric and oxalic acids and solubilization of calcium phosphate by Penicillium bilaii. Appl. Environ. Microbiol. 58, 1451–1458. Darmwal, N. S., Singh, R. B., and Rai, R. (1989). Isolation of phosphate solubilizers from different sources. Curr. Sci. 58, 570 – 571.

146

M. A. WHITELAW

Das, A. C. (1963). Utilisation of insoluble phosphates by soil fungi. J. Indian Soc. Soil Sci. 11, 203 – 207. De La Torre, M. A., Gomez-Alarcon, G., Vizcaino, C., and Garcia, M. T. (1993). Biochemical mechanisms of stone alteration carried out by filamentous fungi living in monuments. Biogeochemistry 19, 129–147. Deming, M. E., and Cate, W. E. (1963). Preparation of variscite. Soil Sci. 95, 206 –207. Downey, J., and Van Kessel, C. (1990). Dual inoculation of Pisum sativum with Rhizobium leguminosarum and Penicillium bilaji. Biol. Fertil. Soils 10, 194 –196. Earl, K. D., Syers, J. K., and McLaughlin, J. R. (1979). Origin of the effects of citrate, tartrate and acetate on phosphate sorption by soils and synthetic gels. Soil Sci. Soc. Am. J. 43, 674 – 678. Eckhardt, F. E. W. (1980). Microbial degradation of silicates. Release of cations from alumino silicate minerals by yeasts and filamentous fungi. In “Biodeterioration.” Proceedings of the Fourth International Symposium, Berlin (T. A. Oxley, D. Allsopp, and G. Becker, Eds.), Pitman & Biodeterioration Society, London. Fox, T. R., Comerford, N. B., and McFee, W. W. (1990). Phosphorus and aluminium release from a spodic horizon mediated by organic acids. Soil Sci. Am. J. 54, 1763 –1767. Gardner, W. K., Parbery, D. G., and Barber, D. A. (1982). The acquisition of phosphorus by Lupinus albus L. I. Some characteristics of the soil–root interface. Plant Soil 68, 19 – 32. Gardner, W. K., Barber, D. A., and Parbery, D. G. (1983). The acquisition of phosphorus by Lupinus albus L. III. The probable mechanism by which phosphorus movement in the soil–root interface is enhanced. Plant Soil 70, 107–124. Gaur, A. C. (1972). Role of phosphate solubilizing microorganism and organic matter in soil productivity. Symp. Soil Prod. Natl. Acad. Sci. 259 –268. Gaur, A. C. (1985). Proc. Soil Biol. Symp. Hisar, p. 125; quoted in Kapoor, K. K., Mishra, M. M., and Kukreja, K. (1989). Phosphate solubilisation by soil microorganisms. Indian J. Microbiol. 29, 119–127. Gaur, A. C., Madan, M., and Ostwal, K. (1973). Solubilization of phosphatic compounds by native microflora of rock phosphates. Indian J. Exp. Biol. 11, 427– 429. Gaur, A. C., Mathur, R. S., and Sadasivam, K. V. (1980). Effect of organic materials and phosphatedissolving culture on the yield of wheat and greengram. Indian J. Agron. 25, 501– 503. Gerretsen, F. C. (1948). The influence of microorganisms on the phosphate intake by the plant. Plant Soil 1, 51–81. Gleddie, S. C. (1993). Response of pea and lentil to inoculation with the phosphate-solubilizing fungus Penicillium bilaii (Provide). In “Proceedings of the Soils & Crops Workshop,” Saskatoon, Saskatchewan, 25 –26, February, 1993, pp. 47– 52. Gleddie, S. C., Hnatowich, G. L., and Polonenko, D. R. (1991). A summary of wheat response to Provide (Penicillium bilaji) in Western Canada. In “Proceedings of the Alberta Soil Science Workshop,” Lethbridge, Alberta, 19 –21 February, 1991, pp. 306 – 313. Goenadi, D. H. (1995). Suitability of selected mixtures of clay minerals with humic substances as carrier of phosphate-solubilizing microbes. Menara Perkebunan 63, 102–113. Goenadi, D. H., and Saraswati, R. (1993). Phosphate-solubilizing capabilities of selected phosphatesolubilizing fungal isolates. Menara Perkebunan 61, 61– 66. Goenadi, D. H., Saraswati, R., and Nganro, N. N. (1995). Nutrient-solubilizing and aggregate-stabilizing microbes isolated from selected humic tropical soils. Menara Perkebunan 63, 60 – 66. Goldstein, A. H. (1986). Bacterial solubilization of mineral phosphates: Historical perspective and future prospects. Am. J. Alternative Agric. 1, 51– 57. Grierson, P. F. (1992). Organic acids in the rhizosphere of Banksia integrifolia L.F. Plant Soil 144, 259–265. Grierson, P. F., and Attiwill, P. M. (1989). Chemical characteristics of the proteoid root mat of Banksia integrifolia L.F. Austr. J. Bot. 37, 137–143.

PHOSPHATE-SOLUBILIZING FUNGI

147

Gupta, R., Rekha, S., Aparna, S., and Kuhad, R. C. (1994). A modified plate assay for screening phosphate solubilizing microorganisms. J. Gen. Appl. Microbiol. 40, 255 –260. Gururaj, R., and Mallikarjunaiah, R. R. (1995). Interactions among Azotobacter chroococcum, Penicillium glaucum and Glomus fasciculatum and their effect on the growth and yield of sunflower (Helianthus annuus L.). Helia 18, 73 – 84. Haynes, R. J. (1982). Effects of liming on phosphate availability in acid soils. Plant Soil 68, 289 – 309. 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. Hedley, M. J., White, R. E., and Nye, P. H. (1982). Plant-induced changes in the rhizosphere of rape (Brassica napus var. Emerald) seedlings. III. Changes in L value, soil phosphate fractions and phosphatase activity. New Phytol. 91, 45 – 56. Hingston, F. J., Posner, A. M., and Quirk, J. P. (1972). Anion adsorption by goethite and gibbsite. I. The role of the proton in determining adsorption envelopes. J. Soil Sci. 23, 177–192. Hinsinger, P., and Gilkes, R. J. (1995). Root-induced dissolution of phosphate rock in the rhizosphere of lupins grown in alkaline soil. Austr. J. Soil Res. 33, 477– 489. Holford, I. C. R. (1983). Differences in the efficiency of various soil phosphate tests for white clover between very acidic and more alkaline soils. Austr. J. Soil Res. 21, 173 –182. Holford, I. C. R., and Mattingly, C. E. G. (1975). Phosphate adsorption and plant availability of phosphate. Plant Soil 44, 377– 389. Hsu, P. H. (1965). Fixation of phosphate by aluminium and iron in acidic soils. Soil Sci. 99, 398 – 402. Hsu, P. H. (1982). Crystallization of iron(III) phosphate at room temperature. Soil Sci. Soc. Am. J. 46, 928–932. Hue, N. V., Craddock, G. R., and Adams, F. (1986). Effect of organic acids on aluminium toxicity in subsoils. Soil Sci. Soc. Am. J. 50, 28 – 34. Illmer, P., and Schinner, F. (1992). Solubilization of inorganic phosphates by microorganisms isolated from forest soils. Soil Biol. Biochem. 24, 389 – 395. Illmer, P., and Schinner, F. (1995a). Solubilization of inorganic calcium phosphates—Solubilization mechanisms. Soil Biol. Biochem. 27, 257–263. Illmer, P., and Schinner, F. (1995b). Phosphate solubilizing microorganisms under nonsterile conditions. Bodenkultur 46, 197–204. Illmer, P., Barbato, A., and Schinner, F. (1995). Solubilization of hardly-soluble AlPO4 with P-solubilizing microorganisms. Soil Biol. Biochem. 27, 265 –270. Johnston, H. W. (1954). The solubilization of “insoluble” phosphates. III. A quantitative and comparative study of the action of chosen aromatic acids on tricalcium phosphate. N. Z. J. Sci. Technol. 36, 281–284. Johnston, H. W. (1956). Chelation between calcium and organic anions. N. Z. J. Sci. Technol. 37, 522– 537. Johnston, H. W. (1959). The solubilization of “insoluble” phosphates. V. The action of some organic acids on iron and aluminium phosphates. N. Z. J. Sci. 2, 215 –218. Johnston, H. W., and Miller, R. B. (1959). The solubilization of “insoluble” phosphate. IV. The reaction between organic acids and tricalcium phosphate. N. Z. J. Sci. 2, 109 –120. Jones, G. D., Lutz, J. A., and Smith, T. J. (1977). Effects of phosphorus and potassium on soybean nodules and seed yield. Agron. J. 69, 1003 –1006. Juo, A. S. R., and Ellis, B. G. (1968). Chemical and physical properties of iron and aluminium phosphates and their relation to phosphorus availability. Soil Sci. Soc. Am. Proc. 32, 216 –221. Kamprath, E. J., and Watson, M. E. (1980). Conventional soil and tissue tests for assessing the phosphorus status of soils. In “The Role of Phosphorus in Agriculture” (F. E. Khasawneh, E. C. Sample, and E. J. Kamprath, Eds.), pp. 433 – 469. ASA-CSSA-SSSA, Madison, WI. Katznelson, H., and Bose, B. (1959). Metabolic activity and P-dissolving capability of bacterial isolates from wheat roots, rhizosphere and non-rhizosphere soil. Can. J. Microbiol. 5, 79 – 85.

148

M. A. WHITELAW

Katznelson, H. E., Peterson, E. A., and Roualt, J. W. (1962). Phosphate dissolving microorganisms on seed and in the roots zone of plants. Can. J. Bot. 40, 1181–1186. Kepert, D. G., Robson, A. D., and Posner, A. M. (1979). The effect of organic root products on the availability of phosphorus to plants. In “The Soil–Root Interface” ( J. L. Harley and R. Scott-Russell, Eds.), pp. 115 –124. Academic Press, London. Khan, J. A., and Bhatnagar, R. M. (1977). Studies on solubilization of insoluble phosphates by microorganisms. Fertil. Technol. 14, 329 – 333. Kim, K. Y., McDonald, G. A., and Jordan, D. (1997). Solubilization of hydroxyapatite by Enterobacter agglomerans and cloned Escherichia coli in culture medium. Biol. Fertil. Soils 24, 347– 352. Kpomblekou-A, K., and Tabatabai, M. A. (1994). Effect of organic acids on release of phosphorus from phosphate rocks. Soil Sci. 158, 442– 452. Kucey, R. M. N. (1983). Phosphate-solubilising bacteria and fungi in various cultivated and virgin Alberta soils. Can. J. Soil Sci. 63, 671– 678. Kucey, R. M. N. (1987). Increased phosphorus uptake by wheat and field beans inoculated with a phosphorus-solubilising Penicillium bilaji strain and with vesicular–arbuscular mycorrhizal fungi. Appl. Environ. Microbiol. 53, 2699 –2703. Kucey, R. M. N. (1988). Effect of Penicillium bilaji on the solubility and uptake of P and micronutrients from soil by wheat. Can. J. Soil Sci. 68, 261–270. Kucey, R. M. N., and Leggett, M. E. (1989). Increased yields and phosphorus uptake by Westar canola (brassica napus L.) inoculated by a phosphate-solubilizing isolate of Penicillium bilaji. Can. J. Soil Sci. 69, 425– 432. Kucey, R. M. N., Janzen, H. H., and Leggett, M. E. (1989). Microbially mediated increases in plantavailable phosphorus. Adv. Agron. 42, 199 –228. Larsen, S. (1967). Soil phosphorus. Adv. Agron. 19, 151–210. Leaver, J. P., and Russell, E. W. (1957). The reaction between phosphate and phosphate-fixing soils. J. Soil Sci. 8, 113–126. Lindsay, W. L. (1979). “Chemical Equilibria in Soils.” Wiley, New York. Lindsay, W. L., Frazier, A. W., and Stephenson, H. F. (1962). Identification of reaction products from phosphate fertilizers in soils. Soil Sci. Soc. Am. Proc. 26, 446 – 452. Lopez-Hernandez, D., Flores, D., Siegert, G., and Rodriguez, J. V. (1979). The effect of some organic anions on phosphate removal from acid and calcareous soils. Soil Sci. 128, 321– 326. Louw, H. A., and Webley, D. M. (1959). A study of soil bacteria dissolving certain mineral phosphate fertilizers and related compounds. J. Appl. Bacteriol. 22, 227–233. Manjunath, A., Mohan, R., and Bagyaraj, D. J. (1981). Interaction between Beijerinckia mobilis, Aspergillus niger and Glomus fasciculatus and their effects on growth of onion. New Phytol. 87, 723–727. Martell, A. E., and Smith, R. M. (1977). “Critical Stability Constants. Vol. 3: Other Ligands.” Plenum, New York. Martin, J. K. (1977). Factors influencing the loss of organic carbon from wheat roots. Soil Biol. Biochem. 9, 1–7. Martinez Cruz, A., Aleman, I., Ferran, J., and Martinez, V. (1990). Phosphate dissolving microorganisms in the Cuban red ferralitic and reddish brown fersialitic soils. Agrokemia,-es-Talajtah 39, 492–494. Mba, C. C. (1996). Rock phosphate solubilizing Streptosporangium isolates from casts of tropical earthworm. Resour. Conservation Recyclling 17, 211–217. McLaughlin, M. J., Alston, A. M., and Martin, J. K. (1988). Phosphorus cycling in wheat-pasture rotations. 1. The source of phosphorus taken up by wheat. Austr. J. Soil Res. 26, 323 – 331. Mehta, A. C., Malavia, D. D., Kaneria, B. B., and Khanpara, V. D. (1996). Response of groundnut (Arachis hypogaea) to farmyard manure, phosphorus and phosphate-solubilizing micro-organism. Indian J. Agron. 41, 172–174.

PHOSPHATE-SOLUBILIZING FUNGI

149

Moghimi, A., and Tate, M. E. (1978). Does 2-ketogluconate chelate calcium in the pH range 2.4 to 6.4? Soil Biol. Biochem. 10, 289 –292. Molla, M. A. Z., Chowdhury, A. A., Islam, A., and Hoque, S. (1984). Microbial mineralization of organic phosphate in soil. Plant Soil 78, 393 – 399. Motekaitis, R. J., and Martell, A. E. (1984). Complexes of aluminium (III) with hydroxy carboxylic acids. Inorganic Chem. 23, 18 –23. Murrmann, R. P., and Peech, M. (1969). Relative significance of labile and crystalline phosphates in soils. Soil Sci. 107, 249 –255. Nagarajah, S., Posner, A. M., and Quirk, J. P. (1970). Competitive adsorption of phosphate with polygalacturonate and other organic anions on kaolinite and oxide surfaces. Nature 228, 83–85. Nahas, E. (1996). Factors determining rock phosphate solubilization by microorganisms isolated from soil. World J. Microbiol. Biotechnol. 12, 567– 572. Nahas, E., Banzatto, D. A., and Assis, L. C. (1990). Fluoroapatite solubilization by Aspergillus niger in vinasse medium. Soil Biol. Biochem. 22, 1097–1101. Narsian, V., Thakkar, J., and Patel, H. H. (1993). Solubilization of natural rock phosphates and pure insoluble inorganic phosphates by Aspergillus awamori. Indian J. Exp. Biol. 31, 747–749. Narsian, V., Thakkar, J., and Patel, H. H. (1994). Isolation and screening of phosphate solubilizing fungi. Indian J. Microbiol. 34, 113 –118. Narsian, V., Thakkar, J., and Patel, H. H. (1995). Mineral phosphate solubilization by Aspergillus aculeatus. Indian J. Exp. Biol. 33, 91– 93. Olsen, S. R., and Watanabe, F. S. (1957). A method to determine a phosphorus absorption maximum of soils as measured by the Langmuir isotherm. Soil Sci. Soc. Am. J. 21, 144 –149. Ortuno, A., Hernansaez, J., Noguera, J., Morales, V., and Armero, T. (1979). Phosphorus solubilization by Aspergillus niger and Pseudomonas fluorescens. Anal. Edafol. Agrobiol. 38, 1311–1316. Ozanne, P. G. (1980). Phosphate nutrition of plants—A general treatise. In “The Role of Phosphorus in Agriculture” (F. E. Khasawneh, E. C. Sample, and E. J. Kamprath, Eds.), pp. 559 – 589. ASACSSA-SSSA, Madison, WI. Parks, E. J., Olson, G. J., Brinckman, F. E., and Baldi, F. (1990). Characterization by high performance liquid chromatography (HPLC) of the solubilization of phosphorus in iron ore by a fungus. J. Ind. Microbiol. 5, 183–190. Petruccioli, M., Piccioni, P., Fenice, M., and Federici, F. (1994). Glucose oxidase, catalase and gluconic acid production by immobilized mycelium of Penicillium variabile P16. Biotechnol. Lett. 16, 939–942. Pikovskaia, R. I. (1948). Mobilization of phosphorus in soil in connection with vital activity by some microbial species. Mikrobiologia 17, 362– 370. Raghu, K., and MacRae, I. C. (1966). Occurrence of phosphate-dissolving micro-organisms in the rhizosphere of rice plants and in submerged soils. J. Appl. Bacteriol. 29, 582– 586. Rajan, S. S. S. (1976). Phosphate reactions with soil constituents and prospects for manipulation. In “Prospects for Improving Efficiency of Phosphorus Utilization” (G. J. Blair, Ed.), pp. 35 – 39. Univ. of New England, Armadale, Australia. Ramos, A., Callao, V., and de Carvalho, P. C. T. (1968). La solubilizacion de fosfatos por hongos del suelo. Microbiol. Española 23, 1–15. Robert, M., and Razzaghe-Karimi, M. (1975). Mise en évidence de deux d’évolution minralogique des micas trioctaédriques en présence d’acides organiques hydrosolubles. C. R. Acad. Sci. Paris 280, 2175–2178. Roos, W., and Luckner, M. (1984). Relationships between proton extrusion and fluxes of ammonium ions and organic acids in Penicillium cyclopium. J. Gen. Microbiol. 130, 1007–1014. Rose, R. E. (1957). Techniques for determining the effect of microorganisms on insoluble phosphates. N. Z. J. Sci. Technol. 38, 773 –780.

150

M. A. WHITELAW

Rosendahl, R. O. (1942). The effect of mycorrhizal and nonmycorrhizal fungi on the availability of difficultly soluble potassium and phosphorus. Soil Sci. Soc. Am. Proc. 7, 477– 479. Rovira, A. D. (1991). Rhizosphere research—85 years of progress and frustration. In “The Rhizosphere and Plant Growth” (D. L. Keister and P. B. Cregan, Eds.), pp. 3– 13. Kluwer, Dordrecht. Rovira, A. D., Foster, R. C., and Martin, J. K. (1979). Note on terminology: Organisms, nature and nomenclature of the organic materials in the rhizosphere. In “The Soil–Root Interface” (J. L. Harley and R. Scott-Russell, Eds.), pp. 1– 4. Academic Press, London. Russell, E. W. (1980). “Soil Conditions & Plant Growth,” 10th ed. Longman, London. Sackett, W. G., Patten, A. J., and Brown, C. W. (1908). The solvent action of soil bacteria upon the insoluble phosphates of raw bone meal and natural rock phosphate. Zentralblatt Bakteriol. 2, 688 – 703. Sainz, M. J., and Arines, J. (1988). P absorbed from soil by mycorrhizal red clover plants as affected by soluble P fertilization. Soil Biol. Biochem. 20, 61– 67. Salih, H. M., Yahya, A. I., Abdul-Rahem, A. M., and Munam, B. H. (1989). Availability of phosphorous in a calcareous soil treated with rock phosphate or superphosphate as affected by phosphatedissolving fungi. Plant Soil 120, 181–185. Sample, E. C., Soper, R. J., and Racz, G. J. (1980). Reactions of phosphate fertilizers in soils. In “The Role of Phosphorus in Agriculture” (F. E. Khasawneh, E. C. Sample, and E. J. Kamprath, Eds.), pp. 263–310. ASA-CSSA-SSSA, Madison, WI. Sauchelli, V. (1951). “Manual on Phosphates in Agriculture.” Davidson Chemical, Baltimore. Sethi, R. P., and Subba-Rao, N. S. (1968). Solubilization of tricalcium phosphate and calcium phytate by soil fungi. J. Gen. Appl. Microbiol. 14, 329 – 331. Singal, R., Gupta, R., and Saxena, R. K. (1994). Rock phosphate solubilization under alkaline conditions by Aspergillus japonicus and A. foetidus. Folia Microbiol. 39, 33 – 36. Singh, C. P., Mishra, M. M., and Kapoor, K. K. (1982). Solubilization of insoluble phosphate by mesophilic fungi. Rev. Écol. Biol. Sol. 19, 17–25. Singh, H. P., and Singh, T. A. (1993). The interaction of rockphosphate, Bradyrhizobium, vesicular– arbuscular mycorrhizae and phosphate-solubilizing microbes on soybean grown in a sub-Himalayan mollisol. Mycorrhiza 4, 37– 43. Singh, H. P., Pareek, R. P., and Singh, T. A. (1984). Solubilization of rock phosphates by phosphate solubilizers in broth. Curr. Sci. 53, 1212–1213. Snoeyink, V. L., and Jenkins, D. (1980). “Water Chemistry.” Wiley, New York. Sperber, J. I. (1957). Solution of mineral phosphates by soil bacteria. Nature 180, 994 – 995. Sperber, J. I. (1958a). The incidence of apatite-solubilizing organisms in the rhizosphere and soil. Austr. J. Agric. Res. 9, 778 –787. Sperber, J. I. (1958b). Solution of apatite by soil microorganisms producing organic acids. Austr. J. Agric. Res. 9, 782–787. Stevenson, F. J. (1967). Organic acids in soil. In “Soil Biochemistry” (A. D. McLaren and G. H. Peterson, Eds.), pp. 130 –146. Dekker, New York. Stevenson, F. J. (1986). “Cycles of Soil, Carbon, Nitrogen, Phosphorus, Sulfur, Micronutrients.” Wiley, New York. Stewart, J. W. B., and Sharpley, A. N. (1987). Controls on dynamics of soil and fertilizer phosphorus and sulfur. In “Soil Fertility and Organic Matter as Critical Components of Production Systems. “Soil Science Society of America, St. Paul, MN. Struthers, P. H., and Sieling, D. H. (1950). Effect of organic anions on phosphate precipitation by iron and aluminium as influenced by pH. Soil Sci. 69, 205 –213. Stumm, W., and Morgan, J. J. (1995). “Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters,” 3rd ed. Wiley, New York. Subba-Rao, N. S., and Bajpai, P. D. (1965). Fungi on the surface of legume root nodules and phosphate solubilization. Experientia 21, 386 – 387.

PHOSPHATE-SOLUBILIZING FUNGI

151

Surange, S. (1985). Comparative phosphate solubilizing capacity of some soil fungi. Curr. Sci. 54, 1134–1135. Swenson, R. M., Cole, C. V., and Sieling, D. H. (1949). Fixation of phosphate by iron and aluminium and replacement by organic and inorganic ions. Soil Sci. 67, 3 –22. Taha, S. M., Mahmoud, S. A. Z., Halim El-Damaty, A., and Abd El-Hafez, A. M. (1969). Activity of phosphate-dissolving bacteria in Egyptian soils. Plant Soil 31, 149 –160. Taylor, A. W., and Gurney, E. L. (1964). Solubility of variscite. Soil Sci. 98, 9 –13. Thimmegowda, S., and Devakumar, N. (1996). Effect of phosphorus management practices on protein and oil yield of soybean. Indian Agriculturist 40, 61– 64. Thomas, G. V., Shantaram, M. V., and Saraswathy, N. (1985). Occurrence and activity of phosphatesolubilizing fungi from coconut plantation soils. Plant Soil 87, 357– 364. Tinker, P. B. (1984). The role of microorganisms in mediating and facilitating the uptake of plant nutrients from soil. Plant Soil 76, 77– 91. Tiwari, V. N., Lehri, L. K., and Pathak, A. N. (1989). Effect of inoculating crops with phospho-microbes. Exp. Agric. 25, 47– 50. Tiwari, V. N., Pathak, A. N., and Lehri, L. K. (1993). Rock phosphate–superphosphate in wheat in relation to inoculation with phosphate solubilizing organism and organic waste. Indian J. Agric. Res. 27, 137–145. Toro, M., Azcon, R., and Herrera, R. (1996). Effects on yield and nutrition of mycorrhizal and nodulated Pueraria phaseoloides exerted by P-solubilizing rhizobacteria. Biol. Fertil. Soils 21, 23 –29. Trolldenier, G. (1992). Techniques for observing phosphorus mobilization in the rhizosphere. Biol. Fertil. Soils 14, 121–125. Vaishya, U. K., Bapat, P. N., and Dubey, A. V. (1996). Phosphate sobulizing efficiency of microorganisms on gram grown on Vertisol. J. Indian Soc. Soil Sci. 3, 524 – 526. Van der Zee, S.E.A.T.M., Fokkink, L. G. J., and van Riemsdijk, W. H. (1987). A new technique for assessment of reversibly adsorbed phosphate. Soil Sci. Soc. Am. J. 51, 599 – 604. Vassilev, N., Baca, M. T., Vassileva, M., Franco, I., and Azcon, R. (1995). Rock phosphate solubilization by Aspergillus niger grown on sugar-beet waste medium. Appl. Microbiol. Biotechnol. 44, 546–549. Vassilev, N., Fenice, M., and Federici, F. (1996). Rock phosphate solubilization with gluconic acid produced by immobilized Penicillium variable P16. Biotechnol. Techniques 10, 585 – 588. Veith, J. A., and Sposito, G. (1977). Reactions of aluminosilicates, aluminium hydrous oxides, and aluminium oxide with o-phosphate: The formation of x-ray amorphous analogs of variscite and montebrasite. Soil Sci. Soc. Am. J. 41, 870 – 876. Venkateswarlu, B., Rao, A. V., and Raina, P. (1984). Evaluation of phosphorus solubilisation by microorganisms isolated from aridisols. J. Indian Soc. Soil Sci. 32, 273 –277. Vidhyasekaran, P., Balaraman, K., Deiveegasundaram, M., and Viswanathan, G. (1973). Phosphatedissolving activity of Aspergillus awamori. Indian J. Microbiol. 13, 51– 53. Whipps, J. M. (1984). Environmental factors affecting the loss of carbon from the roots of wheat and barley seedlings. J. Exp. Bot. 35, 767–773. Whipps, J. M., and Lynch, J. M. (1986). The influence of the rhizosphere on crop productivity. Adv. Microbial. Ecol. 9, 187–244. White, R. E. (1983). The enigma of pH–P solubility relationships in soil. In “Phosphorus: Indispensable Element. Proceedings of the 3rd International Congress on Phosphorus Compounds, Brussels, 4–6 October, 1983,” pp. 53 – 64. Institut Mondial du Phosphate, Casablanca, Morocco. Whitelaw, M. A., Harden, T. J., and Bender, G. L. (1997). Plant growth promotion of wheat inoculated with Penicillium radicum sp. nov. Austr. J. Soil Res. 35, 291– 300. Whitelaw, M. A., Harden, T. J., and Helyar, K. R. (1999). Phosphate solubilisation in solution culture by the soil fungus Penicillium radicum. Soil Biol. Biochem. 31, 655 – 665. Wild, A. (1950). The retention of phosphate by soil. A review. J. Soil Sci. 1, 221–238.

This Page Intentionally Left Blank

HYDROLOGICAL FACTORS FOR PHOSPHORUS TRANSFER FROM AGRICULTURAL SOILS P. M. Haygarth,1 A. L. Heathwaite,3 S. C. Jarvis,1 and T. R. Harrod2 1Institute

of Grassland and Environmental Research and Survey and Land Research Centre Cranfield University North Wyke, Okehampton Devon EX20 2SB, United Kingdom 3Department of Geography University of Sheffield Sheffield S10 2TN, United Kingdom

2Soil

I. Introduction II. Temporal Variables A. Effective Rainfall B. Levels of Hydrological Activity C. Timescales III. Spatial Variables A. Scale B. Pathways IV. Conclusions References

Understanding and managing against phosphorus (P) transfer from agricultural soils to receiving waters is a multidisciplinary task in which the role of hydrology is particularly difficult to simplify. We review current knowledge to define the spatial and temporal controls on P transfer from agricultural soils via the various hydrological pathways. Rainfall intensity and duration and the interval between rainfall events are key temporal variables which influence the discharge (and hence P load) to receiving waters. In terms of understanding mechanisms, we postulate that levels of hydrological activity may be nominally classified at two levels. Level 1 activity occurs during light or little rainfall for a high proportion of time; in contrast, level 2 activity occurs less frequently but is more energetic and has a large capacity for P transfer over a small time period. The range of potential hydrological pathways of P transfer creates confusion because terminology varies. Often, process terms such as leaching or generic terms such as drainage are confused with hydrological pathways per se. Here we define the spatial variation in the hydrological pathways responsible for P transfer at two scales—soil profile and slope/field— which subsequently can help the understanding of P transfers into the wider catch153 Advances in Agronomy, Volume 69 Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved. 0065-2113/00 $30.00

154

P. M. HAYGARTH ET AL. ment, where problems occur. Our aim is to provide a simplified basis for classifying the otherwise complex hydrochemical regimes which result in P transfer. © 2000 Academic Press.

I. INTRODUCTION The consequences of land use on environmental quality have an increasingly high profile. From an agricultural standpoint, one area of critical current interest is the transfer of pollutants from soil to water because soils play a pivotal role in the protection of groundwaters and surface waters (National Rivers Authority, 1992). There is a need to identify and describe sources, media, and pathways of pollutant transfer in order to aid the developing interactions of soil science with hydrology (Boorman et al., 1995) and/or land use specialists with aquatic biologists. Such multidisciplinary links have the potential to create ambiguous, imprecise, and contradictory terminology or to omit key information. Definitions are particularly important when describing environmental phosphorus (P) transfer from agricultural soils to inland water bodies. Most studies on P transfer are from an agronomic standpoint in which the role of hydrology is not fully considered (Haygarth and Jarvis, 1999). We therefore deliberately take this opportunity to describe the hydrological factors in P transfer because, to date, there has been a tendency for such factors to be neglected. In many of the world’s agricultural soils, P accumulates in intensively managed agricultural soils because of farm imports of fertilizer and livestock feeds (Brouwer et al., 1995; Haygarth et al., 1998b). Consequently, agricultural soils are now considered to be the main diffuse source of P reaching freshwaters (Foy and Withers, 1995), in which concentrations as low as 35 –100 g total P liter1 may contribute to eutrophication (Organization for Economic Cooperation and Development, 1982). Previous research on phosphorus transfer (PT) has emphasized measurable “soil factors” such as defining threshold P concentrations using soil extractants (Heckrath et al., 1995; Sharpley et al., 1996), with relatively few studies focusing on “hydrological factors” (Heathwaite, 1995). According to Heathwaite, soil characteristics and factors define the initial chemical form of P export, but the hydrological conditions determine whether or not mobilization occurs and along which pathway. The scarcity of studies on hydrological interactions with PT is perhaps understandable because, unlike soil properties, hydrological factors are less easy to measure, classify, and interpret than many other soil properties. Hydrological conditions are also temporally and spatially dynamic and have the added difficulties of variations with changes in scale.

HYDROLOGICAL FACTORS FOR PHOSPHORUS TRANSFER

155

Conventionally, general-purpose soil classifications have been defined by pedological properties, although broad hydrological status might be implied from class names, as in Avery’s (1980) “stagnogley.” Although pedological classifications place high significance on expression or gleying (indicating duration of waterlogging), it required the hydrology of soil types (HOST) classification of Boorman et al. (1995) to provide a detailed description of the hydrologically driven system in England and Wales. Inferences on overall hydrology, pathways, seasonal behavior patterns, and responses to rainfall can be derived from HOST. Until the Host classification was developed, only the very simple winter rain acceptance potential classification (Farquharson et al., 1978) and U.S. Department of Agriculture (USDA, 1972) hydrologic soil grouping had, in a consistent way, addressed the hydrology within the soil. Studies of the fundamental role played by water flow characteristics on PT are invaluable because hydrology provides both the energy and the carrier for the transfer regime (Haygarth and Jarvis, 1999). Moreover, soils are hydrologically diverse (Boorman et al., 1995), with very different pedological and hydrological properties as well as varied responses to rainfall and land-use practices. Nevertheless, contrasting soils are often found in close proximity on hillslopes and even within individual fields. Such differences are illustrated by the soils in Table I which, along with the hydrologically similar soils they represent, are important in various regions of the United Kingdom. The contrast is shown by the impermeable, clayey Hallsworth series soils (Avery, 1980), which are waterlogged from October to May, whereas the permeable, loamy Denbigh series soils are waterlogged only in very wet weather. Thus, each soil series has different implications for PT. The aim of this review is to help to redress the current shortfalls in information by providing a considered review of the hydrological issues and defining the role of hydrology in context with other aspects of PT. The treatment of hydrological factors in this review is provided in two sections dealing with temporal and spatial variables.

II. TEMPORAL VARIABLES A. EFFECTIVE RAINFALL Precipitation, with its temporal variation, provides the energy source and the physical carrier mechanism within the PT process (Haygarth and Jarvis, 1999). Critical to PT is the relationship between rainfall input and runoff generation. In catchment/watersheds in which the climate is characterized by low-intensity rain-

Table I Hydrological Characteristics Describing the Contrast between the Denbigh and Hallsworth Soil Series Soil classification Soil Survey and Land Research Centre (Avery, 1980) USDA (1972) Food and Agriculture Organization Wetness (Hodgson, 1997) Class Duration of waterlogging (days) 40-cm depth 70-cm depth Saturated hydraulic conductivity (m/day) Porosity Retained water % (moisture content after drainage, i.e., at FC) Air capacity % (drainable pores) Workability/trafficability (Findlay et al., 1984) Good machinery workdays relative to field capacity season HOST (Boorman et al., 1995) General description

Hydrological pathways

HOST class SPR% (standard percentage runoff; from catchment scale interpretation) Base flow index (BFI% 1 total BF 0 no BF; from catchment interpretation)

Denbigh

Hallsworth

Typical brown earths Dystrochrepts Dystric cambisols

Pelostagnogley soils Typic haplaquepts Dystric gleysols

I

IV

0 days 30 days 1 (topsoil), 1 (subsoil)

180 days 180 days 0.8 (topsoil), 0.002 (subsoil)

42 (topsoil), 32 (subsoil) 12 (topsoil), 21 (subsoil)

47 (topsoil), 49 (subsoil) 12 (topsoil), 0.1 (subsoil)

30

55

1:No impermeable or gleyed layer within 1 m 2: No significant aquifer or GW 3: Over impermeable or hard substrate Vertical unsaturated flow; bypass flow in the substrate; some surface runoff

1: Gleyed layer within 40 cm

17 32

2: Over slowly permeable substrate 3: No significant aquifer Surface runoff likely; prolonged seasonal saturated flow; short seasonal bypass flow to the substrate 24 51

0.6

0.31

HYDROLOGICAL FACTORS FOR PHOSPHORUS TRANSFER

157

fall (15 mm h1), the current soil moisture state is more important in controlling the occurrence and magnitude of runoff than rainfall magnitude (Istok and Boersma, 1986). This has implications for PT because most export occurs when the land is at or close to field capacity (FC). For example, in the Slapton catchment, Devon, England, 80% of PT occurred in winter (Heathwaite et al., 1989). Snowmelt has been associated with high rates of PT from soil (Hawkins and Scholefield, 1996; Timmons et al., 1977). The concept of “field capacity” is useful in this context. Field capacity occurs when a soil is thoroughly wetted and drainage starts. For example (see Table I), a freely draining soil such as the Denbigh series would “drain” and reach FC in approximately 2 days from thorough wetting, whereas this takes weeks (Hodgson, 1997) in a poorly drained soil such as the Hallsworth series. The respective hydrological properties of these two soil types are summarized in Table I. Meteorological field capacity (MFC) occurs when received rainfall is greater than evapotranspiration, provided no soil moisture deficit exists. MFC is readily presented in cartographic form (Jones and Thomasson, 1985). Excess precipitation, which occurs when soils are at FC, can be called hydrologically significant precipitation. This effective rainfall is an important driver for the PT process.

B. LEVELS OF HYDROLOGICAL ACTIVITY Temporal variations in rainfall intensity, duration, and intervals between storms (return period) affect the magnitude of discharge along various hydrological pathways (Burgoa et al., 1993; DeWalle and Pionke, 1994; Evans, 1978; Sharpley, 1980a,b; Thornes, 1979). Although flow rate is a continuum from low to high discharge, different hydrological pathways may be triggered at different rates of discharge or rainfall input, with different consequences for PT. However, division into base-flow and storm-flow conditions is a helpful starting point when considering discharges. Storm flows occur infrequently and result in overland and macropore flows, whereas base flows occur more frequently but may only transfer water along predominately subsurface pathways. Pionke et al. (1996) found that storm flow was important for P discharge from a 7.4-km2 agricultural watershed. Dils and Heathwaite (1996) found that individual storm events had different capacities to transport P; moreover, the forms of P varied between events. They suggested that the most important controls on P fractionation were antecedent soil moisture (determining the likelihood of surface runoff ) and the interval between storm events (determining the incidence of “old” and “new” soil water and hence the amount of time for interaction between mobilized P and soil water). Base flow comprises non-storm-flow periods in which groundwater discharge, including springs and near-stream seepage, may form the main component of flow

158

P. M. HAYGARTH ET AL.

(Pionke and DeWalle, 1994). For base flow, subsurface hydrological pathways such as throughflow are most important for PT. Despite the obvious importance of storm events, the significance of subsurface pathways in transporting P during base-flow or low-flow conditions should not be discounted and needs further research in terms of PT. A limitation of the storm-flow/base-flow concept is that it pertains solely to discharge per se, whereas a more generalized temporal hydrological classification, which accounts for precipitation (and thus rainfall erosivity, for example), may be more useful. Haygarth and Jarvis (1997), Haygarth et al. (1998a), and Fraser et al. (1998) identified two populations of data when studying PT, implicating an interaction of rainfall with flow. We therefore suggest that an effective means of classification must recognize at least two levels of hydrological activity. Level 1 activity occurs during light or little rainfall for a high proportion of time (and will incorporate base flows). In contrast, level 2 activities will be of low frequency but high intensity, operating with greater energy than level 1 activity, and have a high propensity for PT during a short period and resulting in storm flows. When level 2 occurs it will produce high-intensity rainfall having a greater erosivity and will therefore result in larger particulate PTs than those of level 1 rain. In reality, a twotier classification may be too simplistic and a sliding scale of increasing activity may be more appropriate: Nevertheless, this system provides useful initial conceptual classification.

C. TIMESCALES Shrinkage of soils such as the Hallsworth series (Table I), with coarsely structured, often clayey horizons (HOST classes 18 –25 of Boorman et al., 1995), during summer soil moisture deficits opens vertical fissures. These permit vertical bypass flow particularly during autumnal rewetting (level 2; Fig. 1a). When rewetted, the horizons revert to an impermeable state, dominated by saturated lateral flow (level 1). The temporal extent of either state depends on the strength of the soil moisture deficit and the duration of the field capacity period. In Table II we present a nonexhaustive list of hydrological pathways which we have attempted to classify in relation to timescales (more detailed discussion of the relevance of this table is given in the following section, which defines scale). At the slope/field scale, the time taken for overland flow to travel 100 m may be of the order of minutes during intense rainfall (Horton, 1945), whereas water in subsurface pathways to underlying aquifers may take months to years (thus giving rise to the terms new and old water; Bohlke and Dener, 1995). Again, many of these variations can be defined by soil class, for example, as expressed in the HOST classification of Boorman et al. (1995) and illustrated by data in Table I for the Denbigh and Hallsworth series.

Figure 1 Soil profile scale pathways. The situation common (a) in late summer/autumn, when soils are dry and prone to bypass flow, and (b) in winter/spring, when soils are saturated and prone to matrix flow.

Table II Terminology Commonly Associated with Hydrochemical Transfer Pathways, Nominally Classified by Discipline, Time, and Spatial Scale Term (sorted alphabetically)

Scale

Generic

160

Arterial drainage

Catchment

Agronomic

Base flow

Slope/field

Hydrological

Bypass flow

Soil

Hydrological

Darcian flow

Soil

Hydrological

Ditching

Slope/field

Agronomic

Interflow Land drainage

Slope/field Subcatchment

Hydrological Agricultural

Leaching

Soil

Chemical

Leakage

Slope/field

Macropore flow

Soil

Hydrological and chemical Hydrological

Matrix flow

Soil

Hydrological

Overland flow

Slope/field

Geomorphological/ pedological

Nominal timescale

Definition Major artificial drainage channel used to remove water after its discharge from field drains Nominally the transfer of water underground in “background” flow conditions, not pathway specific; also used to describe a low magnitude of flow Implies a type of soil water movement—in the case of vertical movement along larger subsoil pathways, e.g., wormholes and fissures, often occurring in unsaturated conditions Not a pathway but describes discharge through the soil as related to the hydraulic gradient and hydraulic conductivity Artificial (human-made) first-order drainage streams, often used on agricultural land to compliment arterial drainage Lateral flows below the soil surface Water and solute ( entrained solids) export to catchment resulting from land drainage practices: anthropogenic Eluviation of chemicals vertically through the soil profile and vadose zone; despite misconceptions, this is a mechanism and not a pathway General nonspecific term describing water and chemical movement As bypass flow; macropores are large enough (60 m) to allow gravitational drainage Implies a type of soil water movement—in this case uniform vertical movement downwards, common in very porous media such as sandy textures; only occurs under saturated conditions Movement of water exclusively over the soil surface, down slope, during heavy rain

Hours/days

Hours/weeks/months Minutes/hours

— Minutes/hours Minutes/hours Minutes/hours Variable

Not applicable Minutes/hours Days

Minutes/hours

161

Percolating water Pipe flow Piston flow Preferential flow Return flow River Roadway

Soil Slope/field Soil Soil Slope/field Subcatchment Subcatchment

Hydrological Geomorphological Hydrological Hydrological Hydrological Hydrological Engineering

Runoff

Slope/field

Hydrological

Saturated (soil) flow Seepage

Soil Slope/field

Hydrological Hydrological

Soil solution

Soil

Chemical

Stream

Subcatchment

Subsurface flow Surface runoff Throughflow

Slope/field Slope/field Soil and Slope/field Slope/field

Hydrological and geomorphological Geomorphological Hydrological Hydrological

Unsaturated flow Vertical saturated flow Vertical unsaturated flow

Hydrological

General nonspecific term describing water movement Lateral subsurface preferential flow As matrix flow As bypass flow Where a subsurface flow pathway emerges at the soil surface Large-order drainage network Human road or path which can assist water transfer from slope/ field to catchment; little studied General hydrological term describing the lateral movement of water off land above and below ground, causing a short-term increase in flow at the catchment outlet; can refer to pathway when qualified (e.g., surface runoff), but also has been used to describe processes and water samples As piston flow, but lateral and not vertical General nonspecific term describing water movement; implies emergence at or near the ground surface Nonspecific term describing water sampled from the soil environment by whatever means; not a pathway Small-order drainage network

Not applicable Minutes/hours Variable Minutes/hours Minutes/days Hours/days Hours/days

Lateral flows below the soil surface As overland flow As percolating water

Minutes/hours Minutes/hours Not applicable Minutes/hours Days Minutes/hours

Soil

Hydrological

As preferential flow, but occurring laterally over capped, compacted, or slowly permeable horizons As piston flow

Soil

Hydrological

As bypass flow

Minutes/hours

Days Not applicable Not applicable Hours/days

162

P. M. HAYGARTH ET AL.

III. SPATIAL VARIABLES A. SCALE Effects of space and scale are important when attempting to understand PT along soil hydrological pathways (Beven et al., 1993; Konikow, 1991). Kirkby (1988) demonstrated the importance of scale (drainage basin area) on hillslope flow processes (Fig. 2). The concepts incorporated in Fig. 2 may be extended to an examination of PT where lag times to peak concentration (Fig. 2a) are impor-

Figure 2 (a) Lag times and (b) peak runoff rates integrating spatial and temporal factors. NB Hortonian  infiltration excess overland flow. (Redrawn from “Hillslope Hydrology,” M. J. Kirkby. Copyright John Wiley & Sons Limited. Reproduced with permission.)

HYDROLOGICAL FACTORS FOR PHOSPHORUS TRANSFER

163

tant in determining exchange and uptake reactions with soil solution and consequently the form of P transported. In contrast, peak runoff rates (Fig. 2b) also aid our understanding of the potential P load reaching the drainage network. Thus, for cultivated land, the higher peak runoff rates associated with Hortonian (infiltration-excess) overland flow (Fig. 2b) will have greater capacity to detach and transport soil particles and sorb P. Although this pathway is restricted in catchment/watersheds in temperate climes such as that in the United Kingdom, where the soil infiltration capacity is rarely exceeded (Kirkby, 1988), poor land management (e.g., overgrazing) will exacerbate its incidence (Heathwaite, 1997). Because soil hydrological and chemical processes operate at different scales, it is essential that their spatial dimensions be identified. The challenge is to identify mechanisms that connect soil profile and slope/field scale process studies with larger scale catchment/watershed effects. Recent attempts to incorporate chemical–physical links initiating PT using a combination of runoff and erosion modeling (Gburek et al., 1996) have advanced knowledge of pathways of P delivery and their interaction with soil profile characteristics. Similarly, knowledge of hydrological pathways of water movement from land to stream has developed with increasingly detailed means of field monitoring. Thus, simple models of pipe-flow channeling of new water via the soil matrix to the stream have been replaced by new concepts incorporating new/old water and bypass flow (Bohlke and Dener, 1995). For the purpose of clarifying approaches to understanding the spatial controls on PT, we have nominally selected two scales smaller than that of catchment/watershed. The size boundaries selected are arbitrary but provide a convenient means for subdivision of landscape units. The smallest is the soil profile scale (centimeters to meters). The second is the slope/field scale; this has also been called the hillslope or the subcatchment/watershed and will range from meters to hectares and is generally restricted to catchment/watershed headwaters and/or first-order streams (Haygarth and Jarvis, 1997; Heathwaite and Johnes, 1996). This scale is likely to be relevant to individual soil mapping units (scales of 1:25,000 and greater). For PT from agricultural land, this scale is perhaps the most critical because it is possible to integrate detailed soil process studies with characterization of the patterns of PT at the slope/field scale (e.g., P fractionation in different hydrological pathways; Dils and Heathwaite, 1996; Haygarth et al., 1998a). The largest scale is the catchment/watershed; although we will not focus on catchment/ watershed pathways, it is necessary to consider some definitions. Catchments (English term) or watersheds (American term) are variable in size and there are many examples of catchment/watershed studies and classifications (Burt et al., 1996; Gburek et al., 1996; Heathwaite and Johnes, 1996; Johnes et al., 1996; Kronvang, 1990; Molden and Cerney, 1994). Catchment/watersheds are appropriate for interpretation and mapping of “reconaissance” maps (1:  M in the United Kingdom, as in appendix D of Boorman et al., 1995). Our scheme for subdividing the spa-

164

P. M. HAYGARTH ET AL.

tial dimensions of PT is similar to that proposed by Kirkby (1988) in order to examine the time lag (in hours from precipitation input to stream output within a catchment/watershed; see Fig. 2). Kirkby argues that “catchments may be thought of as a sequence of moisture stores, some in series, some in parallel.” Temporal rates of water movement (introduced in Section II) clearly interact with the spatial scales and can give rise to the classification of (i) surface detention (0.1–1 h), (ii) infiltration (1–20 h), (iii) unsaturated vertical percolation (1–50 h), (iv) saturated downslope flow (1–12 h), (v) channel flow [depends on catchment/watershed area: 0.5 h (1 km2), 7 h (100 km2), 100 h (104 km2)]. Here, scales i–iv equate with our soil profile scale, scale iv and part of scale v equate with our slope/field scale, and v refers to the wider catchment/watershed. The store with the longest residence time exerts the greatest control on water movement. This has implications for P form and the magnitude of P export because it indicates the amount of time available for equilibration between rainfall and soil water and hence the potential for P mobilization and transfer.

B. PATHWAYS Generalized overviews and definitions of hydrological pathways have been provided by many authors (Anderson and Burt, 1990; Boorman et al., 1995; Cox et al., 1997; Kirkby, 1978; Mangold and Tsang, 1991; McGechan and Wu, 1996; Miyazaki et al., 1993; Youngs and Leeds-Harrison, 1990). When precipitation reaches the soil, it is partitioned between overland flow and subsurface flow (Kirkby, 1988), each having varying potentials to entrain and retain P. Terms used to describe these two pathways (and their various forms) are sometimes confused in the scientific literature and problems arise because 1. Hydrological pathway terms can become confused with process terms. The terms “runoff” and “leaching” are good examples. Runoff has been used in the context of a (rather vaguely defined) pathway (Boorman et al., 1995; Haygarth and Jarvis, 1996, 1997), a water medium (Harms et al., 1974; Loehr, 1974), or, in occasional circumstances, a process (Zobisch et al., 1994). Leaching has also been used in an ambiguous way. It does not describe a pathway, although it has sometimes been used in this context to characterize an amalgam of all pathways of water drainage through soil (Bromfield and Jones, 1972; Heckrath et al., 1995; Jordan and Smith, 1985). Leaching is a process term and describes the elluviation of solutes, such as P, down through soil and is common in porous soils (Wagenet, 1990; Weaver et al., 1988a,b). 2. Pathways tend to be classified according to research background and discipline and may be defined in terms of geomorphological, hydrological, pedological, chemical, or land-use criteria.

HYDROLOGICAL FACTORS FOR PHOSPHORUS TRANSFER

165

Table II provides a synthesis of common soil hydrological pathways and associated terminology. We classified them both generically and by scale. A detailed discussion is provided in the following subsections, which are separated into the soil profile and slope/field scales. 1. Soil Profile Scale Key hydrological pathways at the soil profile scale are illustrated in Fig. 1. Void size (numbers and distribution), number of fissures (pores and macropores), and the degree of structure and aggregation (Hodgson, 1997) all contribute to variability. The modes of hydrological transport at this scale are (i) matrix flow or saturated/piston flow and (ii) preferential bypass or macropore flow. These do not describe pathways per se but rather a mode of flow. Saturated/piston flow refers to the uniform movement of water through soil, which occurs after the soil pores have become saturated. It is dependent on soil porosity and is common in sandy soils (Tindall et al., 1986). For relatively homogeneous soils such as sands, simple models of water movement based on Darcian flow may be applied. Here, discharge through the soil is related to the hydraulic gradient and hydraulic conductivity. The latter is a function of the media (e.g., sand, silt, and clay) and the properties of the fluid flowing through it. Darcian flow may explain the movement of a sharp wetting front between wet and dry soil and models of soil water movement via this pathway are relatively simple. During saturated flow, PT may be delayed through interaction within soil peds en route. Preferential flow, alternatively termed macropore or bypass flow (Thomas and Phillips, 1979), is the rapid and direct transfer of water through soil fissures and macropores (root channels and animal burrows coarser than 60 m; Hodgson, 1977); it may increase the infiltration capacity of soils by one or two orders of magnitude (Burt et al., 1996). Artificial drainage systems, both land drainage pipes and associated permeable fill, and secondary treatments of mole drains and subsoiling are important routes for preferential flow in impermeable agricultural soils (see Section II,B,2 for further discussion on artificial drainage). Preferential flow is important in clay soils, which are particularly vulnerable to shrinkage cracking, often after dry weather. However, after a period of rewetting the cracks close. Macropores, such as earthworm burrows, function year-round. They are often very common in loamy and sandy soil and can result in bypass flow, even in soils with abundant macropores. Figure 1 provides an illustration of the seasonal variations which may lead to preferential vs saturated/piston flow. Preferential flow can occur laterally (i.e., parallel to the surface) or vertically (Beven and Germann, 1982). There has been considerable research on preferential flow (Beven, 1991; Beven and Germann, 1982; Hornberger et al., 1990; Phillips et al., 1995; Singh and Kanwar, 1991; Wopereis, 1994), although it remains difficult to model the spatial complexity of hillslope hydrology in terms of the potentially turbulent nature of flow where continuously

166

P. M. HAYGARTH ET AL.

connected soil voids (macropores or pipes) bypass matrix flow. Three-dimensional macropore–matrix models are needed to further elucidate the hillslope hydrological system (Holden et al., 1995). To date, there has been little work on P movement via preferential flow, with the exception of some initial studies and discussions (Beauchemin et al., 1998; Dils and Heathwaite, 1996; Haygarth et al., 1998a; Heckrath et al., 1995; Stamm et al., 1998). Preferential flow may be particularly effective in transporting P present on the soil surface where it has accumulated after, for example, excreta, manure, or fertilizer deposition/application. Thus, this pathway may be important for PT from grassland soils, in which such practices are common (Haygarth et al., 1998a; Heathwaite, 1997), and for arable soils used for organic waste disposal. The soil hydrological pathways described previously for the soil profile scale may be conveniently summarized by the HOST model (Boorman et al., 1995). HOST assumes two extremes of soil water movement as either (i) predominantly vertical or (ii) predominantly lateral and at or close to the surface. The classification is based on the permeability of substrate and the depth of the water table. There are three settings for the model in terms of relationships to groundwater or aquifers; these are illustrated in Fig. 3. In HOST, soil hydrological pathways are described with respect to • • • • • •

Location in the soil, subsoil, and substrate Vertical or lateral direction of flow Saturated or unsaturated conditions Probability and duration of flow (local conditions) Existence of bypass flow Existence of leakage to substrate

Weaknesses in the HOST approach are that it does not consider capping/crusting soils and the role of slopes. Although the international applicability of the HOST model has yet to be tested, it shows great potential. 2. Slope/Field Scale The pathways at this scale are illustrated in Fig. 4: This is the link between the soil profile and the catchment/watershed, encompassing hydrological pathways which may directly (e.g., by overland or surface flow) or indirectly (e.g., by interflow or lateral subsurface flow) link land to ditch to stream. A significant proportion of PT may be accounted for by erosion mechanisms and overland flow pathways at this scale (Sharpley and Smith, 1990); pathways which arise from artificial land drainage may also be important. There are many studies which summarize water flow on this scale (Atkinson, 1978; Emmett, 1978; Freeze, 1978; Hornberger et al., 1991; Knapp, 1978; Trudgill and Coles, 1988; Whipkey and Kirby, 1978),

Figure 3

The hydrology of soil types (HOST) classification (redrawn from Boorman et al., 1995, Fig. 3.2, p. 27).

Figure 4

Slope/field scale pathways (redrawn from Haygarth and Jarvis, 1999).

HYDROLOGICAL FACTORS FOR PHOSPHORUS TRANSFER

169

and others illustrate techniques for sampling and studying transfers (Cuttle and Mason, 1988; Hoover, 1987). Although PT from agricultural catchment/watersheds is largely derived from nonpoint sources, some parts of the catchment/watershed are more likely to contribute water—and possibly P—flow than others (Gburek and Sharpley, 1998). The link between the spatial variation in PT and hillslope hydrology lies in understanding the role of hillslope morphology in controlling water flow pathways. There is often a close link between hillslope angle and soil (and therefore porosity/infiltration) patterns with, for example, permeable soils such as the Denbigh series (Table I) occupying steeper slopes and impermeable soils such as the Hallsworth series occupying gentler footslopes and interfulvial crests (see Fig. 44 in Findlay et al., 1984). Catchment/watershed morphology is a dominant control on spatial patterns of hydrological response (Beven and Young, 1988), particularly where surface or near-surface flows are the main mechanism by which water reaches a stream. However, other sources of heterogeneity contributing to spatial variation exist, such as local management and the effect of soil characteristics on infiltration and thus overland flow. Work on hillslopes (Anderson and Burt, 1990; Burt et al., 1996) demonstrated the importance of topographical variation in hillslope form on the spatial variation in runoff generation and hence solute transport. Valley side slopes were shown to have high potential for solute export, whereas hillslope hollows form potential point sources and may be important for transferring solutes, such as nitrate, through the soil. Such hollows become important for PT where they coincide with critical source areas (CSAs) of P (Pionke et al., 1997) usually as a result of land use (Gburek et al., 1996). CSAs describe catchment/watershed areas where P sources (e.g., fields with high manure applications and/or high soil P concentrations) coincide with areas with hydrological connectivity to the drainage network (e.g., variable source areas in near-stream zones). Hillslope hollows appear to be activated only when flow converges into hollows and leads to accumulation of soil moisture and activation of saturated overland flow; such activation is dependent on rainfall duration, intensity, and magnitude as well as the interval between rainfall events. For impermeable subsoils, runoff is the most commonly used term for describing water movement at the slope/field scale. As indicated previously, this is potentially ambiguous: Runoff is often used as a general term describing the lateral movement of water (and entrained solids) across or under the surface of a slope and refers to the composite of all soil profile and slope/field pathways. The propensity for runoff to affect PT is described by the effective depth of interaction (EDI) (Ahuja, 1986; Ahuja et al., 1981; Sharpley, 1985). Experiments adding 32P to the soil surface on simulated slopes showed that the top few centimeters of soil play a critical role in the generation of P in runoff. This critical layer was termed the “mixing zone” and under experimental conditions it was calculated that the EDI

170

P. M. HAYGARTH ET AL.

for a 4% slope was 0.2 or 0.3 cm after application of 6.5 cm h1 rainfall for 30 min (Ahuja, 1986; Ahuja et al., 1981; Sharpley, 1985). Although the experiments used extreme rainfall inputs, the research indicated that P residing in the surface 1 mm of soil was most vulnerable to export in runoff. Overland flow, or surface runoff, describes transport which occurs exclusively over the soil surface (Emmett, 1978). It is important in physically transporting P attached to solids following soil erosion and also through incidental losses of fertilizers and manures from the slope (Haygarth and Jarvis, 1999). It is also a pathway with a strong affinity for PT because the surface soil has the greatest EDI (Ahuja et al., 1981) and the highest concentrations of P (Haygarth et al., 1998a), which can interact with mobile water. In the examples in Table I, the Hallsworth series soils are at risk of overland flow during and after rain throughout the field capacity season, whereas Denbigh series soils will be at risk only during heavy rain or after severe surface damage by livestock or machinery. Overland flow occurs through either “infiltration excess” (i.e., Hortonian) or “saturation excess” mechanisms (Fig. 2). Infiltration excess overland flow was first described by Horton (1945) and occurs where rainfall intensity exceeds the infiltration capacity of the soil (Dunne, 1983). The concept is applicable to arid/semiarid environments in which it may occur over wide areas. In temperate zones, infiltration-excess overland flow is more appropriately described as partial-area runoff (Betson, 1964) where only parts of a catchment/watershed contribute flow, for example, overgrazed grassland (Heathwaite et al., 1990) and tracks or tractor wheelings (Heathwaite et al., 1989)—physical effects resulting from land management. Discharge per unit contour length (q) is given by q (I  f )a where I is the rainfall intensity after interception, f is the infiltration rate, and a is the area drained per unit contour length (Thornes, 1979). In Britain, for example, rainfall intensities are generally low and the soil infiltration capacity is unlikely to be exceeded (Kirkby, 1978); therefore, the runoff regime should be dominated by saturation-excess and subsurface storm-flow mechanisms. There is evidence to suggest, however, that land management practices are increasing the incidence of infiltration-excess overland flow (Heathwaite et al., 1990). This has been recorded on a fieldwide basis at Slapton, Devon (Heathwaite and Burt, 1992), where overgrazing of kale by sheep compacted the soil surface to the extent that the infiltration capacity was of the order 0.1 mm h1. In grassland, infiltration capacities ranging from 12 mm h1 for temporary grass to 5 mm h1 for lightly grazed permanent grass and 1 mm h1 for heavily grazed permanent grass were recorded. Snow and frozen ground impair infiltration and therefore increase rates of overland flow with correspondingly high rates of PT (Hawkins and Scholefield, 1996). Other land management activities affecting the soil surface, and thus the propensity for overland flow and PT, are poaching (also called pugging or tram-

HYDROLOGICAL FACTORS FOR PHOSPHORUS TRANSFER

171

pling by livestock), panning (lateral smearing and closing of pores due to the passage of agricultural machinery, often resulting in a subsurface effect), rutting/ wheeling (by agricultural vehicles), and capping/slaking/crusting (the degradation of the structure of the soil surface), often by rain impact to impair infiltration and general compaction. Saturation-excess overland flow describes soil saturation via a high water table or lateral water movement above an impeding horizon. It may incorporate return flow (Table II and Fig. 4) and is a seasonally variable and potentially important component of storm runoff, especially in catchment/watersheds with thick vegetation, thin soils, high water tables, and long, gentle concave slopes (Dunne, 1983). Dunne and Black (1970) demonstrated that the major proportion of storm runoff was produced as overland flow from a small proportion of the catchment/watershed. The authors identified partial areas contributing quick runoff; such areas were controlled by rainfall intensity and expanded/contracted seasonally or during an individual storm. Geology, topography, soils, and rainfall characteristics defined their location. Under steady rainfall, lower rainfall intensities are required to maintain saturation overland flow in comparison with infiltration-excess overland flow. Whereas infiltration-excess overland flow may occur over wide areas such as whole fields, the areas of a hillslope or subcatchment/watershed contributing saturation-excess overland flow are both spatially and temporally variable. These are called variable source areas (VSAs), which are zones of saturation within a catchment/watershed or hillslope where saturation excess overland flow is likely to occur: this is a hydrological term and does not relate to P per se. VSA should not be confused with CSAs, which combine VSAs with the source of P and may be a particularly important vehicle for PT (Gburek et al., 1996). Saturation-excess overland flow is generally confined to areas subject to saturation, such as (i) areas adjacent to perennial streams (base of hillslopes), (ii) areas of concave upward slopes (slope profile concavities), (iii) hollows (areas of concave outward contours), and (iv) areas with thin or impermeable soils. In addition to the surface flow pathways described previously, subsurface mechanisms of PT are also important. Throughflow describes the transport of water below the soil surface (Thornes, 1979), along or above an impermeable subsoil horizon, such as clay or bedrock, often in a vector parallel or nearly parallel with the slope surface (Whipkey and Kirby, 1978). Throughflow is common under many soils in the United Kingdom and occurs either as saturated flow or as preferential flow (at this scale, preferential flow may include pipe flow) (Table II and Fig. 4). Return flow occurs where a subsurface flow pathway emerges at the soil surface (Miller et al., 1977). Transmission of water through the soil produces different rates of throughflow from different soil layers where, in general, saturated permeability decreases with depth (Kirkby, 1978). Dunne (1983) suggests that throughflow is an important component of storm runoff, in which soil conductivity is high. It enters the drainage network (i) via groundwater, (ii) via lateral flow where soil

172

P. M. HAYGARTH ET AL.

layers have vertical conductivity  rainfall intensity, and (iii) where concave topographic contours create contributing areas (high water table and/or subsurface impedance causes convergent flow). Zones generating subsurface flow are temporally and spatially variable. The hydrological pathways are continuous and dynamic: A hillslope may generate only subsurface flow during a gentle rainstorm, infiltration-excess overland flow during a deluge, or subsurface flow only during a short rainstorm and saturation-excess overland flow during a long event (Dunne, 1983). Finally, base flow, introduced earlier to contrast levels of hydrological activity with storm flow, can also be used in a related manner as a spatial term describing the deep percolation of water from the soil into groundwater and subsurface pathways. Deep percolation is probably only important where well-defined aquifers are present. Thus, a large part of base flow results from percolation or throughflow in the soil—a time lag of the order of months may describe water movement in this instance from interfluve (catchment/watershed boundary) to stream. Artificial land drainage provides an additional pathway of PT at the slope/field scale. This involves placing outlets for water in the subsoil to supplement the natural downward flux of water (Armstrong et al., 1984; Armstrong and Garwood, 1991; Armstrong and Harris, 1996; Tyson et al., 1993; Youngs, 1983) and involves both pipe drains and, in the United Kingdom, mole drains on heavy textured soils (Armstrong and Harris, 1996). Under grazed pasture slopes, land drainage has been shown to reduce PT (Haygarth et al., 1998a). Agricultural land drainage will inevitably influence pathways of PT depending on (i) depth and spacing of pipe drains, combined with the presence or absence of permeable fill; (ii) secondary treatment such as mole drainage or subsoiling; (iii) ditching (human-made streams to assist land drainage); and (iv) arterial drainage (a major artificial drainage channel used to remove water after its discharge from field drains). Other examples of research examining PT and land drainage are given in Roberts et al. (1986) and Sharpley and Syers (1979). The previous discussion refers to situations at the slope/field scale in which the subsurface soil and/or geology is impermeable. Where subsoils and bedrock are permeable, most water (and associated P) will be transferred to groundwater and aquifers (Denver, 1991), with negligible surface pathway activity. Where well-defined aquifers are present at shallow depths, such as in The Netherlands (Chardon and Oenema, 1995; Chardon et al., 1997) and on the east coast of the United States (Delmarva Peninsula, Chesapeake Bay, and the Atlantic Coastal Plain in Maryland) (Bachman and Phillips, 1996; Bohlke and Dener, 1995; Phillips and Bachman, 1996), PT to groundwater becomes an important environmental problem, especially where it is associated with livestock intensification (Sims, 1996). The behavior of P in deep groundwater has been a neglected area of study, in contrast to N (Bachman and Phillips, 1996), and more studies of P in throughflow and percolation pathways are required.

HYDROLOGICAL FACTORS FOR PHOSPHORUS TRANSFER

173

IV. CONCLUSIONS Water provides the energy and hydrological pathways offer the carrier route for PT from land to stream. The spatial and temporal variabilities in water movement along hydrological pathways are difficult to define and describe. However, the temporal variation in precipitation amount and intensity is clearly important in governing the magnitude of PT, whereas spatial variables control the pathway and perhaps the form of P entering the drainage network. We believe that the definitions of the spatial and temporal variability of hydrological pathways of PT at various scales given here will eliminate some of the misunderstandings that have occurred in the past. If advances are to be made in quantifying and controlling PT from diffuse agronomic sources, interpretations of PT processes must always be qualified and defined in these spatial and hydrological terms.

ACKNOWLEDGMENTS We are grateful to Rachael Dils, Andy Fraser, William Gburek, and Ben Turner for continuing inspiring “discussion” sessions. This work is funded by the Ministry of Agriculture, Fisheries and Food, London. The Institute of Grassland and Environmental Research is supported by the Biotechnology and Biological Sciences Research Council.

REFERENCES Ahuja, L. R. (1986). Characterization and modelling of chemical transfer to runoff. Adv. Soil Sci. 4, 149–188. Ahuja, L. R., Sharpley, A. N., Yamamoto, M., and Menzel, R. G. (1981). The depth of rainfall–runoff interaction as determined by 32P. Water Resour. Res. 17, 969 – 974. Anderson, M. G., and Burt, T. P. (1990). Process studies in hillslope hydrology: An overview. In “Process Studies in Hillslope Hydrology” (M. G. Anderson and T. P. Burt, Eds.), pp. 1– 8. Wiley, New York. Armstrong, A. C., and Garwood, E. A. (1991). Hydrological consequences of artificial drainage of grassland. Hydrol. Processes 5, 157–174. Armstrong, A. C., and Harris, G. L. (1996). Movement of water and solutes from agricultural land: The effects of artificial drainage. In “Advances in Hillslope Processes” (M. G. Anderson and S. M. Brooks, Eds.), pp. 187–211. Wiley, Chichester, UK. Armstrong, A. C., Atkinson, J. L., and Garwood, E. A. (1984). Grassland drainage economics experiment, North Wyke, Devon, Report No. RD/FE/23 TFS 7442. Ministry of Agriculture, Fisheries and Food, London. Atkinson, T. C. (1978). Techniques for measuring subsurface flow on hillslopes. In “Hillslope Hydrology” (M. J. Kirkby, Ed.), pp. 73 –120. Wiley, New York. Avery, B. W. (1980). “Soil Classification for England and Wales (Higher Categories).” Lawes Agricultural Trust, Harpenden, UK. Bachman, L. J., and Phillips, P. J. (1996). Hydrologic landscapes on the Delmarva Peninsula. Part 2:

174

P. M. HAYGARTH ET AL.

Estimates of base-flow nitrogen load to Chesapeake Bay. Journal of the American Water Resources Association 32, 779 –791. Beauchemin, S., Simard, R. R., and Cluis, D. (1998). Forms and concentration of phosphorus in drainage water of twenty-seven tile-drained soils. J. Environ. Qual. 27, 721–729. Betson, R. (1964). What is watershed runoff? J. Geophys. Res. 69, 1541–1552. Beven, K. (1991). Modeling preferential flow: An uncertain future? In “Proceedings ASAE Conference on Preferential Flow.” American Society of Agricultural Engineers, Chicago. Beven, K., and Germann, P. (1982). Macropores and water flow in soils. Water Resour. Res. 18, 1311– 1325. Beven, K. J., and Young, P. C. (1988). An aggregated mixing zone model of solute transport through porous media. J. Contam. Hydrol. 3, 129 –143. Beven, K. J., Henderson, D. E., and Reeves, A. D. (1993). Dispersion parameters for undisturbed partially saturated soil. J. Hydrol. 143, 19 – 43. Bohlke, J. K., and Dener, J. M. (1995). Combined use of groundwater dating, chemical, and isotopic analyses to resolve the history and fate of nitrate contamination in two agricultural watersheds, Atlantic coastal plain, Maryland. Water Resour. Res. 31, 2319 –2339. Boorman, D. B., Hollis, J. M., and Lilly, A. (1995). “Hydrology of Soil Types: A Hydrologically-Based Classification of the Soils of the United Kingdom.” Institute of Hydrology, Natural Environment Research Council, Wallingford, UK. Bromfield, S. M., and Jones, O. L. (1972). The initial leaching of hayed-off pasture plants in relation to the recycling of phosphorus. Austr. J. Agric. Res. 23, 811– 824. Brouwer, F. M., Godeschalk, F. E., Hellegers, P. J. G. J., and Kelholt, H. J. (1995). “Mineral Balances at Farm Level in the European Union,” Agricultural Economics Research Institute (LEI-DLO), Onderzoekverslag 137, The Hague. Burgoa, B., Hubbard, R. K., Wauchope, R. D., and Davis-Carter, J. G. (1993). Simultaneous measurement of runoff and leaching losses of bromide and phosphate using tilted beds and simulated rainfall. Community Soil Sci. Plant Anal. 24, 2689 –2699. Burt, T. P., Heathwaite, A. L., and Johnes, P. J. (1996). Stream water quality and nutrient export in the Slapton catchments. Field Studies 8, 613 – 627. Chardon, W. J., and Oenema, O. (1995). “Leaching of Dissolved Organically Bound Phosphorus.” DLO Research Institute for Agrobiology and Soil Fertility (AB-DLO), Wageningen, The Netherlands. Chardon, W. J., Oenema, O., del Castilho, P., Vriesema, R., Japenga, J., and Blaauw, D. (1997). Organic phosphorus solutions and leachates from soils treated with animal slurries. J. Environ. Qual. 26, 372–378. Cox, J. W., Kirkby, C. A., and Chittleborough, D. J. (1997). Pathways for losses of phosphorus from rainfed pastures in South Australia. Conference Proceedings from XVIII International Grasslands Congress, Winnipeg and Saskatoon, No. 2, pp. 35 – 36. (Full proceedings in preparation.) Cuttle, S. P., and Mason, D. J. (1988). A flow-proportional water sampler for use in conjunction with a V-notch in small catchment studies. Agric. Water Management 13, 93 – 99. Denver, J. M. (1991). Groundwater sampling network to study agrochemical effects on water quality in the unconfined aquifer, southeastern Delaware. In “Groundwater Residue Sampling Design” (R. G. Nash and A. R. Leslie, Eds.), pp. 139 –149. American Chemical Society, Washington, DC. DeWalle, D. R., and Pionke, H. B. (1994). Streamflow generation on a small agricultural catchment during autumn recharge: II. Stormflow periods. J. Hydrol. 163, 23 – 42. Dils, R. M., and Heathwaite, A. L. (1996). Phosphorus fractionation in hillslope hydrological pathways contributing to agricultural runoff. In “Advances in Hillslope Processes” (M. Anderson and S. Brooks, Eds.), pp. 229 –252. Wiley, Chichester, UK. Dunne, T. (1983). The relation of field studies and modelling in the prediction of storm runoff. J. Hydrol. 65, 25–48.

HYDROLOGICAL FACTORS FOR PHOSPHORUS TRANSFER

175

Dunne, T., and Black, R. D. (1970). Partial area contributions to storm runoff in a small New England watershed. Water Resour. Res. 6, 1296 –1311. Emmett, W. W. (1978). Overland flow. In “Hillslope Hydrology” (M. J. Kirkby, Ed.), pp. 145–176. Wiley, Chichester, UK. Evans, R. (1978). Soil erosion in north Norfolk. J. Agric. Sci. Cambridge 90, 185 –192. Farquharson, F. A. K., Mackney, D., Newson, M. D., and Thomasson, A. J. (1978). “Estimation of Runoff Potential of River Catchments from Soil Surveys.” Lawes Agricultural Trust, Harpenden, UK. Findlay, D. C., Colborne, G. J. N., Cope, D. W., Harrod, T. R., Hogan, D. V., and Staines, S. J. (1984). “Soils and Their Use in South West England.” Lawes Agricultural Trust, Harpenden, UK. Foy, R. H., and Withers, P. J. A. (1995). The contribution of agricultural phosphorus to eutrophication. Fertilizer Soc. Proc. 365, 1– 32. Fraser, A. I., Harrod, T. R., and Haygarth, P. M. (1998). The effect of rainfall intensity and hydrological pathways on phosphorus transfer from agricultural soils—Importance de l’intensité de la pluie et des chemins hydrologiques sur le transfert du phosphore provenant des sols agricoles. In “Proceedings of the 16th World Congress of Soil Science.” ISSS, Montpellier. [CD-Rom Format] Freeze, R. A. (1978). Mathematical models of hillslope hydrology. In “Hillslope Hydrology” (M. J. Kirkby, Ed.), pp. 177–225. Wiley, Chichester, UK. Gburek, W. J., and Sharpley, A. N. (1998). Hydrologic controls on phosphorus loss from upland agricultural catchments. J. Environ. Qual. 27, 267–277. Gburek, W. J., Sharpley, A. N., and Poinke, H. B. (1996). Identification of critical source areas for phosphorus export from agricultural catchments. In “Advances in Hillslope Processes” (M. G. Anderson and S. M. Brooks, Eds.), pp. 263 –282. Wiley, Chichester, UK. Harms, L. L., Dornbush, J. N., and Anderson, J. R. (1974). Physical and chemical quality of agricultural runoff. J. Water Pollution Control Federation 46, 2460 –2470. Hawkins, J. M. B., and Scholefield, D. (1996). Molybdate-reactive phosphorus losses in surface and drainage waters from permanent grassland. J. Environ. Qual. 25, 727–732. Haygarth, P. M., and Jarvis, S. C. (1996). Pathways and forms of phosphorus losses from grazed grassland hillslopes. In “Advances in Hillslope Processes” (M. G. Anderson and S. M. Brooks, Eds.), pp. 283–294. Wiley, Chichester, UK. Haygarth, P. M., and Jarvis, S. C. (1997). Soil derived phosphorus in surface runoff from grazed grassland lysimeters. Water Res. 31, 140 –148. Haygarth, P. M., and Jarvis, S. C. (1999). Transfer of phosphorus from agricultural soils. Adv. Agron. 66, 195–249. Haygarth, P. M., Hepworth, L., and Jarvis, S. C. (1998a). Forms of phosphorus transfer in hydrological pathways from soil under grazed grassland. Eur. J. Soil Sci. 49, 65 –72. Haygarth, P. M., Jarvis, S. C., Chapman, P., and Smith, R. V. (1998b). Phosphorus budgets for two contrasting grassland farming systems in the UK. Soil Use Management 14, 160 –167. Heathwaite, A. L. (1995). Sources of eutrophication: Hydrological pathways of catchment nutrient export. Int. Assoc. Hydrol. Sci. 230, 161–176. Heathwaite, A. L. (1997). Sources and pathways of phosphorus loss from agriculture. In “Phosphorus Loss to Water from Agriculture” (H. Tunney, O. T. Carton, P. C. Brookes, and A. E. Johnston, Eds.), pp. 205–224. CAB. International, Cambridge, UK. Heathwaite, A. L., and Burt, T. (1992). The evidence for past and present erosion in the Slapton Catchment. In “Soil Erosion, Past and Present: Archeological and Geographical Perspectives” (M. Bell and J. Boardman, Eds.), Vol. 22, pp. 89 –100. Oxbow Monograph, Oxford, UK. Heathwaite, A. L., and Johnes, P. J. (1996). Contribution of nitrogen species and phosphorus fractions to stream water quality in agricultural catchments. Hydrol. Processes 10, 971– 983. Heathwaite, A. L., Burt, T. P., and Trudgill, S. T. (1989). Runoff, sediment and solute delivery in agricultural drainage basins—A scale dependent approach. Int. Assoc. Hydrol. Sci. 182, 175 –191.

176

P. M. HAYGARTH ET AL.

Heathwaite, A. L., Burt, T. P., and Trudgill, S. T. (1990). The effect of land use on nitrogen, phosphorus and suspended sediment delivery to streams in a small catchment in southwest England. In “Soil Erosion on Agricultural Land” ( J. Boardman, L. D. L. Foster, and J. A. Dearing, Eds.), pp. 161–177. Wiley, Chichester, UK. Heckrath, G., Brookes, P. C., Poulton, P. R., and Goulding, K. W. T. (1995). Phosphorus leaching from soils containing different phosphorus concentrations in the Broadbalk Experiment. J. Environ. Qual. 24, 904–910. Hodgson, J. M. (1997). “Soil Survey Field Handbook—Describing and Sampling Soil Profiles,” 3rd ed. Cranfield University, Silsoe, UK. Holden, N. M., Scholefield, D., Williams, A. G., and Dowd, J. F. (1995). The simultaneous use of three common hydrological tracers for fine spatial and temporal resolution investigation of preferential flow in a large soil block. In “Tracer Technologies for Hydrological Systems,” pp. 77– 86. International Association of Hydrological Sciences, Boulder, CO. Hoover, J. R. (1987). Instrumentation system for determining watershed hydrologic characteristics. Trans. ASAE 30, 1051–1056. Hornberger, G. M., Beven, K. J., and Germann, P. F. (1990). Inferences about solute transport in macroporous forest soils from time series models. Geoderma 46, 249–262. Hornberger, G. M., Germann, P. F., and Beven, K. J. (1991). Throughflow and solute transport in an isolated sloping soil block in a forested catchment. J. Hydrol. 124, 81– 99. Horton, R. E. (1945). Erosional development of streams and their drainage basins: Hydrophysical approach to quantitative morphology. Bull. Geol. Soc. Am. 56, 275 – 330. Istok, J. D., and Boersma, L. (1986). Effect of antecedent rainfall on runoff during low-intensity rainfall. J. Hydrol. 88, 329 – 342. Johnes, P. J., Moss, B., and Phillips, G. (1996). The determination of total nitrogen load and total phosphorus concentrations in freshwaters from land use, stock headage and population data: Testing of a model for use in conservation and water quality management. Freshwater Biol. 36, 451– 473. Jones, R. J. A., and Thomasson, A. J. (1985). “An Agroclimatic Databank for England and Wales.” Lawes Agricultural Trust, Harpenden, UK. Jordan, C., and Smith, R. V. (1985). Factors affecting leaching of nutrients from an intensively managed grassland in County Antrim, Northern Ireland. J. Environ. Management 20, 1–15. Kirkby, M. J. (1978). “Hillslope Hydrology.” Wiley, Chichester, UK. Kirkby, M. (1988). Hillslope runoff processes and models. J. Hydrol. 100, 315 – 339. Knapp, B. J. (1978). Infiltration and storage of soil water. In “Hillslope Hydrology” (M. J. Kirkby, Ed.), pp. 43–72. Wiley, Chichester, UK. Konikow, L. F. (1991). “Flow and Transport Problems in Groundwater Systems.” Miljoministeriet Miljostyrelsen, Elinsore, Denmark. Kronvang, B. (1990). Sediment associated phosphorus transport from two intensively farmed catchment areas. In “Soil Erosion on Agricultural Land” ( J. Boardman, I. D. L. Foster, and J. A. Dearing, Eds.), pp. 131– 330. Wiley, Chichester, UK. Loehr, R. C. (1974). Characteristics and comparative magnitude of non point sources. J. WPFC 46, 1849–1872. Mangold, D. C., and Tsang, C.-F. (1991). A summary of subsurface hydrological and hyrochemical models. Rev. Geophys. 29, 51–79. McGechan, M. B., and Wu, L. (1996). Modelling pollutant transport processes following the land spreading of farm wastes. In “AgEng.” Universidad Politecnica de Madrid, Madrid. Miller, W. W., Guitjens, J. C., and Mahannah, C. N. (1977). Quality of irrigation water and surface return flows from selected agricultural lands in Nevada during the 1974 irrigation season. J. Environ. Qual. 6, 193. Miyazaki, T., Hasegawa, S., and Kasubuchi, T. (1993). “Water Flow in Soils.” Dekker, New York.

HYDROLOGICAL FACTORS FOR PHOSPHORUS TRANSFER

177

Molden, B., and Cerney, J. (1994). Small catchment research. In “Biogeochemistry of Small Catchments” (B. Molden and J. Cerney, Eds.), pp. 1–29. Wiley, New York. National Rivers Authority (1992). “Policy and Practice for the Protection of Groundwater.” Stanley Hunt, Rushden, UK. Organisation for Economic Cooperation and Development (OECD) (1982). “Eutrophication of Waters, Monitoring, Assessment and Control.” OECD, Paris. Phillips, P. J., and Bachman, L. J. (1996). Hydrologic landscapes on the Delmarva Peninsular. Part 1: Drainage basin type and base flow chemistry. J. Am. Water Resour. Assoc. 32, 767–778. Phillips, R. E., Quisenberry, V. L., and Zeleznik, J. M. (1995). Water solute movement in an undisturbed, macroporous column: Extraction pressure effects. Soil Sci. 59, 707–712. Pionke, H. B., and DeWalle, D. R. (1994). Streamflow generation on a small agricultural catchment during autumn recharge: I. Nonstormflow periods. J. Hydrol. 163, 1–22. Pionke, H. B., Gburek, W. J., Sharpley, A. N., and Schnabel, R. R. (1996). Flow and nutrient export patterns for an agricultural hill–land watershed. Water Resour. Res. 32, 1795 –1804. Pionke, H. B., Gburek, W. J., Sharpley, A. N., and Zollweg, J. A. (1997). Hydrologic and chemical controls on phosphorus losses from catchments. In “Phosphorus Loss to Water from Agriculture” (H. Tunney, O. T. Carton, P. C. Brookes, and A. E. Johnston, Eds.), pp. 225–242. CAB International, Cambridge, UK. Roberts, G., Hudson, J. A., and Blackie, J. R. (1986). Effect of upland pasture improvement on nutrient release in flows from a “natural” lysimeter and a field drain. Agric. Water Management 11, 231–245. Sharpley, A. N. (1980a). The effect of storm interval on the transport of soluble phosphorus in runoff. J. Environ. Qual. 9, 575 – 578. Sharpley, A. N. (1980b). The enrichment of soil phosphorus in runoff sediments. J. Environ. Qual. 9, 521–526. Sharpley, A. N. (1985). Depth of surface soil runoff as affected by rainfall, soil slope and management. Soil Sci. Soc. Am. J. 49, 1010 –1015. Sharpley, A. N., and Smith, S. J. (1990). Phosphorus transport in agricultural runoff: The role of soil erosion. In “Soil Erosion on Agricultural Land” ( J. Boardman, L. D. L. Foster, and J. A. Dearing, Eds.), pp. 351–366. Wiley, Chichester, UK. Sharpley, A. N., and Syers, J. K. (1979). Loss of nitrogen and phosphorus in tile drainage as influenced by urea application and grazing animals. N. Z. J. Agric. Res. 22, 127–131. Sharpley, A. N., Daniel, T. C., Sims, J. T., and Pote, D. H. (1996). Determining environmentally sound soil phosphorus levels. J. Soil Water Conserv. 51, 160 –166. Sims, J. T. (1996). “The Phosphorus Index: A Phosphorus Management Strategy for Delawares Agricultural Soils.” Univ. of Delaware, Newark. Singh, P., and Kanwar, R. S. (1991). Preferential solute transport through macropores in large undisturbed saturated soil columns. J. Environ. Qual. 20, 295 – 300. Stamm, C., Fluhler, H., and Gachter, R. (1998). Preferential transport of phosphorus in drained grassland soils. J. Environ. Qual. 27, 515 – 522. Thomas, G. W., and Phillips, R. E. (1979). Consequences of water movement in macropores. J. Environ. Qual. 8, 149–152. Thornes, J. B. (1979). Fluvial processes. In “Process in Geomorphology” (C. Embleton and J. Thornes, Eds.), pp. 213–271. Edward Arnold, London. Timmons, D. R., Verry, E. S., Burwell, R. E., and Holt, R. F. (1977). Nutrient transport in surface runoff and interflow from an aspen–birch forest. J. Environ. Qual. 6, 188 –192. Tindall, T. A., Morrill, L. G., and Henderson, L. J. (1986). The influence of soil structure on movement rate of salts through soils. Soil Sci. 142, 108 –113. Trudgill, S. T., and Coles, N. (1988). Application of simple soil–water models to the transfer of surface-applied solutes to streams. J. Contam. Hydrol. 3, 367– 380.

178

P. M. HAYGARTH ET AL.

Tyson, K. C., Hawkins, J. M. B., and Stone, A. C. (1993). “Final Report on the AFRC–ADAS Drainage Experiment 1982–1993.” Institute of Grassland and Environmental Research, Okehampton, UK. U.S. Department of Agriculture (1972). “National Engineering Handbook, Section 4, Hydrology.” U.S. Government Printing Office, Washington, DC. Wagenet, R. J. (1990). Quantitative prediction of the leaching of organic and inorganic solutes in soil. Philos. Trans. R. Soc. London 329, 321– 330. Weaver, D. M., Ritchie, G. S. P., and Anderson, G. C. (1988a). Phosphorus leaching in sandy soils. II. Laboratory studies of the long-term effects of the phosphorus source. Austr. J. Soil Res. 26, 191– 200. Weaver, D. M., Ritchie, G. S. P., Anderson, G. C., and Deeley, D. M. (1988b). Phosphorus leaching in sandy soils. I. Short term effects of fertiliser applications and environmental conditions. Austr. J. Soil Res. 26, 177–190. Whipkey, R. Z., and Kirkby, M. J. (1978). Flow within the soil. In “Hillslope Hydrology” (M. J. Kirkby, Ed.), pp. 121–144. Wiley, Chichester, UK. Wopereis, M. C. S. (1994). Reducing bypass flow through a dry, cracked and previously puddled rice soil. Soil Tillage Res. 29, 1–11. Youngs, E. G. (1983). The contribution of physics to land drainage. J. Soil Sci. 34, 1–21. Youngs, E. G., and Leeds-Harrison, P. B. (1990). Aspects of transport processes in aggregrated soils. J. Soil Sci. 41, 665 – 675. Zobisch, M. A., Richter, C., Heiligtag, B., and Schlott, R. (1994). Nutrient losses from cropland in the central highlands of Kenya due to surface runoff and soil erosion. Soil Tillage Res. 33, 109 –116.

CASSAVA, Manihot esculenta Crantz, GENETIC RESOURCES: THEIR COLLECTION, EVALUATION, AND MANIPULATION Nagib M. A. Nassar Departamento de Genética e Morfologia Universidade de Brasília Brasília 70919, Brazil

I. Wild Taxa of Cassava Manihot Species A. Taxonomy B. Determination of Wild Manihot Species Localities with Emphasis on Probable Origin C. Relationships between Manihot Species D. Genetic Variation of Wild Manihot Species II. Broadening the Genetic Base of Cassava, M. esculenta Crantz, and Development of Interspecific Hybridization A. Production of Cassava Interspecific Hybrids B. Development of Cassava Interspecific Hybrids C. Development of Cassava Interspecific Hybrids for Savanna (Cerrado) Conditions D. Overcoming Crossing Barriers between Cassava, M. esculenta Crantz, and a Wild Relative, M. pohlii Warwa III. Development and Selection for Apomixis in Cassava, M. esculenta Crantz A. Genetic Study of Apomixis in Cassava B. Molecular and Embryonic Evidence of Apomixis in Cassava IV. Production of Polyploid Types A. Prospects of Polyploidizing Cassava, M. esculenta Crantz, by Unreduced Microspores B. Induction of a Productive Aneuploid in Cassava M. esculenta Crantz C. Production of Triploid Cassava, M. esculenta Crantz, by Hybrid Diploid Gametes V. Protein Contents in Cassava Cultivars and Its Hybrid with Wild Manihot Species References

179 Advances in Agronomy, Volume 69 Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved. 0065-2113/00 $30.00

180

NAGIB M. A. NASSAR

Wild species of Manihot are progenitors of cassava. They constitute valuable genetic reservoirs with genes of new characters. Screening these species showed some of them to have a notably high percentage of protein combined with a low percentage of hydrocyanic acid. Study of natural habitats revealed toxicity and adaptation to cool temperature. Hybridizations between different wild Manihot species and cassava have been carried out and hybrids were obtained, some of which showed high root productivity and resistance to stem borers. Polyploid types were produced by manipulation of 2n gametes. Apomixis was discovered in the wild and transferred successfully to the cultivate. © 2000 Academic Press.

I. WILD TAXA OF CASSAVA Manihot SPECIES A. TAXONOMY Cassava Manihot esculenta Crantz does not grow wild. However, it is known that about 98 species belong to the genus Manihot (Rogers and Appan, 1973). These species include subshrubs, shrubs, and trees. A majority of them produce latex and contain cyanogenic glucoside. Roots of wild species, contrary to the cultigen, are fibrous and slender, but some species exhibit a limited number of tuberous roots. The root surface is smooth or rought. The subepidermis varies in color from red or yellow to white, and the cortexes of tuberous rooted species are white, cream, or yellow. The stem varies in height from almost acaulescent in subshrubs to about 20 m in tree species. Frequently, the stems of shrubs native to the Brazilian savanna die back to the crown in the dry season. The color of the stem varies from gray or brown to red. The stem normally branches dichotomously or trichotomously. The branching point exhibits a terminal inflorescence. In wild species frequently the young stem has a varying degree of pubescence. This characteristic is rarely encountered in the cultigen (Grattapaglia et al., 1986). Leaves vary from subsessile to long petiolated. All species with the exception of three have palmately lobed leaves. Infloresence is terminal and monoecious with the exception of the acaulescent species native to central Brazil. Flowers have a single perianth composed of five petals; their length ranges from 0.5 to 2.0 cm. Buds of staminate flowers are ovoid or spheric, whereas pistillate buds are conic. The fruit is capsule with three locules. The caruncle of the seed varies in size. The chromosome number in all species investigated is 2n 36 (Nassar, 1978a). All species of the genus Manihot are native to the New World; in Brazil and Mexico they form distinct centers of diversity (Nassar, 1978b). They normally grow sporadically in their habitat, rarely becoming dominant in the vegetation. Due to the monoecious or dioecious structure of the inflorescence, wild Manihot

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

181

species are typically alogamous plants. However, in cultivated cassava a shift toward autogamous plants has occurred. Nassar and O’Hair (1985) attributed this to the monoclonal system of cultivation and domestication history. Observations of frequent hybridization of the wild species and the cultigen and between wild species suggest weak barriers in the genus (Nassar, 1980a). This is probably due to the polyploid origin on the genus level.

B. DETERMINATION OF WILD Manihot SPECIES LOCALITIES WITH EMPHASIS ON PROBABLE ORIGIN From May to July 1975, I collected seeds of the wild species of Manihot in northeastern Brazil in three states: Pernambuco, Ceará, and Bahia. the geographical distribution of the Manihot species was studied by Rogers and Appan (1973). Manihot specimens collected by the expedition of Reading University and deposited at IPA herbarium, Recife, were also examined. Table I lists the wild species of Manihot that were collected from different localities of northeastern Brazil. It is apparent that western Pernambuco and central Bahia had the greatest variability in Manihot. It should be noted that certain species reported by the Reading University expedition occur in some localities from which they no longer can be collected (e.g., specimens of M. glaziovii collected about 12 km west of Ibimirim, PE). Unfortunately, it was found that vegetation in that place had been cleared and the land cultivated by mamona. Unlike most Manihot species, M. glaziovii grows in large numbers and not as sporadic plants. Extinction of some wild Manihot species from their natural habitats may be due to another factor: The majority of these species are poisonous to grazing animals because of the presence of HCN.

Table I Wild Species of Manihot Collected from Different Localities in Northeastern Brazil Species

Locality

M. caerulescens Pohl M. heptaphylla Ule M. cichotoma Ule M. catingae Ule M. brachyandra Pax et Hoffmann M. maracasensis Ule M. epruinosa Pax et Hoffmann M. glaziovii Mueller M. jacobinensis Mueller M. quinquefolia Pohl

Aparipina, PE Seabra, BA Jequie, BA Itaberaba, BA Petrolina, PE Itambé, BA Bentecoste, Fortaleza, CE Arcoverde, Ouricure, Serratatlada, PE Virtoria da Conquista, BA Senhor do Bonfim, Juazeiro, BA

182

NAGIB M. A. NASSAR

They are known among people of northeastern Brazil as “maniçoba,” i.e., the poisonous cassava. Therefore, many plants are exterminated by farmers. By studying the geographic distribution of Manihot species provided by Rogers and Appan (1973) combined with localities determined on my trip, it became possible to create a map of concentration of wild species. It shows that in central Brazil (southern Goias and eastern Minas Gerais) there are approxiimately 38 wild species of the 98 recognized. Thus, this region includes a large number of wild Manihot species and represents the highest diversity of these species. In this region the following species occur: M. acuminatissima Mueller M. sparsifolia Pohl M. pruinosa Pohl M. alutacea Rogers et Appan M. divergens Pohl M. cecropiaefolia Pohl M. triphylia Pohl M. pentaphylla Pohl M. anomala Pohl M. procumbens Mueller M. crotalariaeformis Pohl M. Pusilla Pohl M. logepetiolata Pohl M. tomentosa Pohl M. purpureo-costata Pohl M. attenuata Mueller M. orbicularis Pohl M. tripartita (Sprengel) Mueller M. pilosa Pohl M. sagittato-partita Pohl M. falcata Rogers et Appan M. quinqueloba Pohl M. violacea Pohl M. irwinii Rogers et Appan M. mossamedensis Taubet M. fruticulosa (Pax) Rogers et Appan M. gracilis Pohl M. warmingii Mueller M. reptans Pax M. stipularis Pax M. oligantha Pax M. nana Mueller

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

183

M. stricta Baillon M. salicifolia Pohl M. weddelliana Baillon M. peltata Pohl M. janiphoides Mueller M. handroana N. D. Cruz The second largest center of diversity is southwestern Mexico, which includes M. pringlei Watson M. aesculifolia Pohl M. oaxaca Rogers et Appan M. rhomboidea Mueller M. walkarae Croizat M. divisiae Croizat M. michaelis McVaugh M. websterae Rogers et Appan M. auriculata Mcvaugh M. rubricaulis I. M. Hohnson M. chlorosticta Standley et Goldman M. subspicata Rogers et Appan M. caudata Greenman M. angustiloba (Torrey) Mueller M. tomatophylla Standley M. foetida Pohl The third largest center of diversity is northeastern Brazil, which includes M. zenhtneri Ule M. surinamensis Rogers et Appan M. quinquefolia Pohl M. pseudoglaziovii Pax et Hoffmann M. maracasensis Ule M. quinquepartita Huber M. caerulescens Pohl M. marajoara Chermont de Miranda M. tristis Mueller M. glaziovii Mueller M. epruinosa Paz et Hoffmann M. brachyandra Pax et Hoffmann M. dichotoma Ule M. leptophylla Pax M. reniformis Pohl M. heptaphylla Ule

184

NAGIB M. A. NASSAR

Finally, the fourth largest center of diversity is western South Mato Grosso and Bolivia. It includes the following species: M. guaranitica Choda et Hassier M. pruinosa Pohl M. jacobinsis Mueller M. condesata Rogers et Appan M. xavantinensis Rogers et Appan M. flemingiana Rogers et Appan Vavilov (1951) showed that variation in cultivated plants is confined to relatively few restricted areas or centers. In 1920, he set up 6 main geographic centers for cultivated plants and in 1935 increased the number of centers to about 10. He assumed that the center of diversity for cassava was in the Brazilian–Bolivian center. Vavilov proposed that centers of diversity are places of origin of cultivated plants. Since this exposition of the center of diversity in the 1920s, much more information has been gathered and it has become clear that not all centers of diversity represent centers of origin. Harlan (1961) showed that more than one center of diversity may be formed for a given crop through introgression. This phenomenon explained why in many cases there are centers of diversity for a given crop far from areas of much diversity of wild relatives. Since Harlan proposed this theory (giving a convincing example of the evolved species of Helianthus) much evidence has supported it. Dobzhansky (1973) stated many conspicuous cases, such as the formation of species of Iris, Eucalyptus, Liatris, Penstemon, and Tragopogon. Thus, this phenomenon serves as a model for what apparently occurred during the formation of these four centers of diversity of Manihot. Assuming that cassava was domesticated for the first time in one place and then carried by Indians through immigrations, there could then result an extensive hybridization between the cultivated species and local wild ones, giving rise to numerous new species through introgression. Cassava does not grow wild. The large variation of cassava cultivars due to maintaining them by vegetative reproduction over hundreds of years makes it difficult to designate definite characteristics for M. esculenta. Thus, it is believed that this species did not arise by natural selection. Hybrids between some wild species may have been domesticated and maintained afterwards through vegetative reproduction. Surely if these cultivars were left to sexual reproduction and subjected to natural selection, this would have led to different populations with specific gene pools depending mainly on local environments. Our assumption is that domestication included some natural hybrids and that the selected plants were maintained by vegetative reproduction for hundreds of years. This assumption is supported by the fact that many experimental crosses and observations led to frequent hybridity of cultivars of M. esculenta and local wild species (Abraham, 1975; Bolhuis, 1953; Cruz, 1968; Jennings, 1959; Lanjouw, 1939; Magoon et al., 1966;

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

185

Nichols, 1947). In this genus, systems of genetic and cytologic barriers are not well established. Additional support may come from Schmidt’s (1951) statement about the very rapid response of selection in different wild species to increase starch content in tubers and tuber formation through only a few generations. It seems that many different wild species have the potential to increase tuber formation and starch content. I observed two tree species of Manihot (M. epuinosa and M. brachyandra) frequently grown in dooryards at Goiania with considerable tuber production. These two species are natives of Bahia. It seems that they were carried by people of this state. Immigration was common during the past 30 years due to the rapid development of Goias. The assumption that domestication included hybrids and did not include a certain wild species has been discussed by Rogers (1963) using the expression “species complexity.” The place of domestication still needs much discussion. I prefer to use “place of domestication” and not “center of origin” because it is obvious that this crop was not brought to existence as a wild species by means of natural selection. Studying the history of ethnological groups in Brazil and their immigrations throws light on the subject. It is reported that the “Aruak,” who lived in north Amazonia more than 1000 years ago (Schmidt, 1951), knew cassava and practiced a developed agriculture. Aruak in the Indian language means “people who eat tubers.” Numerous reports indicate that they cultivated cassava many centuries before the arrival of Columbus. The Aruak were obliged to immigrate in the eleventh century to Central America, crossing the Caribbean and establishing themselves in the West Indies. Many reasons were given to explain their immigration, but the most likely are that they were escaping from enemies or possibly looking for a place where man does not die. However, the most important reason given was that they were searching for a better soil to cultivate cassava. However, this immigration coincides with the formation of a center of diversity of Mexico. Cassava carried by the Aruak to Mexico would be expected to hybridize with local wild species, thus creating a center of diversity. The fact that the Aruak continued on to Planalto Boliviano and to central Brazil is in agreement with the existence of the two centers of diversity in these regions. The northeastern Brazilian center of diversity is believed to be the result of immigration of the Tupi-Guarani group. We must still determine which of these four centers constitutes the primary center of diversity of Manihot. In other words, Manihot as a “biological group” must have passed their differentiation in a certain region from which species spread to other regions. Central Brazil, with its enormous number of species of Manihot, is the primary center. Indeed, this region is an ancient area long available for growth of the angiosperms. Considering Stebbins (1950) explanations of Vavilov’s interpretation of diversity patterns may be useful here: Vavilov’s concept is an elaboration of Willy’s age-and-area hypothesis (i.e., the longer a given biological entity occupies an area, the more variability of Manihot species), and this area might constitute its primary center of diversity. This assumption is supported by the fact

186

NAGIB M. A. NASSAR

that species which exhibit the most primitive characteristics are restricted to this region: M. stipularis Pax, M. pusilla Pohl, M. longipetiolata Pohl, with their dioecious inflorescences, and M. stricta Baillon, M. purpureo-costata Pohl, and M. salicifolia Pohl, with their nonlobed and sessile leaves.

C. RELATIONSHIPS BETWEEN Manihot SPECIES According to Rogers and Appan (1973), 98 Manihot species have been recognized. Only 1 species, Manihotoides pauciflora, is known in the closest related genus, Manihotoides. Several of its attributes are not found in any Manihot species, including uniflorous inflorescences, which is a primitive characteristic compared with the multiflowered inflorescence in Manihot, and leaves born at the apex of short, condensed stems arising from branchlets. Rogers and Appan classified Manihot species into 19 sections, varying from trees in the section Glazioviannae to subshrubs, nearly acaulescent, in the section Stipularis. The species in this latter section are also characterized by being more dioecious than monoecious, a condition reversed in all other Manihot species. Other sections, such as Tripartitae and Graciles, are perennial subshrubs with large woody roots; their stems frequently die back to the root crown in response to dry periods or fires. All Manihot species are native to tropical regions of the New World, particularly Brazil and Mexico. Nassar (1978b) defined four centers of diversity for these species: Mexico and northeast, central, and southwest Brazil. Microcenters of diversity of these species exist within central Brazil, where large numbers of species are concentrated in small areas (50 km in diameter) (Nassar, 1978b,c,d,e,f, 1979a,b, 1980a,b, 1982, 1984, 1985, 1986). Nassar attributes the formation of these microcenters to the frequent hybridization between species and the heterogenic topography of their habitats, which help isolate fragmented gene pools that lead to speciation. Tree-like species, such as M. glaziovii and M. pseudoglaziovii, are found in northeastern Brazil, whereas short species and subshrubs are found in central Brazil. Natural hybridization occurs between wild Manihot species and between these and cassava (Nassar, 1984, 1989). Barriers within the genus appear to be weak due to recent evolution of the group. All wild Manihot species examined cytogenetically have a chromosome number of 2n 36 (Nassar, 1978a) (Table II). Despite this high chromosome number, Manihot species behave meiotically as diploids. Therefore, they are believed to be allopolyploids and this seems to have anticipated the emergence of the whole group and is responsible for their rapid speciation and their weak interspecific barriers, leading to interspecific hybridization. An extremely heterozygous gene pool is thus created, followed by differentiation; this begins a sequence of hybridization followed by speciation. Nassar (1980a) reported frequent hybridization between M. reptans Pax and m. alutacea Rogers et

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

187

Table II Chromosome Number in Wild Manihot Species Species M. handroana M. jolyana M. tripartita M. tweedieana M. humilis M. pedicellaris M. gracilis M. dichotoma M. giaziovii M. anomala M. zehntneria M. olighanta M. nana M. tomentosa

Growth habitat Shrub Shrub Shrub Shrub Shrub Subshrub Shrub Subshrub Subshrub Tree Tree Tree Shrub Shrub Subshrub Subshrub Subshrub

n

2n

— — — 18 — — — — 18 — 18 — 18 18 18 18 18

36 36 36 — 36 36 36 36 — 36 — 36 — — — — —

Appan in sympatric natural habitats in which their population boundaries overlap. Morphological marker gene leaf color and bract size were used to identify this interspecific hybridization. The range of M. reptans has expanded during the past 100 years (Nassar, 1984) and this is attributed to the continuing gene introgression of Manihot species. Introgression of M. reptans with germplasm from other species allowed its ecotypes to penetrate and colonize areas where M. reptans (pure) had previously been unable to do so. This phenomenon was also noted in other species such as M. cearulescens (Nassar, 1980a). From a plant breeding viewpoint, the value of these hybrids lies in their ability to cross with the cultigen. Marker genes lobe shape, the presence of stem nodes, flower disc color, fruit color, and fruit shape were discovered in controlled crosses between cassava and wild Manihot species as well as in natural hybrids between cassava and different species. These genes were used by Nassar to identify hybridization. Interspecific hybrids of cassava with M. glaziovii, M. pseudoglaziovii, M. aesculifolia, M. pilosa, M. corymbilora, M. dichotoma, M. pohlii, M. neusana, and M. anomala were obtained by Nassar through controlled crosses, although their frequency was low. The meiotic behavior of several hybrids (cassava with M. neusana and cassava with M. pseudoglaziovii) was studied by Nassar (1992), and results indicated low hybrid fertility between these species and cassava. Grattapaglia et al. (1986) conducted a biosystematic analysis of wild Manihot

188

NAGIB M. A. NASSAR

species based on soluble seed protein pattern. Nineteen species were analyzed electrophoretically (Table III). A similarity matrix between species, which considered differences in band density and number, was established (Table IV). Several species were found to be highly similar, for example, M. fruticolosa and M. pentaphylla and M. pilosa and M. corymbiflora. These results correlate well with the taximetric analysis made by Rogers and Appan (1973). Manihot pilosa and M. corymbiflora are the most similar species to cassava. Profile analysis confirmed the introgression between M. cearulescens and cassava. The electrophoresis of 0.1% of SDS was conducted according to Laemilli (1970). The concentrator gel containing 5.5% of acrylamide Tris–HCl (pH 6.8) was prepared and fixed during 12 h in 5% trichloroacetic acid. The bands were colored by Coomassie brilliant blue (0.65%). Every species had its profile revealed in four different gels. The approximate molecular weights (AMWs) were determined according to Webber and Osborn (1969).

Table III Wild Manihot Species and Their Identification Number in the Germplasm Bank at the Universidade de Brasília

Species

Section

M. esculenta Crantz (var. EAB) I. (A) Manihot M. esculenta Crantz (var. RB) (B) Manihot M. zehntnieri Ule II. (C) Heterophyllae M. grahami Hooker (D) Heterophyllae M. pilosa Pohl (E) Heterophyllae M. corymbioflora Pax (F) Heterophyllae M. pohlii Wawra (G) Heterophyllae M. glaziovii Muell III. (H) Glaziovinae M. pseudoglaziovii Pax et Hoffmann (I) Glaziovinae M. epruinosa Pax et Hoffmann (J) Glaziovinae M. brachyandra Pax et Hoffmann (K) Glaziovinae M. reptans Pax IV. (L) Crotalariaeformes M. alutacea Rogers et Appan V. (M) Quinquelobae M. fruticulosa Rogers et Appan VI. (N) Graciles M. pentaphylla Pohl (O) Graciles M. stipularis Pax VII. (P) Stipulares M. salicifolia Pohl VIII. (Q) Brevipetolatae M. caerulescens subsp. Caerulescens IX. (R) Caerulescentes M. caerulescens (no classification) (S) Caerulescentes M. caerulescens (no classification) (T) Caerulescentes M. leptophylla Pax X. (U) Peuvianae M. neuzana Nassar —. (V) —

Habitat Brasília (DF) Brasília (DF) Goiânia (GO) Maringá (PR) São Miguel de Antes (MG) São Miguel de Antes (MG) Lençóis (BA) Pentocoste (CE) Remigio (PB) Serra Talhada (PE) Currais Novos (RN) Corumbá (GO) Goiás Velho (GO) Alexânia (GO) Goiás Velho (GO) Alexânia (GO) Xavantina (MT) Picos (PI) Morro do Chapéu (BA) Jequié (BA) Barra do Corda (MA) Maringá (PR)

No. 01 02 173 375 601 605 139 221 545 554 524 602 115 162 103 184 195 258 567 269 517 360

No. herbarium collection 01 01/a 02 03 04 05 06 08 09 10 11 13 07 10,938 11,755 14 — 15 16 17 12 18

Table IV Classification of Bands in Wild Manihot Species According to Approximate Molecular Weight (AMW)a Section I No. AMW (kDaltons) 81–75 75–66 66–62 62–50 50–37.5 37.5–33 33–30 30–27 27–25 25–24 24–21 21–20 20–18 18–13 No. bands No. reference bands No. total bands aFor

II

III

IV

V

VI

VII

VIII

IX

X

—

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

R

S

T

U

V

— 2 1 3 5 1 1 2 1 — 1 — 1 3 21 15

— 2 1 4 5 1 1 2 1 — 1 — 1 1 20 15

— 2 1 4 6 2 2 2 — 1 — 2 1 2 24 15

— 2 1 3 6 1 2 2 — — — 2 2 3 24 14

— 2 1 3 6 1 1 2 — — 1 — — 3 20 15

— 2 1 4 6 1 1 2 — — — — — 3 20 15

— 2 1 4 6 1 — 2 — — — 1 1 3 21 14

— 2 — 3 5 1 2 1 — — 1 1 — 3 19 15

— 2 1 4 6 1 2 1 — — 1 — — 2 20 15

— 2 1 4 6 1 — 1 — — — — 1 3 19 15

— 2 1 3 6 1 2 1 — 1 1 — — 3 20 15

— 2 1 3 4 1 3 2 — — 1 1 — 3 21 15

— 2 1 3 5 1 1 3 — — 1 1 1 3 20 15

— 2 1 3 3 1 2 1 — — 1 — — 3 17 14

— 2 1 3 4 1 2 1 — — 1 1 — 2 18 14

— 2 — 1 4 1 3 3 — — 1 — — 2 18 14

— 2 1 5 5 1 1 1 — — 1 — 1 2 20 14

— 1 — 1 4 1 1 1 — — — — — 2 11 15

— 1 — 1 4 1 1 1 — — — — — 3 12 15

— 3 1 4 5 1 2 1 1 — — 1 1 2 22 15

— 2 — 3 3 1 1 1 — — — — — 3 15 15

— 2 1 3 4 2 2 1 — 1 1 — — 3 20 14

36

35

39

38

35

35

35

34

35

34

36

36

35

31

32

32

34

26

27

37

30

34

identification species see Table II.

190

NAGIB M. A. NASSAR

To proceed with the analysis of profiles, 15 bands were selected as references; their aspects were evaluated in all four profiles obtained for every species (Table V). They were classified into four categories of intensity: A, absent; B, slightly visible; C, visible; D, dense band; and E, very dense band. To compare quantitatively the protein patterns, the total values of every band were calculated and expressed in numbers. The protein profiles varied in band intensity, and 15 bands were selected as references. The variability of wild Manihot species in morphology, growth habit, and geographic distribution was reflected in the electrophoretic profiles as differences in number and intensity of visible bands (Table VI). The two studied varieties of cassava showed a 78% similarity with different species of the section Glaziovinae. Manihot glaziovii Mueller and M. pseudoglaziovii Pax et Hoff showed a high index of similarity based on electrophoresis analysis. A high similarity was also found in species of the section Gracilis in which its level reached 78%. The same high similarity was found in species of the section Heterophyllae. Within this section, the species M. pilosa and M. corymbiflora showed the highest similarity to the cultivated M. esculenta Crantz, coinsidereding at the same time with morphological affinities between them and cassava. They are probably part of the complex from which the cultigen had originated (Nassar, 1978b). The high similarity between species in various sections indicates their recent speciation and accords with the taxonomic classification. Genetically, they probably share the same gene pool.

D. GENETIC VARIATION OF WILD Manihot SPECIES Through the program at the Universidade de Brasília, wild Manihot species were collected from South and Central America, evaluated, reproduced, and maintained in a living collection (Nassar, 1986). The description and identification of the wild Manihot species was made according to Rogers and Appand (1973) and Nassar (1978b). Natural habitats were described, and 30 accessions of each species were examined for the following characteristics: (i) tuber formation and their protein and hydrocyanic acid (HCN) content—tubers were taken 1 year after planting and analyzed according to the Association of Official Analytic Chemists (AOAC) methods (1970); (ii) adaptation to various soil types—chemical analysis of soil was carried out according to Black et al. (1965); (iii) data regarding adaptation to various habitats were extracted from records of federal meteorological stations; and (iv) seeds, cuttings, or whole plants of the collected species were planted in a living collection at the Universidade de Brasília. The collected wild Manihot species were screened rapidly for tuber formation and growth habit. Results of this screening are given in Table VII.

Table V Distribution of Reference Bands According to Density in Studied Wild Manihot Species Profilesa Section I

II

III

Reference band No.AMW

A

B

C

D

E

F

G

H

I

J

01/81 02/75 03/66 04/62 05/50 06/37.5 07/33 08/30 09/27 10/25 11/24 12/21 13/20 14/18 15/13

B B C C C C E C C B B D D D D

B B C C D D E B C B B D D D D

B B C C C C B B C B B B B D D

B B C C C C B C C B C B A C C

B B C C D C D C C B B C C C C

B B C C C C B C C B B B C D C

B B C C C C C C C C B C A C C

B B C C B C D C B B B C C D D

B B C C C C D C B B D C C D C

B B C C C C C E D B B C C D D

IV

V

VI

K

L

M

N

B B C C C C C D D B B C C C C

B B C C C C B C C C D C C C C

B B C C B B B B B B B B B C C

B B C C C C D C B B B E E B A

Note. Abbreviations used: A, absent band; B, slightly visible band; C, visible band; D, dense band; E, very dense band. aSee Table III for species identification.

IX

VII

VIII

O

P

Q

R

S

B B C C B B D C B C B E E B A

B B C C B B D B A B B E E B C

B B C C C C D C C C C C B D A

B B C C B B B C C D C B B B C

B B C C B B B C C D C B C B C

X

—

T

U

V

B B C C D D C C C C C C C D C

B B C C C C D C C B D B B D C

B B C C C C D C C C C C A B C

Table VI Matrix of Similarity between Studied Manihot Speciesa Section I Section species A B C D E F G H I J K L M N O P Q R S T U aSee

II

III

IV

V

VI

VII

VIII

IX

X

—

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

R

S

T

U

V

—

78 —

54 49 —

45 38 62 —

67 68 51 47 —

64 68 65 53 75 —

58 68 48 65 61 58 —

66 61 51 40 70 67 51 —

64 60 49 45 75 71 54 74 —

58 56 51 40 63 67 51 71 59 —

58 54 54 50 74 70 65 60 64 74 —

58 50 54 54 66 70 65 55 70 55 71 —

50 52 59 47 62 71 52 61 45 52 59 59 —

45 42 31 30 46 42 38 49 45 43 41 38 40 —

43 41 30 29 44 40 39 50 43 41 39 39 43 78 —

43 44 32 30 45 41 38 48 43 42 38 46 50 55 —

54 52 45 39 60 58 50 56 62 52 52 56 50 50 51 36 —

30 28 33 32 36 41 33 36 36 34 35 41 45 32 33 32 35 —

32 31 33 33 39 44 35 39 38 36 37 45 44 34 36 35 35 88 —

54 53 50 50 66 58 59 50 58 50 64 70 36 37 38 37 51 38 39 —

47 43 44 39 53 56 41 54 63 49 47 51 47 43 37 38 49 49 50 43 —

50 50 40 59 62 56 78 52 60 48 54 59 50 42 43 41 56 36 38 58 45

Table III for species identification.

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

193

Table VII Tuber Formation, Growth Habit, and Other Characteristics of the Wild Manihot Species Tuber formation

Growth habit

Other characteristics

M. oligantha Pax subsp. M. tripartita Mueller Arg. M. anomala Pohl

  

Subshrub Subshrub Tall shrub

M. zehntneri Ule M. gracilis Pohl

 

Tall shrub Subshrub

M. paviaefolia Pohl



Subshrub

M. pruinosa Pohl M. falcata Rogers et Appan

 

Subshrub Subshrub

M. reptans Pax



Shrub

M. alutacea Rogers et Appan



Subshrub

M. pentaphylla Pohl M. caerulescens Pohl

 

Subshrub Tall shrub

M. procumbens Mueller Arg.



Subshrub

M. stipularis Pax



Subshrub

Abundant cylindric tubers Abundant spheric tubers Abundant spheric tubers with strong HCN smell Abundant conic tubers Rare tubers; grows throughout central Brazil Rare tubers; occurs in poor sandy soil Grows in limestone soil Grows on slopes and well-drained soil Adapted to large range of soil; exhibits different leaf shapes; hybridizes easily with other Manihot species occurring in its natural habitat Grows on rocky slopes at ⬃ 1200-m altitude Grows in limestone soil Adapted to dry areas of northeastern Brazil Grow in poor soil, with high concentration of aluminum Adapted to high altitude, ⬃1450 m

Species

1. Tuber Formation Pattern and Protein Content Among the wild species collected from Goias state, Brazil, four species formed abundant tubers (Nassar, 1978d) and these were screened for tuber formation pattern and fiber and protein content (Nassar, 1978e): M. oligantha Pax emend. Nassar subsp. nestili, collected from Cristalina; M. tripartita Mueller, collected from Serra Dourada, municipal Goiania; M. zehntneri Ule, collected from Goianesia; and M. anomala Pohl, collected from Goiania–Inhumas road. These species differed greatly in tuber formation pattern and tuber composition. Manihot oligantha subsp. nestili forms abundant cylindrical tubers superficially (10–30 cm below the soil surface) which are dark brown with a rough surface. Manihot tripartita forms spherical tubers deep in the ground (50 cm) which are bright brown and smooth with a creamy cortex. Manihot anomala forms superficial tubers at a depth of about

194

NAGIB M. A. NASSAR

20–30 cm that are oval shaped with a rough surface and light brown color with a creamy cortex. Manihot zehntneri forms cylindrical to oval tubers at a depth of about 50–70 cm which are dark brown with a rough surface and a white cortex (Nassar, 1978f). Protein and fiber contents are shown in Table VIII. The composition of cassava as reported in the literature is variable. This variation results from the fact that bitter cultivars differ from sweet ones not only in the amount of HCN they contain but also in the proportions of nutrients. According to Bolhuis (1953), cultivars with roots containing 50 mg of HCN per kilogram are considered sweet. However, many reports state that crude protein content ranges from 2.2 mg/ kg in sweet to 2.7 mg/kg in bitter cultivars and fiber ranges from 3.1 to 10.3% (Anonymous, 1968). Notably high percentages of protein occur in wild species in comparison to cultivated cassava. Some reports have referred to a protein percentage in some cassava cultivars as high as 6 or 7% (Anonymous, 1968). Since this estimation of protein was based on total nitrogen, it must be viewed with caution because it is not certain whether the breakdown products of cyanogenic glucosides enhance the total nitrogen content. Narty (1969) showed that the hydrolytic products of glucosides are incorporated into amino acids for protein synthesis in cassava. Therefore, it is not unlikely that the reported cultivars with high nitrogen content were simply bitter cultivars with a high glucoside content. A wild species attracting attention is M. oligantha subsp. nestili due to its high protein content combined with a very low level of HCN (Nassar, 1978d). The author observed cows and horses eating the vegetative parts and tubers of this species when grazing in its natural habitat. In the literature, two other wild Manihot species have been reported to have high protein content: M. melanobasis (Jennings, 1959) and M. saxicola (Lanjouw, 1939). However, because there is no reference to their HCN content, it is not possible to determine to what extent crude protein estimates were affected by hydrolytic products of glucosides. It seems logical that there are wild cassava with a high protein content since selection for cultivation is aimed at increased tuber size and decreased fiber content and not concerned with protein

Table VIII Average Protein and Fiber Content of Wild Manihot Species on a Percentage Dry Matter Basisa Species M. oligantha Pax subsp. nesteli M. tripartita Mueller Arg. M. anomala Pohl M. zehntneri Ule aFour

Crude protein (%)

Crude fiber (%)

7.10 & 0.58 6.88 & 1.48 3.74 & 0.63 3.06 & 0.82

26.67 & 4.86 33.48 & 6.36 23.44 & 4.82 21.52 & 4.82

replicates of 20 tubers of each species were analyzed.

195

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

content. This could have led to the discarding of protein-producing genes from the cultivated varieties. 2. Hydrocyanic Acid Content Tubers of M. tripartita, M. anomala, M. zehntneri, and M. oligantha subsp. nestili, which form abundant tubers, and M. gracilis, which grows widely in central Brazil, were analyzed for HCN content (Table IX). HCN in fresh unpeeled tubers was within a range of 238 mg/kg in M. tripartita to 62 mg/kg in M. oligantha subsp. nesteli. On a dry matter basis, the results were similar, except that M. anomala had the highest HCN content. HCN content in cassava tubers is reported to vary between cultivars. The analysis of about 100 cultivars for HCN content by Raymond et al. (1941) indicated an average of 158 mg/kg fresh whole tuber with a maximum value of 438 mg/kg. Little information is available on HCN content in roots of wild Manihot species. Bolhuis (1953) reported 430 mg/kg HCN in fresh roots of M. saxicola, but this is probably the only wild Manihot species in which HCN content has been estimated. He considered that the high HCN content represented an obstacle to the use of this species in breeding cassava, despite its high protein content. Bolhuis stated that the minimum lethal dose of HCN for a human being is 50 –60 mg. However, chronic toxicity due to the continuous intake of small amounts of HCN is considered more important than acute toxicity because of its association with many diseases (Oke, 1973). The occurrence of species with low HCN levels is a valuable discovery. The species M. oligantha Pax emend. Nassar subsp. nestili, with its notably low HCN content, can be considered a useful parent.

Table IX Hydrocyanic Acid Content of Unpeeled Tubers of Wild Manihot Species

Species M. tripartita Mueller Arg. M. anomala Pohl M. zehntneri Ule M. gracilis Pohl M. oligantha Pax emend. Nassar subsp. nestili

HCN content in fresh root (mg/kg)

HCN content on dry matter basis (mg/kg)

281.1a 192.2a 125.8b 97.2c 62.3d

357.2b 1026.3a 504.2b 291.2c 183.2d

Note. Means followed by the same letter are not significantly different using Duncan’s multiple range test (p 0.5)

196

NAGIB M. A. NASSAR

3. Growth Habitat and Natural Habitats Results of screening wild Manihot species for growth habitat and natural habitat are presented in Table VII. Of particular interest is M. paviaefolia, which forms tubers, has limited vegetative growth, and is adapted to very poor soil. As a parent in cassava breeding programs, it offers potential for additional adaptation to poor soil conditions. Manihot reptans readily forms hybrids with other species in its natural habitat, producing intermediate forms. Collections made by Ule in 1892 were restricted to the northern border of Minas Gerais, near Goias (Rogers and Appan, 1973), but Rogers and Appan found it widespread over most of Goias. In the past 80 years this species may have expanded its geographical distribution and ecological range through genetic variation and interspecific hybridization. In our samples of M. reptans, leaf shape was found to vary widely, reflecting the extent of hybridization with other Manihot species. For example, M. reptans from Goias Velho was distinguished by bright red leaf veins, a characteristic of the native M. alutacea. Manihot reptans was identified by its characteristic growth habit, flower, and inflorescence morphology. Donations of genes from different species adapted to different environments could allow this species to expand rapidly over the whole state of Goias. Harlan (1961) used an example of Helianthus annuus (the annual sunflower) which has acquired a vast gene pool due to the formation of hybrids with at least six other Helianthus species. Manihot pruinosa forms tubers with about 3.8% protein by dry weight compared to cassava with 2%. As seen in Table X, M. pruinosa and M. pentaphylla may represent a source of adaptation to limestone soils. The adaptation of M. alutacea to high altitude makes it a good candidate for breeding programs concerned with producing cultivars tolerant to low temperature. Manihot falcata may provide the potential for breeding cultivars with limited vegetative growth which are adapted to well-drained soil.

Table X Analysis of Soil from Natural Habitats of Six Manihot Species

Species

pH

M. paviaefolia M. pruinosa and M. pentaphylla M. caerulescens M. procumbens M. stipularis

4.9 5.5 5.2 4.9 5.0

P (ppm)

K (ppm)

Al3 (meq/kg)

2

—

16

4

191 1 2 3

1 1 0 1

136 9 18 28

— 2 5 6

Ca2  Mg2 (meq/kg)

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

197

4. Adaptation to Climatic Conditions Study of annual rainfall, evaporation, and temperature ranges in natural habitats of wild cassavas showed a species of particular interest: M. caerulescens Pohl collected from Araripina, state of Pernambuco, and from Posse, state of Goias (shrubs 1–3 m tall, with a deeply extended root to 2 m underground). They rarely form tubers: Tubers were intermittent, at depths exceeding 50 cm; the external color was brown, the surface was smooth, and the cortex was white. Protein content was 3.9% on a dry matter basis and HCN content was 125 mg/kg unpeeled fresh root. Chemical analysis of soil showed it to be very poor (Table X). A fascinating aspect of the ecology of M. caerulescens is its habitats in the western part of Pernambuco and South Caera, which are among the most semiarid of the world’s tropics. The mean average rainfall in this region is about 500 mm, with a high evaporation capacity and high temperature (Table XI). This unfavorable climate, coupled with poor soil, suggests that this species is capable of affording a potential source of resistance to drought. It seems likely that adaptation of M. caerulescens to this arid region depends on its deeply growing root system. However, this species has some characteristics that distinguish it from other Manihot species. For example, it has very large ribbed fruit,—four to six times the normal size of Manihot fruit. I was informed by local inhabitants that seeds are eaten in times of

Table XI Mean Monthly Precipitation, Evaporation, and Temperature in Natural Habitats of M. caerulecens Precipitation (mm)

Evaporation potential (mm)

Temperature (°C)

Month

Picos

Posse

Picos

Posse

Picos

Posse

January February March April May June July August September October November December Total

98.8 168.3 130.3 31.4 12.6 4.1 1.1 1.3 3.2 17.9 32.2 61.7 556.7

286.2 89.4 68.5 117.5 16.4 17.6 0.7 20.2 26.3 195.2 204.3 344.1 1386.4

149.1 138.4 124.8 120.1 122.4 121.9 121.2 137.7 147.0 160.6 150.3 145.8

129.0 129.3 264.9 167.9 224.3 230.9 364.1 352.3 269.1 202.4 135.6 128.0

26.3 25.9 25.5 25.4 24.5 24.3 24.1 26.4 27.8 26.1 26.0 26.2

22.5 22.2 25.3 22.5 22.5 22.0 21.9 22.8 24.5 24.1 23.9 24.9

198

NAGIB M. A. NASSAR

famine. It has a large range extending from northeastern to central Brazil. Some biotypes of this species have apparently spread through this area. They tolerate a wide range of environmental conditions from severe drought in the regions of Araripina, Picos, and Crato in Pernambuco, Piaui, and Ceara states, respectively, to a considerable amount of moisture at Posse in Goias state (Table X). Manihot stipularis Pax collected from Brasília, a very short subshrub approximately 20 cm tall, does not form tubers, has a woody root, and grows on rocks banks. Soil analysis indicates a poor soil (Table X). This species is characterized by dioecious flowers. This characteristic, together with its very short height, distinguishes it from other Manihot species. Manihot stipularis was collected from an altitude of about 1450 m in one of the highest regions of Brazil. This species may offer genes for adaptations to coolness. Manihot procumbens Mueller Arg. collected from Columba is a procumbent subshrub (⬃ 40 cm high) with a large woody root and yellow latex and grows in very poor soil. It shows potential for tolerance to soil aluminum toxicity and the absence of major elements. The fact that wild Manihot species hybridize easily because of their weak interspecific barriers (Nassar, 1984) must have contributed to the vast variation within this group and the evolution of the large number of species in the genus Manihot.

II. BROADENING THE GENETIC BASE OF CASSAVA, M. esculenta Crantz, AND DEVELOPMENT OF INTERSPECIFIC HYBRIDIZATION A. PRODUCTION OF CASSAVA INTERSPECIFIC HYBRIDS Cassava cultivars are deficient in many economic characteristics such as resistance to insects, diseases, and drought and have low protein content (Nassar and Dorea, 1982; Nassar and Grattapaglia, 1986). This can be attributed to the mode of evolution in the species and modifications of the allogamy system of the plant (Nassar and O’Hair, 1985). Lost genes can be restored to the gene pool of the cultigen by interspecific hybridization with wild relatives which possess these genes (Nassar et al., 1986). Wild species of cultivated crops have frequently been used as an important source of genetic diversity and have been employed effectively in a variety of breeding programs (Stalker, 1980). Controlled introgression of genes could alleviate stress problems in cassava given the availability of wild relatives which exhibit diversity in adaptations and attributes. There are interspecific barriers to hybridization (Nassar et al., 1986), but these can be broken in different ways. Nassar, (1989) reported production of interspecific hybrids of two Manihot species (M. neusana Nassar and M. anomala Pax) with cassava through controlled crosses by vector insects. Two wild Manihot species, M. anomala and M. neusana,

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

199

maintained in the living collection at the Experimental Biology Station, Universidade de Brasília, were used for this experiment. In October 1982 the species were planted in three rows alternated with cassava. In June 1983, 200 seeds were collected from each species and replanted in October 1984 for identification of possible natural hybridization. The following marker genes were used to identify interspecific hybrids: variegated color of fruit dominant to smooth, red color of flower disk dominant to yellow, setaceous bracteole dominant to foliaceous, and noded stem dominant to smooth. Observations of growth habit, height, stem texture, and tuber formation were also recorded. In addition to open pollination for the previously mentioned species, 400 manual crosses with pollen of cassava cultivar Catelo were realized. Of 200 seeds of M. neusana, only 43 seedlings emerged, of which 2 hybrids were identified. Interspecific hybrids were identified by dominant markers from cassava; noded stem, setaceous bracteoles, ribbed fruit, and tuberculated root (Table XII). Other characters provided indirect evidence of hybridization. The 200 seeds collected from M. anomala gave rise to 112 seedling. Of these, 3 seedlings showed characteristics of interspecific hybridization. Only 1 seedling survived to maturity. This hybrid plant exhibited dominant phenotypes from cassava, namely, ribbed fruit, red color of the flower disk, noded stem, and tuberous root (Table XII). These results show that glabrous stem, setaceous–foliaceous bracteoles, red-creamy color of flower disks, variegated-green color of fruit, and ribbed– nonribbed fruit are simple marker genes that can be used to recognize interspecific hybridization. It is evident that interspecific barriers between Manihot species can be broken by the use of an abundant diversity of pollinator gametes transmitted by insect vectors. Interspecific crosses were difficult to fertilize manually in the Table XII Comparison of Morphological Characteristics of M. neusana, Cassava, and Their Hybrid Characteristic

M. neusana

Cassava

Hybrid

Growth habit Young stem texture Bracteoles Fruits

Erect shrub (1.5–2 m) Glabrous Setaceous Ovoid, ribbed, green

Tuber formation

Procumbent shrub (1.5–2 m) Hairy Foliaceous Globose, without ribs, variegated None

Forms tubers

Erect shrub (1.5–2 m) Hairy Setaceous Ovoid, ribbed, variegated Forms tubers

Growth habit Young stem texture Bracteoles Flower disk color Leaf form Fruit Tuber formation

Erect shrub (2–2.5 m) Hairy Semifoliaceous Creamy Anomala Globose, without ribs Scarely forms tubers

Erect shrub (1.5–2 m) Glabrous Setaceous Red Lobed; 5 lobes Ovoid, ribbed Forms tubers

Erect shrub (1.5–2 m) Hairy Setaceous Red Anomala Ovoid, ribbed Forms tubers

200

NAGIB M. A. NASSAR

present and in previous crosses (Nassar, 1980a). The evidence suggests that barriers between cassava and other Manihot species are weak and recently evolved. It seems they have arisen not as a primary isolating event but rather secondarily after geographic isolation. Nassar (1978b) postulated that cassava is an interspecific hybrid that appeared by domestication approximately 2000 years ago or less.

B. DEVELOPMENT OF CASSAVA INTERSPECIFIC HYBRIDS The wild Manihot species of M. neusana Nassar was hybridized with the cassava clone Catelo through controlled hybridization with the help of pollinating insects (Nassar, 1989). An interspecific hybrid that combined marker genes of both parents was obtained. The marker genes were ribbed fruit, acquired from cassava, and variegated fruit color from M. neusana. This hybrid (HN) was backcrossed with cassava and used as a pollinator in the first trial and as a fruit carrier in the second trial. Seeds were obtained from both crosses, but only one plant could be raised from each; HO1 was the result of the interspecific hybrid (HN) as maternal plant (seed carrier), and HO4 resulted from crosses in which the interspecific hybrid (HN) was used as pollinator. The three hybrid plants (HN, H1, and H4) were studied cytogenetically for both meiotic and mitotic behavior. For the study of meiosis, inflorescences were fixed in a mixture of three parts absolute alcohol and one part glacial acetic acid and kept in a refrigerator for 24 h. The anthers were smeared with acetic carmine. Chromosome configurations in the metaphase, chromosome distribution in anaphase I, and tetrad formation were also studied. Pollen viability had been determined by using acetocarmine and iodine stain (Nassar, 1978a). For the mitotic study, root tips were left in 0.2% colchicine for 2 h and then fixed in acetic alcohol for 24 h. Before smearing with acetocarmine, they were treated with 1 N HCl for 10 min. 1. Meiotic Behavior of the F1 Hybrids (HN) One hundred pollen mother cells (PMCs) were studied in metaphase I of the interspecific hybrid M. neusana with cassava; 30 PMCs in metaphase II and 1000 tetrads of the same material were also investigated. The study of metaphase I showed different chromosome configurations, as shown in Table XIII. The average bivalent frequency in all cells of metaphase I was 16.13 per cell. The high frequency of univalents was attributed to lack of synapses between chromosomes or failure of the two species to remain associated. Virtually the only other report on this subject is that of Magoon et al. (1970), in which chromosome pairing in the interspecific hybrid M. glaziovii (rubber tree) and cassava was studied, and a regular synapsis led the authors to conclude that there is a strong relationship between this species and cassava. Nassar et al. (1986) suggested that the material used by Magoon et al. was not a pure M. glaziovii but rather a natural interspecific hybrid between this species and cassava. If this is true, the supposed interspecific hybrid

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

201

Table XIII Frequency of Chromosome Configuration of Metaphase I in Interspecific Manihot Hybrids and Their Parents

M. neusana Cassava GN HO1 HO4

PCMs (No.)

Trivalents

Mean Bivalents

Univalents

20 20 100 30 100

— — — 1.86 1.63

18.00 18.00 17.00 16.13 12.41

— — 1.58 0.13 8.84

would be a backcrossed progeny. The study of anaphase I showed that of 40 PMCs studied, 38 cells exhibited laggards, which were attributed to the occurrence of univalents resulting from nonhomologous chromosomes. Anaphase II showed meiotic restitution. Of 33 PMCs studied in this phase, 5 cells exhibited a second meiotic restitution (SMR), forming 36 chromosomes on each pole. Apparently this phenomenon is a consequence of meiotic disturbance in the hybrid. An example of this disturbance was the breakdown of anaphase I. This was probably due to disharmony between the two different genomes. Nassar (1991) documented this phenomenon in the interspecific hybrids of cassava with M. pseudoglaziovii. The presence of such restitution was confirmed in the following tetrad stage, in which the formation of both dyads and tetrads was observed. In various crops, interspecific hybridization has led to the disturbance of meiotic division, with consequent meiotic restitution, e.g., in Trifolium pratense by Parrot and Smith (1984) and in Medicago spp. by Vorsa and Bingham (1979). In manioc species, Hahn et al. (1990) reported 2n pollen formation in wild species in addition to certain clones of cassava. The detection of this phenomenon enabled these researchers to isolate triploid and tetraploid types from progeny that came from crosses of cassava with certain wild Manihot species, namely, M. glaziovii and M. epruinosa. These types proved much more productive than commercial clones used in Nigeria. Nassar (1991) manipulated the meiotic restitution occurring in interspecific hybrids of M. pseudoglaziovii with cassava to produce triploid types that showed very good productivity under semiarid conditions. The discovery of the frequent occurrence of this phenomenon in interspecific hybrids of cassava offers an effective tool for the production of polyploid types by sexual means instead of the traditional method of colchicine applied to vegetative parts, which normally induces unstable, chimeral types (Abraham et al., 1964). An additional advantage is that production of triploid types may lead to production of trisomics among their progeny. If genes which control productivity in cassava are polygenes with addictive model action, as is the case for many crops, certain trissomics of this crop may be more productive than their diploid ancestors. In general, the pro-

202

NAGIB M. A. NASSAR

duction of polyploidy via sexual means is advantageous from both genetic and evolution standpoints because it offers a vigorous heterotic effect and releases useful genetic variability adaptations. For the study of the tetrad stage, 1065 PMCs were investigated. Of these, 62 cells formed dyads and 62 cells formed micronuclei. The presence of dyads and tryads at this stage confirms that observed earlier in the anaphase: the occurrence of first- and second-divisioin meiotic restitution (FMR and SMR). Both types are capable of producing 2n gametes. However, the FMR is more valuable than the SMR since it preserves the major part of its heterosis and epistatic interaction (Mendiburu and Peloquin, 1977; Parrot and Smith, 1983; Vorsa and Bingham, 1979). 2. Cytogenetic Behavior of the Backcrossed Generation (HO1) Table XIV shows the frequency of chromosome configurations at metaphase I. Bivalents, trivalents and univalents were present, with a mean of 16.1 bivalents, 1.86 trivalents, and 0.13 univalents. The presence to trivalents indicated aneuploidy in this hybrid. This was confirmed by mitotic counting, which showed 2n 38, i.e., 2n  2. In anaphase I, the presence of laggards with different frequencies was recorded. The study of 900 tetrads showed 644 normal ones, 218 micronuclei, 12 dyads, and 26 tryads. Analysis of pollen viability revealed that only 35.8% were viable (Table XV). 3. Cytogenetic Behavior of the Backcrossed Generation (HO4) One hundred PMCs at metaphase I were studied. Again, bivalents, trivalents, and univalents with averages of 12.4, 1.6, and 8.8, respectively, were observed. The total chromosome count for the different configurations was 38. This showed a constitution of 2n  1  1, which was confirmed by root tip mitotic counting. In anaphase I, of 32 PMCs studied, 31 proved to have laggards. In anaphase II, 35 PMCs were examined; of these, 7 cells appeared as restitution nuclei, and this was later confirmed in the tetrad stage. In the tetrad stage, 1196 sporocytes were obTable XIV Diploid Pollen in Interspecific Manihot Hybrids and Their Parents Diploid pollen

M. neusana Cassava HN HO1 HO4

Pollen grains examined (No.)

No.

%

818 1162 1235 1128 1007

3 3 20 8 6

0.36 0.26 1.62 0.71 0.60

203

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES Table XV Viability of Pollen of the Interspecific Manihot Hybrids and Their Parents Viable pollen Pollen analyzed (No.)

No.

%

Nonviable pollen

1001 1235 1830 1542

818 1162 655 273

81.72 94.09 35.80 17.70

183 73 1175 1269

M. neusana HN HO1 HO4

served. Of these, 326 were normal, 826 contained micronuclei, 25 were tryads, and 19 were dyads. The study of pollen viability showed a very low viable grain percentage of 17.7% (Table XV). 4. Cytogenetic Behavior of the Parents The cassava clone EB 01 showed a regular meiotic division in all of the 20 PMCs studied, with 18 bivalents formed. Of the 950 tetrads that were examined, 942 were normal, 5 contained micronuclei, and 3 were tryads (Table XVI). The study in which M. neusana was used as a parent showed it to have a regular meiosis with 18 bivalent formations. Of 1011 tetrads studied, 1003 were normal, whereas 6 had micronuclei and 2 were dyads. The pollen viability was 81.72% (Table XV). Mitotic counting of root tips showed 2n 36. This was the first report of a chromosome number for this new species described by Nassar (1985). 5. Evolutionary and Plant Breeding Significance The fertility of the hybrid HO1 indicates the possibility of further manipulation of this hybrid through backcrosses to transfer useful genes of M. neusana to cassava. The backcrossed generations produced were aneuploid 2n  1  1 in both Table XVI Analysis of Tetrads of Interspecific Manihot Hybrids and Their Parents Tetrads

M. neusana Cassava HN HO1

PMCs

Tryads

Dyads

Total

Normal

%

No.

%

No.

%

No.

%

1011 9502 1065 900

1003 942 694 644

99.22 99.15 65.15 71.55

6 5 262 218

0.59 0.53 24.60 24.22

2 3 62 26

0.19 0.31 5.82 2.88

— — 47 12

— — 4.41 1.33

204

NAGIB M. A. NASSAR

cases studied (HO1 and HO4). In the case of hybrid HO4, the plant was completely sterile, having a chromosomal constitution of 2n  1  1. Obviously, this hybrid resulted from fertilization of a pollen gamete n  1  1 of the parent hybrid (HN), with a cassava ovule of “n.” On the other hand, when the interspecific hybrid HN was used as a maternal plant, a fertile progeny was obtained. When it was used as a pollen parent in the backcross with cassava, it resulted in the production of a sterile progeny (HO4). This was probably due to the elimination of fertile embryo genotypes in the progeny because of incompatibility or disharmony between them and the endosperm. The partial fertility of the backcrossed generation (HO1) shows that the species M. neusana may be classified within the secondary gene pool of cassava according to the concept of Harlan and de Wet (1971). Other Manihot species that may be included in this category are M. melanobasis (Jennings, 1959), M. glaziovii (Magoon, 1970), M. reptans, M. zenhtneri, M. anomala, M. oligantha, M. pohlii (Nassar et al., 1986), and M. dichotoma, M. epruinosa, and M. leptophylla (Hahn et al., 1990). It was concluded that the cassava hybrid with M. neusana showed irregular meiotic behavior in the lack of complete chromosome pairing, formation of univalents in metaphase I, chromosome retardation in anaphase I, micronuclei in the tetrad, and meiotic restitution. When backcrossed to cassava, the interspecific hybrid produced two aneuploids 2n  1  1. These showed irregular meiosis, partial chromosome pairing, and the presence of meiotic restitution. The two backcrossed F2 hybrids differed in regard to pollen-viable genetic constitutions. The meiotic restitution continued to occur in F2 hybrids, which showed that it must be correlated with interspecific meiotic irregularity.

C. DEVELOPMENT OF CASSAVA INTERSPECIFIC HYBRIDS FOR SAVANNA (CERRADO) CONDITIONS Interspecific hybrids of cassava with M. glaziovii and M. pseudoglaziovii were produced by Nassar in 1991 (Nassar et al., 1996a). They were propagated by cuttings and planted in November 1992 alternately with clone Sonora. This clone was chosen because of its high resistance to bacteriose. Seeds were collected from cassava and planted in November 1993. Out of 182 seeds whole plants. In March 1994, these plants were reproduced by cuttings; six of each were planted for evaluation of productivity and survival during the drought season of June to October (5 months). In November 1994, plants which survived were evaluated for root formation. Ten clones were selected and given to the Semi Arid Centre at Pernambuco for propagation and cultivation under semiarid conditions of northeastern Brazil. The selected clones were characterized morphologically according to Rogers and Appan (1973) and Nassar and Grattapaglia (1986). This characterization was aimed at detecting the association of different morphological characters

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

205

with tolerance to drought. Since polyploidy offers another alternative for this tolerance, the previously mentioned interspecific hybrids and several others available in our living collection were investigated cytologically to detect formation of diads and triads as evidence of unreduced 2n microspore production (Vasquez and Nassar, 1994). This will enable the selection of possible progenitors of polyploid types. For cytogenetic study, 10 interspecific hybrids and/or their progenies were used: F1 M. glaziovii cassava (3 genotypes), F2 M. epruinosa cassava, F2 M. anomala cassava (3 genotypes), F3 M. pseudoglaciovvi cassava (2 genotypes), and F4 M. pseudoglaciovvi cassava. Chromosome association in meiotic metaphase I and the occurrence of dyads, tryads, and pollen viability were studied. For the meiotic study, inflorescence was fixed in a 3:1 mixture of absolute alcohol and glacial acetic acid and kept at 5C for 24 h. The anthers were smeared in acetocarmine stain. Chromosome configurations at metaphase I and sporad formation were studied. For the pollen viability study, one or two flowers per plant were selected, and their pollen was crushed from anthers in acetocarmine preparations and scanned for viability. Pollen counts and percentage of stained normal pollen were calculated. Of the 182 seeds planted, only 35 germinated, established, and developed to the flowering phase. This is due to dormancy of wild seed retained in F2 progeny. These 35 plants were multiplied by cuttings to evaluate their performance for root production. Only those plants which gave more than 2-kg roots with a medium 2.3 kg per plant by 8 months were selected for multiplication for further cultivation. Morphological characterization showed that certain characters were associated with tolerance to drought. As seen in Tables XVII and XVIII, all selected clones have a notable brown, thick, and rough superficial epiderm. It seems that the brown-colored thick epiderm is associated with tolerance to drought because of its isolative nature, which impedes evaporation. All the wild species investigated by the author have fibrous roots with brown external color and their epidermic layer is thick. It is believed, therefore, that this character is inherited from the wild. Graner (1942) reported that this character behaves dominant to white. Anatomically, the distinct portion of the enlarged root is composed of three sections. First, a layer referred to as the phelloderm which is generally composed of the previously mentioned epidermis, a subepidermis, and a thicker inner layer. The phelloderm is thick and easily separated from the next inner layer. Second, a layer of parenchymatous cells that constitutes the bulk of the root and is the carbohydrate storage region. Third, a portion called the cortex of flesh at the center of the root is a well-defined central vascular core. As noted previously, the outer epidermis is so fine that it is difficult to measure, but it is possible to evaluate its thickness using the naked eye. It is about 0.5 mm in the thickest types. The second interesting case in selected clones is the prominence of leaf scars on stems. All selected clones have a prominent enlarged leaf scar. This character

Table XVII Characters Used in Clone Characterization Root characters 1. Surface of root A. Smooth B. Rough 2. Thickness of epiderm layer A. Thick (>. 0.3 mm) B. Thin (<. 0.3 mm) 3. External color of root A. Light brown-yellow B. Brown, dark brown, reddish brown C. Light brown, tan, light tan D. Pinkish brown, pinkish tan E. Pinkish white, light pink, pink 4. Root flesh color A. White to cream B. White to cream with pink C. Cream-yellow to yellow D. Cream-yellow to yellow with pink 5. Color of stem A. Silver B. Silver-brown C. Brown D. Yellow 6. Nature of scars on stem A. Smooth B. Slightly raised C. Moderately raised D. Very large 7. Branching of plant A. One branch at top or no branches B. One or two branches but not one branch at top C. More than two branches Leaf characters 8. Leaf lobe shape A. Obovate B. Linear 9. Sinuosity of leaf lobes A. Pandurate B. Some sinuosity C. Simple (not sinuous) D. Logical (linear) 10. Petiole color A. Red B. Greenish red C. Reddish green D. Green 11. Color of young foliage A. Reddish blue B. Bluish green C. Green

207

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES Table XVIII Morphological Characteristics of Selected Clones Characteristic

1

2

3

4

5

6

8

9

10

11

1. Surface of root 2. Thickness of epiderm layer 3. External color of root 4. Root flesh color 5. Color of stem 6. Nature of scars on stem 8. Leaf lobe shape 9. Sinuosity of leaf lobes 10. Petiole color 11. Color of young foliage

B A B A B C B A B C

B A B B B C A A D C

B A B B B D A A A A

B A B C B C A D A C

B A B A B C B A A A

B A B B B C A A C A

B A B A B C A C D C

B A B A B C B B B B

B A B A B C A A B C

B A B A B D A D B B

Note. Numbers and letters correlate with those used in Table XVII.

which seems to be well associated with enlarged root formation in the hybrid progeny, apparently came from cassava. All wild species studied by the author have a smooth stem without any leaf scar. All selected clones gave deep fibrous roots in addition to enlarged ones. It seems that this character is inherited from the wild. This appears to be a mechanism of cassavas to tolerate drought since they are capable of capturing water from long distance. Both wild species and their interspecific hybrids produced long, deep roots from the fourth month onwards, reaching 4 or 5 m when the plants were 1 year old. Another mechanism of tolerance to drought is the thick epiderm layer, probably because of its structure, it impedes evaporation. The dieback of the vegetative parts to the crown in dry season was the third common character shown by all the selected clones. Presumably this habit helps plants to reduce respiration and consumption of carbohydrate deposits. From this study, it is obvious that breeders can make use of morphological characterization as a criterion to detect the association of morphological characters and drought tolerance and consequently selection of genotypes complying with this objective.

D. OVERCOMING CROSSING BARRIERS BETWEEN CASSAVA, M. esculenta Crantz, AND A WILD RELATIVE, M. pohlii WARWA Manihot pohlii has a high resistance to stem borers, Coelosternus spp., and bacteria Xanthomonas Manihotis. Hybrids of M. pohlii and other species with cassava can be obtained but only at a very low frequency due to interspecific crossing barriers (Nassar et al., 1986). However, one technique for overcoming such barri-

208

NAGIB M. A. NASSAR

ers involves the use of pollen mixes which combine the pollen intended for syngamy with “mentor” pollen of the maternal species. This technique has been reported for many plant taxa, including Populus (Knox et al., 1972), Malus (Dayton, 1974), Cosmos (Howlett et al., 1975), and Petunia (Sastri and Shivanna, 1976). The role of the mentor pollen is to facilitate fertilization by foreign pollen. Apparently, the mentor pollen supplies proteinaceous substances which permit the foreign incompatible pollen grains to germinate (Knox et al., 1972). A treatment destroys the generative function of the pollen grain without affecting germination and growth of the pollen tube and consequently does not affect stimulation. The stimulating effect of destroyed pollen is due to protein recognition substances. They are liberated by pollen grains while germinating and have enzymatic and antigenic properties. These substances are localized in the internal layer of the pollen surface and are correlated with pollen germination and growth of its tube on the stigma surface. They have been localized by cytochemical means in the cellulose intine (Knox et al., 1972). Many studies using this technique have been successful in overcoming an incompatibility barrier (Brewer and Henstra, 1974; Williams and Church, 1975). Subsequent work indicated the need to refine the preparation of the mentor pollen by using freeze–thaw cycles or methanol treatment (Knox et al., 1972). Nassar et al. (1996b) successfully used freezed mentor pollen in hybridizing cassava with M. pohlii. In addition to the mentor effect, other techniques seem to have potential for overcoming interspecific barriers. One of these may be the use of a bridge species. Nassar et al., inspired by the capacity of M. neusana Nassar to cross easily with all Manihot species growing in the vicinity, used it as a bridge species. Very little information is available regarding the bridge technique. Probably the only case is that of Dionne (1963), who used Solanum acaule Bitter as a bridge between Solanum tuserosum L. and Solanum bulbocostanum Dunal. A natural hybrid between M. pohlii and M. neusana Nassar was used by Nassar et al. (1996b) as a bridge species. The hybrid was identified by fruit marker genes, which produce variegation of fruit color in M. neusana and a straight white line in M. pohlii fruit. Crosses of this hybrid (named HNP) and cassava (clone EB 05) were carried out from January 1993 to January 1994. Flowers were taped shut for 2 days until they had been pollinated manually. Pollination of both the hybrid HNP and cassava was done with pollen mixes of cassava and M. pohlii. Manihot pohlii pollen used as mentor pollen was sucessively frozen for 5 min at 4C and thawed for 30 min for a period of 105 min. The purpose of this treatment was to kill the mentor pollen and increase the chance of obtaining interspecific hybrid seed. To verify the presence of any autoincompatibility that would interfere in the controlled crosses in M. polii and cassava, a controlled autopollination was undertaken. Crosses between cassava and M. pohlii (POH) were carried out using mentor pollen in one trial and untreated pollen in the second trial. Seeds were collected from both crosses and planted during the next growing season.

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

209

The pollination of M. pohlii by untreated cassava pollen did not produce any fruit set (Table XIX), whereas crosses with mixed pollen of cassava and mentor resulted in the production of 21 seeds, which is 4.9% of the possible maximum (assuming that every fruit has three ovules). Using cassava as a maternal parent, pollination by untreated pollen M. pohlii did not result in any seed. This clearly demonstrates the effect of mentor pollen in the crosses of cassava with M. pohlii. It likewise means that the freeze–thawing treatment administered to M. pohlii pollen, although it did kill the pollen, did not affect its stimulatory function so that all the seed produced by the mentor effect had embryos and endosperm. This result indicates that only the stigmatic barrier functions in preventing crossing in these species. Postfertilization mechanisms fail to prevent crossing. Only one plant germiseeds to severe dormancy. This plant bore fruits carrying the marker genes of both M. pohlii and cassava: a straight white line from M. pohlii and winged fruit from cassava. The mentor effect has also been successfully used in Populus and Cosmos. In these genera, interspecific incompatibilities have been overcome by using compatible but dead pollen (Knox et al., 1972). These studies suggest that this phenomenon is due to the release of proteinaceous recognitioin factors from the wall of the killed compatible pollen, masking the rejection reaction of the recipient stigma. Nassar et al.’s report (1996b) represents the first case of obtaining hybrid seed of M. pohlii and cassava and its further reproduction. Despite the useful characters of M. pohlii, no successful breeding program has been carried out due to a lack of hybrids between this species and cassava. The use of M. neusana as a bridge species through the hybrid M. neusana–M. pohlii has improved seed set. When used as a maternal parent pollinated by cassava, it gave seed in 3.5% of cases, whereas the reciprocal crosses had greatly improved seed set, yielding 25.9% (Table XX). The combined treatment of both species (bridge and mentor) produced 3.5% seed production. The success of M. neusana as a bridge species between M. pohlii and cassava may occur because the two genomes of M. neusana carry different genetic mechanisms of cross incompatibility. This hypothesis is confirmed by observations on crossing behavior of this species in the living collection of wild Manihot species. Manihot neusana has Table XIX Crosses Attempted between M. pohlii and M. esculenta (Cassava) and Numbers of Fruit and Seed Produced Treatmenta

Flowers pollinated

No. Fruit

No. Seed

145 142

0 10

0 21

1. Cassava M. pohlii 2. M. pohlii cassava a1,

Without mentor; 2, with mentor.

210

NAGIB M. A. NASSAR Table XX Crosses Attempted between Cassava and M. neusana Hybrids and Numbers of Fruit and Seed Produced

Treatment HNP cassava Cassava HNP HNP cassava  mentor

Flowers pollinated

No. Fruit

161 90 208

14 14 198

No. Seed 17 70 22

hybridized naturally with several Manihot species grown in the vicinity, e.g., M. pseudoglaziovii Pax et Hoffmann, M. glaziovii Mueller Arg., M. tripartita Mueller Arg., M. caerulescens Pohl, M. salicifolia Pohl, M. pohlii, and M. esculenta (Nassar, 1989). Its hybrids with the previously mentioned species were easily identified by the marker dominant gene of variegated fruit which came from M. neusana. Manihot neusana is a newly emerging species, described and identified by Nassar (1985).

III. DEVELOPMENT AND SELECTION FOR APOMIXIS IN CASSAVA, M. esculenta Crantz A. GENETIC STUDY OF APOMIXIS IN CASSAVA Cassava is a labor-intensive crop. It is propagated vegetatively by stem cuttings which are often responsible for the spread of diseases and pests in the tropics. Nassar and O’Hair (1985) hypothesized that the use of seed to replace stem cuttings for cassava production would eliminate these problems and potentially reduce production costs. One limiting factor, though, is the lack of rapid and uniform seed germination. Another difficulty is genetic segregation and lack of true breeding lines. Apomixis, which is synonymous with agamospermy or asexual seed formation (Asker, 1979), may provide a solution to these problems. It can be used as a means for fixing heterozygosity responsible for vigor, high productivity, and wide adaptation of cassava clones to different conditions of soil and climate (Nassar, 1992). Nassar began a project in 1981 with the objectives of selecting for easy germinating seed of cassava clones (Nassar and O’Hair, 1985) and of examining for apomictic occurrence within the selected populations. Three clones (011, 040, and 060) were selected from an easy germinating population developed by Nassar. The offsprings of these clones were studied through several generations for transference of combinations of the following characters: abundant fruit formation in

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

211

plants which have sterile pollen, uniform progenies identical to the maternal types, and the occurrence of multiple seedlings per seed as described by Asker (1979); and segregation of several distinct characters of root [external color and surface morphology (smooth or rough)], stem [color and surface morphology (smooth or rough)], stem [size of scars and storey length (distance between two ramifications)], flower (petal color), and fruit (color) as defined by Rogers and Fleming (1973) and Nassar and Grattapaglia (1986). The presence of apomixis was confirmed experimentally by controlled crosses made between clones 011, 040, and 060, which carry the recessive gene for green, young shoots, and clone 051 used as a pollinator. Clone 051 carries the dominant, homozygous gene for dark red, young shoots which serves as a marker gene. An F2 clone, 031, was included in the crosses because it behaved as if it were apomictic. Cytogenetical studies were conducted with mature plants of these selected clones to examine their meiotic behavior as follows: Flower buds were fixed in acetic acid: ethanol 3:1 (vol/vol) for 24 h, preserved in 70% ethanol, smeared and stained with 1% acetocarmin, and then examined microscopically. Fifteen pollen mother cells of each clone were examined at metaphase I for chromosome association (Nassar, 1978a). Pollen viability was estimated by staining with 1% acetocarmin. Round, regular-shaped, and regular-sized pollen were considered viable. Crosses of clone 021, an interspecific hybrid of cassava (M. esculenta M. dichotoma used as a pollinator), and cultivated clone Branca Santa Caterina were carried out to transfer apomictic genes from the wild species M. dichotoma. This species, maintained in the living collection by the author, has been observed to have apomictic behavior. One hundred F2 seeds resulted in 72 seedlings from which the 3 most vigorous were selected and allowed to open pollinate and bear fruit. One hundred seeds from each plant were collected and planted in rows, and the raised offspring were examined for apomixis using the criterion for transference of maternal combination of characters as described previously. This enabled selection of clone 031 which was included in the controlled crosses. There were variations in transference of maternal type to the progeny showing different degrees of apomixis. Meiotic study (Table XVIII) revealed that clones 011, 031, and 040 were aneuploids having 37 chromosomes compared with 36 chromosomes (18 bivalents) in normal meiosis of cassava clone 060 and cultivar Branca Santa Caterina. The chromosome associations in clones 011, 031, and 040 consisted of 17 bivalents and 1 trivalent (Table XXI). Pollen viability from offsprings of the easy germinating clones was low (17.8–26.0%) compared with that of the normal clone 060 (89.9%) and cultivar Branca Santa Caterina (86.5%). Gustafsson (1946) reported a correlation between apomixis and polyploidy, specifically euploidy. Brown and Emery (1958) found that apomixis in the Gramineae subfamily Ponicoideae was restricted to euploid species. Hybridization between cassava and its wild relatives was reported to occur frequently in nature (Nassar, 1992). Wild Manihot species, con-

212

NAGIB M. A. NASSAR Table XXI Pollen Viability, Chromosome Number, and Predominant Chromosome Association of an Easy-Germinating F2 Hybrid and Cultivar Branca Santa Caterina Easy-germinating F2

Clone Pollen viability Chromosome No. Chromosome association

011

040

060

031

Branca Santa Caterina

20.4 37 17  1

17.8 37 17  1

89.9 36 18

26.0 37 17  1

86.5 36 18

sidered the source of many useful characters (Nassar, 1978b), seemed to have conferred apomixis to cultivared cassava. Specifically, the interspecific hybridization of Brazilian cassava indigenous clones (Nassar, 1978b, 1992) may be responsible since their crossings resulted in partial or complete sterility as well as formation of aneuploid forms. Although obligate apomixis is clearly most favorable for use in cassava breeding, a low percentage of seeds produced by gamete formation and fertilization have been known to occur in different taxa of different crops which were considered obligate in the past (Asker, 1979). Asker questions if a 100% obligate exists.

B. MOLECULAR AND EMBRYONIC EVIDENCE OF APOMIXIS IN CASSAVA To confirm the presence of apomixis and study its anatomic nature in cassava, Nassar et al. (1997, 1998a) selected two putative apomictic cassava clones. The first was identified by number 031 based on vigor in an F2 population resulting from a cross between an interspecific hybrid (M. dichotoma M. esculenta) and cultivated clone Branca Santa Catarina. Clone 200 is an F1 individual from hybridization between cultivated cassava and M. glaziovii. Progenies from both clones were obtained by open pollination. Thirty-seven progeny plants of each clone were used in the anatomic and molecular studies. 1. Embryo Sac Analysis Morphostructural development of embryo sacs was studied histologically. Unpollinated pistillate buds at presumably 1 day preanthesis and pollinated pistils at postanthesis were sampled and immediately fixed in Farmer’s fixative (1:3 glacial acetic acid:95% ethanol) in the field between 7:30 a.m. and 12:00 p.m. Fixed pistils were dissected under a dissection microscope (magnification 40,

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

213

transmitted light). Dissected ovules were dehydrated in ethanol series and cleared overnight in the benzyl-benzoate-four-and-a-half fluid (lactic acid:chloral hydrate:phenol:clove oil:xylene:benzyl benzoate 2:2:2:2:1:1, w/w) devised by Herr (1982) and treated in a modified Herr’s fluid as previously reported by Ogburia and Adachi (1994). Transparent ovules were then observed using an Olympus BX50 microscope equipped with Nomarski’s differential interference contrast (DIC) optics and a 100-W high-pressure mercury lamp with appropriate filters for optimal viewing and photography. Both megasporangia and megagametophyte components were photographed and printed by a Sony color video printer Mavigraph UP-1200. One hundred twenty-seven ovules from clone 031 and 134 ovules of clone 200 were used in the embryo sac analysis. 2. DNA Extraction and RAPD Assay Total genomic DNA was isolated from 200 mg of fresh leaf tissue ground in liquid nitrogen using the CTAB protocol of Doyle and Doyle (1987), modified by the addition of 1% PVP and 1% 2-mercaptoethanol to the isolation buffer. DNA concentration was estimated by gel electrophoresis comparing the fluorescence intensities of the ethidium bromide-stained samples to those of lambda DNA standards. For the RAPD assay, working stocks of genomic DNA were diluted in water at a concentration of 2.5 ng/ l. Arbitrary 10-base primers (kits OP-A–OP-Z) were obtained from Operon Technologies, Inc. (Alameda, CA). Amplification reactions (13 l) were carried out according to William et al. (1990) with the following modifications: 0.4 mM 10-base primer, 10 g/ul nonacetylated bovine serum albumin (New England Biolabs), 5–10 ng genomic DNA, and 1 unit of Taq DNA polymerase. Amplifications were performed in 96-well microwell plates using a MJ Research PT-100 thermal controller. RAPD products were analyzed by electrophoresis in 1.5 or 2.0% agarose gels containing 0.2 mg/ml ethidium bromide. Gel images were captured and digitalized with an Eagle-Eye II system (Stratagene). Gel scoring was performed directly from the gel images on a computer screen and images were stored electronically on laser CD. A set of 24 arbitrary primers selected earlier for high multiplex content and discrimination power in cassava (Grattapaglia et al., 1996) were used. The presence and absence of RAPD fragments was scored by visual inspection of the gel images. Informative RAPD markers were identified as described previously (Grattapaglia and Sederoff, 1994). Two replicate RAPD analysis experiments, including DNA extractions, RAPD assays, and marker scoring, were carried out with the set of select primers on the putatively apomictic individuals to confirm the patterns of bands observed. The two maternal parents and 67 offspring individuals were genotyped with 24 selected arbitrary 10-base primers. Each selected primer amplified an average of 8.25 clearly interpretable RAPD fragments with a range of 5 to 14 fragments. A total of 198 clear-

214

NAGIB M. A. NASSAR

ly interpretable and reproducible RAPD markers were surveyed in this study. This number of markers was considered to provide a representative genome coverage for the objective of this study. On average, 81% of the scored RAPD bands were monomorphic between maternal parent and progeny set in the two families. The remaining bands were polymorphic and thus useful for testing the hypothesis proposed herein. Progenies of clones 031 and 200 displayed a high uniformity of DNA fingerprints. However, it was possible to find markers that readily showed that individuals were not derived from apomixis, except for one individual in each progeny. In the progeny of clone 031, individual 4 showed a pattern of RAPD bands identical to that observed in individual 5 in the maternal one for all primers examined. The same pattern was observed in individual 5 in the maternal one for all primers examined. The same pattern was also observed in individual 5 in the progeny of clone 200. Although it seems very unlikely that the maternal parent and a zygotic progeny individual could have an identical combination of more than 100 RAPD fragments, it could be argued that it is possible. As described in a previous report (Grattapaglia et al., 1996), to exclude this possibility the statistical procedure described by Novy et al. (1994) was used to show that the complete similarity between the two samples in not an artifact resulting from a limited number of RAPD markers surveyed and rather has a biological basis. Given the number of markers surveyed, the probability that complete uniformity in RAPD markers between the maternal parent and the respective progeny individuals occurred due to change was equal or 103 in both clones. Therefore, putative apomixis was detected in progenies of both clones (031 and 200) at a rate of 3.13 and 2.70, respectively. These results clearly indicate that the type of apomixis detected in this study is facultative and occurs at very low frequency in cassava. A total of 261 ovules were analyzed histologically. Both clones studied indicated the presence of aposporic sacs inside the sexual embryo sacs. The presence of two embryo sacs in an ovule is an indicator of the apospory nature of apomixis. It is supposed that one of them is derived from somatic cells at its location in the ovule, whereas the second embryo sac is derived from a normal megaspore mother cell. At a certain stage, before the complete maturation of the sexual embryo sac, it may abort and be replaced by the developing apospporous sac; otherwise, it continues to develop, giving rise to two embryos sacs and two embryos in the ovule. This abnormality was verified in 2.36 and 1.49% of the ovules of clones 031 and 200, respectively. Similar results using the same histological clarification technique were reported in Cenchrus ciliaris (Young et al., 1979). Using a DIC microscope, it was possible to see in the cleared pistils details of the cassava embryo sac. The normal sacs showed an egg, two polar nuclei, and three antipodals. Synergids were occasionally seen, and the egg was often inconspicuous. The antipodals were distinguished by a swollen, tear-drop shape, dense cytoplasm, chalazal position, and the absence of a wall separating them from the cavity of the sac. The

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

215

aposporous sacs lacked antipodals and had only one nuclei per sac. Sometimes there was a single polar nucleus and an egg. These results strongly suggest that the mechanism responsible for apomixis in cassava is apospory (the development of aposporic embryo sacs). Apospory in the angiosperms is the most common mechanism responsible for apomixis. In this type of apomixis the apospory embryo sac originates from one or more somatic cells of the ovule followed by its enlargement and vaculation (Asker, 1980). This type of apomixis explains the multiple plants per seed found in clone 031 by Nassar (1994). The presence of apomixis in clones 031 and 200 shows the potential of the utilization of wild species as a source of genes for apomixis in cassava since clone 031 represents generation F2 of a cross between cassava and M. dichotoma and clone 200 is an F1 of cassava and M. glaziovii. Sources of genes controlling apomixis in wild species that are relatives of corn, sugar beet, wheat, and several forage grasses have already been reported (Asker, 1979). Nassar et al. (1998) further validated the occurrence of apomixis in cassava by presenting a combination of molecular and embryonic evidence from two putative apomictic clones.

IV. PRODUCTION OF POLYPLOID TYPES A. PROSPECTS OF POLYPLOIDIZING CASSAVA, M. esculenta Crantz, BY UNREDUCED MICROSPORES One of the most promising approaches to improve cassava is to produce triploid types which may offer resistance to drought and insects, combined with high productivity (Hahn et al., 1990; Nassar, 1992). The triploid tree produced by Nassar (1992) showed a doubling of root production under semiarid conditions. Polyploid types were obtained by Nassar and were found to be produced by unreduced gamete fertilization (Nassar, 1992). It is believed that if parents with a high frequency of unreduced gametes could be selected, this would enhance the possibility of obtaining polyploid types in their progeny. 1. Unreduced Microspores in Cassava, M. esculenta Crantz, Clones The formation of unreduced microspores (gametes with somatic chromosome number) appears to be a common phenomenon in angiosperms (Harlan and de Wet, 1975; de West, 1980). They most likely have a major role in the evolution and origin of polyploids. From a plant breeding viewpoint, these gametes may lead to the development of highly productive tetraploids and triploids by sexual means for preserving their heterozygosity (Mendiburu and Peloquin, 1977). Probably the most convincing cytological evidence of their occurrence in higher plants has come from

216

NAGIB M. A. NASSAR

microspores. Prakken and Swaminathan (1952) observed that diads (with two reduced microspores) and tetrads (with four reduced microspores) occur together in the same plants of several Solanum species. This was confirmed later by several workers using different crop plants (Roads and Dempsey, 1966). Nassar (1992) reported for the first time in cassava interspecific hybrids as a consequence of meiotic irregularity. Now, there is agreement that diads form due to spindle abnormalities which may be visible at meiotic metaphases I and II (Nassar, 1996). Cassava clones grown in the germplasm bank of the Centro Agronomico Tropical de Investigacion y Ensenanza were used to prove the previously mentioned hypothesis (Vasquez and Nassar, 1994). Floral buds were collected and fixed for 24 h in 3:1 ethanol:acetic acid mixture, transferred to 70% ethanol, and stored at 5C. The anthers were squashed in a drop of 1% acetocarmine. Metaphase I was used for chromosome counting and study of chromosome associations. The tetrad stage was observed to determine the frequency of diads and triads. About 300 tetrads were counted in each clone. Photomicrographs were taken from temporary preparations using the Zeiss standard research microscope. Of the nine cassava clones studied, eight showed regular metaphase with complete pairing and formation of 18 bivalents. The ninth clone (No. 6477), known as Chioriqui, showed a sectorial chimera in the inflorescence. One sector of inflorescence developed flowers with normal metaphase I and complete pairing of chromosomes. No laggards at anaphase I and no micronuclei at the tetrad stage were observed. The other sector of the inflorescence had flowers with extreme abnormal metaphase I in all the flowers investigated. This was accompanied by the presence of empty anthers. The abnormality at metaphase I was due to lack of chromosome pairing asynapsis. In the 20 metaphases examined the chromosome association was 11 or 12 bivalents instead of 18 normal bivalents. The remaining chromosomes formed only univalents. This resulted in the formation of micronuclei at the tetrad stage. There were 1–15 micronuclei per tetrad (Table XXII). Apparently, this chimeral sector was due to a gene mutation in the second layer (LII) of the apex of the growing shoot. Since only a lateral sector exhibits this abnormality, the chimera must be sectorial rather than mericlinal or periclinal. This is the first report of a mutation which affects chromosome pairing in cassava. Since cassava reproduces vegetatively, it is likely that this mutation has been preserved for a long time in this indigenous Costa Rican clone. To trace unreduced microspore formation, diads and triads were checked among 300 tetrads of each clone. The most interesting case was found in clone No. 11,965 which formed 11 diads per 300 sporocytes studied, indicating first meiotic restitution (Table XXIII). No other clone showed diad formation. Regarding triad formation, four clones, namely, 9959 (Mangi), 6429, Vegna Mochera, and 6399, formed triads in the range of 1–1.3%. The clone 11,965, the only diad producer among the clones studied, also produced triads with a frequency of 1%. As can be seen from Table XXIII there is variation for 2n gamete production

217

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES Table XXII

Chromosome Association and Diad and Triad Frequencies in Investigated Germplasm Diads Cultivar 9,959 (Mangi) 6,417 (White) 10,861 (Num-4-RB) 6,429 (Negra Muchera) 11,965 (Sim Nombre) 6,379 (Amarilla-1) 6,473 (Vagana 4208) 6,477 (Chiriqui) Chimera a Chimera b

Triads

Chromosome association

Frequency

2%

Frequency

2%

18II 18II 18II 18II 18II 18II 18II

— — — — 11 — —

— — — — 3.7 — —

4 — — 4 3 3 —

1.33 — — — 1 1 —

18II 11II  14I

among the cassava clones. Several workers reported that in species with a tendency to form unreduced microspores, the frequency of such gametes may vary from one line to another (Mok and peloquin, 1975). In the meantime, stable high 2n gamete production has been observed in 2n gamete producer lines (Ballington and Galleta, 1976). Thus, the variation in the frequency of 2n gametes may be attributed mainly to their genotypic differences (Veilleux and Lauer, 1981). It seems that unreduced microspore formation is gene controlled and not due to the disturbance of chromosome asynapsis. However, it is agreed that this is due to the occurrence of nonfunctional spindle, resulting in all the metaphase chromosomes remaining in the center instead of separating to poles (Ramana, 1974; Vorsa and Bingham, 1979). 2. Unreduced Microspores in Interspecific Hybrids The causes of unreduced microspores in plants are variable, ranging from simple recessive genes (Mok and Peloquin, 1975) to disturbed spindle function in in-

Table XXIII Frequency of Diads and Five Kinds of Tetrads in the Sectorial Chimera of Clone 6477 (300 Tetrads Studied)

Parameter

Diad

Normal tetrads

Frequency %

1 0.3

18 6

Tetrads with different Nos. of micronuclei 26 8.7

41 13.9

26 8.9

188 62.6

218

NAGIB M. A. NASSAR

terspecific hybrids. Nassar et al. (1995) suggested that the disturbance of meiotic division in interspecific hybrids may lead to a higher frequency of aneuploid gametes, consequently making it possible to select polyploids among their progeny. Brazilian Manihot species are believed to be progenitors of cassava, and Brazil contains its centers of diversity (Nassar, 1978b). Interspecific hybridizations have been carried out systematically by Nassar since 1980 and many interspecific hybrids have been produced (Nassar, 1989). Some of these, which represent different interspecific hybrids and cover a vast array of variation, were used by Nassar in the current investigation. Ten interspecific hybrids and/or their progenies were used in this investigation: F1 M. glaziovii cassava (3 genotypes), F2 M. epruiinosa cassava, F2 M. anomala cassava (3 genotypes), F3 M. pseudoglaziovii

cassava (two genotypes), and F4 M. pseudoglaziovii cassava. These hybrids and progenies were chosen because they provide large genetic variation. Chromosome associations in meiotic metaphase I and the occurrence of dyads and triads and also pollen viability were studied. For the meiotic study, inflorescences were fixed in a 3:1 mixture of absolute alcohol and glacial acetic acid and kept at 5C for 24 h. The anthers were smeared in an acetocarmine stain. Chromosome configurations at metaphase I, together with spared formation, were studied. For the pollen viability study, one to three flowers per plant were selected, and their pollen was crushed in acetocarmine and scanned. Pollen counts and the percentage of stained normal pollen were calculated. Chromosome associations and their frequency in meiotic metaphase I of PMCs of all interspecific hybrids have shown regular pairing of 18 bivalents except the F2 progeny of M. glaziovii cassava (first genotype), which revealed the presence of two univalents in 5% of the cells examined. This genotype also showed a high frequency (3.7%) of dyad formation and low pollen viability (42%). The study of PMCs in the tetrad phase revealed the formation of abnormal tetrads having one to three micronuclei in all hybrids used in the experiment, except F1 M. glaziovii cassava (second genotype). The abnormal tetrads ranged from 0.8% in f1 M. glaziovii cassava (first genotype) to 94% in the F1 hybrid M. epruinosa

cassava (Table XXIV). The high frequency of unreduced microspores in the F2 progeny of M. glaziovii

cassava will facilitate selection of this genotype and its progeny as a possible progenitor of polyploids in the future. Apparently, the dyad formation is due to meiotic restitution. The low pollen viability may be due to univalent formation and the consequent irregular chromosome distribution leading to unbalanced gametes. Cassava is a natural allopolyploid judging by its chromosome number and the complete pairing of its meiotic metaphase chromosomes (Nassar, 1978b; Vásquez and Nassar, 1994). If unreduced gametes were responsible for its natural polyploidization in the past, it is possible that one can detect them among wild relatives and their hybrids with cassava. The presence of unreduced microspores in the hybrid progeny confirms this hypothesis.

Table XXIV Frequency and Percentage of Diads and Triads in Different Sporads

Manihot hybrids

Triad

Tetrad

Abnormal tetrad

n (%)

n (%)

n (%)

n (%)

Pollen examined

n (%)

1217 1168

10 (0.82) 43 (3.70)

8 (0.66) 8 (0.70)

1158 (95.15) 1107 (94.80)

41 (3.37) 10 (0.80)

2979 1713

1197 (40.18) 723 (42.21)

1044

0 (0.00)

4 (0.40)

1040 (99.60)

0 (0.00)

1463

1123 (76.76)

1252 1374 1145 1136 1416 1210 1138

0 (0.00) 1 (0.07) 1 (0.00) 1 (0.09) 0 (0.00) 0 (0.00) 1 (0.09)

0 (0.00) 0 (0.00) 0 (0.00) 0 (0.00) 0 (0.00) 0 (0.00) 0 (0.00)

1252 (100.00) 1326 (96.51) 1117 (95.56) 1106 (97.36) 1134 (95.62) 1207 (98.80) 1125 (98.80)

0 (0.00) 47 (3.42) 28 (2.44) 29 (2.55) 62 (4.38) 3 (0.20) 12 (1.05)

603 836 1297 801 1427 1130 1704

567 (94.03) 183 (21.89) 441 (34.00) 452 (56.43) 873 (61.18) 1034 (94.50) 862 (50.59)

491

2 (0.40)

3 (0.60)

476 (98.86)

10 (2.04)

1273

299 (23.49)

Microspores examined

M. glaziovii cassava F1 M. glaziovii cassava cassava (1st genotype) F1 M. glaziovii cassava cassava (2nd genotype) M. epriminoso cassava F2 M. anomala cassava (1st genotype) F2 M. anomala cassava (2nd genotype) F2 M. anomala cassava (3rd genotype) F3 M. pseudoglaziovii cassava F4 M. pseudoglaziovii cassava F1 M. dichotoma cassava (1st genotype) F1 M. dichotoma cassava (2nd genotype)

Pollen viability

Diad

220

NAGIB M. A. NASSAR

Vásquez and Nassar (1994) reported a high frequency of unreduced microspores among cassava clones. It is believed, therefore, that this character is heritable and genetically controlled. It was probably acquired by cassava from its wild ancestors (Nassar et al., 1995), representing an evolutionary remnant in cassava. The significance of this phenomenon is that it provides direct evidence on polyploidization from lower ploidy levels through the mechanism of unreduced gametes and not through other types of somatic doubling (Harlan and de Wet, 1975; de Wet, 1980). According to the genes of wild ancestors they have probably had the opportunity to combine and produce larger rooted cassava (Nassar, 1992; Nassar et al., 1996a). It seems that this character is correlated with meiotic irregularities provided there is some univalent formation in the genotype producing unreduced microspores. It is worth mentioning that this unreduced microspore-producing genotype is a progeny of a natural hybrid of M. glaziovii with cassava. This natural hybrid has been maintained by farmers for hundreds of years through vegetative reproduction. The unreproduced microspore gene has probably arisen through recurrent mutation and been maintained through vegetative reproduction. In other germplasm that reproduces sexually, it would be eleminated by natural selection due to the abortion of gametes which carry it.

B. INDUCTION OF A PRODUCTIVE ANEUPLOID IN CASSAVA, M. esculenta Crantz Within the wild Manihot species which have been evaluated and hybridized with cassava by Nassar to transfer their useful genes (Nassar, 1978a, 1980, 1986), two interspecific hybrids between cassava and wild Manihot species have been obtained and/or maintained in the Nassar program at the Universidade de Brasília. One of the two interspecific hybrids was between cassava and M. pseudoglaziovii Pax (Nassar, 1982); the second hybrid was between cassava and M. neusana Nassar (Nassar, 1989). These hybrids were cloned by vegetative reproduction and seed produced by open pollination. Ten seeds were collected from each hybrid for germination studies. Four seeds from progenies of cassava with M. pseudoglaziovii germinated, of which one plant was selected for vigor. Only two seeds of progenies of the hybrid of cassava with M. neusana germinated, of which one plant was selected for vigor. These three plants were cloned by vegetative propagation of cuttings. When the plants flowered, the three clones and their parents were studied meiotically for chromosome association in metaphase I, segregation in anaphase I, and pollen viability. Mitotic analysis was also performed on root tips taken from cuttings. After 9 months of growth, these clones were examined for root productivity. For the study of meiosis, buds were fixed in a 3:1 mixture of ethyl alcohol and glacial

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

221

acetic acid for 24 h, transferred to 70% alcohol, and then smear stained with 1% acetic carmin. For mitotic counting of chromosomes, root tips were treated with 0.2% colchicine for 2 h, washed in distilled water, transferred to the previously mentioned fixative fluid for 24 h, treated with 1 N HCl for 10 min, and smear stained with 1% acetic carmin. Meiotic examination in metaphase I of the clone, progeny of the interspecific hybrid of cassava with M. pseudoglaziovii (HPS), showed a mean of 16–19 bivalents, 154 trivalents, and 1.00 univalent for the 15 sporocytes examined. These results suggested an aneuploid having a 2n  2 chromosome constitution. The parental hybrid showed 2n 36, with a mean association of 17.42 bivalents and 1.58 univalents. Estimation of pollen viability revealed a viability percentage of 54% in 2917 aneuploid grains and 40% in 1350 of the parental hybrid. The hybrid of cassava and M. neusana (HO1) had a mean of 1.86 trivalents, 16.13 bivalents, and 0.13 univalents among 30 PCMs examined. The chromosome constitution was 2n 38. This was confirmed by chromosome counting at mitosis of root tip cells. The second clone of this cross (HO4) showed a mean chromosome association of 1.63 trivalents, 12.41 bivalents, and 8.84 univalents in metaphase I. This clone also had 2n 38. The parent had a mean of 17.42 bivalents and 1.58 univalents, which means 2n 36. Pollen viability was found to be 35.8% in HO1, 17.7% in HO4, and 36.8% in their parent. The progeny of the interspecific hybrid of cassava with M. pseudoglaziovii had a very large starched root with a mean of 6.1 kg per plant at 10 months for the 15 plants evaluated, compared to 3.7 kg for the cultivated cassava clones. The other two aneuploids had fibrous roots. These aneuploids apparently resulted from disturbance of meiotic division of their interspecific hybrid parents. This disturbance was interpreted by Nassar et al. (1995) as an inhibition of spindle function and chromosome asynapsis. Both have led to laggard formation and unbalanced gametes. Thus, the origin of these aneuploids can be attributed to the fertilization of an n  1 egg by an n  1 male gamete based on the fact that gametes with this structure are more viable than gametes with n  2. The other possibility is the fertilization of an n  2 egg with an external n male gamete because the plants were open pollinated. It was not possible to determine whether the male gamete came from selfing or from another pollinator plant. In various crops, interspecific hybridization has led to the disturbance of meiotic division. Similar findings were reported in Trifolium pratense by Parrot and Smith (1984) and in Medicago spp. by Vorsa and Bingham (1979). No aneuploids, however, were obtained in these crops. The two additioinal chromosomes in clone HPS resulted in the production of a large tuber root that is superior to normal root production in cultivated cassava clones. In all likelihood this size increase in due to additive polygenes. These polygenes may be distributed on more than one chromosome. Nassar (1978b) postulated the origin of cassava as a product of hybridization

222

NAGIB M. A. NASSAR

between two wild species, both having fibrous nonstarchy root as a result of an accumulation of additive polygenes derived from ancestral wild parents.

C. PRODUCTION OF TRIPLOID CASSAVA, M. esculenta Crantz, BY HYBRID DIPLOID GAMETES Through the Nassar program of cassava germplasm collection and utilization in Brazil, a natural hybrid of M. pseudoglaziovii and cassava was collected in 1978 from Remigo county, Paraíba state, and grown in the living collection at the Biological Experimental Station, Universidade de Brasília (Nassar, 1982). This natural hybrid was identified by the marker genes of winged fruit which came from cassava and peltated, pendurated leaf, two characteristic genes of M. pseudoglaziovii overlaps cassava plantations. The hybrid was multiplied vegetatively and cloned 20 times to individuals maintained in the living collection and used in this experiment. This clone was left for open pollination “entre si.” Seeds collected were planted in 1980 and the progeny of 22 plants were produced. Evaluation of the progeny for root production was carried out in the following years. The natural hybrid and its selected progeny were studied meiotically. The anthers were fixed for 24 h in a 3:1 solution of ethanol and acetic acid, transferred to 70% ethanol, and stored at 5C. The anthers were squashed in 1% acetocarmine. Pollen analysis was conducted by collecting pollen from undehisced natural anthers of a flower onto a microscope slide and staining with 1% acetocarmine. Pollen grain diameter was assessed at 450x using an eyepiece micrometer. A minimum of 1000 stained grains per plant were counted. Pollen stainability was used as a criterion for viability. Well-filled, darkly stained pollen grains were considered fertile and partially filled, and unstained ones were considered sterile. The frequency of mature 2n pollen grains was based on counts of 1000 pollen grains stained with 1% acetocarmine. Classification of n vs 2n pollen was made by visual size discrimination of stained pollen, given that the latter should have twice the nuclear volume of the former. To obtain data on ploidy status, mitosis in root tips was examined and chromosome number was counted to confirm the plant’s chromosome constitution. Screening the hybrid cloned plants for tuber formation showed them to have complete fibrous roots but they extended deep in the ground, almost 5 or 6 m in length. The presumed progenitor, M. pseudoglaziovii grows in one of the driest pockets of Paraíba state, Remigio county, where the wild species and the hybrid were collected. Apparently, the wild species had acquired this deep-rooted character as a mechanism for removal of subterranean water. This character may be useful in subsequent improved generations for production of drought-tolerant cassava cultivars. Two years after planting, the hybrid reached 3 or 4 m in height. Its leaves

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

223

were peltate and frequently pendulous. These characters are typical of M. pseudoglaziovii and one of the striking features that distinguish this species from the other 98 species recognized by Rogers and Appan (1973). The hybrid carried few fruit (4 or 5 fruits compared to 60–70 fruits in the pure M. pseudoglaziovii of the same age). The fruit is winged, a marker gene characterizing cassava fruit (Nassar, 1989). Cytological analysis of PMCs revealed a regular occurrence of 18 bivalents at metaphase I. No multivalents were observed in 50 metaphase I cells studied. However, disturbed anaphases I was noted in 6 of the 30 anaphase I cells studied. Dyads and triads were noted totalling 14, compared to 86 normal tetrads. Pollen viability as measured by stainability with acetocarmine was as low as 41%. Of these viable pollen, 21% were small. The high percentage of sterility despite regular pairing may be attributed to the maintenance of the hybrid through vegetative reproduction by farmers in Paraiba state. This mode of reproduction may have led to the accumulation of sterility mutations which have not been eliminated by natural selection through the sexual cycle (Nassar, 1978c). Of the 20 plants cloned from the hybrid, 82 fruits were collected. They gave 48 seeds with an average of 0.59 seeds per fruit. The average number of fruits collected per plant is 4.1 compared to 60 –70 fruits in the wild species plant. This low fertility is expected in view of the high sterility of pollen grains. By planting the seeds, 25 seedlings were obtained but all of them, with the exception of four plants that survived, were chlorotic or deformed and died. The plants that survived were identified in the living collection by numbers 120 –123. Of the four plants, three gave fibrous roots, but the fourth (No. 121) was very vigorous, resistant to stem borers, and produced large tuberated roots by the end of the year. When multiplied vegetatively and compared with other cassava clones it showed superiority in tuber formation by the age of 12 months. This superiority reached approximately 28% more tubers than the Sonora clone, one of the best commercial cultivars. Its production was 2.13 kg/plant compared with 1.67 for Sonora. An evaluation of this selection under arid conditions was carried out during the years 1985 –1989 by the National Center for Tropical Semi-arid Region and confirmed its superiority (Table XXV). The meiotic study of this plant showed formation of trivalents, bivalents, and univalents in metaphase I with different frequencies (Table XXVI). The meiotic division triploid resulted from the fertilization of a 2n gamete with n gamete in the progenitor hybrid. Viability of pollen of this triploid is as low as 19%. Seeds rarely formed. However, five seeds were collected and have been reproduced in our program. The 2n pollen formation which led to production of this triploid is reported in cassava by Nassar for the first time. The cytological mechanism of its formation is probably inhibition of spindle at metaphases I and II judging from the triad formation. In the literature, several mechanisms were reported. Clement and Stanford (1961) and Tyagi (1988) attributed it to abnormal cytokinesis, whereas

224

NAGIB M. A. NASSAR Table XXV

Average Productivity per Plant of the Triploid (Selection 121) and Other Cassava Clones under Savanna and Semiarid Conditions Clone Triploid 121 Sonora Triploid 121 Nagib 01/84 Nagib 02/84 Nagib 03/84 Nagib 04/84 Cigana

No. plants

Productivity

30 30 29 27 29 28 22 28

2.13 1.67 5.9 1.8 3.7 1.1 4.1 2.9

Region, location of trial, and harvest age (kg/plant) Central Brazil, conducted at the experimental station, Univ. of Brasília; harvest age 12 months Under semiarid conditions; trial conducted by the CPATS, Petrolina, northeastern Brazil; harvest age 18 months

Mendiburu and Peloquin (1977) proposed parallel spindle as an additional mechanism. The 2n pollen mutations were described in several plant species: Zea mays (Roads and Dempsey, 1966), Solanum (Hogland, 1970), Medicago sativa (Vorsa and Bingham, 1979), and Lolium (Sala et al., 1989). The vigor of Nassarselected triploid as seen in its productivity both under central Brazil conditions and in the semiarid tropics adds evidence of the usefulness of this 2n gamete phenomenon as a powerful mechanism for transferring variability and fitness to polyploid offspring. If a partially fertile type of triploid could be produced, it will serve in establishing a founder population of a new chromosome race. Its progeny may rehybridize with new polyploids and diverse genotypes, producing additive heterotic potentialities. Since trivalent occurrence in this triploid is predominantly demonstrating gene exchange between wild and cultivated genomes, it is likely that this will generate more variability in the progeny. The wild parent and its interspecific hybrid progeny are highly tolerant to drought as indicated by their deep roots. They normally survive the frequent years of drought in their natural habitat. This triploid is a potential progenitor of cultivars adapted to these conditions.

Table XXVI Chromosome Associations at Metaphase I in the Triploid and Its Parent Average chromosome association Type Natural hybrid Triploid

No. PMCs examined 50 30

III

II

I

— 10.2

18 9.2

— 3.6

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

225

V. PROTEIN CONTENTS IN CASSAVA CULTIVARS AND ITS HYBRID WITH WILD Manihot SPECIES Tubers of 16 cassava clones, 8 months old and maintained in the germplasm collection at the Experimental Station, University of Brasília, were analyzed chemically for protein content. Total nitrogen and dry matter basis were determined by the 1970 AOAC procedures (Nassar and Dorea, 1982). Percent protein was obtained by multiplying percent N by the factor 6.25. For every clone two samples were analyzed, one from a small tuber (50 g) and the second from older tubers (200 g). Tubers were peeled, and protein was estimated in the coth peel and pulp. Tubers of hybrid between cassava clone Catelo and M. oligantha were analyzed in the same way. Seed of cassava and wild Manihot species maintained in the living collection at the University of Basília were analyzed for protein content by the same procedures. The concentration of protein in tubers of cassava clones and tubers of the hybrid is presented in Table XXVII. It can be seen that protein percent (N% 6.25 protein percent) ranged from 0.7 to 1.2% for larger tubers (200 g), whereas the range was 0.9 to 1.4% for tubers 50 g. In the same clone, protein content of the pulp was higher in small tubers than in larger ones. These data agree with those found by Akinrele (1964), who reported 0.7% for protein content in peeled tuber on a dry matter basis, and Chada (1961), who found 1.2%. However, these researchers did not pay attention to the effect of tuber size on protein content. Jennings (1959) stated that protein in cassava root tends to be concentrated in the outer zone of the root. This may explain why the small tubers have higher protein content since they have a larger proportion of the outer zone than do the older and larger ones. The very little variation of protein content among the 16 clones collected throughout Brazil shows that selection for this characteristic in cassava clones will not bring any notable improvement. From Table XXVII it can also be seen that protein is higher in peel than in pulp in all the examined samples and in all clones. This may be explained by the previously mentioned statement of Jennings that protein in cassava roots tends to be concentrated in the outer layers. The analysis of hybrid tubers showed a notable increase in the hybrid of cassava with M. oligantha at 4.6%. In an earlier paper (Nassar and Costa, 1978), the protein content in M. oligantha was reported to be 7.1% on a dry matter basis. Moreover, crosses of this species with cassava were highly fertile (Nassar, 1980). Nassar and Fitchner (1978) also showed a low HCN content in the wild species M. oligantha. This may exclude any possibility that high protein content in the species is due to HCN nitrogen. Table XXVIII shows the results of seed analysis of some wild Manihot species maintained in our living collection. The highest protein content is that of M. Brachyandra followed by M. alutacea.

226

NAGIB M. A. NASSAR Table XXVII Protein Content of Tubers of Cassava Clones

Clone CBM 0206 EAB 348 BGM 188 CPM 0231 CPM 2002 CPM 0232 BGM 808 CPM 0225 BGM 204 CPM 1805 EAB 1156 EAB 484 BGM 048 BGM 020 CPM 1060 EAB 675 Hybrid

Approximate size (g)

Protein in peel (%)

Protein in pulp (%)

200 50 200 50 200 50 200 50 200 50 200 50 200 50 200 50 200 50 200 50 200 50 200 50 200 50 200 50 200 50 200 50 200 50

2.13 2.09 1.41 1.69 — 1.68 — 1.56 — 2.08 2.00 1.82 1.63 — 1.38 1.25 1.24 — 1.14 1.37 1.58 1.28 1.96 — 1.41 1.11 1.80 1.53 — 1.58 1.36 1.51 6.63 8.06

0.90 1.22 0.85 1.04 — 1.45 — 1.26 — 0.99 1.02 1.15 0.93 — 0.89 0.95 1.06 — 0.72 1.00 0.84 1.16 1.07 — 0.82 1.17 0.98 1.23 — 1.19 0.70 0.93 4.56 4.56

Manihot brachyandra is native to western Pernambuco and northern Bahia, two of the driest areas in Brazil. Nassar (1980) reported that seed of wild cassava is eaten by the population of these regions particularly in times of famine. Jones (1959) reported that cassava seed is eaten in several parts of west and central Africa. Thus, the discovery of the high protein content in native cassava hybrids may open a new door to better protein-balanced food for people of the tropical world.

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

227

Table XXVIII Protein Content in Wild Manihot Species Seed on Dry Matter Basis Species

Protein%

M. glaziovii M. caerulescens M. brachyandra M. pseudoglaziovii M. alutacea M. zenhtneri M. dichotoma M. reptans M. esculenta

30.09 27.91 35.35 31.15 37.33 28.99 29.24 33.25 26.81

ACKNOWLEDGMENT The living collection of wild Manihot species was established at the Universidade de Brasília with the help of the International Development Research Center Ottawa, Ontario, Canada, for which the author is grateful.

REFERENCES Abraham, A. (1975). Breeding of tuber crops in India. J. Genet. Plant Breeding 17, 212–217. Abraham, A., Panicher, P. K., and Mathew, P. M. (1964). Polyploids in relation to breeding in tuber crops. J. Indian Bot. Soc. 43, 278 –283. Akinrele, I. A., Collins, C., Cook, A. S., Hogate, R. A., Junaid, Y., and Baumer, G. (1962). Pilot plant (1 ton a day) results of 3 month trial run, Research Reprints, Vol. 13, pp. 1– 30. Federal Institute of Industrial Research, Nigeria. Anonymous (1968). Tabla de composicion de pastes y otros alimentos de Centro America, Publication No. E-440. Instituto Nacional de Centro America e Panama. Asker, S. (1979). Progress in apomixis research. Hereditas 91, 231–240. Asker, S. (1980). Gametophytic apomixis: Elements and genetic regulation. Hereditas 93, 277–293. Bolhuis, G. G. (1953). A survey of some attempts to breed cassava varieties with a high content of proteins in the roots. Euphytica 20, 107–112. Brewer, J. L., and Henstra, S. (1974). Pollen of Pyrethrum. Fine structure recognition. Euphytica 23, 567–663. Brown, W. V., and Emery, W. H. P. (1958). Apomixis in the Gramineae: Panicoideae. Am. J. Bot. 45, 253–263. Chada, Y. R. (1961). Sources of starch in Commonwealth territories. III. Cassava. Trop. Sci. 3, 101–113. Clement, W. M., Jr., and Stanford, E. M. (1961). A mechanism for the production of tetraploid and pertoploid progeny from diploid and tetraploid cross alfafa. Crop Sci. 1, 11–14. Cruz, N. D. (1968). Citologia no gênero Manihot Adans. 1 Determinação do número de cromossomo em algumas espécies. anal. Acad. Bras. Ci. 40, 81– 95.

228

NAGIB M. A. NASSAR

de Wet, J. M. J. (1980). Origin of polyploids. In “Polyploidy: Biological Relevance” (W. H. Lewis, Ed.). Plenum, New York. Dione, L. A. (1963). Studies on the use of Solanum acaule as a bridge between Solanum tuberosum and species in the series Bulbocastana, cardiophylla and Pinnatisecta. Euphytica 12, 263 –269. Dobzhansky, T. (1973). Genética do processo evolutivo. In “Tradução de Celso Abbade Mourão.” Poligono, São Paulo, Brazil. Doyle, J. J., and Doyle, J. L. (1987). Isolation of plant DNA from fresh tissues. Focus 12, 13 –15. Food and Agriculture Organization (FAO) (1994). “Production Yearbook.” FAO, Rome. Graner, E. A. (1942). Genet. Manihot Braganita 2, 13 –19. Grattapaglia, D., and Sederoff, R. R. (1994). Genetic linkage maps of Eucalyptus grandis and E. uroplylla using a pseudo-testcross strategy and RAPD markers. Genetics 137, 1121–1137. Grattapaglia, D. E., Nassar, N. M. A., and Dianese, J. (1986). Biossistemática de espécies brasileiras do Gênero Manihot baseada em padrões de proteína da semente. Ciencia Cultura (J. Brazilian Assoc. Advancement Sci) 19(3), 294 – 300. Grattapaglia, D., Costa e Silva, C., and Nassar, N. M. A. (1996). Strict maternal inheritance of RAPD fingerprints confirm apomixis in cassava (Manihot esculenta Crantz). Can. J. Plant Sci. 76, 379–382. Gustafsson, A. (1946). Apomixis in higher plants. Part 1. The mechanisms of apomixis. Lunds Univ. Arsskr. Adv. 42, 1– 66. Hahn, S. K., Bai, K. V., and Asiedu, R. (1990). Tetraploids, triploids and 2n pollen from diploid interspecific crosses with cassava. Theor. Appl. Genet. 79, 433 – 439. Harlan, J. (1961). Geographic origin of plants useful to agriculture. Am. Assoc. Adv. Sci. 66, 3 – 9. Harlan, J. R., and de Wet, J. M. J. (1975). On O wings and a prayer; The origins of polyploids. Bot. Rev. 41, 361–390. Hogland, M. (1970). Meiosis in Solomum phureja. Jereditas 66, 183 –188. Jennings, D. L. (1959). Manihot melanobasis Muell. Arg.—A useful parent for cassava breeding. Euphytica 8, 157–162. Knox, R. B., Willing, R. R., and Prior, L. D. (1972). Interspecific hybridization in populus using recognitioin pollen. Silvae Genet. 21, 263 –298. Laemilli, U. K. (1970). Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature 227, 680 – 685. Lanjouw, J. (1939). Two interesting species of Manihot L. from Surinam. recueil Travaux Botaniques Neerlandais 36, 542– 549. Magoon, M. L., Jos, J. S., and Appan, S. G. (1966). Cytomorphology of interspecific hybrid between cassava and ceara rubber. Chromosome Information Service (Japan), pp. 8 –10. Mendiburu, A. O., and Peloquin, S. J. (1977). The significance of 2n gametes in potato breeding. Theor. Appl. Genet. 49, 53 – 61. Mok, D. W. K., and Peloquin, S. S. (1975). The inheritance of three mechanisms of diploandroid (2n pollen) formation in diploid potatoes. Heredity 35, 295 – 302. Narty, F. (1969). Studies on cassava, Manihot utilissima Pohl. 1. Cyanogenesis: The biosynthesis of linamarin and lutaustalin in etioled seedlings. Phytochemistry 7, 1307–1312. Nassar, N. M. A. (1978a). Genetic resources of cassava: Chromosome behavior in some Manihot species. Indian J. Genet. Plant Breeding 38, 135 –137. Nassar, N. M. A. (1978b). Conservation of the genetic resources of cassava (Manihot esculenta); Determination of wild species localities with emphasis on probable origin. Econ. Bot. 32, 311– 320. Nassar, N. M. A. (1978c). Microcenters of wild cassava Manihot spp. diversity in central Brazil. Turrialba 28(4), 345 – 347. Nassar, N. M. A. (1978d). Hydrocyanic acid content in some wild Manihot (cassava) species. Can. J. Plant Sci. 58, 577– 578. Nassar, N. M. A. (1978e). Some further species of Manihot with potential value to cassava breeding. Can. J. Plant Sci. 58, 915 – 916.

CASSAVA, M. esculenta Crantz, GENETIC RESOURCES

229

Nassar, N. M. A. (1978f ). Wild Manihot species of central Brazil for cassava breeding. Can. J. Plant Sci. 58, 257–261. Nassar, N. M. A. (1979a). A study of the collection and maintenance of the germoplasm of wild cassavas Manihot spp. Turrialba 29(3), 221. Nassar, N. M. A. (1979b). Three brazilian Manihot species with tolerance to stress conditions. Can. J. Plant Sci. 59, 533– 555. Nassar, N. M. A. (1980a). Attempts to hybridize wild Manihot species with cassava. Econ. Bot. 34, 13–15. Nassar, N. M. A. (1980b). The need for germplasm conservation in wild cassava. Indian J. Genet. Plant Breeding 39(3), 465 – 470. Nassar, N. M. A. (1982). Collecting wild cassava in Brazil. Indian J. Genet. Plant Breeding 42, 405–411. Nassar, N. M. A. (1984). Natural hybrids between Manihot reptans Pax and M. alutaced Rogers & Appan. Can. J. Plant Sci. 64, 423 – 425. Nassar, N. M. A. (1985). Manihot neusana Nassar: A new species native to Parana, Brazil. Can. J. Plant Sci. 65, 1097–1100. Nassar, N. M. A. (1986). Genetic variation of wild Manihot species native to Brazil and its potential for cassava improvement. Field Crop Res. 13, 177–184. Nassar, N. M. A. (1989). Broadening the genetic base of cassava, Manihot esculenta Crantz, by interspecific hybridization. Can. J. Plant Sci. 69, 1071–1073. Nassar, N. M. A. (1991). Production of triploid cassava, Manihot esculenta Crantz, by hybrid diploid gametes. Field Crop Res. 13, 173 –182. Nassar, N. M. A. (1992). Cassava in South America: A plant breeder view. Ciëncia Cultura (J. Brazilian Assoc. Advancement Sci.) 44, 25 –27. Nassar, N. M. A. (1994a). Selection and development of apomitic cassava clone. Ciëncia Cultura (J. Brazilian Assoc. Advancement Sci. 41(6), 168 –171. Nassar, N. M. A. (1994b). Relationship between Manihot species: A review of the proceedings of the 2nd Internacional Scientific Meeting of the Cassava Biotechnology Network. Bogor, 90 –100. Nassar, N. M. A. (1995). Development of cassava interspecific hybrids for savanna (cerrado) conditions. J. Root Crops 22, 9 –17. Nassar, N. M. A., and Costa, C. P. (1978). Tuber formation and protein content in some wild cassava (mandioca) species native to central Brazil. Experiëncia 33, 1304 –1306. Nassar, N. M. A., and Dorea, G. (1982). Protein contents of cassava cultivars and its hybrid with Manihot species. Turrialba 32(4), 429 – 432. Nassar, N. M. A., and Grattapaglia, D. (1986). Variabilidade de clones de mandioca em relação ä fertilidade e aspectos morfológicos. Turrialba 36(4), 555 – 559. Nassar, N. M. A., and O’Hair, S. K. (1985). Variation among clones in relation to seed germination. Indian J. Genet. 45(2), 429 – 432. Nassar, N. M. A., Silva, J. R., and Vieira, C. (1986). Hibridação insterespecifica entre mandioca e espécies silvestres do Manihot. Ciëncia Cultura ( J. Brazilian Assoc. Advancement Sci.) 33(6), 1050–1055. Nassar, N. M. A., Nassar, H. N. M., Vieira, C., and Saraiva, S. L. (1995). Cytogenetic behavior of interspecific hybrids of cassava and M. neusansa Nassar. Can. J. Plant Sci. 75, 675 – 678. Nassar, N. M. A., Nassar, H. N. M., Carvalho, C. G., and Vieira, C. (1996a). Induction of a productive aneuploid in cassava, Manihot esculenta Crantz. Brazilian J. Genet. 19, 123 –125. Nassar, N. M. A., Carvalho, C. G., and Vieira, C. (1996b). Overcoming barriers between cassava Manihot esculenta Crantz and wild relative M. pohili Warwa. Brazilian J. Genet. 19, 617– 620. Nassar, N. M. A., Vieira, M. A., Vieira, C., and Grattapaglia, D. (1998a). Molecular and embionic evidence of apomix in cassava interspecific hybrids (Manihot spp). Can. J. Plant Sci. 78, 348 – 352. Nassar, N. M. A., Vieira, M. A., Vieira, C., and Grattapaglia, D. (1998b). Molecular and embryonic evidence of apomix in cassava (Manihot esculenta Crantz). Euphytica 102, 9 –13.

230

NAGIB M. A. NASSAR

Nichols, R. F. W. (1947). Breeding cassava for resistance. East Afr. Agr. J. 12(3), 184 –194. Novy, R. G., Kobak, C., Gofreda, J., and Vorsa, N. (1994). RAPDs identify varietal misclassification and regional divergence in cranberry (Vaccinium macrocarpon (Ait.) Pursh. theor. Appl. Genet. 88, 1004–2020. Oke, O. L. (1973). The mode of cyanide detoxification. In “Chronic Cassava Toxicity. Proceeding of Inderdisciplinary Workshop, London, 29– 30 January 1973” (B. Nestel and R. MacIntyte, Eds), pp. 97–104. IDRC-OLOe, Ottawa, Ontario, Canada. Parrot, W. A., and Smith, R. R. (1984). Production of 2n pollen in red clover. Crop. Sci. 24, 467– 472. Prakken, R., and Swaminathan, M. S. (1952). Cytological behaviour of some interspecific hybrids in the genus Solanum. Genetic 54, 505 – 522. Raymond, W. D., Jojo, W., and Zicodemus, Z. (1941). The nutritive value of some Tanganyka foods: Cassava. East Afr. Agric. J. 6, 154 –159. Roads, M. M., and Dempsey, E. (1966). Induction of chromosome doubling at meiosis by elongated gene in maize. Genetics 54, 505 – 522. Rogers, D. J. (1963). Studies on Manihot esculenta Crantz and related species. Bull. Torrey Bot. Cl. 99, 43–54. Rogers, D. J., and Appan, C. (1973). Manihot, Manihotoides, Euphorbiaceae. In “Flora Neotropica.” Hafner, New York. Rogers, D. J., and Fleming, H. S. (1973). A monograph of Manihot esculenta with an explanation of the taximetrics method used. Econ. Bot. 27, 1–113. Sastri, D. C., and Shivanna, K. R. (1976). Attempts to overcome interspecific incompatibility in Sesanum by using recognition pollen. Ann. Bot. 40, 891– 893. Schmidt, C. B. (1951). A mandioca, contribuição para o conhecimento de sua origem. Boletim Agric. 1, 1. Stebbins, G. L. (1950). “Variation and Evolution in Plants.” Columbia Press, New York. Vavilov, N. I. (1951). Phytogeographic basis of plant breeding. The origin, variation, immunity and breeding of cultivated plants. Chronica Bot. 13, 1– 366. Vásquez, N., and Nassar, N. M. A. (1994). Unreduced microspores in cassava, Manihot esculenta Crantz clones. Indian J. Genet. Plant Breeding 54, 436 – 441. Vorsa, N., and Bingham, E. T. (1979). Cytology of pollen formation in diploid alfalfa, Medicago sativa. Can. J. Genet. Cytol. 21, 526 – 530. Webber, K., and Osborn, M. J. (1969). The reliability of molecular weight determination by sulfate– polyacrilamide gel electrophoresis. J. Biol. Chem. 244, 4406 – 4412. Williams, J. G. K., Kubelik, A. R., Livak, K. J., Rafalski, J. A., and Tingey, S. V. (1990). DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Res. 18, 6531–6535. Young, B. A., Sherwood, R. T., and Bashaw, E. C. (1979). Cleared-pistil and thick-sectioning techniques for detecting aposporous apomixis in grasses. Can. J. Bot. 57, 1668 –1672.

Index A Absorption defined, 3–4 of phosphate in soil, 101 Acidolysis, 130 Adsorption defined, 3 on mineral surfaces, 7 of phosphate in soil, 101, 102 pH and, 104–105 Advection-dispersion equation, 50 Agricultural chemicals, see Herbicides; Pesticides Agricultural management adverse environmental effects, 76, 77 effects on soil organic matter, 87– 88 Agroecosystem Resources Group, 77 Agroecosystems environmental degradation and, 76 environmental indicators biological, 79–82 chemical, 83–84 of greatest significance, 78 –79 landscape, 83, 84–85 physical, 78, 82–83 ranking of, 90, 91t recommendations for using, 92 soil organic matter, 85 – 90 environmental monitoring and, 77–78 overview of, 76–77 Aluminum chelated by organic acids, 129 –130 in soil phosphorus dynamics, 102, 103 Aluminum phosphate solubility in liquid media, 125 Ammonium effects on phosphate solubilization by fungi, 132 Aneuploidy in cassava, 220–222 Apomixis in cassava, 210–215

Apospory in cassava, 214 –215 Aquifer remediation modeling, 65 Arrhenius equation, 32 Aruak Indians, 185 Aspergillus, see Fungi, phosphate-solubilizing

B Bacteria phosphate solubilizing, 141–142 Base flow described, 160t, 172 phosphorus transfer and, 157–158 Batch solution-solid techniques, 45 – 49 Bioavailability coupled sorption-degradation kinetic models, 58 – 65 overview of, 56 – 58 of soil chemicals, 83 Biodegradation sorption-degradation kinetic models, 58 – 65 Biporous diffusion model, 36 – 37 “Black carbon” sorption by, 13 Bolivia wild species of Manihot in, 184 Brazil domestication of cassava in, 185 wild species of Manihot in, 181t, 182–184 Bridge species, 208, 209 –210 Bypass flow, 160t, 165

C Calcium in soil phosphorus dynamics, 102, 103 Calcium phosphate solubility in liquid media, 125 Carbon, atmospheric soil organic matter and, 86– 87 Carboxylic acids chelation of phosphate metal cations, 129

231

232

INDEX

Cassava, see also Manihot apomixis in as apospory, 214 –215 embryo sac analysis of, 212–213, 214 –215 genetic study of, 210 –212 molecular analysis of, 213 –215 genetic origins of, 184 –185, 221–222 interspecific hybrids, 184 –185, 187 aneuploidy in, 221–222 apomixis and, 211–212 drought tolerance and, 204 –207 with M. anomala, 198 –200 with M. neusana, 198 –204 overcoming crossing barriers to M. pohlii, 207–210 polyploidy and, 217–220 protein content, 225 –226 unreduced microspores in, 217–220, 223 – 224 place of domestication, 185 polyploidy in, 218 advantages of, 215 aneuploids, 220–222 through interspecific hybridization, 201– 202, 217–220 triploids, 222–224 unreduced microspores and, 215 –220, 223–224 protein content, 194, 196, 225, 226t relationship to Manihot species, 190 sectorial chimera in, 216 Catchment/watershed hydrological pathways, 163, 164, 166, 168 –172 Charge-transfer interactions, 6 Chelation by organic acids, 127–130 Chemisorption, 5 Chimeras in cassava, 216 Citric acid excreted by phosphate-solubilizing fungi, 126 Clays hindered diffusion and, 20 Coal sorption by, 13 Condensation defined, 4 Continuous-flow stirred tank reactor, 54 – 56 Coordination complexes chemisorption and, 5

Crop diversity as environmental indicator, 81– 82 Crop yield as environmental indicator, 78 soil organic matter and, 86

D Darcian flow, 160t, 165 Darken equation, 31– 32 Delphi technique, 90 Denbigh soil hydrological characteristics, 155, 156t Desorption, see also Sorption defined, 3 nonequilibrium mechanisms in, 17–27 uptake and release profiles, 16 –17 rate of, 14 zero-length columns, 56 Diffusion defined, 17–18 intraorganic matter diffusion, 20–23 soil pore diffusion, 18 –20 Diffusion equations, 31– 32 Diffusion models combined organic matter/pore diffusion model, 39 – 41 coupled with biodegradation kinetics, 62– 63 for fixed-pore systems, 32– 37 general considerations in, 30 – 32 for multiple particle sizes, 41– 43 for organic matter, 37– 39 for soil columns, 52– 54 Drought tolerance cassava interspecific hybrids and, 204–207 Dual resistance diffusion model, 36 – 37

E Earthworms as environmental indicators, 80 Electrodynamic thermogravimetric analyzer, 49 Environmental indicators agricultural chemical use, 79 biological, 79 crop diversity, 81– 82 crop productivity, 78 earthworms, 80 fecal pathogens, 82

233

INDEX genetic diversity, 79 honeybees, 80–81 insects, 78, 81 pesticide resistance, 81 soil microbial status, 80 landscape, 83, 84–85 physical, 78, 82–83 ranking of, 90, 91t recommendations for using, 92 soil chemicals, 83–84 soil organic matter, 85 – 90 Environmental monitoring assessment endpoints, 77–78 federal agencies in, 77 pressure-state-response framework, 78 Environmental Protection Agency ranking of landscape metrics, 85 Equilibrium partition bioavailability model, 58 Erosion effects on soil organic matter, 87– 88

F Fava-Eyring model, 30 Fecal pathogens as environmental indicators, 82 Ficks’s laws, 31–32 Field capacity, 157 Film resistance sorption model, 30 First-order solute transport model with biodegradation term, 61 with sorption kinetic term, 52 Freundlich equation, 34 Fulvic acids, 8 Fungi vesicular mycorrhizal, 100, 141–142 Fungi, phosphate-solubilizing, 100, 144 agar studies, 107 liquid medium studies chelation of cations by organic acids, 127– 130 nitrogen source effects, 132 pH effects, 109, 124, 130–131 production of organic acids, 126 –127 results from, 110–123t solubility of phosphate compounds, 109, 124–126 time-related soluble phosphate fluctuations, 132–133 titratable acidity and, 130

promotion of plant growth by, 133, 134 –140t, 141–143

G Genetic diversity as environmental indicator, 79 Gluconate excreted by phosphate-solubilizing fungi, 126 –127 Graciles, 186 Groundwater phosphorus transfer and, 172

H Hallsworth soil hydrological characteristics, 155, 156t Herbicides environmental health and, 79 Honeybees as environmental indicators, 80 – 81 HOST classification, see Hydrology of soil types classification Humic acids, 8 Humic substances, 8 Humin, 8 Hydrocyanic acid in Manihot tubers, 194, 195 Hydrogen bonding in organic molecules, 5 Hydrological pathways phosphorus transfer and overview of, 154 –155 scale effects, 162–164 at the slope/field scale, 166, 168 –172 at the soil profile scale, 165 –166 temporal aspects of, 158, 159f terminology and, 164 timescales in, 158 –161 types of, 160 –161t variable source areas, 171 Hydrology of soil types (HOST) classification, 155, 166, 167f Hydrophobic effect, 13 –14 Hydroxylated mineral surfaces, 6 Hysteresis isotherm, 24 –26 kinetic, 26 –27 overview of, 24

234

INDEX I

Infiltration-excess overland flow, 170 –171 Insects as environmental indicators, 78 Intraorganic matter diffusion overview of, 20–22 structure activity relationships in, 22–23 Ion-exchange forces, 6 Iron in soil phosphorus dynamics, 102, 103 Isotherm hysteresis, 24 –26

K Kerogen sorption by, 13 -Ketogluconic acid, 130 Kinetic hysteresis, 26–27

L Land drainage phosphorus transfer and, 172 Landscape in environmental assessments, 83, 84 – 85 National Resource Inventory and, 83 Langmuir kinetic model, 27–28 Leaching described, 160t use of term, 164 Linear driving force sorption models, 29 – 30 with biodegradation term, 61– 62

M Macropore flow, 160t, 165 Macropores modeling diffusion in, 33 – 35 size of, 18 Manihot, see also Cassava adaptation to climatic conditions, 197–198 centers of diversity, 184, 185 –186 chromosome number, 186, 187t genetic variability, 190, 193 –198 growth habit, 193t, 196 hybridization with cassava, 184 –185, 187 aneuploidy in, 221–222 apomixis and, 211–212 characterization of hybrids, 200 –204

drought tolerance and, 204 –207 overcoming crossing barriers, 207–210 polyploidy and, 217–220 production of hybrids, 198 –200 protein content, 225 –226 unreduced microspores in, 217–220, 223 – 224 interspecific hybridization in, 186 –187, 196 natural habitats, 193t, 196 relationships between species, 186 –190, 191t, 192t species in Brazil, 181t, 182–184 taxonomy of, 180 –181 tuber formation patterns, 193 –194 tuber hydrocyanic acid content, 194, 195 tuber protein content, 194 –195, 196, 225 – 226, 227t M. anomala, 198 –200 M. brachyandra, 225, 226 M. caerulescens, 197–198 M. dichotoma, 211 M. falcata, 196 M. glaziovii, 181, 200, 204, 218, 220 M. oligantha, 225 M. oligantha subsp. nestili, 194, 195 M. paviaefolia, 196 M. pohlii, 207–210 M. procumbens, 198 M. pruinosa, 196 M. pseudoglaziovii, 201, 204, 220, 221, 222– 223 M. reptans, 186 –187, 196 M. saxicola, 195 M. stipularis, 198 Manihot esculenta, see Cassava Manihot neusana as bridge species in hybridization, 208, 209 – 210 hybridization with cassava, 198 –200, 220 cytogenetic behavior of backcrossed generation, 202–203 cytogenetic behavior of parents, 203 evolutionary and breeding significance of, 203 –204 meiotic behavior of F1 hybrids, 200–202 Manihotoides pouciflora, 186 Meiotic restitution in cassava interspecific hybrids, 201, 202 Mentor pollen, 208 Mercuric chloride, 47

INDEX Mesopores modeling diffusion in, 33 – 35 size of, 18 Meteorological field capacity, 157 Mexico wild species of Manihot in, 183 Micropores modeling diffusion in, 35 size of, 18 Microspores, unreduced in cassava, 216–217 interspecific hybrids, 217–220, 223 –224 overview of, 215–216 Mineral surfaces sorption and, 6–7 types of, 6

N National Resource Inventory, 83 Nitrate effects on phosphate solubilization by fungi, 132 Nitrogen effects on phosphate solubilization by fungi, 132 Nonaqueous phase liquids sorption by, 13 Nonlinear driving force sorption models, 29 – 30

O Organic acids excreted by phosphate-solubilizing fungi, 126–127 chelation of phosphate metal cations by, 127–130 excreted by roots, 106 stability constants for, 128t Organoclays sorption in, 20 Overland flow described, 160t, 164 phosphorus transfer and, 170 –171 Oxalic acid excreted by phosphate-solubilizing fungi, 126

P Partial-area runoff, 170 Penicillium, see Fungi, phosphate-solubilizing

235

Pesticide resistance as environmental indicator, 81 Pesticides environmental health and, 79 pH, see also Soil pH effects on phosphate solubility in liquid media, 109, 124, 130 –131 Phenathrene, 17 Phosphate, see also Fungi, phosphate-solubilizing; Phosphorus transfer; Soil phosphorus in liquid medium studies chelation of metal ions by organic acids, 127–130 pH effects and, 109, 124, 130–131 solubility of, factors affecting, 109, 124 – 126 time-based fluctuations in, 132–133 titratable acidity and, 130 precipitation in soils, 102–103 sorption in soils, 101–102 Phosphate fertilizers soil accumulation of phosphate, 100, 103, 154 Phosphorus transfer effective rainfall and, 155, 157, 158 hydrological pathways scale effects, 162–164 at the slope/field scale, 166, 168 –172 at the soil profile scale, 165 –166 temporal aspects of, 158, 159f levels of hydrological activity and, 157–158 overview of, 154 –155, 173 Physisorption on mineral surfaces, 7 overview of, 5 – 6 rates of, 14 –15 Piston flow, 161t, 165 Plants uptake of soil phosphate, 105 –106 Pollen mentor, 208 Pollutant transfer, 154, see also Phosphorus transfer Polycyclic aromatic hydrocarbons sorption by soot, 13 Polyploidy, in cassava, 218 advantages of, 215 aneuploids, 220 –222 through interspecific hybridization, 201–202, 217–220 triploids, 222–224

236

INDEX

Polyploidy (continued) unreduced microspores and, 215 –220, 223 – 224 Precipitation phosphorus transfer and, 155, 157, 158 Preferential flow, 161t, 165 –166 Pressure-state-response framework, 78 Probability density functions, 43 – 44 Propylene oxide, 47

R Radial diffusion laws coupled with biodegradation kinetics, 62– 63 Rainfall phosphorus transfer and, 155, 157, 158 Return flow, 161t, 171 Rhizosphere solubilization of phosphates in, 105 –106 Roots of drought tolerant cassava hybrids, 205 –207 Runoff described, 161t phosphorus transfer and, 169 –170 use of term, 164

S Saturated flow, 161t, 165 Saturation-excess overland flow, 171 Self-diffusion, 31–32 Shale sorption by, 13 Siloxane mineral surfaces, 6 Slope/field hydrological pathways, 162, 164, 166, 168–172 Sodium azide, 47 Soil accumulation of phosphate in, 100, 103, 154 bioavailability of chemicals in, 56 – 58 soil organic matter in factors controlling content of, 87– 88 functions of, 86 sorption in heterogeneous soils, 23 –24 mineral surfaces, 6 –7 other carbonaceous material, 13 soil organic matter, 8 –12 types of hydrological pathways through, 158 – 161

typing by hydrological characteristics, 155, 156t Soil aggregates, 33 Soil chemicals bioavailability of, 83 as environmental indicators, 83 – 84 Soil columns advection-dispersion equation, 50 overview of, 49 – 51 transport models with sorption kinetic terms, 51– 54 zero-length, 56 Soil degradation as environmental indicator, 82 Soil microorganisms as environmental indicators, 80 phosphate solubilizing, 106 –108 (see also Fungi, phosphate–solubilizing) soil phosphorus dynamics and, 100 Soil nutrients as environmental indicators, 83 – 84 Soil organic matter atmospheric carbon and, 86 – 87 components of, 8 defined, 85 – 86 diffusion models for, 37– 41 as environmental indicator, 85 – 90 functions of, 86 hindered diffusion in, 20 –22, 23 hysteresis and, 25 –26 measures of, 88 models of sorption in, 8 –12 quantities of, measuring and expressing, 89 – 90 rubber-glassy polymer concept of, 9 –12 in soil aggregates, 33 soil content of, factors controlling, 87– 88 sorption rates, 15 Soil particles description of, 32– 33 diffusion models and for multiple particle sizes, 41– 43 for particles with fixed pore sizes, 32– 37 Soil pH phosphate solubility and, 104 –105 Soil phosphorus, see also Fungi, phosphate-solubilizing cycle, 103 –104 deficiency in, 104 overview of, 100

237

INDEX phosphate sorption, 101–102 plant uptake of, 105–106 precipitation of phosphate compounds, 102–103 soil pH and, 104–105 Soil pores condensation and, 4 diffusion models for biporous particles, 36 – 37 combined pore diffusion/organic matter model, 39–41 for macropores, 33– 35 for mesopores, 33– 35 for micropores, 35 diffusion through, 18–20 size classes, 18 water concentration and bioavailability relationship, 58 Soil productivity as environmental indicator, 78 Soil profile hydrological pathways, 165 –166 Soil solution phosphorus dynamics and, 103 –104 Solute transport models advection-dispersion equation, 50 with biodegradation term, 63 – 65 with sorption kinetic term diffusion model, 52– 54 first-order model, 52 two-region model, 51– 52 Soot sorption by, 13 Sorbate, 2 Sorbent, 2 Sorption, see also Sorption, nonequilibrium; Sorption kinetics categories of, 3–5 defined, 3 experimental methods batch techniques, 45 – 49 column techniques, 49 – 54 stirred-flow cell techniques, 54 – 56 time frames, 45 zero-length columns, 56 factors affecting, 2 on mineral surfaces, 6 –7 of phosphate in soil, 101–102 pH and, 104–105 steps in, 3 thermodynamics of, 13 –14 types of molecular interactions in, 5 – 6

Sorption, nonequilibrium retardation mechanisms hysteresis, 24 –27 intraorganic matter diffusion, 20–23 overview of, 17–18 pore diffusion, 18–20 soil heterogeneity and, 23 –24 uptake and release profiles, 16 –17 Sorption kinetics bioavailability of soil chemicals and, 56 – 58 models based on bond energetics, 27–29 coupled sorption-degradation models, 58 – 65 diffusion models, 30 – 43, 62– 63 linear driving force models, 29 – 30, 61– 62 solute transport models, 50 – 54, 63 – 65 stochastic models, 43 – 45 rates of elementary processes in, 14 –15 significance of, 2 Static batch reactor, 49 Stirred-flow cells, 54 – 56 Stochastic sorption models, 43 – 45 Storm flow phosphorus transfer and, 157, 158 Storm runoff, 171 Subsurface flow, 161t, 164 Successive-dilution technique, 48 Surface runoff described, 161t phosphorus transfer and, 170 –171

T Throughflow described, 161t phosphorus transfer and, 171–172 Tillage effects on soil organic matter, 87, 88 Titratable acidity, 130 Tortuosity in diffusion through pores, 18 Transgenic crops pest resistance and, 81 Transport diffusion, 31– 32 Tripartitae, 186 Triploidy in cassava, 222–224

238

INDEX W

Tupi-Guarani Indians, 185 Two-region solute transport model, 51– 52

V Van Der Waals forces in physisorption, 5 Vesicular mycorrhizal fungi, 100, 141–142

Water films sorption and, 5

Z Zero-length columns, 56