Advances in Agronomy, Volume 110

V O LU M E O N E ADVANCES H U N D R E D IN AGRONOMY T E N ADVANCES IN AGRONOMY Advisory Board PAUL M. BERTSCH ...

Author: Donald L. Sparks

82 downloads 1370 Views 3MB Size Report

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

Report copyright / DMCA form

DOWNLOAD PDF

V O LU M E

O N E

ADVANCES

H U N D R E D

IN

AGRONOMY

T E N

ADVANCES IN AGRONOMY Advisory Board

PAUL M. BERTSCH

RONALD L. PHILLIPS

University of Kentucky

University of Minnesota

KATE M. SCOW

LARRY P. WILDING

University of California, Davis

Texas A&M University

Emeritus Advisory Board Members

JOHN S. BOYER

KENNETH J. FREY

University of Delaware

Iowa State University

EUGENE J. KAMPRATH

MARTIN ALEXANDER

North Carolina State University

Cornell University

Prepared in cooperation with the American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America Book and Multimedia Publishing Committee DAVID D. BALTENSPERGER, CHAIR LISA K. AL-AMOODI

CRAIG A. ROBERTS

WARREN A. DICK

MARY C. SAVIN

HARI B. KRISHNAN

APRIL L. ULERY

SALLY D. LOGSDON

V O LU M E

O N E

ADVANCES

H U N D R E D

T E N

IN

AGRONOMY EDITED BY

DONALD L. SPARKS Department of Plant and Soil Sciences University of Delaware Newark, Delaware, USA

AMSTERDAM • BOSTON • HEIDELBERG • LONDON NEW YORK • OXFORD • PARIS • SAN DIEGO SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO Academic Press is an imprint of Elsevier

Academic Press is an imprint of Elsevier 525 B Street, Suite 1900, San Diego, CA 92101-4495, USA 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA 32 Jamestown Road, London, NW1 7BY, UK Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands First edition 2011 Copyright # 2011 Elsevier Inc. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email: [email protected]. Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made ISBN: 978-0-12-385531-2 ISSN: 0065-2113 (series) For information on all Academic Press publications visit our website at elsevierdirect.com Printed and bound in USA 11 12 13 10 9 8 7 6 5 4 3 2 1

CONTENTS

Contributors Preface

ix xi

1. Dissolved Organic Matter: Biogeochemistry, Dynamics, and Environmental Significance in Soils

1

Nanthi S. Bolan, Domy C. Adriano, Anitha Kunhikrishnan, Trevor James, Richard McDowell, and Nicola Senesi 1. Introduction 2. Sources, Pools, and Fluxes of Dissolved Organic Matter in Soils 3. Properties and Chemical Composition of Dissolved Organic Matter in Soils 4. Mechanisms Regulating Dynamics of Dissolved Organic Matter in Soils 5. Factors Influencing Dynamics of Dissolved Organic Matter in Soils 6. Environmental Significance of Dissolved Organic Matter in Soils 7. Summary and Research Needs Acknowledgments References

3 5

20 30 37 60 62 62

2. Genomic Selection in Plant Breeding: Knowledge and Prospects

77

13

Aaron J. Lorenz, Shiaoman Chao, Franco G. Asoro, Elliot L. Heffner, Takeshi Hayashi, Hiroyoshi Iwata, Kevin P. Smith, Mark E. Sorrells, and Jean-Luc Jannink 1. Introduction 2. Important Population and Trait Characteristics 3. Single Nucleotide Polymorphism Marker Discovery and Genotyping 4. Statistical Methods 5. GS Prediction Accuracies 6. Impact of Statistical Model on GEBV Accuracy 7. Modeling Epistasis and Dominance 8. GS in the Presence of Strong Subpopulation Structure 9. Long-Term Selection 10. Summary and Conclusions References

78 80 82 84 94 103 107 109 111 114 116 v

vi

Contents

3. Differences of Some Leguminous and Nonleguminous Crops in Utilization of Soil Phosphorus and Responses to Phosphate Fertilizers

125

Sheng-Xiu Li, Zhao-Hui Wang, and Bobby Alton Stewart 1. Introduction 2. The Difference of P Uptake Amounts of Leguminous and Nonleguminous Crops 3. Leguminous and Nonleguminous Crop Responses to Powdered Rock Phosphates 4. The Relation of Plants’ Root Morphology to Their Capacity of Using Soil Sparingly Soluble P and Responses to P Fertilizers 5. Microorganisms in Rhizosphere Soil and Their Function in Supplying P to Plants 6. Root Exudates (Substances Secreted from Roots) and the Plants’ Capacity to Use Sparingly Soluble P in the Soil and Crop Responses to P Fertilizers 7. Effects of Root Cation Exchange Capacity and Calcium Uptake Amount of Crops on Soil P Absorption and Crop Responses to P Fertilizer 8. The Responses to P Fertilizer Between Leguminous and Cereal Crops with Their Biological Characteristics 9. Factors Affecting the Responses of Leguminous and Nonleguminous Crops to P Fertilizer 10. Conclusions Acknowledgments References

130

203 222 227 227

4. The Role of Mineral Nutrition on Root Growth of Crop Plants

251

141 146 163 173

176

189 193

N. K. Fageria and A. Moreira 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.

Introduction Root-Induced Changes in the Rhizosphere Root Systems of Cereals and Legumes Contribution of Root Systems to Total Plant Weight Rooting Depth and Root Distribution Root Growth as a Function of Plant Age Root–Shoot Ratio Root Growth Versus Crop Yield Genotypic Variation in Root Growth Root Oxidation Activity in Oxygen-Deficient Soils Root Growth in Conservation Tillage Systems Mineral Nutrition Versus Root Growth Management Strategies for Maximizing Root Systems

252 255 256 260 263 265 268 270 271 274 276 278 312

Contents

14. Conclusions Acknowledgment References

5. Physiology of Spikelet Development on the Rice Panicle: Is Manipulation of Apical Dominance Crucial for Grain Yield Improvement?

vii

317 318 318

333

Pravat K. Mohapatra, Rashmi Panigrahi, and Neil C. Turner 1. 2. 3. 4. 5.

Introduction Panicle Structure and Development Panicle Architecture and Grain Yield Physiological Factors Regulating Spikelet Development Is Manipulation of Apical Dominance Crucial for Grain Yield Improvement? 6. Suggestions for Modification of Apical Dominance Acknowledgments References

Index

334 335 337 340 348 351 352 352 361

This page intentionally left blank

CONTRIBUTORS

Domy C. Adriano (1) University of Georgia, Savannah River Ecology Laboratory, Drawer E, Aiken, South Carolina, USA Franco G. Asoro (77) Department of Agronomy, Iowa State University, Ames, Iowa, USA Nanthi S. Bolan (1) Centre for Environmental Risk Assessment and Remediation (CERAR), and Cooperative Research Centre for Contaminants Assessment and Remediation of the Environment (CRC CARE), University of South Australia, Australia Shiaoman Chao (77) Biosciences Research Laboratory, USDA-ARS, Fargo, North Dakota, USA N. K. Fageria (251) Rice and Bean Research Center of Embrapa, Santo Antoˆnio de Goia´s, GO, Brazil Takeshi Hayashi (77) Data Mining and Grid Research Team, National Agricultural Research Center, Tsukuba, Ibaraki, Japan Elliot L. Heffner (77) Department of Plant Breeding and Genetics, Cornell University, Ithaca, New York, USA Hiroyoshi Iwata (77) Department of Agricultural and Environmental Biology, Graduate School of Agriculture & Life Sciences, University of Tokyo, Bunkyo, Tokyo, Japan Trevor James (1) AgResearch, Ruakura Research Centre, Hamilton, New Zealand Jean-Luc Jannink (77) R.W. Holley Center for Agriculture and Health, USDA-ARS, Ithaca, New York, USA Anitha Kunhikrishnan (1) Centre for Environmental Risk Assessment and Remediation (CERAR), and Cooperative Research Centre for Contaminants Assessment and Remediation of the Environment (CRC CARE), University of South Australia, Australia

ix

x

Contributors

Sheng-Xiu Li (125) College of Resources and Environmental Sciences, Northwest Science and Technology University of Agriculture and Forestry, Yangling, Shaanxi, PR China Aaron J. Lorenz (77) R.W. Holley Center for Agriculture and Health, USDA-ARS, Ithaca, New York, USA Richard McDowell (1) AgResearch, Invermay Agricultural Centre, Mosgiel, New Zealand Pravat K. Mohapatra (333) School of Life Science, Sambalpur University, Sambalpur, India A. Moreira (251) Western Amazon Research Center of Embrapa, Manaus, AM, Brazil Rashmi Panigrahi (333) School of Life Science, Sambalpur University, Sambalpur, India Nicola Senesi (1) Department of Agroforestal and Environmental Biology and Chemistry, University of Bari, Bari, Italy Kevin P. Smith (77) Department of Agronomy and Plant Genetics, University of Minnesota, St. Paul, Minnesota, USA Mark E. Sorrells (77) Department of Plant Breeding and Genetics, Cornell University, Ithaca, New York, USA Bobby Alton Stewart (125) Dryland Agriculture Institute, West Texas A&M University, Canyon, Texas, USA Neil C. Turner (333) Centre for Legumes in Mediterranean Agriculture and UWA Institute of Agriculture, The University of Western Australia, Crawley, WA, Australia Zhao-Hui Wang (125) College of Resources and Environmental Sciences, Northwest Science and Technology University of Agriculture and Forestry, Yangling, Shaanxi, PR China

PREFACE

Volume 110 contains five excellent reviews dealing with crop and soil sciences. Chapter 1 is a detailed review on the biogeochemistry, dynamics, and environmental significance of dissolved organic matter in soils. Chapter 2 deals with prospects and current efforts in using genomic selection in plant breeding. Chapter 3 is a comprehensive overview on the use of phosphorus and response to phosphate fertilizers by leguminous and nonleguminous crops. Chapter 4 deals with the role of mineral nutrition on root growth of crop plants. Chapter 5 covers the physiology of spikelet development on the rice panicle and the role that apical dominance plays in grain yield improvement. I am grateful to the authors for their first-rate contributions. DONALD L. SPARKS Newark, Delaware, USA

xi

C H A P T E R

O N E

Dissolved Organic Matter: Biogeochemistry, Dynamics, and Environmental Significance in Soils Nanthi S. Bolan,*,† Domy C. Adriano,‡ Anitha Kunhikrishnan,*,† Trevor James,§ Richard McDowell,} and Nicola Senesi# Contents 1. Introduction 2. Sources, Pools, and Fluxes of Dissolved Organic Matter in Soils 3. Properties and Chemical Composition of Dissolved Organic Matter in Soils 3.1. Structural components 3.2. Fulvic acid—The dominant component 3.3. Elemental composition 4. Mechanisms Regulating Dynamics of Dissolved Organic Matter in Soils 4.1. Sorption/complexation 4.2. Biodegradation 4.3. Photodegradation 4.4. Leaching 5. Factors Influencing Dynamics of Dissolved Organic Matter in Soils 5.1. Vegetation and land use 5.2. Cultivation 5.3. Soil amendments 5.4. Soil pH 6. Environmental Significance of Dissolved Organic Matter in Soils 6.1. Soil aggregation and erosion control 6.2. Mobilization and export of nutrients 6.3. Bioavailability and ecotoxicology of heavy metals

3 5 13 13 15 20 20 23 27 28 29 30 31 32 33 36 37 37 38 43

* Centre for Environmental Risk Assessment and Remediation (CERAR), University of South Australia, Australia { Cooperative Research Centre for Contaminants Assessment and Remediation of the Environment (CRC CARE), University of South Australia, Australia { University of Georgia, Savannah River Ecology Laboratory, Drawer E, Aiken, South Carolina, USA } AgResearch, Ruakura Research Centre, Hamilton, New Zealand } AgResearch, Invermay Agricultural Centre, Mosgiel, New Zealand # Department of Agroforestal and Environmental Biology and Chemistry, University of Bari, Bari, Italy Advances in Agronomy, Volume 110 ISSN 0065-2113, DOI: 10.1016/B978-0-12-385531-2.00001-3

#

2011 Elsevier Inc. All rights reserved.

1

2

Nanthi S. Bolan et al.

6.4. Transformation and transport of organic contaminants 6.5. Gaseous emission and atmospheric pollution 7. Summary and Research Needs 7.1. Macroscale (landscape to global) 7.2. Microscale (water bodies and soil profile) 7.3. Molecular scale (carbon fractions, organic acids, and microorganisms) Acknowledgments References

50 58 60 61 61 61 62 62

“Dissolved organic matter comprises only a small part of soil organic matter; nevertheless, it affects many processes in soil and water including the most serious environmental problems like soil and water pollution and global warming.” (Kalbitz and Kaiser, 2003)

Abstract Dissolved organic matter (DOM) is defined as the organic matter fraction in solution that passes through a 0.45 mm filter. Although DOM is ubiquitous in terrestrial and aquatic ecosystems, it represents only a small proportion of the total organic matter in soil. However, DOM, being the most mobile and actively cycling organic matter fraction, influences a spectrum of biogeochemical processes in the aquatic and terrestrial environments. Biological fixation of atmospheric CO2 during photosynthesis by higher plants is the primary driver of global carbon cycle. A major portion of the carbon in organic matter in the aquatic environment is derived from the transport of carbon produced in the terrestrial environment. However, much of the terrestrially produced DOM is consumed by microbes, photo degraded, or adsorbed in soils and sediments as it passes to the ocean. The majority of DOM in terrestrial and aquatic environments is ultimately returned to atmosphere as CO2 through microbial respiration, thereby renewing the atmospheric CO2 reserve for photosynthesis. Dissolved organic matter plays a significant role in influencing the dynamics and interactions of nutrients and contaminants in soils and microbial functions, thereby serving as a sensitive indicator of shifts in ecological processes. This chapter aims to highlight knowledge on the production of DOM in soils under different management regimes, identify its sources and sinks, and integrate its dynamics with various soil processes. Understanding the significance of DOM in soil processes can enhance development of strategies to mitigate DOM-induced environmental impacts. This review encourages greater interactions between terrestrial and aquatic biogeochemists and ecologists, which is essential for unraveling the fundamental biogeochemical processes involved in the synthesis of DOM in terrestrial ecosystem, its subsequent transport to aquatic ecosystem, and its role in environmental sustainability, buffering of nutrients and pollutants (metal(loid)s and organics), and the net effect on the global carbon cycle.

Dissolved Organic Matter

3

1. Introduction The total organic matter (TOM) in terrestrial and aquatic environments consists of two operationally defined phases: particulate organic matter (POM) and dissolved organic matter (DOM). For all practical purposes, DOM is defined as the organic matter fraction in solution that passes through a 0.45 mm filter (Thurman, 1985; Zsolnay, 2003). Some workers have used finer filter paper (i.e., 0.2 mm) in an effort to separate “true” DOM from colloidal materials, but 0.45 mm filtration appears to be standard (Buffle et al., 1982; Dafner and Wangersky, 2002). In some literature, the term dissolved organic carbon (DOC) is used, which represents total organic carbon in solution that passes through a 0.45 mm filter (Zsolnay, 2003). Since carbon represents the bulk of the elemental composition of the organic matter (ca. 67%), DOM is often quantified by its carbon content and referred to as DOC. In the case of studies involving soils, the term water-soluble organic matter (WSOM) or water-extractable organic matter (WEOM) is also used when measuring the fraction of the soil organic matter (SOM) extracted with water or dilute salt solution (e.g., 0.5 M K2SO4) that passes through a 0.45 mm filter (Bolan et al., 1996; Herbert et al., 1993). Recently, the distinction between POM and DOM in the marine environment is being replaced by the idea of an organic matter continuum of gel-like polymers, replete with colloids and crisscrossed by “transparent” polymer strings, sheets, and bundles, from a few to hundreds of micrometers—referred to as oceanic “dark matter” (Dafner and Wangersky, 2002). Dissolved organic matter is ubiquitous in terrestrial and aquatic ecosystems, but represents only a small proportion of the total organic matter in soil (McGill et al., 1986). However, it is now widely recognized that because DOM is the most mobile and actively cycling organic matter fraction, it influences a myriad of biogeochemical processes in aquatic and terrestrial environments as well as key environmental parameters (Chantigny, 2003; Kalbitz et al., 2000; McDowell, 2003; Stevenson, 1994; Zsolnay, 2003). Dissolved organic carbon has been identified as one of the major components responsible for determining the drinking water quality. For example, DOM leads to the formation of toxic disinfection byproducts (DBPs), such as trihalomethanes, after reacting with disinfectants (e.g., chlorine) during water treatment. Similarly, DOM can be related to bacterial proliferation within the drinking water distribution system. Therefore, the control of DOM has been identified as an important part of the operation of drinking water plants and distribution systems (Volk et al., 2002). In aquatic environments, the easily oxidizable compounds in the DOM can act as chemical and biological oxygen demand compounds, thereby depleting the oxygen concentration of aquifers and influencing

4

Nanthi S. Bolan et al.

aquatic biota ( Jones, 1992). Dissolved organic carbon can act as a readily available carbon source for anaerobic soil organisms, thereby inducing the reduction of nitrate (denitrification) resulting in the release of green house gases, such as nitrous oxide (N2O) and nitric oxide (NO), which are implicated in ozone depletion (Siemens et al., 2003). Organic pesticides added to soil and aquifers are partitioned preferentially onto DOM, which can act as a vehicle for the movement of pesticide residues to groundwater (Barriuso et al., 1992). Similarly, the organic acids present in the DOM can act as chelating agents, thereby enhancing the mobilization of toxic heavy metals and metalloids [metal(loid)s] (Antoniadis and Alloway, 2002). The release and retention of DOM are the driving forces controlling a number of pedological processes including podzolization (Hedges, 1987). Biological fixation of atmospheric CO2 by higher plants during photosynthesis is the primary driver of global carbon cycle. A major portion of the carbon in aquatic environments is derived from the transport of carbon produced on land. It has been estimated that worldwide about 210 Mt DOM and 170 Mt POM are transported annually to oceans from land. Carbon in the ocean is recognized as one of the three main reservoirs of organic material on the planet, equal to the carbon stored in terrestrial plants or soil humus (Hedges, 1987). The terrestrially produced DOM is subject to microbial- and photodegradation and adsorption by soil and sediments. The majority of DOM in terrestrial and aquatic environments is returned to the atmosphere as CO2 through microbial respiration, thereby ultimately replenishing the atmospheric CO2 reserve for photosynthesis and reinvigorating the global carbon cycle. Dissolved organic carbon can be envisioned both as a link and bottleneck among various ecological compartments. Combined with its dynamic nature, this enables DOM to serve as a sensitive indicator of shifts in ecological processes, especially in aquatic systems. Recently, the significance of DOM in the terrestrial environment has been realized and attempts have been made to extend this knowledge to DOM dynamics in aquatic environments. However, DOM dynamics on land are fundamentally different from those in water, where biomass of primary producers is relatively small, allochthonous sources of DOM are dominant, the surface area of reactive solid particles (i.e., sediments) is smaller, and the fate of DOM is strongly influenced by photolysis and other light-mediated reactions. In contrast, the dynamics of DOM on land are largely controlled by its interactions with abiotically and biotically reactive solid components. Although there have been a number of reviews on the individual components of DOM in soils (e.g., sources and sink—Kalbitz et al. (2000); microbial degradation—Marschner and Kalbitz (2003); sorption by soils—Kaiser et al. (1996)), there has been no comprehensive review linking the dynamics of DOM to its environmental significance. This chapter aims to elaborate on the production and degradation of DOM in

Dissolved Organic Matter

5

soils under different landscape conditions, identify its sources and sinks, and integrate its dynamics with environmental impacts. Understanding the long-term control on DOM production and flux in soils will be particularly important in predicting the effects of various environmental changes and management practices on soil carbon dynamics. Improved knowledge on the environmental significance of DOM can enhance the development of strategies to mitigate DOM-induced environmental impacts. It is hoped that this chapter will encourage greater interaction between terrestrial and aquatic biogeochemists and ecologists and stimulate the unraveling of fundamental biogeochemical processes involved in the synthesis and transport of DOM in terrestrial and aquatic ecosystems.

2. Sources, Pools, and Fluxes of Dissolved Organic Matter in Soils Nearly all DOM in soils comes from photosynthesis. This represents the various C pools including recent photosynthates, such as leaf litter, throughfall and stemflow (in the case of forest ecosystems), root exudates, and decaying fine roots, as well as decomposition and metabolic by-products and leachates of older, microbiologically processed SOM (Figure 1) (Guggenberger, et al., 1994a; McDowell, 2003; McDowell, et al., 1998). The majority of DOM in soils and aquifers originates from the solubilization of SOM accumulated through vegetation and the addition of biological waste materials (Guggenberger, et al., 1994b; McDowell, 2003; McDowell, et al., 1998; Tate and Meyer, 1983). The addition of biological waste materials, such as poultry and animal manures and sewage sludges, increases the amount of DOM in soils either by acting as a source of DOM or by enhancing the solubilization of the SOM. Most biological waste materials of plant origin contain large amounts of DOM (Table 1) and the addition of certain organic manures such as poultry manure increases the pH and thereby enhances the solubilization of SOM (Schindler et al., 1992). The concentrations of DOM in soils and aquifers are highly susceptible to changes induced by humans, such as cultivation, fire, clear-cutting, wetland drainage, acidic precipitation, eutrophication, and climate change (Kreutzweiser et al., 2008; Laudon et al., 2009; Martinez-Mena et al., 2008; Mattsson et al., 2009; Yallop and Clutterbuck, 2009). Dissolved organic matter in environmental samples, such as soils and manures, is often extracted with water or dilute aqueous salt solutions. Various methods have been used to measure the concentration of DOM in extracts (Table 2). These methods are grouped into three categories (Moore, 1985; Sharp et al., 2004; Stewart and Wetzel, 1981; Tue-Ngeun et al., 2005). The most frequently used method involves the measurement of

6

Nanthi S. Bolan et al.

Photosynthesis

Photosynthesis

CO2

CO2 1

7

7

6

6 2

3

4

5

Litter layer

5

Crop residue

A horizon

DOM

4

3

2

8

8

DOM

9

9

B horizon 10

10

C horizon

11

11

Parent/geologic material

Aquifer

Forest soil

Agricultural soil

Figure 1 Pathways of inputs and outputs of dissolved organic matter (DOM) in forest and agricultural soils. Inputs: 1, throughfall and stemflow; 2, root exudates; 3, microbial lysis; 4, humification; 5, litter/and crop residue decomposition; 6, organic amendments; outputs; 7, microbial degradation; 8, microbial assimilation; 9, lateral flow; 10, sorption; 11, leaching.

absorption of light by the DOM using a spectrophotometer (Stewart and Wetzel, 1981). The second method involves wet oxidation of samples containing DOM and the subsequent measurement of the CO2 released or the amount of oxidant consumed (Ciavatta et al., 1991). This method is often referred to as chemical oxygen demand (COD). Dichromates or permanganates are the most common oxidizing agents used in the wet oxidation of DOM, and the amount of oxidant consumed in the oxidation of DOM is measured either by titration with a reducing agent or by calorimetric methods. The third method involves dry oxidation of DOM to CO2 at high temperature in the presence of a stream of oxygen. The amount of CO2 produced is measured either by infrared (IR) detector or by titration after absorbing in an alkali, or by weight gain after absorbing in ascarite (Bremner and Tabatabai, 1971). The most commonly used dry combustion techniques include LECOTM combustion and total organic carbon (TOC) analyzer.

7

Dissolved Organic Matter

Table 1

Sources of dissolved organic matter input to soils Dissolved organic matter

Sources

Pasture leys Brome grass Clover Crowtoe Lucerne Cv. Longdong Lucerne Cv. Saditi Sainfoin Sweet pea Soil Forest soil—litter leachate Arable soil Soil under bermuda grass turf Pasture soil Pasture soil Organic amendments Sewage sludge Sewage sludge Paper sludge Poultry manure Poultry litter a Mushroom compost Fresh spent mushroom substrate Composted spent mushroom substrate Separated cow manure Poultry manure Pig manure a

Total organic matter (g C kg 1)

(g C kg 1)

(% of total organic matter)

Reference

13.3 15.1 10.4 11.4 10.9 13.8 10.2

0.041 0.039 0.036 0.038 0.036 0.040 0.034

0.31 0.26 0.35 0.32 0.33 0.29 0.33

Shen et al. (2008) Shen et al. (2008) Shen et al. (2008) Shen et al. (2008) Shen et al. (2008) Shen et al. (2008) Shen et al. (2008)

60.0

0.026

0.04

12.0 8.10

0.150 0.300

1.25 3.70

Jaffrain et al. (2007) Gonet et al. (2008) Provin et al. (2008)

32.0 82.5

1.02 3.12

3.18 3.80

Bolan et al. (1996) Bolan et al. (1996)

420 321 281 425 377 385 288

2.42 6.00 7.19 8.18 75.7 7.10 133

0.58 1.87 2.56 1.92 20.1 1.84 46.2

274

43.4

15.8

456

9.80

2.15

425 296

8.18 6.13

1.92 2.07

Hanc et al. (2009) Bolan et al. (1996) Bolan et al. (1996) Bolan et al. (1996) Guo et al. (2009) Bolan et al. (1996) Marin-Benito et al. (2009) Marin-Benito et al. (2009) Zmora-Nahuma et al. (2005) Bolan et al. (1996) Bolan et al. (1996)

Bisulfate amended, phytase-diet Delmarva poultry litter.

Plant litter and humus are the most important sources of DOM in soil, which is confirmed by both field and laboratory (including greenhouse) studies (Kalbitz et al., 2000; Kalbitz et al., 2007; Muller et al., 2009;

Table 2

Selected references on methods of extraction and analysis of DOM in environmental samples

Samples

Extraction of DOM

Measurement of DOM

Reference

Volcanic ash soils

Soil solutions collected by centrifugation of cores at 7200 rpm; filtration (0.45 mm filters) Soil samples were crushed an passed through a 1 mm sieve, then heated in a redistilled water at 100 C for 2 h under a reflex condenser; filtration (0.45 mm filters) Extraction with 0.5 mol L 1 K2SO4 solution 1:5 (w/v); filtration (Advantec MFS Nº 5C paper). Aqueous samples were estimated for DOC by oxidation of the sample with a sulfochromic mixture (4.9 g dm 3 K2Cr2O7 and H2SO4, 1:1, w/w) with colorimetric detection of the reduced Cr3þ Samples were filtered through a Whatman G/F glass fiber filters.

DOC by Shimadzu TOC5000 analyzer

Kawahigashi et al. (2003)

DOC by Shimadzu TOC 5050A analyzer

Szajdak et al. (2007)

Peat—moorsh soil

Soils (medial, amorphic thermic, Humic Haploxerands) Moss, litter and topsoil (0–5 cm)

TOC by combustion at 675 C Undurraga et al. (2009) in an analyzer (Shimadzu— model TOC-V CPN) Colorimeter KFK-3 at 590 nm Prokushkin et al. (2006)

Wet combustion persulfate digestion followed by TOC analyzer Extracted DOC by 0.01 M CaCl2 solution Shimadzu TOC-5000A TOC analyzer with a solid to solution ratio of 1:10 (w/v), mixed for 30 min at 200 rpm; filtration (0.45 mm filter) Soil solution and stream waters Soil solution collected by tension-free DOC by infrared detection along a natural soil catena lysimeters following persulfate oxidation Soil solutions from forested watersheds of North Carolina Organic fertilizer

Qualls and Haines (1991) Li et al. (2005)

Palmer et al. (2004)

Liquid and solid sludge, farm Water extraction followed by centrifugation (40,000 g) and filtration (0.45 mm filter) slurry, fermented straw, soil, and drainage water Soils, peat extract, sludge, pig Extracted with water (1:3 solid:solution ratio); centrifugation (12,000 rpm) and filtration and poultry manure and (0.45 mm filter) mushroom compost Soil (Entic Haplothord) Extraction with deionized water (1:10 solid: solution ratio); filtered through 0.45 mm polysulfore membrane Pig manure Extracted with water (1:3 solid:solution ratio); shaken at 200 rpm for 16 h at 4oC; centrifugation (12,000 rpm) and filtration (0.45 mm filter) Cow manure slurry filtered through 0.45 mm polysulfore membrane Sewage sludge DOC was extracted in a soil:water ratio of 1:10 (w/v) after 1 h agitation. River water Peat water

River water

Natural water from river filtered by 0.22 mm filter Peat water filtered through 0.45 mm membrane filters

Filtered through 0.7 mm glass fiber filter

Dry combustion (Dhormann Carbon Analyzer DC-80)

Barriuso et al. (1992)

Wet chemical oxidation with Baskaran et al. (1996) dichromate followed by back titration Kaiser et al. (1996) Dry combustion (TOC analyzer Shimadzu 5050) DOC by Shimadzu TOC5000A TOC analyzer

Cheng and Wong (2006)

TOC analyzer using UV absorbance Wet combustion with chromate followed by back titration DOC by wet oxidation TOC analyzer DOC was analyzed using a high-temperature catalytic oxidation method (Dohrman DC-190 analyzer) In situ optical technology using fluorescence

Aguilera et al. (2009) Gasco´ and Lobo (2007)

Krachler et al. (2005) Rixen et al. (2008)

Spencer et al. (2007) (continued)

Table 2

(continued)

Samples

Extraction of DOM

Measurement of DOM

Reference

Sea water

Filtered through 0.45 mm polysulfore membrane

Lang et al. (2007)

Freshwater

Filtered through 0.7 mm glass fiber filter

Effluent water Groundwater, lake water, and effluent

– –

Sea water and effluent

Filtered through 0.7 mm glass fiber filter

High-temperature combustion instrument to measure isotope composition of DOC Acid-peroxydisulfate digestion and hightemperature catalytic oxidation (HTCO) with UV detection In situ UV spectrophotometer High-performance liquid chromatography-size exclusion chromatographyUVA fluorescence system Measurement of carbon atomic emission intensity in inductively coupled plasma atomic emission spectrometry (ICP-OES) TOC analyzer (TOC 5000; Shimadzu) DOC by Shimadzu TOC 5050A analyzer

Lake water

Water samples filtered using precombusted GF/F filters Soil solution and stream water Samples were filtered through from forested catchments 0.45 mm filters

Tue-Ngeun et al. (2005)

Rieger et al. (2004) Her et al. (2003)

Maestre et al. (2003)

Ishikawa et al. (2006) Vestin et al. (2008)

Dissolved Organic Matter

11

Sanderman et al., 2008). In forest ecosystems, which are the most intensively studied with regard to C cycling and its associated DOM dynamics, the canopy and forest floor layers are the primary sources of DOM (Kaiser et al., 1996; Kalbitz et al., 2007; Park and Matzner, 2003). However, it is still unclear whether DOM originates primarily from recently deposited litter or from relatively stable organic matter in the deeper part of the organic horizon (Kalbitz et al., 2007). In a temperate, deciduous forest, the source of DOM leaching from the forest floor (O layer) is generally a water-soluble material from freshly fallen leaf litter and throughfall (Kalbitz et al., 2007; Qualls et al., 1991). Apparently all of the DOM and dissolved organic N (DON) could have originated from the Oi (freshly fallen litter) and Oe (partially decomposed litter) horizons. They further observed that, while about 27% of the freshly shed litter C was soluble, only 18.4% of the C input in litterfall was leached in solutions from the bottom of the forest floor. Virtually all the DOM leached from the forest floor appeared to have originated from the upper forest floor, with none coming from the lower forest floor—an indication of the role of this litter layer as a sink. The role of freshly deposited litter as DOM source was further corroborated by laboratory studies (Magill and Aber, 2000; Moore and Dalva, 2001; Muller et al., 2009; Sanderman et al., 2008). Michalzik and Matzner (1999) found high fluxes of DOM from the Oi layer than from the Oe and Oa layers and indicated that the bottom organic layers acted instead as a sink rather than as a source of DOM. Logically, however, because of the more advanced state of decomposition, the bottom litter layers could produce more DOM than the surface layer. Indeed, Solinger et al. (2001) measured greater DOM fluxes out of the Oa than out of the Oi layer. Recently, Froberg et al. (2003) and Uselman et al. (2007) confirmed with 14C data that the Oi layer is not a major source of DOM leached from the Oe layer. In a comprehensive synthesis of 42 case studies in temperate forests, Michalzik et al. (2001) observed that, although concentrations and fluxes differed widely among sites, the greatest concentrations of DOM (and DON) were generally observed in forest floor leachates from the A horizon and were heavily influenced by annual precipitation. However, somewhat surprisingly, there were no meaningful differences in DOM concentrations and fluxes in forest floor leachates between coniferous and hardwood sites. The flux of soluble organic compounds from throughfall and the litter layer could amount to 1–19% of the total litterfall C flux and 1–5% of the net primary productivity (Froberg et al., 2007; McDowell and Likens, 1988; Qualls et al., 1991). Nearly one-third of the DOM leaving the bottom of the forest floor originated from throughfall and stemflow (Qualls et al., 1991; Uselman et al., 2007). Values for the potential solubility of litter in the field and in laboratory studies are in the 5–25% range of the litter dry mass and 5–15% of the litter C content (Hagedorn and Machwitz, 2007; McDowell

12

Nanthi S. Bolan et al.

and Likens, 1988; Muller et al., 2009; Sanderman et al., 2008; Zsolnay and Steindl, 1991). In typical soils, DOM concentrations may decrease by 50–90% from the surface organic layers to mineral subsoils (Cronan and Aiken, 1985; Dosskey and Bertsch, 1997; Worrall and Burt, 2007). Similarly, fluxes of DOM in surface soil range from 10 to 85 g C m 2 yr 1, decreasing to 2–40 g C m 2 yr 1 in the subsoils (Neff and Asner, 2001). In cultivated and pastoral soils, plant residues provide the major source of DOM, while in forest soils, litter and throughfall serve as the major source (Ghani et al., 2007; Laik et al., 2009). In forest soils, DOM represents a significant proportion of the total C budget. For example, Liu et al. (2002) calculated the total C budgets of Ontario’s forest ecosystems (excluding peat lands) to be 12.65 Pg (1015g), including 1.70 Pg in living biomass and 10.95 Pg in DOM in soils. Koprivnjak and Moore (1992) determined DOM concentrations and fluxes in a small subarctic catchment, which is composed of an upland component with forest over mineral soils and peat land in the lower section. DOM concentrations were low (1–2 mg L 1) in precipitation and increased in tree and shrub throughfall (17–150 mg L 1), the leachate of the surface lichens and mosses (30 mg L 1), and the soil A horizon (40 mg L 1). Concentrations decreased in the B horizon (17 mg L 1) and there was evidence of strong DOM adsorption by the subsoils. Khomutova et al. (2000) examined the production of organic matter in undisturbed soil monoliths of a deciduous forest, a pine plantation, and a pasture under constant temperature (20 C) and moisture. After 20 weeks of leaching with synthetic rain water at pH 5, the cumulative values of DOM production followed: coniferous forest > deciduous forest > pasture, the difference being attributed to the nature of carbon compounds in the original residues. The residues from the coniferous forest were found to contain more labile organic components. Among ecosystems types, Zsolnay (1996) indicated that DOM tends to be greater in forest than agricultural soils: 5–440 mg L 1 from the forest floor compared with 0–70 mg L 1 from arable soils. Other studies have also indicated greater concentrations of DOM and concentrations in grasslands than in arable soils (Ghani et al., 2007; Gregorich et al., 2000; Haynes, 2000). In general, DOM concentration decreases in the order: forest floor > grassland A horizon > arable A horizon (Chantigny, 2003). The rhizosphere is commonly associated with large C flux due to root decay and exudation (Muller et al., 2009; Uselman et al., 2007; Vogt et al., 1983). Microbial activity in the rhizosphere is enhanced by readily available organic substances that serve as an energy source for these organisms (Paterson et al., 2007; Phillips et al., 2008). Because of their turnover, soil microbial biomass is also considered as an important source of DOM in soils (Ghani et al., 2007; Steenwerth and Belina, 2008; Williams and Edwards, 1993). Thus, microbial metabolites may represent a substantial proportion

Dissolved Organic Matter

13

of the soil’s DOM. It may well be that the rate of DOM production and extent of DOM dynamics in soil is regulated by the rate of litter/residue incorporation in soils, kinetics of their decomposition, and various biotic and abiotic factors (Ghani et al., 2007; Kalbitz et al., 2000; Michalzik and Matzner, 1999; Zech et al., 1996). In summary, the various C pools in an ecosystem represent the sources of DOM in soils. Due to their abundance, recently deposited litter and humus are considered the two most important sources of DOM in forest soils. Similarly, recently deposited crop residues and application of organic amendment such as biosolids and manures are the most important sources of DOM in arable soils. However, the role of root decay and/or exudates and microbial metabolites cannot be downplayed in both forested and arable ecosystems.

3. Properties and Chemical Composition of Dissolved Organic Matter in Soils 3.1. Structural components Because DOM is a heterogeneous composite of soluble organic compounds arising from the decomposition of various carbonaceous materials of plant origin, including soluble microbial metabolites from the organic layers in the case of forest ecosystem, DOM constituents can be grouped into “labile” DOM and “recalcitrant” DOM (Marschner and Kalbitz, 2003). Labile DOM consists mainly of simple carbohydrate compounds (i.e., glucose and fructose), low molecular weight (LMW) organic acids, amino sugars, and LMW proteins (Guggenberger et al., 1994b; Kaiser et al., 2001; Qualls and Haines, 1992). Recalcitrant DOM consists of polysaccharides (i.e., breakdown products of cellulose and hemicellulose) and other plant compounds, and/or microbially derived degradation products (Marschner and Kalbitz, 2003) (Table 3). Soil solution DOM consists of LMW carboxylic acids, amino acids, carbohydrates, and fulvic acids—the first comprising less than 10% of total DOM in most soil solutions and the last (i.e., fulvic acid) being typically the most abundant fractions of DOM (Strobel et al., 1999, 2001; Thurman, 1985; van Hees et al., 1996). Dissolved organic matter is separated into fractions based on solubility, molecular weight, and sorption chromatography. Fractionation of DOM by molecular size and sorption chromatography separate DOM according to properties (hydrophobic and hydrophilic) which regulate its interaction with organic contaminants and soil surfaces. The most common technique for the fractionation of aquatic DOM is based on its sorption to non-ionic and ion-exchange resins (Leenheer, 1981).

Table 3

Components identified in specific fractions of dissolved organic matter

Fraction

Compounds

Reference

Hydrophobic neutrals

Hydrocarbons Chlorophyll Carotenoids Phospholipids Tannins Flavonoids Other polyphenols Vanillin Fulvic acid and humic acid Humic-bound amino acids and peptides Humic-bound carbohydrates Aromatic acids (including phenolic carboxylic acids) Oxidized polyphenols Long-chain fatty acids Humic-like substances with lower molecular size and higher COOH/C ratios Oxidized carbohydrates with COOH groups Small carboxylic acids Inositol and sugar phosphates Simple neutral sugars Non-humic-bound polysaccharides Alcohols Proteins Free amino acids and peptides Aromatic amines Amino-sugar polymers (such as from microbial cell walls)

Polubesova et al. (2008) Albrechtova et al. (2008) Leavitt et al. (1999) Yoshimura et al. (2009) Suominen et al. (2003)

Weak (phenolic) hydrophobic acids

Strong (carboxylic) hydrophobic acids

Hydrophilic acids

Hydrophilic neutrals

Bases

Hernes et al. (2007) Suominen et al. (2003) Christensen et al. (1998) Lytle and Perdue (1981) Volk et al. (1997) Gigliotti et al. (2002) Serrano (1994) Jandl et al. (2002)

Obernosterer et al. (1999) Monbet et al. (2009) Borch and Kirchman (1997) Rosenstock et al. (2005) Chefetz et al. (1998) Schulze (2004) Yamashita and Tanoue (2004) Yamashita and Tanoue (2004) Jones et al. (2005)

Dissolved Organic Matter

15

Bolan et al. (1996) examined the distribution of various molecular weight fractions in the DOM extracts of various sources including soil, manures, composts, and sewage sludge. The DOM samples varied in the relative distribution of molecular weight fractions. The DOM from sewage sludge and poultry manure has a greater proportion of DOM in LMW fractions than DOM from soil or stream water. These results are consistent with the results for chemical oxidation, indicating that LMW fractions are more readily oxidized than the high molecular weight fractions. Dai et al. (1996) examined the structural composition and fractions (hydrophobic and hydrophilic acids and hydrophilic neutrals) of DOM from forest floor leachates over a 2-year period using 13C NMR spectroscopy. Total DOM in forest floor leachates ranged from 7.8 to 13.8 mmol L 1 with an average of 8.6 mmol L 1. These solutions were enriched with organic acids that averaged 92% of the total DOM. The 13C NMR data suggested that alkyl, carbohydrate, aromatic, and carboxylic C were the primary constituents of DOM fractions. Compositional changes of C with depth were observed, aromatic and carbohydrate decreased, whereas alkyl, methoxy, and carbonyl moieties increased with depth. Hydrophobic acids contained high contents of aromatic C, whereas hydrophilic acids primarily comprised carboxylic C. Hydrophilic neutrals were rich in carbohydrate C. Engelhaupt and Bianchi (2001) noticed that DOM from soils and leaf litter was dominated by aliphatic (41%), carbohydrate (33%), and carboxyl (16%) carbon, with relatively low aromatic carbon (10%). This study demonstrated that lignin and other compounds from terrestrially derived organic matter in sediments and adjacent soils were not a significant source of soluble moieties that enter the HMW DOM pool of tidal streams. Maurice et al. (2002) observed that the contribution of soil pore water relative to groundwater controlled not only the concentration, but also the average physicochemical characteristics of the DOM in streams. Combined field and laboratory experiments suggested that preferential adsorption of HMW and aromatic DOM components to mineral surfaces within the lower soil horizons resulted in more aliphatic groundwater DOM pool. Low flow periods resulted in an aliphatic dominated DOM in streams, whereas higher flow periods resulted in more aromatic downstream surface water DOM pool.

3.2. Fulvic acid—The dominant component The fractions of soil humic substances that are water soluble at any pH above 1–2, that is, fulvic acids (FAs), are very abundant and important components of soil DOM and a large number of studies have focused on the structure and chemical composition of FAs in DOM (Plaza and Senesi, 2009; Senesi and Loffredo, 1999; Senesi and Plaza, 2007). The FAs feature composition, structure, and chemical and biochemical properties that

16

Nanthi S. Bolan et al.

definitely distinguish them from the other typical components of DOM, all of which belong to definite organic chemical classes. On the contrary, soil FAs are not defined by a unique chemical formula and do not belong to any of the known chemical classes of organic compounds. The FAs consist of a physically and chemically heterogeneous mixture of relatively low molecular weight (500–2000 Da), yellow-to-light-brown/reddish organic molecules of mixed aliphatic and aromatic nature, and bearing acidic functional groups (mainly carboxylic and phenolic OH), which are formed by secondary synthesis reactions of recalcitrant compounds with products of microbial and chemical decay and transformation of biomolecules originated from organisms during life and after death (Senesi and Loffredo, 1999). These distinctive features confer to the FA fraction of soil DOM unique behavior and performances in soil chemical and biological reactivity, especially toward metal(loid) ions and organic contaminants. The major oxygen-containing functional groups in FA are COOH and phenolic OH groups, whereas alcoholic OH and carbonyl and methoxyl groups are found in smaller amounts. During humification, COOH and carbonyl groups have been found to increase, whereas phenolic and alcoholic OH and methoxyl groups decreased. FAs behave like weak acid polyelectrolytes whose acidic properties have been studied by base titration using potentiometric, conductometric, highfrequency, and thermometric techniques. Such occurrence has been found in FAs of a continuous and complex spectrum of nonidentical acidic functional groups with pKa values that span a very wide range as a function of FA concentration and presence of neutral salts. Although there is disagreement about the pKa values of soil FAs that are recorded in the literature, the pKa provides a convenient means of comparing the strengths of acidic groups in FAs from various sources and for any given FA as affected by neutral salts and an indication of the expected degree of ionization at various pHs. Several mathematical models have been applied to describe proton binding by FAs including continuous distribution models and affinity spectrum models. FAs are a variable-charge soil component with a low-point-of-zero charge of about 3. Thus, FAs are negatively charged at pH >3, and COOH and phenolic OH groups of FAs are among the major contributors to the negative charge of soil. In general, the cation-exchange capacity of FAs increases with increasing the degree of humification and soil pH. The molecular weight (MW), size, and shape are very important basic properties of FA. However, several problems have been encountered when dealing with the measurement of these properties that are greatly dependent on the physical state and concentration of the FA, and pH and ionic strength of the medium. FAs are polydisperse materials, that is, they exhibit a range of MW that may vary from a few hundred to a couple of thousand Daltons. Because of the polydispersed nature of FAs, methods that could provide

Dissolved Organic Matter

17

distribution patterns of MW for FA have been applied. Furthermore, the average MW of polydispersed systems can be expressed in several ways depending on the physical method of determination. These include the number-average MW, the weight-average MW, the z-average MW, and viscosity-average MW. The weight-average MW is generally considered the most representative average MW value because it better correlates with the molecular properties of FA. The sizes and shapes of FAs, that is, their morphological conformation, can be directly observed by the use of transmission and scanning electron microscopes (TEM and SEM). However, sample preparation, especially drying procedure, was found to affect markedly the morphological features of FAs and thus represent the most critical aspect of electron microscopy application to the study of FAs. Furthermore, the pH of the medium and FA concentration were found to be crucial for determining the conformation of FAs. In particular, at acidic to neutral pH (from 2 to 7), FA exhibited the shape of elongated, linear, or curved fibers that tended to become thinner with increasing pH, and of bundles of fibers that tended to become predominant at pH 6, and to give a fine network at pH 7. At higher pH (8 and 9), the FA assumed a sheet-like structure of increasing thickness, whereas at pH 10, a fine homogeneous gramlike shape was apparent. At low concentrations, the FA particles assumed an almost spheroidal shape with tendencies to coalesce to round-shaped aggregates or linear, chain-like shapes. At intermediate concentrations fiber-like shapes were formed by FA, whereas at the highest concentrations parallel arrays of filaments tended to coalesce to sheet-like shapes. A number of chemical and structural information on FAs could be provided by the use of chemical and thermal degradation methods including hydrolysis; reduction with sodium amalgam and by zinc dust distillation and fusion; oxidation with alkaline permanganate, alkaline cupric oxide, and peracetic acid; degradation with sodium sulfide and phenol; thermogravimetry; differential thermal analysis; and differential thermal calorimetry (e.g., Chen et al., 1978b). However, the most modern and powerful pyrolysis techniques have provided the most interesting results. Pyrolysates of FA contain a rich mixture of products in various proportions that can be related to their constituent building blocks, lateral chains, and functional groups. These include high levels of polysaccharides, phenolic constituents, n-alkanes, fatty acids, diols, sterols, alkyl mono- and di-esters, and furan rings; low levels of polypeptide products, lignin products, microbially synthesized polyphenols, and aromatic hydrocarbons; and various levels of substituted polycarboxylic acids, amino sugars, lipids, and other aliphatic constituents. Advanced methods of analytical pyrolysis, especially Curiepoint pyrolysis–gas chromatography–mass spectrometry and pyrolysis– FIMS, made possible the identification of chemical building blocks in FA and provided a molecular chemical basis for modeling a structural network for FAs in which aromatic rings are joined by alkyl chains.

18

Nanthi S. Bolan et al.

Spectroscopic techniques, such as infrared (IR), nuclear magnetic resonance (NMR), fluorescence, and electron spin resonance (ESR) spectroscopies, have had wide applications to the study of FAs that have enhanced our knowledge of their chemical structure and properties (Senesi, 1990a,b, c; Senesi et al., 1989). IR spectroscopy has been the most used classical spectroscopic technique in the study of FAs and has allowed the qualitative and semiquantitative identification of several typical components present at various levels in the structure of FAs. These include short- and long-chain aliphatic CH bonds, COOH and other carbonyl groups, variously substituted aromatic structures, amide groups, nonaromatic double bonds, conjugated ketones and quinones, phenolic and alcoholic groups, aryl ethers, and polysaccharides. The rapidly advancing powerful NMR techniques are among the most useful tools currently available for the qualitative and quantitative study of structural components and functional groups of FAs. The 1H NMR has allowed the identification of several hydrogenated components (protons) present in FAs. These include terminal methyl and methylene groups, methyl and methylene groups bound to alicyclic or aromatic rings, olefins, phenols, and COOH groups. The dominant peak area of the CP MAS 13C NMR spectra of FAs is the C–O chemical shift region primarily due to polysaccharides. Other well-resolved peaks are assigned to (a) unsubstituted aliphatic C comprising methyl, methylene, and methine groups; (b) C in C– O of methoxyl groups; (c) C in all other aliphatic and C–O and C–N groups; (d) anomeric C; (e) aromatic C; and (f) carbonyl C in carboxyl, ester, and amide groups. Additional two poorly resolved peaks are assigned to (a) aromatic C in phenolic groups, aromatic amine groups, and aromatic ethers and (b) carbonyl C in ketonic groups. For quantitative analysis, peak areas of the spectrum corresponding to the various chemical shift zones are often measured by integration, thus providing the distribution of various C types in FA. 15N, 31P, and other nuclei NMR has also been applied in FA studies with various success. Fluorescence monodimensional spectroscopy in the emission, excitation, and synchronous scan modes and bi- and tri-dimensional fluorescence spectroscopies have also been widely applied in the study of FAs. The fluorescence emission spectra of FAs generally consist of a unique broadband with a maximum wavelength that ranges from 445 to 465 nm. Fluorescence excitation spectra of FAs generally feature one main peak in the intermediate region of the spectrum (around 390 nm) with additional minor peaks and shoulders at longer and/or shorter wavelengths. FAs generally exhibit fluorescence synchronous scan spectra that are more structured than the corresponding emission and excitation ones, featuring two main peaks at long (450–460 nm) and intermediate (390–400 nm) wavelengths, often with some less intense peaks and/or shoulders at both sides (Miano and Senesi, 1992). Bi- and tri-dimensional fluorescence has

Dissolved Organic Matter

19

also been applied with success in the study of FAs. Analysis of fluorescence spectra has provided some useful and unique information on the structure and functionalities of FAs. For example, hydroxyl- and methoxy-coumarinlike structures, such as esculetin and scopoletin, originated from lignin, chromone, and xanthone derivatives; Schiff-base fluorophores derived from polycondensation reactions of carbonyls with amino groups; benzene rings bearing an hydroxyl conjugated to a carbonyl, methylsalicylate moieties, and dihydroxybenzoic acid units such as protocatechuic, caffeic, and ferulic acids have all been suggested as possibly responsible for fluorescence of FAs at various wavelengths. Fluorescence properties and intensity of FAs have been shown to be extensively affected by some molecular parameters and conditions of the medium. These include origin and nature, molecular weight and concentration of FA in solution, and pH and ionic strength of the medium. Application of ESR spectroscopy has provided important information on the existence, nature, and concentration in FAs of indigenous organicfree radicals and complexed paramagnetic metal(loid) ions such as Cu, Fe, Mn, and V, which may be involved at various stages in several important chemical, biochemical, and photochemical processes occurring in soil and water systems. ESR data are consistent with the existence in FAs of indigenous semiquinone radical units extensively conjugated to aromatic rings. The concentration of organic free radicals (between about 1016 and 1018 spins g 1) is probably the most important datum that can be obtained from the ESR spectrum of FA and has been shown to depend on numerous measurement conditions and environmental factors (Senesi et al., 1977a,b). A marked increase in free radical concentration of FAs is caused by raising the pH or temperature, chemical reduction, UV–Vis light irradiation, and acid hydrolysis. However, the increase was shown not to be sustained in time but followed by a gradual decrease soon after the maximum value was attained. On the contrary, mild chemical or electrochemical oxidation, methylation, and an increase in neutral electrolyte concentration often produced a time- and pH-dependent decrease of free radical concentration in FA. The 10-fold decrease of free radical concentration measured for some FAs confirmed that phenolic OH groups are the most important electron donors responsible for the formation and existence of free radicals in FA. The effect of oxidation could be reversed, however, by treatment with a reductant or by light irradiation of the FA sample. The accumulated ESR evidence supports the existence of a quinone–hydroquinone electron donor–acceptor (or charge transfer) system for the reversible generation and maintenance of free radicals of semiquinonic nature in FAs. Two classes of free radicals of similar nature, but of different stability, were suggested to exist in FAs. Besides indigenous or “native” semiquinone radicals, which are stable over long time spans and survive in any conditions of the system, “transient” or short-lived semiquinone radicals can be generated by reaction

20

Nanthi S. Bolan et al.

of quinone and hydroquinone moieties in FA, which can only persist in relatively short time spans. The free radical concentration in FAs was also shown to be directly related to their color, degree of aromaticity, and molecular size and complexity.

3.3. Elemental composition The elemental composition of DOM depends on its origin (Table 4). The major elements accompanying carbon include oxygen, hydrogen, nitrogen, phosphorus, sulfur, and trace amounts of various cations including calcium, potassium, magnesium, and metal(loid)s including aluminum, iron, zinc, and copper. For example, Kaiser (2001) found that the organic forest floor layers were large sources for DOC, DON, DOP, and DOS. The dissolved organic nutrients were mainly concentrated in the hydrophilic DOM fraction, which proved to be more mobile in mineral soil pore water than the hydrophobic one. Consequently, the concentrations and fluxes of dissolved organic nutrients (DON, DOP, and DOS) decreased less with depth than those of DOC. The average elemental composition (in percentage) of soil FA is C, 45.7; O, 44.8; H, 5,4, N, 2.1; S, 1.9 (Senesi and Loffredo, 1999). However, the composition range of FAs varies at some extent as a function of several factors including climate, parent material, vegetation, soil age, and pH (Chen et al., 1978a; Senesi et al., 1989). Methods used for soil FA extraction may also affect the analytical results and may cause lack of reproducibility. Typical O/C and H/C ratios of soil FAs are 0.7 and 1.4, respectively. High O/C ratios reflect high amounts of oxygenated functional groups such as COOH and carbohydrates, whereas low H/C ratios would indicate a high contribution of aliphatic components in FA. The elemental composition of DOM in relation to mobilization of nutrients is discussed in Section 6. Thus, the chemical composition and structural properties of various components in DOM are influenced by sources and their decomposition stage and play a vital role in the interactions of DOM with heavy metal(loid)s, nutrients, and pesticides.

4. Mechanisms Regulating Dynamics of Dissolved Organic Matter in Soils The net pool of DOM in soils is the result of various biogeochemical processes, resulting in a balance between the input and output of organic C in the forest floor (or surface soils in arable and grassland soils). These biological (biodegradation/decomposition, biotransformation), chemical (sorption, complexation, photodegradation), and physical (leaching,

Table 4 Elemental composition of dissolved organic matter Elemental composition (mg L 1) Source

Carbon

Pasture soil Arable soil Forest soil Rhizosphere soil (Grassland) Grassland soil Grassland soil Wetland soil Cattle manure Forest floor Forest floor-derived from litter Forest floor Sewage sludgeamended soil

28.8

Nitrogen Phosphorus

Sulfur Metals

7.9–13.9 0.9–1.2 11–32 2.5–9 2.5–10 0.017–0.133 5–140 1807.2

0.03–2.4

45,1000

17,000

1000

0.253 3400

23 61.7

1.18

0.06

0.25

Sewage sludge

277.7

Sewage sludge

4395

Stream

11–46

0.2–0.6

Measured in

Reference

Soil extract Soil extract Soil solution Soil solution

Stumpe and Marschner (2010)

Soil solution Soil solution Soil solution

Jones et al.(2004) McDowell (2005) D’Amore et al. (2009) Stumpe and Marschner (2010) Kaiser and Guggenberger (2005b) Kaiser (2001)

Seepage water Soil solution Soil solution Sludge soil Cd-0.13 solution Ni-271.42 Zn-145.31 Liquid sewage sludge Cu-0.905 Sludge solution Ni-2.215 Zn-2.315 Stream water

Mo¨ller et al. (2005) Khalid et al. (2007)

Kaiser and Guggenberger (2005a) Antoniadis et al. (2007)

Zhaohai et al. (2008) Ashworth and Alloway (2004)

D’Amore et al. (2009) (continued)

Table 4 (continued) Elemental composition (mg L 1) Source

Carbon

Surface water

38.2

Groundwater

10.5

Poultry litter

16,600

Nitrogen Phosphorus

2160

Sulfur Metals

Measured in

Cu-0.009 Stream water Pb-0.018 Zn-0.371 Cd-0.0004 Cu-0.012 Stream water Pb-0.029 Zn-0.506 Cd-0.0009 Poultry litter extract

Reference

Karlik and Szpakowska (2001)

Karlik and Szpakowska (2001)

Goyne et al. (2008)

23

Dissolved Organic Matter

eluviation) processes are in turn moderated by biotic and abiotic factors that include soil pH, organic carbon and clay contents, microbial activity, and environmental factors including temperature and moisture content (Table 5). The role of these factors in controlling the dynamics of DOM is discussed in Section 5.

4.1. Sorption/complexation Like any other solute in soils, DOM undergoes both sorption and complexation reactions (Guggenberger and Kaiser, 2003; Kothawala et al., 2009; Remington et al., 2007; Vandenbruwane et al., 2007; Yurova et al., 2008). While sorption results in the retention of DOM with soil components and subsequent retardation of its mobility and degradation, complexation can result in the formation of both soluble and insoluble DOM–metal(loid) complexes, thereby affecting both movement and degradation. While soluble DOM–metal(loid) complexes enhance the movement of DOM in soils, insoluble complexes result in the retardation of DOM movement (Guggenberger and Kaiser, 2003; Jansen et al., 2005; Martin and Goldblatt, 2007). Complexation of DOM with metal(loid) ions controlling the movement and bioavailability of both DOM and metal(loid)s is Table 5 Mechanisms and factors regulating the dynamics of dom in different land use systemsa Agricultural lands

a b

Forest lands

Wetlands

Mechanism/Factor

Arable Pasture/Prairie Upland Savannah Rice Swamps

Sorption Complexation Bidegradation Biotransformationb Photodegradation Leaching Vegetation Cultivation Soil amendments Soil pH Clay mineralogy Metal oxides Organic matter

xx x xx NA x xx xx xxx xxx xx xx xx xx

xx xx xx NA x xx xx x xx xx xx xx xxx

xx x xx NA x xx xx NA x xx xx xx xxx

xx xx xx xx x xx xx NA NA x x x xxx

xx xx xx xxx xx xxx xx NA xxx xx xx xx xx

x xxx xx xxx xxx xxx x NA NA x x x x

Degree of importance: x, low; xx, medium; xxx, high; NA, not applicable. This Mechanism refers primarily to methane formation in reducing conditions such as rice paddy, swamps and to some extent savannahs.

24

Nanthi S. Bolan et al.

discussed in detail in Section 6.3. Briefly, when DOM percolates in the soil profile, it may interact with metal oxide surfaces, thereby forming a “shield” against microbial attack. In acid forest soils, Al and Fe can form relatively stable complexes with DOM, which can enhance solubility and transport, as might be the case during podzolization (Blaser, 1994; Jansen et al., 2005). However, complexation of potentially toxic metal(loid)s may not result in diminished biodegradability of DOM, but may even enhance microbial activity by sequestering the toxic effects from free metal(loid) ion activity (Apte et al., 2005; Marschner and Kalbitz, 2003). Similarly, formation of stable complexes between DOM and certain heavy metal(loid)s ions such as Cu, Hg, and Pb can alter the metal(loid) toxicity to fish and other aquatic organisms (Adriano, 2001; Alberts et al., 2001; Martin and Goldblatt, 2007). In temperate soils, the greatest concentrations of organic C typically occur in the organic layers and the mineral topsoil (A) horizon. However, based on total soil mass in the various horizon depths in the soil profile, subsoil (B and C) horizons could account for greater amounts of organic C ( Johnson et al., 2009; Paul et al., 2002; Schulze et al., 2009; Ziegler, 1991). In investigating two temperate acid forest soils, Kaiser et al. (2002) observed that the organic forest floor layer and B and C horizons contained 40–50% of the total DOM. The ultimate fate of DOM in the soil profile is largely influenced by the nature and extent of soil mineral—organic carbon—microbe interactions (Huang et al., 2005a; Young et al., 2008). In essence, partitioning of the DOM between the aqueous (i.e., soil solution) and the solid phase (i.e., soil matrix) is controlled by the properties and composition of DOM, microbial population, and mineralogical and chemical properties of the soil (Adriano, 2001; Guggenberger and Kaiser, 2003; Kothawala et al., 2009; Stevenson, 1994). For example, clays may interact directly with microbes, thereby altering the rate and pathways of microbial metabolism; modify the aqueous phase environment (e.g., buffering the pH that affects microbial and enzyme activity and chemical speciation of contaminant chemicals); and bind extracellular enzymes altering their activity (Grandy et al., 2008; Huang et al., 2005b; Sollins et al., 1996). Clay is a generic term that includes layer and amorphous aluminosilicates and the sesquioxides (i.e., oxides, hydroxides, and oxyhydroxides of Al and Fe) that provide the majority of surface area for the sorption of DOM and other solutes in soil. Organic–mineral interactions range in strength from strong ligand exchange to weaker anion-exchange reactions (McBride, 1994). The bonding mechanisms of DOM onto the soil solid phase have already been elucidated by Gu et al. (1994) and Sollins et al. (1996). This includes bonding of negatively charged organics by ligand exchange especially in oxide-rich and allophanic soils, positively charged organics into negative surfaces by cation exchange, anion exchange onto subsoils and variably charged soils, and the less important mechanisms such as cation bridging, water bridging, hydrogen bonding, and van der Waals forces.

Dissolved Organic Matter

25

The affinity of soils for DOM is influenced by several properties. Correlations between the extent of partitioning of DOM and surface area of clay, organic C, dithionite–citrate–bicarbonate-extractable Fe, and oxalate-extractable Fe and Al have been reported (Donald et al., 1993; Guggenberger and Kaiser, 2003; Kaiser et al., 1996; Kothawala et al., 2009; Nelson et al., 1993). Tipping (1981) reported that the surface area is the main factor influencing DOM sorption to Fe oxides/hydroxides. Donald et al. (1993) measured the sorption of DOM and its fractions by soil horizons from a catenary sequence. Variation in DOM sorption among the soil horizons was related to differences in the clay content and citrate– dithionate-extractable Fe, Al, and Mn. The hydrophobic acid and the hydrophilic acid fractions were the most abundant in the soil solution (72% of the total DOM) and accounted for most of the sorption of DOM in the Bt and C horizons. Moore et al. (1992) obtained DOM sorption isotherms for 48 soil samples derived from Humaquepts, Inceptisols, and Spodosols in southern Quebec using a DOM solution derived from a swamp peat. Forty-six samples had DOM sorption adequately represented by the linear initial mass isotherm. Null-point DOM concentrations (DOMnp), where there is zero net DOM sorption, ranged from 6.7 to 85.4 mg L 1. Distribution coefficients (kd) averaged 1.00 10 2 m3 kg 1, suggesting that DOM sorption by soils is of moderate strength compared with inorganic anions. DOMnp values were positively correlated to organic C content and negatively correlated to oxalate-extractable Al and dithionite-extractable Fe, which explained 70% of the variation in DOMnp. Recently, Kothawala et al. (2009) noticed that poorly crystalline Al oxides exerted a stronger influence than Fe oxides on maximum sorption capacity of DOM for 52 mineral soil samples from 17 temperate and boreal soil profiles. Kaiser and Zech (1997) obtained DOM sorption isotherms for 135 soil horizons from 36 profiles of the major forest soils of the temperate zones (Leptosols, Vertisols, Cambisols, Luvisols, Podzols, Stagnosols, and Gleysols). When solutions containing no DOM were added, the release of DOM was greatest for horizons rich in organic C. In subsoil horizons, DOM release was much less. Most of the topsoil horizons showed weak DOM sorption. This was caused by poor concentrations of sorbents (clay and sesquioxides) and/or high concentrations of organic C. Organic C apparently decreased DOM sorption by occupying binding sites. Subsoils rich in clay and sesquioxides showed a strong retention of DOM. The majority of the soils preferentially sorbed hydrophobic DOM—caused by the greater affinity of hydrophobic DOM to oxide/hydroxide soil constituents. From microcalorimetric, FTIR, and 13C NMR analyses, Gu et al. (1994) concluded that ligand exchange between carboxyl/hydroxyl formational groups of the SOM and iron oxide surfaces were the dominant sorption mechanisms, especially under acidic or slightly acidic pH conditions.

26

Nanthi S. Bolan et al.

In deeper mineral soil horizons of forest lands, DOM fluxes declined from 10–40 g C m 2 yr 1 translocated from the organic surface layer into the mineral soil horizons to about 1–10 g m 2 yr 1 in deeper mineral horizons, indicating substantial retentions of DOM in subsoil horizons (Guggenberger and Kaiser, 2003). This observation and other similar observations prompted a general hypothesis that retention of DOM in the soil (or sediment) solid phase is a mechanism that promotes stability and conservation SOM in soils (Hedges and Oades, 1997; Kaiser and Guggenberger, 2000; Kaiser et al., 1996; Michalzik and Matzner, 1999; Sollins et al., 1996). However, Guggenberger and Kaiser (2003) estimated a mean subsidence time of sorbed SOM of about 4–30 years, inferring that instead of the “preservation” role of sorbed DOM, such DOM may enhance bioavailability to microbe causing subsequent biodegradation. Investigative consensus indicates that high organic C concentrations of the soil decrease DOM sorption, especially the hydrophilic fraction. In soils free of carbonates, sorption is related to oxalate-extractable Al and dithionate-extractable Fe; however, in carbonitic soils, DOM sorption is correlated with dithionate-extractable Fe only. The sorption of DOM by topsoil is always less than in subsoil samples. Sorption is generally high in B horizons of Alfisols, Inceptisols, and Spodosols with low organic C content and high contents of oxalate-extractable Fe and Al and dithionate-extractable Fe, whereas little or no sorption is noticed in soils with high contents of organic C and/or low contents of metal oxides alone, much as those in the A and E horizons ( Jin et al., 2008; Kaiser et al., 1996; Kothawala et al., 2009; Muller et al., 2009) Dissolved organic C concentrations in soil solutions can be as low as 0.1–3.6 mmol dm 3 in forest soils often in contact with subsurface horizons (Cronan and Aiken, 1985; Guggenberger and Zech, 1993; Laik et al., 2009; Laudon et al., 2009; McDowell and Likens, 1988; Sanderman et al., 2008). The decrease in DOM concentrations is characterized by a change in DOM composition, indicated by a preferential decrease of the hydrophobic fraction (Guggenberger and Zech, 1993). This was subsequently confirmed by Kaiser et al. (1996) where the majority of the soils studied preferentially sorbed hydrophobic DOM, apparently caused by higher affinity of the hydrophobic fraction for metal oxides/hydroxides in the soil matrix. Indeed, sorption of hydrophobic DOM by some soils was accompanied by the release of hydrophilic substances (Moore and Matos, 1999; Ussiri and Johnson, 2004). The formation of soil organo-mineral complexes is a key reaction in the carbon cycle in soil, since organic materials acquire a resistance to decomposition due to the formation of the complexes. Adsorption of DOM onto soil minerals provides a model of this important process. Adsorption of DOM onto samples from Andisols, Inceptisols, and Entisols in batch experiments was compared in terms of the quantitative relationship

Dissolved Organic Matter

27

between the soil properties and the adsorption behavior of DOM (Nambu and Yonebayashi, 2000). Andisols showed a particularly high efficiency of adsorption compared with those from other soils that contained a comparable amount of organic carbon. Although the adsorption mechanisms varied among soils, two soil variables, the degree of carbon accumulation in the soil sample (or total carbon/specific surface area ratio), and the amount of ligand exchange sites on labile aluminum accounted for the variation in DOM adsorption. In general, DOM components that are low in molecular weight, organic N, acidic groups, and aromatic structures can be expected to remain soluble in the soil’s aqueous phase, whereas constituents that are rich in organic N, acidic groups, and with high aromaticity can be preferentially sorbed (Gu et al., 1995; McKnight et al., 1992). While the sorption of DOM in soils increases with increasing levels of Fe and Al oxides in soils, it decreases with increasing concentrations of organic matter.

4.2. Biodegradation Biological assimilation of organic carbon and subsequent generation of DOM plays an important role in controlling DOM dynamics in soils (Figure 1). Thus, DOM originates primarily from the decomposition of SOM that had accumulated through vegetation, the addition of biological waste materials (e.g., biosolids and livestock manures), the release of root exudates, and the lysis of microorganisms. The decomposer community in soil consists of a wide range of bacteria, fungi, protista, and invertebrates (Dilly et al., 2004; Kalbitz et al., 2000; Swift et al., 1979). Considerable emphasis has focused on microorganisms because of their dual roles as decomposition agent and as a sink of labile organic C. Microbial assimilates represent an important source of DOM released from the forest floor, while microbial biomass serves as an important reservoir of DOM. Soil fauna, including earthworms, can facilitate the turnover rate of microbial biomass in soil (Aira et al., 2008; Kalbitz et al., 2000; Osler and Sommerkorn, 2007; Siira-Pietikainen and Haimi, 2009). Dissolved organic carbon is an important substrate for microorganisms (Marschner and Bredow, 2002; Michelsen et al., 2004; Qualls, 2005). Laboratory incubation studies of varying length have indicated that 10–44% of DOM in soil solution is microbiologically degradable ( Jandl and Sletten, 1999; Kalbitz et al., 2000; Qualls, 2005; Sachse et al., 2001). The more labile fraction of DOM is more readily mineralized or assimilated into microbial biomass (Nelsen et al., 1994; Qualls, 2005). It is likely that DOM production is controlled by the same factors controlling biological activity. The decomposition rate of DOM is influenced by soil depth, land use, soil fertility, etc. It decreased with increasing soil depth and is less in forest than in arable soils (Ludwig et al., 2000; Muller et al., 2009).

28

Nanthi S. Bolan et al.

Simply, microbial activity with depth is limited by the bioavailability of organic C as a substrate (Celerier et al., 2009; Ghiorse and Wilson, 1988; RodriguezZaraqoza et al., 2008; Zablotowicz et al., 2009) or the supply of essential nutrients such as N and P. It is well known that C decomposition rate decreases with decreasing available N (Chantigny et al., 1999; Enowashu et al., 2009; Frank and Groffman, 2009; Sirulnik et al., 2007). Most of the DOM in soils is the end product of microbial metabolism of organic residues. Fresh litter also contributes significantly to the production of DOM, indicating the presence of DOM in the original litter. Ludwig et al. (2000) studied the production of DOM in soils from two sites with different microbial activities using C13-depleted plants of differing decomposability (Epilobium angustifolium and Calamagrostis epigeios). Cumulative DOM production was markedly greater in the readily decomposing Epilobium experiment (2% of the added C) than in the slow decomposing Calamagrostic experiments (0.1% of the added C). The rate of biodegradation of DOM varies among sources, which has been attributed to the difference in its composition (Kalbitz et al., 2000). Some of the hydrophobic compounds extracted in the DOM are less accessible to microbial degradation than hydrophilic compounds (Amon et al., 2001; Kalbitz et al., 2003; Qualls, 2005). Based on biodegradation kinetics, DOM in soils is grouped into various categories such as labile and nonlabile fractions (Marschner and Kalbitz, 2003; Saadi et al., 2006). Microbes selectively degrade the less recalcitrant compounds, thereby gradually increasing the average recalcitrance of the remaining organic carbon (Bowen et al., 2009; Sollins et al., 1996; Waldrop and Firestone, 2004). 14C-dating has indicated that organic C in deeper horizons had longer residence times, indicating lower bioavailability to microbes (Chiti et al., 2009; Favilli et al., 2008; Oades, 1984). In summary, while microbial degradation of SOM, followed by desorption of organic substances from the soil matrix and leaching of soluble organic compounds from fresh litter are viewed as the most important processes causing the release of DOM (Currie et al., 1996; Guggenberger et al., 1994a; Marschner and Kalbitz, 2003; Qualls and Haines, 1991), microbial assimilation of readily available carbon in DOM results in the ultimate degradation of DOM to carbon dioxide.

4.3. Photodegradation Although DOM undergoes photochemical and microbial degradation, the former process dominates in aquatic systems and the latter on land (Marschner and Kalbitz, 2003; Minor et al., 2007; Mostofa et al., 2007). In Fe-rich surface waters, light-induced redox cycling of Fe and DOM photo-oxidation are strongly coupled (Norton et al., 2008; Shiller et al., 2006). Iron can catalyze DOM photo-oxidation via ligand-to-metal(loid)

Dissolved Organic Matter

29

charge transfer reactions of Fe(III)–DOM complexes and through DOM oxidation by the hydroxyl radical (HOo) formed via a Fenton reaction (Giroto et al., 2006; Voelker et al., 1997). Photo-oxidation can enhance the turnover of DOM in aquatic systems, transforming labile into more recalcitrant (less bioavailable) components and vice versa (Benner and Biddanda, 1998; Obernosterer et al., 1999). For example, photo-cleavage and photo-oxidation of HMW DOM resulted in the release of bioavailable LMW that stimulated bacterioplankton activity (Keiber et al., 1989; Mostofa et al., 2007; Wetzel, et al., 1995). Pullin et al. (2004) observed that photo-irradiation of DOM decreased the abundance of HMW components and formed new LMW components, including LMW carboxylic acids, that is, acetic, formic, and malonic acids. This can alter the complexation potential of DOM with metal(loid)s such as Fe. For example, it has been shown that intermediate and/or HMW, more aromatic, components of DOM sorb preferentially onto Fe(III) oxyhydroxide surfaces (Gu et al., 1995; Kaiser and Zech, 1997; Kothawala et al., 2009; Meier et al., 1999; Vandenbruwane et al., 2007; Zhou et al., 2001). Thus, by decreasing the concentration of DOM and degrading larger DOM components into smaller organic molecules, photo-irradiation can alter the chemical speciation of metal(loid)s such as Fe, thereby modifying the DOM and/or Fe mobility or sorptive ability. Photo-irradiation decreases the UV- and visible light-absorbing properties of DOM (Gao and Zepp, 1998; Goldstone et al., 2002; Moran et al., 2000), a process that may lead to decreased attenuation of photo-irradiation in lakes. Photo-irradiation may also decrease the metal(loid)-binding affinity of DOM for metal(loid)s through the oxidation of carboxylic acid groups (Faust and Zepp, 1993; Gao and Zepp, 1998; Shiller et al., 2006). Friese et al. (2002) suggested that photochemical reactions are responsible for depletion of DOM and formation of ferrous iron (Fe(II)) in acidic soft-water lakes. They noticed the coupling of DOM degradation with Fe(II) release when water samples from an acidic mining lake were irradiated with natural sunlight. Up to 50% of the DOM was transformed to CO2 after 300 min of irradiation, and the light-induced production of Fe(II) was increased by 450%. Thus, photodegradation is an important process in regulating the dynamics of DOM in aquatic environment. Although photodegradation has minimal significance on the DOM dynamics in forest, grassland, and arable ecosystems, it may exert more effect in wetland soils or in soils prone to flooding.

4.4. Leaching Dissolved organic matter is a reactive constituent in aquifer and soil media and equilibrates between the mobile aqueous phase and the immobile solid phase (Marschner and Kalbitz, 2003; Stutter et al., 2007). Since DOM is

30

Nanthi S. Bolan et al.

believed to accelerate the transport of associated contaminants, a number of studies have examined cotransport of contaminants and nutrients with DOM (Section 6). Adsorption of DOM to soil components is an important property controlling the leaching of DOM (Kalbitz et al., 2000; Marschner and Kalbitz, 2003). For example, Kaiser et al. (1996) examined the processes governing DOM adsorption to porous media that may affect the cotransport of contaminants. Column displacement experiments of DOM through aquifer sediments were modeled with convective–dispersive equation, which considered time-dependent adsorption reactions and linear or nonlinear adsorption processes. Observed DOM breakthrough curves (BTCs), with influent concentrations >10 mg C L 1, were adequately modeled as two-site, nonlinear adsorption processes, with DOM interactions with both types of sites being time dependent. The extended tailing of the BTCs was influenced more by the slow, time-dependent adsorption of DOM during transport than to the nonlinear features of the adsorption isotherms. Observed BTCs with influent concentrations <10 mg C L 1 did not exhibit extensive tailing and were modeled as a one-site, linear or nonlinear, time-dependent adsorption process. The leaching of DOM from the forest canopy (i.e., throughfall) and forest floor (i.e., litterfall) could also be an important source of N, and to a lesser extent of P, to the A horizon. Vinther et al. (2006) examined the leaching of DOC and DON, with focus on the period after cultivating grass-clover swards. Grass-clovers were plowed in the spring prior to sowing cereals followed by either catch crops or bare soil. The total percolation was 218 and 596–645 mm in the sandy loam soil and coarse sandy soil, respectively, which resulted in an annual leaching of 22–40 kg DOC ha 1 and 3–4 kg DON ha 1 and 174–310 kg DOC ha 1 and 10– 31 kg DON ha 1 in the respective soil. It was shown that higher amounts of DOC were lost by leaching under the catch crops than from bare soil.

5. Factors Influencing Dynamics of Dissolved Organic Matter in Soils The nature and amount of DOM in soils depends on a number of factors including type of vegetation, land management practices (cultivation; crop rotation; and lime, fertilizer, and organic manure applications), edaphic factors (soil type, clay mineralogy, and metal oxides), and environmental factors (temperature and rainfall). Changes in DOM upon management practices are generally of short duration, whereas long-term effects are more related to vegetation type and the amount of plant litter (including organic amendments) returned to the soil (Chantigny, 2003; Gibson et al., 2009; Ward et al., 2007).

Dissolved Organic Matter

31

5.1. Vegetation and land use Conversion of native forest to arable and pasture farming decreases SOM, which has been attributed mainly to net loss of carbon resulting from decomposition (Celik, 2005; Oades, 1984). A decrease in SOM is likely to cause a decrease in DOM, which may be one of the reasons for low DOM in cultivated land when compared to native forest (Chen et al., 2009; Khomutova et al., 2000; Ward et al., 2007; Zhang et al., 2007). For example, Jinbo et al. (2006) examined land use effects on the distribution of labile fraction organic C through soil profiles in Northeast China. Four land use types were selected: Deyeuxia angustifolia wetland, upland forest, two farmlands of soils previously under D. angustifolia wetland, and an abandoned cultivated soil. Soil TOC, DOM, microbial organic C, and hot waterextractable C were measured. The intact D. angustifolia wetland soil had significantly greater labile fraction organic C concentrations in the topsoil than upland forest, abandoned cultivated, and cultivated soils; however, there was no significant difference in these fractions among subsoils. The effects of land use on labile fraction organic C occurred mainly in the topsoil (0–20 cm). Xiao-gang et al. (2007) measured total organic C, total N, organic C associated with particle-size fractions, DOM and microbial biomass and activity at three soil depths under annual oats for 28 years, introduced perennial pasture (8 years after establishment) and native pasture at an alpine site in western China. Compared with native pasture, the soil under annual oats had 26–42% less total organic C and total N at different depths. In perennial pasture, total organic C and total N decreased by 10–18%, but only at 0–10 cm depth. Dissolved organic matter decreased by 40% (0–10 cm depth) and 11–16% (below 10 cm) under annual oats and by 16% (0–10 cm) under perennial pasture. The significant decrease in many of the soil organic C pools in native alpine pastures was caused by frequent soil disturbance. Cilenti et al. (2005) and Provenzano et al. (2008) studied the effect of soil salinity on some chemical properties of the hydrophilic (HI) and hydrophobic (HO) fractions of DOM by using FTIR and mono- and tri-dimensional fluorescence spectroscopies. With increasing soil salinity, a higher percentage of HI fractions were extracted as compared with their corresponding HO fractions, and a general decrease in polysaccharide components and a marked fluorescence intensity quenching were measured in all fractions, which suggests a different configuration assumed by DOM molecules at different ionic strengths of soil solution. Traversa et al. (2008) evaluated the influence of different forest tree covers on the amount and properties of DOM isolated from the plant litters and underlying soil layers. The carbon contents and humification indexes of litter DOM were much higher than those of the DOM of corresponding

32

Nanthi S. Bolan et al.

underlying soils. Chemical and spectroscopic (FTIR and fluorescence) analyses showed that litter DOM was impoverished of easily degradable components, such as carbohydrates, and enriched in low molecular size lignin-derived components. The molecular size, aromatic character, and polycondensation degree of DOM decreased with soil depth, whereas aliphaticity increased. These results suggest that the aromatic portion of partially degraded lignin-derived compounds might be gradually adsorbed by mineral soil components and protected from microbial degradation.

5.2. Cultivation Cultivation influences DOM in soils by providing oxygen, thereby accelerating the decomposition of organic matter releasing DOM and subsequent degradation and leaching of the released DOM (Bueno and Ladha, 2009). The net effect of cultivation on DOM concentration in soils depends on the rate of decomposition of SOM and rates of degradation and leaching of DOM. Recently, Dou et al. (2008) examined the impacts of conventional tillage, no tillage, and cropping sequence on the depth distribution of DOM, SOM, and total N for a silty clay loam soil after 20 years of continuous sorghum cropping. Conventional tillage consisted of disking, chiseling, ridging, and residue incorporation into soil, while residues remained on the soil surface for no tillage. Tillage effects on DOM and total N were primarily observed at 0–5 cm, whereas cropping effects were observed up to 55 cm depth. Dissolved organic matter, SOM, and total N were 37, 36, and 66% greater, respectively, under no tillage than conventional tillage at the 0–5 cm depth. The DOM decreased with soil depth and averaged 18% greater under a sorghum–wheat–soybean rotation than a continuous sorghum monoculture. Similarly, Shao et al. (2009) noticed that a paddy soil under conservation tillage had higher easily oxidizable organic carbon, DOM, POC, and microbial biomass carbon than traditional tillage. Linn and Doran (1984) noticed a decrease in DOM in cultivated soils compared with no-till and direct drilled plots. In examining the effect of cultivation on DOM concentration and leaching in grass-clover swards, these were plowed in the spring prior to sowing cereals followed by either catch crops or bare soil (Vinther et al., 2006). The concentrations of DOM decreased with soil depth and at 90-cm soil depth ranged between 7 and 21 mg C L 1 and between 16 and 63 mg C L 1 in sandy loam and coarse sandy soils, respectively. It resulted in an annual leaching of 22–40 kg DOC ha 1 year 1 and 3–4 kg DON ha 1 year 1 in the sandy loam soil, and 174–310 kg DOC ha 1 year 1 and 10– 31 kg DON ha 1 year 1 in the coarse sandy soil. It was shown that higher amounts of DOM were lost by leaching under the catch crops than from bare soil.

33

Dissolved Organic Matter

Delprat et al. (1997) examined the impact of cultivation on the nature of DOM. During the first stage, cultivation intensified mineralization of initial organic matter, thereby leading to an increase of 2- to 5-fold in the quantity of DOM as fulvic and humic acids and humic metal(loid) complexes. During the second stage, once the original organic matter was stabilized, DOM concentrations decreased with time of cultivation. Both SOM and DOM were enriched in carbon originating from maize with time of cultivation. Drying of soils could increase the solubility of SOM (Haynes and Swift, 1989; Hentschel et al., 2007). For example, Bolan et al. (1996) extracted DOM from soil, manure, and sludge samples after drying at various temperatures. The amount of DOM, as measured by total C analysis using LECOTM combustion, increased with drying (Table 6). There was no difference in the amount of DOM between the field-moist and the freeze-dried samples. Similarly, Hentschel et al. (2007) noticed an increase in DOM input from organic horizons to the mineral soil owing to drying and wetting of a Norway spruce soil. They also noticed that the composition of DOM markedly differed within each wetting period, pointing to enhanced release of easily decomposable substrates in the initial wetting phases and the release of more hardly decomposable substrates in the final wetting phases. An increase in DOM during drying has been attributed to the breaking of hydrogen bonds in the organic matter or to the lysis of microbes (Magid et al., 1999). Removal of water from soil through drying will result in the macromolecules of SOM changing into a highly condensed state. Such shrinkage of organic compounds may result in the disruption of organo-mineral association and subsequent release of some low molecular weight humic components.

5.3. Soil amendments Application of lime, inorganic fertilizers, and organic amendments can affect DOM in soils. While lime and organic amendments have often been shown to increase DOM in soils, the effect of inorganic fertilizers on DOM can vary with the amount and type of fertilizer added (Bolan et al., 2003; Brown Table 6 Effect of drying on dissolved organic matter Dissolved organic matter (mg C kg 1) Materials

Field moist

Air dry

Oven dry

Freeze dried

Pasture soil Poultry manure Sewage sludge

1.02 8.18 6.00

2.12 10.8 9.7

3.56 19.5 17.2

1.21 7.98 6.12

34

Nanthi S. Bolan et al.

et al., 2008; Hildebrand and Schack-Kirchner, 2000). For example, while the application of urea fertilizer has been shown to increase the concentration of DOM, mainly due to an initial increase in soil pH resulting from ammonification reactions, application of superphosphate fertilizers has been shown to decrease DOM due to soil acidification (Ouyang et al., 1999; Ruark et al., 2009). The effect of liming on DOM is discussed in the next section. Most biological waste materials of plant origin contain large amount of DOM and the addition of certain organic manures, such as poultry manure, increases the pH and thereby enhances the solubilization of SOM (Goyne et al., 2008; Martin-Olmedo and Rees, 1999; Schindler et al., 1992). However, DOM in these organic amendments appears to be rapidly degraded in soil mainly due an increase in microbial activity resulting from carbon input, and this increase in microbial activity (priming effect) may sometimes result in the immobilization of soil carbon, thereby resulting in a net decrease in DOM (Boddy et al., 2007; Chantigny, 2003; Gullberg et al., 1997; Murphy et al., 2000). Even though the increase in DOM after the addition of organic amendment might be short-lived, long-term field trials have indicated that, compared with conventional farming, organic farming, which relies heavily on organic amendment input, may result in a net increase in DOM (Herencia et al., 2008; McDonald et al., 2001). Provenzano et al. (2006) characterized the hydrophilic (HI) and hydrophobic (HO) fractions of DOM extracted from urban waste and sewage sludge composts, anaerobically digested municipal sewage sludge, and pig slurry. Results showed the dominance of aliphatic structures in all DOM samples, especially in the HI fractions with low aromatic polycondensation, and of conjugated chromophores, whereas the HO fractions were rich in polysaccharides and showed a greater molecular complexity with high aromatic polycondensation. Garcia-Gil et al. (2007, 2008) examined the compositional, structural, and acid–base properties of the FA fractions of DOM of a sewage sludge (SL), municipal waste compost (MWC), and SL- and MWC-amended Calcaric Regosol and Calcic Luvisol cropped with barley. The SL–FA and MWC–FA were characterized by aliphatic compounds with higher levels of C, H, N, and S, but smaller levels of O, carboxyl and phenolic OH, total acidity, and C/N and C/H ratios. The SL and MWC additions to soils increased aliphaticity, proton affinity of the carboxylic and phenolic groups, and N, H, and S contents, but decreased O and acidic functional group contents and C/N ratios. Furthermore, SL and MWC applications induced a decrease of heterogeneity of carboxylic-type groups, whereas the heterogeneity of phenolic-type groups increased substantially. In another study, Fernandez et al. (2007) investigated the acid–base properties of FAs isolated from composted (CS) and thermally dried (TS) sewage sludge, soils amended with either CS or TS at a rate of 80 t ha 1 y 1

Dissolved Organic Matter

35

for 3 years, and the corresponding unamended soil. With respect to unamended soil FA, the FAs from CS, and especially TS, were characterized by smaller acidic functional group contents, larger proton-binding affinities, and smaller heterogeneity of both carboxylic- and phenolic-type groups. Amendment with CS or TS led to a decrease of acidic functional group contents and a slight increase of proton-binding affinities of carboxylic- and phenolic-type groups of amended soil FAs. These effects were more evident in the FA fractions from CS-amended soil than in those from TS-amended soil. In examining DOM fractions from two Brazilian Oxisols amended with biosolids consisting of aerobically and anaerobically digested and limed SL at a rate of 390 Mg ha 1, Bertoncini et al. (2005) noticed that components of relatively small molecular size, with a low level of aromatic polycondensation and low degree of humification present in the biosolid, were partially and variably incorporated into amended soil FA, thus modifying its structure and functions. In general, these modifications seemed to be smaller in FA of clayey soil than in that of sandy soil, possibly because of protective effects exerted by clay minerals against disturbances caused by biosolid application. The compositional, structural, and acid–base properties of the FA fractions of DOM isolated from a pig slurry (PS) and Calcic Luvisols cropped with barley after consecutive annual additions of PS at rates of 90 and 150 m3 ha 1 per year over a 4-year period under semiarid conditions have been elucidated (Plaza et al., 2003, 2005a). With respect to the unamended soil FA, PS–FA was characterized by large contents of S- and N-containing groups and polysaccharide components; smaller COOH and phenolic OH contents; larger affinities for proton binding by COOH groups and similar affinities for proton binding by phenolic OH groups; a prevalent aliphatic character; extended molecular heterogeneity; and small degrees of aromatic polycondensation, polymerization, and humification. The soil FAs were affected by PS amendment, showing a decrease of O content and proton affinity by phenolic OH groups and an increase of C, N, and S contents, C/N ratios, proton affinity of COOH groups, and aliphaticity. These effects generally increased with increasing cumulative amount of PS applied to soil over time, apparently modifying the content and properties of the FA fraction of DOM, including its effects on bioavailability and mobilization/ transport of micronutrient/microtoxic metal(loid) ions in soil. Using chemical and spectroscopic (FTIR and fluorescence) methods, the evolution of the composition, structure, and acid–base properties of the FA fraction of DOM isolated during co-composting of a mixture of olive oil mill sludge and tree cuttings was probed (Plaza et al., 2005b,c; 2007). At the initial stage of composting, the FA was characterized by a prevalent aliphatic character; large contents of C- and S-containing groups; proteinaceous materials and polysaccharide components; extended molecular heterogeneity; small O and acidic functional group contents; and small degrees of aromatic ring polycondensation, polymerization, and humification.

36

Nanthi S. Bolan et al.

However, as composting proceeded, C, H, and S contents, C/N ratio, aliphaticity, and proton affinity of COOH and phenolic OH groups decreased, whereas N, O, C/H, and O/C ratios, COOH and phenolic OH contents, and aromaticity increased, suggesting that the chemical and structural properties of FA approached those typical of native soil FA. Thus, co-composting was suggested to represent a suitable treatment for enhancing the agronomic quality of DOM in these materials when used as a soil amendment.

5.4. Soil pH Changes in soil pH resulting from soil acidification and lime application can affect both the concentration and chemical characteristics of DOM in soils (Clark et al., 2006; Guggenberger, 1994; Menneer et al., 2001; Nilsson et al., 2001; Oste et al., 2002). For example, while Anderson et al. (1994) noticed an increase in the leaching of DOM with increasing lime application to a Mor humus, Zech et al. (1994) and Guggenberger (1994) noticed increased solubilization of SOM with increasing rates of acid deposition. The effect of pH on DOM in soils has been attributed to pH-induced changes in the chemical solubilization and microbial decomposition of SOM and sorption and coagulation reactions of DOM (Guggenberger, 1994). Anderson et al. (1994) and Andersson and Nilsson (2001) noticed an increase in bacterial activity in the limed Mor humus, and chemical fractionation indicated that DOM from the limed treatment was more decomposed than DOM from the unlimed treatment, as the percentage of hydrophilic acids compared with hydrophobic acids was greater in the limed treatment. Anthropogenic soil acidification has been shown to result in an increase in DOM output of the mineral soil due to (i) greater DOM release in the forest floor, (ii) competition of DOM with sulfate for sorption sites, and (iii) loss of Al and Fe oxides/ hydroxides due to buffer processes (Guggenberger, 1994; Zech et al., 1994). Bolan et al. (1996) extracted DOM from soil, manure, and sludge samples amended to different pH values using HCl, NaOH, and Ca(OH)2 (Table 7). There was an increase in the amount of DOM with increasing pH. The effect was more pronounced with NaOH addition than Ca(OH)2 Table 7 Effect of pH on dissolved organic matter Dissolved organic matter (mg C kg 1) Carbon sources

Control

HCl

NaOH

Ca(OH)2

Pasture soil Poultry manure Sewage sludge

1.02 8.18 6.00

0.58 0.73 0.79

3.69 21.8 12.9

2.12 15.4 9.8

Dissolved Organic Matter

37

addition. Various reasons have been attributed to the decrease in the extractability of organic matter in the presence of polyvalent cations. These include formation of insoluble compounds and subsequent flocculation of organic molecules, complexation of organic molecules with polyvalent cations, and adsorption of organic molecules onto clay through cation bridging (Luo et al., 2008; Mayer and Xing, 2001). David et al. (1989) observed that acid treatment of an organic horizon of a spodosol has resulted in a decrease in DOM concentration, which they attributed to changes in the proportion of hydrophobic and hydrophilic compounds in the DOM. However, acidification of lakes has been shown to increase the concentration of Al and Fe (Dillon et al., 1988; Lydersen et al., 2002; Nilsson, 1985; Wallstedt et al., 2008), and the increase in Al input causes flocculation and subsequent removal of DOM, and thus the clarity of acidic lakes. At pH > 4, DOM may also co-precipitate with Fe(OH)3. Similarly, Schindler et al. (1992) and Lydersen et al. (2002) have obtained a negative correlation between DOM and Hþ concentration in lakes. The surface functional groups in DOM and the change in surface charge with pH have indicated that DOM is likely to coagulate at low pH values. This may be one of the reasons for the decrease in the concentrations of DOM at acidic pH values. Decreasing pH also increases the lipophilic nature of humic substances, which may partly explain the decline in DOM.

6. Environmental Significance of Dissolved Organic Matter in Soils 6.1. Soil aggregation and erosion control Nonliving and living organic matter components varying in chemical composition, decomposition state, and size promote and stabilize aggregations of soil particles at sizes ranging several orders of magnitude (Denef et al., 2002; Deurer et al., 2009; Zhang and Horn, 2001). The primary organic matter components contributing to the aggregation stability are believed to be free and occluded POC fractions and biomolecules synthesized and exuded by roots, mycorrhizal fungi, and other microorganisms in soils (Golchin et al., 1997). Thus, organic matter is essential in stabilizing the entire soil matrix. The mechanisms of soil structure stabilization occur over microscale to coalesce microaggregates into macroaggregates. Such microaggregation involves nonliving POC capable of spanning distances >100 mm or involving a network of a fungal hyphae and plant roots that literally enmeshes the microaggregates. Upon the death of roots and hyphae assemblages, their residue serves as a biochemical binder for macroaggregates (Annabi et al., 2007; Elmholt et al., 2008; Liu et al., 2005).

38

Nanthi S. Bolan et al.

The organic matter responsible to maintain soil aggregation is subject to biodegradation by microbial activity (Baldock, 2002). Thus, maintaining organic matter is essential to ensure the availability of the key binding agents that may promote in decreasing erosion. McDowell and Sharpley (2003) found that addition of up to 50 kg P ha 1 y 1 as dairy manure to a finely textured Hagerstown soil in Pennsylvania resulted in less particulate and total P and sediment loss in surface runoff compared with an untreated control. This was traced back to an improvement in soil aggregation that was correlated with hydrolyzable carbon from the manure. While inputs of C may enhance soil aggregation, cultivation may impart the opposite effect. For example, Chan et al. (1992) monitored soil aggregation for three tillage treatments: continuous cultivation, reduced cultivation, and direct drilling on a Wagga red earth under wheat–lupin rotation. Aggregation was 88, 72, and 58% for these three tillage treatments, respectively—the disparity being directly linked to cultivation causing a decrease in various organic matter components. Similarly De-Campos et al. (2009) noticed a diminution in aggregate stability of both cultivated and uncultivated soils with reducing conditions, which they attributed to the release of DOM and redox-sensitive elements such as Fe and Mn that act as a binding material for soil aggregation.

6.2. Mobilization and export of nutrients Dissolved organic matter influences the mobility and bioavailability of those nutrients that interact with organic matter and undergo biological transformation processes (Table 8). To elaborate on the role of DOM on biological transformation of nutrients and their subsequent mobilization in soils, it is prudent to understand, firstly, the distribution of nutrient elements in organic form and, secondly, the biochemical reactions involved in the biological transformation processes (i.e., mineralization/immobilization). 6.2.1. Distribution of Nutrients in Soils Nutrient elements occur in various fractions in soils that usually include (Shuman, 1991) a. Structural components of primary and secondary minerals (e.g., K in feldspar); b. Precipitated in inorganic forms, including those occluded by Fe, Al, and Mn oxides (e.g., occluded P); c. Complexed by organic matter (e.g., complexed Cu); d. Incorporated into organic matter including microbial biomass (e.g., biomass N, P, and S); e. Specifically adsorbed onto silicate clay minerals and Fe, Al, and Mn oxides (e.g., adsorbed P and Zn);

Table 8

Example of mean concentrations of dissolved organic C, N and P in relation to scale, land- use, management, and flow regime

Scale

Land use Treatment

Reference

Parfitt et al. (2009)

Pasture Pasture Catchment Pasture Forest

–a 41 kg P ha 1 y 1 – –

Seepage Seepage Streamflow Streamflow

38.0 20.0 9.6 13.8

0.70 1.30 1.40 1.20

Catchment Forest Forest Catchment Forest Forest

– – – –

Baseflow Stormflow Baseflow Stormflow

2.2 5.7 0.4 1.9

0.05 0.38 0.05 0.09

Europe

Mixed

–

Streamflow 7.6

0.58

Global

Mixed

–

Streamflow 5.9

0.43

Small plot

a

Flow regime DOC DON DOP Finding

Not measured or not applicable.

0.020 Increased fertility enriched losses of DOP and DON, but not DOC 0.120 0.150 DOP and DON accumulated during 0.060 summer and flushed in autumn, but DOC showed no seasonality – Flushing of DOC and DON in stormflow – 0.003 Flushing of DOC and DON in 0.002 stormflow, but DOP relatively immobile or pool exhausted 0.020 DOC controlled by land use and hydrological processes 0.018 Hydrologic processes dominant but human inputs of DON and DOP, and wetlands for DOC, important

Vink et al. (2007)

Inamdar et al. (2009) Zhang et al. (2007)

Mattsson et al. (2009) Harrison et al. (2005)

40

Nanthi S. Bolan et al.

f. Exchangeable form on the exchange sites of silicate clays and organic matter (e.g., exchangeable anions and cations); and g. Water soluble as free and complexed inorganic and organic ions (e.g., H2PO4, SO42, NO3, and NH4þ ions in soil solution). Significant quantities of N, P, S, B, Co, Mo, and Se occur in organic form (fraction d) that are subject to biological transformation in soils (Chepkwony et al., 2001; McDowell, 2003; Prietzel et al., 2001). Soil organic N constitutes 90–98% of total N in most soils. The organic N pool includes crude protein, simple protein, amino acids, amino sugars, and nucleic acids ( Jarvis et al., 1996; Stockdale et al., 1997). Soil organic P constitutes 30–50% of total P in most soils (Dalal, 1977), although some of the Andisols have up to 75 % of their total P content occurring mainly as mono-ester phosphate (Bricen˜o et al., 2004; Escudey et al., 2004). Organic P occurs mostly as esters of orthophosphoric acid, although direct C–P compounds called phosphonates are also important in some soils (McDowell and Stewart, 2005). Five classes of esters have been identified by 31P NMR (Condron et al., 1996). These include inositol phosphates, phospholipids, nucleic acids, nucleotides, and sugar phosphates. The main forms of organic S in soils include S-containing amino acids (cysteine, cystine, and methionine) and peptides (glutathione and g-glutamylcysteine), sulfonates (C–S), sulfate ester (C–O–S), and sulfamates (N–O– SO3 and N–SO3) (Saggar et al., 1998). Some of the B is associated with organic matter, partly by adsorption and partly through the reactions of boric acid with hydroxyl aliphatic acids and aromatic compounds containing o-dihydroxy groups (Stevenson, 1985). Molybdenum is present partly as organically bound Mo (Harter, 1991). Selenium occurs in organic forms partly as organic complexes and partly as Se-amino acid synthesized by microorganisms and plants (Gissel-Nielsen et al., 1984; Mayland, 1994). 6.2.2. Mineralization/Immobilization The mineralization–immobilization process regulates the release of nutrients that are associated with SOM, thereby affecting DOM-induced mobility and bioavailability. The mineralization process involves the conversion of plant unavailable organic forms of nutrients such as N, S, and P and some of the trace elements into plant available inorganic forms by soil microorganisms. Immobilization is the reverse process in which the plant available nutrients are converted to plant unavailable organic compounds. While N cycling is controlled mainly by biotic-induced reactions, both abiotic and biotic soil processes control P and S dynamics. Abiotic processes include physicochemical reactions such as phosphate and sulfate sorption– desorption and precipitation–dissolution. Biotic processes include the microbial assimilation of phosphate and sulfate into organic form and the mineralization of plant and animal residues by soil microorganisms through

Dissolved Organic Matter

41

extracellular hydrolytic enzymes, such as phosphatase and sulfatase enzymes that hydrolyze organic P such as ester phosphate into inorganic P and organic S such as sulfate esters to inorganic S (Ganeshamurthy and Nielsen, 1990; Sinsabaugh et al., 2009). 6.2.3. DOM-Induced Mobilization of Nutrients By definition, DOM constitutes not only C but also nutrients. Hence, DOM dynamics is an integral factor in the mobilization and loss of nutrients such as N, P, and S. For example, Qualls et al. (1991) found that about half of the dissolved N and P in throughfall of a forested catchment was in organic form. Although soil and management factors can influence the mobility of DOC, DON, and DOP, coupling them with soil physical and hydrologic processes facilitates losses to surface water and groundwater (Neff et al., 2000). Soils under more intensive management tend to maintain greater dynamic stocks of DON and DOP—touted as a decoupling of faster nutrient cycling from that of DOM (McDowell, 2003). Analysis of 93 sites of contrasting land use indicated that intensive dairy pastures contained the most DON, followed by sheep and beef pastures, cropping land, and lastly, forest (Ghani et al., 2007). Concentrations of DOC followed a similar pattern, but were not nearly as enriched or variable. The enrichment of DON can be attributed to the input of N via fertilizer, but can also be attributed to large N input and alkaline pH of soil beneath urine patches. The extreme pH and N enrichment beneath urine patches contributes to localized solubilization of organic matter and leaching losses of DON commonly > 100 kg N ha 1 y 1 (Wachendorf et al., 2005). Enhanced soil fertility via addition of superphosphate can not only increase the loss of DOP (McDowell et al., 2004), but also cascade into enriched DON losses in mixed pastures due to atmospheric fixation of N by legumes (Parfitt et al., 2009; Table 8). Another management scenario sees that enhanced nutrient losses occur when soils are limed. Increasing the pH of acid soils stimulates microbial diversity and functioning by mitigating metal(loid) toxicity and increasing concentrations of bioavailable carbon and nutrients via desorption and solubilization of SOM (Bolan et al., 2003). For example, addition of liming materials such as calcium carbonate (CaCO3), dolomite (CaMg(CO3)2), and magnesium carbonate (MgCO3) increased mineralization and the leaching of DOM and accumulated sulfur with decreasing Ca2þ concentration in the lime (Valeur et al. 2000). In addition to the mobilization of dissolved organic N and P via direct input, Kaiser et al. (2001), among others, noted that different forms (i.e., hydrophobic and hydrophilic fractions) can affect mobility in soil. Soil water under a Scots pine (Pinus sylvestris L.) and a European beech (Fagus sylvatica L.) forest has most of the DON and DOP concentrated in the

42

Nanthi S. Bolan et al.

hydrophilic DOM fraction, which was more mobile in pore water than the hydrophobic fraction. Changes can also occur as dissolved organic nutrients moving down the soil profile. Since DON is used as a substrate by soil microbes (Lajtha et al., 2005), the hydrophilic DON tends to be utilized faster than the hydrophobic DON. Coupled with sorption of hydrophobic DON to soil, these processes decreased DON concentrations with depth (van Kessel et al., 2009). In contrast, decreased sorption of DOP compounds compared with orthophosphate can promote the movement of DOP down the soil profile (e.g., Donald et al., 1993; Leytem et al., 2002; McDowell and Koopmans, 2006). While the enhanced mobility of some dissolved organic compounds has been established, this phenomenon may not translate into enriched losses when scaled up to hillslopes and catchments. At this larger scale, availability is morphed into an issue of supply when faced with hydrologic controls (Cooper et al., 2007). For example, at the plot-scale concentrations, dissolved organic nutrients tend to be much greater than at the catchment scale (Table 8) when their supply becomes exhausted and diluted through plant uptake and sorption during transport. Concentrations of DOC, DON, and DOP could be greatest during runoff events in late summer and autumn, but rapidly decrease as discharge increases during winter and spring (Vink et al., 2007). However, the issue of supply and dilution can be pathway and nutrient dependent. For example, Zhang et al. (2007) found that DOM and DON increased 10-fold during stormflow compared with baseflow, but DOP remained constant. This was attributed to both the high availability of DOC and DON in the forested catchment and the highly permeable soil allowing for pathways to explore deeper layers during storms. On DOC and DON loss, Cooper et al. (2007) offered a view that DOC losses are only significant through shallow organic soil horizons where DOM is enriched (McDowell and Wood, 1984) could be misleading. They concluded that losses may simply reflect the dominant soil water flowpath during runoff, rather than any direct influence on the release of DOM. In large catchments it may be difficult to distinguish any single dominant process or factor affecting dissolved nutrient loss. For example, in a survey of major European rivers, Mattsson et al. (2009) found that the concentration of DOM was controlled by a combination of factors including the proportion of wetland and forest cover, precipitation, and hydrological processes that led to greater loads in northern latitudes than near the Mediterranean Sea. Despite this mixture, they were able to show that concentrations and that DON and DOP loads increased with the proportion of agricultural land. Globally, Harrison et al. (2005) made a similar conclusion when highlighting the response to precipitation and hydrological processes for DOC, DON, and DOP loads in a survey of major rivers, but noted clear regional differences in the rate of loss due to anthropogenic inputs of DON and DOP and wetland influences on DOC.

Dissolved Organic Matter

43

Recently, increased effort has been expended on the study of DOC and DON losses with time, largely due to the potential impact of climate change and increasing atmospheric N deposition. Attention has largely focused on upland and forested areas where it is thought that increasing temperatures (Stottlemeyer and Toczydlowski, 2006), changing rainfall patterns (Worrall et al., 2003), and enriched atmospheric CO2 concentration (Hagedorn et al., 2008) may stimulate the release of C stored in these systems to air and water. Indeed, long-term records indicate increased loss of DOC and DON in Finland (Lepisto¨ et al., 2008), Michigan (Stottlemeyer and Toczydlowski, 2006), and Scotland (Cooper et al., 2007). Adjunct to this are heightened concerns such as enhancement of surface water eutrophication (Mitchell and Baldwin, 2005).

6.3. Bioavailability and ecotoxicology of heavy metals The transport and bioavailability of metal(loid)s can be strongly influenced by forming soluble and insoluble complexes with DOM (Table 9). Such interactions can alter the chemical speciation of the metal(loid)s modifying their affinity for sorptive surfaces in the soil matrix or their uptake, accumulation, and eventual toxicity to organisms (Arnold et al., 2010; Boyd et al., 2005; Vulkan et al., 2000). A number of studies have obtained positive correlation between DOM and metal(loid) concentrations in leachates, indicating that DOM is acting as a vehicle for the movement of metal (loid)s (Bhatt and Gardner, 2009; Zhao et al., 2007a,b). For example, Kalbitz and Wennrich (1998) noticed that the concentration of Cr, Hg, Cu, and As in the soil percolates was positively correlated with DOM. Strobel et al. (2001) investigated the kinetics of Cd and Cu release from an arable soil applied with forest floor soil solution isolated from Norway spruce (Picea abies (L.) Karst). Cadmium release rates were very low at pH >5 and increased exponentially as pH decreased to <5 and was not affected by DOM. In experiments without DOM, the release rate of Cu was slightly less at high pH than at low pH. In experiments above pH 5, the presence of 5 mM DOM in the solution increased the release rate of Cu. However, the Cu release was retarded by DOM in the pH range 3.8–5.0, which coincided with a maximum retention of DOM in the soil. Many have viewed DOM as an important contributor to the elevated mobility of trace metal(loid)s in soils amended with biosolids (Al-Wabel et al., 2002; Haynes et al., 2009; Peckenham et al., 2008). Ashworth and Alloway (2004) examined the extent of mobility of DOM and co-mobility of Cu, Ni, and Zn released from a surface application of anaerobically digested sewage sludge in a sandy loam. Leaching of DOM through the soil column was found to be almost unimpeded. Similar behavior was exhibited by Ni, suggesting that it migrated as organic complexes. Although Cu was also leached, significant retardation occurred. The presence of

Table 9 Selected references on DOM-induced mobilization of metal(loid)s DOM source

Metals

Urea-treated pine bark

Zn, Ni, and Pb

Observations

In the presence of DOM, sorption of Zn, Ni, and Pb by pine bark decreased by up to 75% depending on the DOM concentration Pig manure amendment Cu, Cd, Pb, and Zn Pig manure amendment increased DOM in leachates, thereby in mine soils increasing the release of metals from mine soil Hg The presence of DOM reduced Hg maximum adsorption capacity Humus soil (DOMH), with up to 40% over the control in Xanthi-Udic Ferralosols and rice straw (DOMR), sorption followed: DOMH (250.00 mg kg 1) < DOMR (303.03 and pig manure (DOMP) mg kg 1) < DOMP (322.58 mg kg 1) < control (416.67 mg kg 1) Sewage sludge Cu, Ni, and Pb The solubility of the heavy metals Cu, Ni, and Pb showed a strong amended soil positive relationship to the solubility of organic matter, particularly at high pH. Cd and Zn The addition of DOM significantly reduced the Cd and Zn Anaerobically sorption capacity with a maximum inhibition on metal sorption digested dewatered occurring at pH 7–7.5. sludge The kd values for acidic sandy loam soil in the absence and presence of DOM were 22.2 and 9.54 for Cd and 3.86 and 1.84 for Zn. The kd values for calcareous sandy loam soil in the absence and presence of DOM were 329 and 107 for Cd and 212 and 40.2 for Zn Poultry litter (PL) Cu and As Increased levels of DOM from the PL treatments enhanced As and Cu solubility through competitive sorption and complexation

Reference

Khokhotva and Waara (2010) Carmona et al. (2008) Yongkui et al. (2008)

Ashworth and Alloway (2008) Wong et al. (2007)

Jackson et al. (2006)

Sewage sludge

Cu and Zn

Purified peat humic As acid and peat drainage Untreated wastewater Cu and Cd

Biosolid

Cu, Zn, and Pb

Municipal solid waste compost

Cd

Wood ash

Cu and Ni

Cu and Zn sorption capacity decreased in the presence of DOM. The kd values for Cu without and with DOM were 121.20 and 36.88 and for Zn the values were 33.58 and 14.825 for Zn, respectively Sorption of DOM induced the mobilization of As from soil and sediments Long-term irrigation with wastewater increased the mineralizable carbon fraction and DOM concentrations. The water-extractable Cu and Cd concentrations also increased and correlated with DOM Biosolid application increased DOM, Pb, and Cu concentrations in the leachates, resulting in a positive correlation between DOM and Cu and Pb, indicating the formation of soluble DOM–metal complexes. Zn mobility was facilitated by the formation of inorganic complexes. Anion stripping voltammetry measurements indicated that only a small percentage of the total dissolved metals existed as free ions and the remainder was complexed Humic acid (HA) and fulvic acid (HA) in compost achieved a high sequestration of Cd. The metal complexing capacity (CC) of HA and FA was CCHA (Cd) ¼ 2386, and CCFA (Cd) ¼ 2468 mmol Cd g 1 DOM The land application of wood ash increased soil pH, solubilizing soil organic matter and enhancing metal mobility. Although the mobility of both Cu and Ni increased at low level of wood ash application by forming soluble DOM–metal complexes, both these metals formed insoluble complexes at high rates of wood ash application resulting from high ionic strength

Mesquita and Carranca (2005)

Bauer and Blodau (2006) Herre et al. (2004)

Al-Wabel et al. (2002)

Kaschl et al. (2002)

Chirenje et al. (2002)

(continued)

Table 9

(continued)

DOM source

Metals

Observations

Reference

Soil

Cu, Pb, Cd, Ni, and Zn

Impellitteri et al. (2002)

Sewage sludge

Zn, Cd, Cu, Ni, and Cr

Sewage sludge

Cd, Ni, and Zn

Sewage sludge

Cu, Zn, Sr, Rb, Mo, Cd, As, Cr, Ni, Sb, W, Ag, Hg, and Sn

DOM increased with increasing pH with a corresponding increase in Cu and Pb concentrations. Correlation analysis strongly suggests that HA plays a major role in increasing the concentration of solution Cu and Pb with increasing pH. In acidic environments, FA plays a larger role than HA in governing organo-metallic interactions. The effect of DOM on metal mobilization is more pronounced at high pH than at low pH The concentrations of Zn, Cd, Cu, Ni, and Cr in the saturation extract closely correlated with the concentrations of DOM. Considerable amounts of Zn and Cd from sewage sludge were found in the mobile fractions of the soil with Cu, Ni, and Pb in organic particles DOM applications significantly increased the extractability of metals and also their uptake by ryegrass The distribution coefficients, kd, for the metals in newly prepared sludge/soil mixtures were lower than kd values of the field-aged sludge-treated soil, suggesting that metal mobility is likely to be higher shortly after sludge application than many years later, which is attributed to formation of soluble DOM–metal complexes in the former. The kd values of Cd, Cu, and Zn in the presence of 20% sludge are 139, 221, and 293 L kg 1 and the kd values for the control site are 2500, 360, and 4800 L kg 1.

Schaecke et al. (2002)

Antoniadis and Alloway (2002) McBride et al. (1999)

Biosolids

Cu

Alkali-stabilized sludge

Cu, Ni, Mo, Cd, Zn, Pb, Ag, and Hg

Wetland soils

Cr, Hg, Cu, and As

Han and Thompson The Cu-binding ability of DOM decreased with increasing (1999) molecular weight, indicating that LMW DOM had more metalbinding sites than HMW DOM. The hydrophilic and the hydrophobic components, within each MW fraction, also exhibited differences in Cu-binding ability. For the DOM with MW 500–3500 Da, the hydrophilic fraction showed a greater Cu-binding capacity than did the hydrophobic fraction Alkaline stabilization of sludge increased pH, thereby resulting in McBride (1998) the release of metals; 40–50% of the total Cu, 15–25% of the total Ni, and 11–13% of the Mo dissolved out of the sludge product into water, whereas much smaller fractions of the Cd, Zn, Pb, Ag, and Hg were immediately water soluble The concentration of Cr, Hg, Cu, and As in the soil percolates was Kalbitz and Wennrich positively correlated with DOM (1998)

48

Nanthi S. Bolan et al.

DOM did not prevent Zn from becoming completely adsorbed by the soil solid phase. Al-Wabel et al. (2002) noticed that biosolid application increased both DOM and Cu concentrations in the leachates, resulting in a positive correlation between Cu and DOM across application rates. Both Cu and Pb were mobilized under conditions of low EC. This may be the result of the release of a strong metal(loid)-binding component of DOM under these conditions. Conversely, Zn mobility was positively correlated with EC, suggesting that either cation exchange or the formation of inorganic complexes influences Zn mobility. Anodic stripping voltammetry measurements indicated that only a small percentage of the total dissolved metal(loid)s existed as free ions or inorganic complexes; the remainder appears to be complexed with DOM. Han and Thompson (1999) noticed that Cu-binding ability of DOM decreased with increasing molecular weight, indicating that LMW DOM had more metal(loid)-binding sites than HMW DOM. Within each MW fraction, the hydrophilic and the hydrophobic components also exhibited differences in Cu-binding ability. For the DOM with MW 500–3500 Da, the hydrophilic fraction showed a greater Cu-binding capacity than did the hydrophobic fraction, whereas the hydrophobic acid components were most important in binding Cu for DOM with MW > 3500 Da. The maximum Cu-binding capacities of different biosolid-derived DOM fractions ranged from 1.85 to 14.3 mmol Cu mol 1 DOM. Kaschl et al. (2002) and Titeux et al. (2002) estimated the metal(loid) complexation capacity (CC) of various DOM fractions such as humic acid (HA), fulvic acid (FA), and DOM. The complexation capacity (CC) was greatest for Cu: CCHA ¼ 3357 and CCFA ¼ 5221 mmol Cu g 1 DOM at pH 5. Zinc and Cd were bound (at pH 7) in smaller concentrations: CCHA (Zn) ¼ 2167, CC FA (Zn) ¼ 2809, CCHA (Cd) ¼ 2386, and CCFA (Cd) ¼ 2468 mmol metal g 1 DOM. Sposito et al. (1988) studied Cu complexation by a chestnut leaf litter extract in the pH range 4–7 using fluorescence, IR, and ESR spectroscopies. The increasing fluorescence intensity quenching measured with increasing pH suggested increasing Cu complexation. The IR spectra gave evidence for Cu complexes involving COO groups, whereas the much more sensitive ESR spectra showed that Cu(II) could displace Fe(III) and Mn(II) from the litter extract binding sites and that inner-sphere Cu(II) complexes were formed involving COO groups and H2O at pH<6 and O- and N-containing ligands at pH 6 and 7. Senesi et al. (1977a,b) applied ESR and Mossbauer spectroscopies combined with chemical treatments including acid hydrolysis, reduction with hydrazine, and complexation with ammonium thiocyanate to study the oxidation states and symmetries of Fe bound by soil FAs. At least two, possibly three, different binding sites for Fe(III) were found to occur in FAs:

Dissolved Organic Matter

49

(a) tetrahedral and/or octahedral sites exhibiting considerable resistance to chemical replacement and reduction and (b) external surface octahedral weak sites where Fe(III) could be easily reduced and removed. Most of Fe (III) ions additionally reacted with FA were bound to surface octahedral sites. Senesi and Sposito (1984) and Senesi et al. (1985a,b) have studied the complexation of Cu(II) ions to the FA fraction of DOM isolated from soil and sewage sludges (SL) by using ESR spectroscopy. The ESR spectra of purified soil and SL–FAs indicated the presence of residual Cu(II) ions bound to both O- and N-ligands of FA in two or three different sites as inner-sphere polydentate complexes with a tetragonal (distorted octahedral) arrangement. The principal binding sites of soil FA involved 4O-ligands, whereas that of SL–FA involved 3O-1N sites. Both soil and SL–FAs were shown to possess a high and increasing residual binding capacity for additional Cu(II) ions in solutions at increasing Cu/FA molar ratios by forming inner-sphere complexes involving less and less reactive functional groups. Plaza et al. (2005a) and Hernandez et al. (2007) evaluated the effects of annual additions of pig slurry (PS) at rates of 90 and 150 m3 ha 1 y 1 for four and seven consecutive years, respectively, on the binding capacities for Cu(II) and Zn(II) ions of the FA fraction of DOM from PS and PSamended Calcic Luvisol. The binding capacities of FAs and the strengths of metal(loid) ion–FA complexes were larger for Cu(II) than for Zn(II). The binding capacities and stability constants of PS–FA were smaller than those of unamended soil FAs, whereas the values of PS-amended soil FAs were intermediate between those of unamended soil FA and PS–FA, but tended to decrease with increasing amount of PS applied to soil and the number of years of PS application, thus suggesting an increasing partial incorporation of FA fractions of PS into native soil FA. Certain soil amendments, by virtue of their ability to change soil pH, may impact on DOM and DOM-induced metal(loid) dynamics. Land application of wood ash increased soil pH, thereby solubilizing SOM and enhancing metal(loid) mobility (Chirenje and Ma, 1999; Chirenje et al., 2002). Although the mobility of both Cu and Ni increased at low rates of wood ash application by forming soluble DOM–metal(loid) complexes, both these metal(loid)s formed insoluble complexes at high rates of wood ash application resulting from high ionic strength. Xiao et al. (1999) demonstrated that alkaline ash material increased the leaching of metal(loid)s (Cd, Cr, Cu, Pb, Se, and Zn) due to high pH resulting in the solubilization of organic matter, thereby resulting in the release of DOM, whereas organic residues that are enriched in solid phase organic matter resulted in an increase in metal(loid) retention in soils. Addition of organic matter-laden soil amendments can enhance the reduction of metal(loid)s such as Cr and Se (Herbel et al., 2007; KnotekSmith et al., 2006; Kumpiene et al., 2008). Adding cattle manure to soil

50

Nanthi S. Bolan et al.

enhanced the reduction of Cr(VI) to Cr(III) (Cifuentes et al., 1996; Higgins et al., 1998; Losi et al., 1994). Similarly, powdered leaves (Suseela et al., 1987) and Pinus sylvestris bark (Alves et al., 1993) have been used to remove Cr(VI) from industrial effluents. The reduction of Cr(VI) in the presence of the organic manure composts is caused primarily by the supply of carbon and protons and the stimulation of microorganisms (Losi et al., 1994). For example, the hydroquinone group in organic matter is a major source of electron donor for the reduction of Cr(VI) to Cr(III) in soils (Elovitz and Fish, 1995). Easily oxidizable organic carbon in DOM also provide the energy source for soil microorganisms involved in the reduction of N (i.e., nitrate (NO3) to gaseous nitrogen (NO, N2O, and N2) via denitrification) ( Jardine et al., 1999; Paul and Beauchamp, 1989). Although the addition of manure induces the remediation of Cr-contaminated soils by reducing mobile toxic Cr(VI) to nontoxic, less mobile Cr (III), it is likely that biotransformations of aromatic arsenic and mercuric compounds could also occur, but may result in the production of more toxic inorganic species (Effler and Matthews, 2008; Ghosh and Bhattacharyya, 2004; Kumpiene et al., 2008).

6.4. Transformation and transport of organic contaminants Known for its high affinity for non-ionic hydrocarbons and organic pesticides, DOM can alter their sorption onto the soil matrix, transformation/ degradation, and mobility (Table 10). Addition of DOM affects pesticide sorption, degradation, and transport. It can form complexes with pesticides and/or can compete with pesticides for sorption sites enhancing pesticide mobility. However, these two processes can be counteracted by DOM– pesticide complexes adsorbing to the soil matrix (co-sorption) and/or by DOM adsorbing to soil surfaces, which in turn leads to the formation of an additional sorption domain (cumulative sorption), reducing pesticide mobility. Dissolved organic matter has been implicated in the degradation and transport of pesticides since the 1970s (Ballard, 1971). Initial emphasis was on DOM that was derived from naturally occurring organic matter, but more recently, with the advent of organic waste disposal on land, both endogenous and exogenous DOM have been demonstrated to play a role in these processes. Because of the environmental importance of exogenous DOM, the two types of DOM will be treated separately here. 6.4.1. Endogenous DOM The initial eminent role of DOM in pesticide mobility occurred in early 1970 when urea solubilized SOM causing greater DDT movement in soils (Ballard, 1971). By the early 1980s, Means and Wijayratne (1982) had clearly identified the increased sorptive capability of small organic particles.

Table 10

Selected references on DOM-induced mobilization and degradation of organic pesticides

Pesticide

DOM source

Key finding

Reference

Atrazine, linuron

Natural DOM

On an OC basis DOM 10–35 times more sorption

DDT, lindane

Natural DOM

Bromacil, diuron, chlorotoluron, simazine, diquat, glyphosate Napropamide, lindane, prometryn, DDT

Natural DOM

DOM enhanced solubility controlled by DOM size and polarity DOM with mass in range of 700–5000 Da was involved in sorption process

Means and Wijayratne (1982) Chiou et al. (1986)

Aldicarb, carbofuran, carbaryl Bromacil, metribuzin, alachlor, paraquat, diquat Dimefuron, atrazine, carbetamide

Atrazine, simazine, diazinon, methylparathion, chlorpyrifos

Humic acid from peat þ other sources Natural DOM DOM from four mineral soils Endo- and exogenous DOM, many sources Natural DOM

Madhun et al. (1986)

Sorption to DOM ranged from 6% to 70% depending Lee and Farmer (1989) on pesticide chemical properties DOM enhanced photodegradation of pesticides

de Bertrand and Barcelo´ (1991) Pennington et al. (1991)

Non-ionic pesticides did not bind with DOM, ionic pesticides did but it did not affect their binding to the soil OC A new, empirical, two-compartment model provided Barriuso et al. (1992) good description of desorption isotherms

DOM facilitated hydrolysis of organophosphates in water but not the triazines

Noblet et al. (1996)

(continued)

Table 10 (continued) Pesticide

DOM source

Key finding

Reference

Carbofuran, carbaryl, aldicarb

Natural DOM from forest to stream Natural DOM

DOM from the forest (top of the drainage sequence) had largest capacity to bind carbofuran, binding of other pesticides much less Pesticides bound to DOM giving higher concentration of pesticide in pore water and decreased kd DOM-bound herbicide was not extractable but still bioavailable Herbicide mobility greater in amended soil

Fang et al. (1998)

Atrazine, bifenox metabolites of both Amitrole Napropamide

Natural DOM

Isoproturon, simazine

Sewage sludge amended soil Natural DOM

Atrazine, prometryn

Effluent DOM

Napropamide

DOM from three soils Exogenous DOM The organic amendments modified the mobility and persistence of both herbicides Various animal Increased residence time on the treatment lagoons effluent DOM decreased the amount of DOM and its sorptive capacity Treated effluent Alkaline effluent enhanced dissolution of OC but DOM decreased flow rates which allowed increased desorption time with the coupled effect generally enhancing the movement of triazine herbicides

Simazine, 2,4-D Chlorpyifos

Atrazine, prometryn

Gao et al. (1998) Klaus et al. (2000) Nelson et al. (1998)

Lysimeter studies showing pesticide leaching occurred Worrall et al. (1999) in the first few samples following rainfall Seol and Lee (2000) For moderately polar triazines, sorption data and modeling indicate DOM having little effect on mobility DOM facilitated the transport of napropamide Williams et al. (2000) Cox et al. (2001) Huang and Lee (2001)

Seol and Lee (2001)

Imidacloprid MCPA

Atrazine, 2,4-D, isoproturon, paraquat Atrazine

Atrazine, terbuthylazine, simazine, prometryn

Atrazine

Exogenous DOM Mobility of this pesticide was increased in the presence of the exogenous DOM Humic substances Humic acids increased the movement of MCPA, while fulvic acids had the opposite effect extracted from other soils Natural DOM DOM had little or no effect on the sorption and transport of these pesticides in five soils with low-tohigh OC content Natural DOM The higher the soil OC content the higher the atrazine affinity to the soil phase to the extent that for soils with > 7% OM the atrazine–DOM complexes appeared to play a negligible role and the atrazine– DOM complex decreased the mobility of atrazine if the lower horizons have less OC than the upper ones Natural DOM Light had little effect on the removal of the herbicides from river water but had a marked effect on their removal from sea water and groundwater, with pesticide removal inversely proportional to DOM concentration Sewage sludge Low concentrations of DOM increased sorption of atrazine, while higher concentrations of DOM decreased sorption. Further, the hydrophobic fraction of the DOM promoted sorption, while the hydrophilic fraction decreased atrazine sorption

Flores-Cespedes et al. (2002) Haberhauer et al. (2002)

Spark and Swift (2002)

Ben-Hur et al. (2003)

Navarro et al. (2004)

Ling et al. (2005)

(continued)

Table 10

(continued)

Pesticide

DOM source

Various

Effluent irrigation Reports on the inconsistency of findings in the literature with both reduced and increased mobility on the presence of effluent-derived DOM; this was partly attributed to spatial variation and heterogeneity of the effluent quality Olive waste Two types of waste product from olive processing decreased diuron sorption in a clay soil but increased sorption in a sandy soil. However, the formation of complexes with the DOM resulted in earlier breakthrough of diuron in the sandy soil in column leaching studies Cow manure Very high application rates of liquid manure decreased the sorption of atrazine due to competition for bonding sites Lakebed sludge Both types of DOM reduced the sorption capacity of and rice straw the soil and increased the migration of prometryn in soil columns Fresh and mature Increased leaching of diuron occurred with higher concentrations of DOM, while the degree of organic maturity of the DOM had variable effects amendments

Diuron

Atrazine

Prometryn

Diuron

Key finding

Reference

Mu¨ller et al. (2007)

Cox et al. (2007)

Briceno et al. (2008)

Jiang et al. (2009)

Thevenot et al. (2009)

Dissolved Organic Matter

55

They found that organic particles, which passed through a 0.45 mm filter, were 10–35 times better, on a weight basis, at sorbing atrazine and linuron as particles > 0.45mm. Chiou et al. (1986) later added the relatively waterinsoluble DDT to the list. The DOM-normalized equilibrium-partitioning coefficient (kDOM) values for humic acids were approximately four times greater than those for fulvic acids, although molecular size and source of the organic matter were also important factors. However, they also found that the effect of DOM on the solubility of the relatively water-soluble pesticide lindane was minimal. Working with DOM extracted from a high organic carbon (27%) peat soil, Madhun et al. (1986) demonstrated for the first time the binding of pesticides to DOM. The DOM involved was quantified as having a mass range of 700–5000 Da and to closely resemble fulvic acids. For the herbicides bromacil, diuron, simazine, and chlorotoluron, they reported sorption about 70 times greater in DOM soil solution compared with soil. They concluded that this would likely play an important role in the mobility and transport of pesticides. Conversely, Pennington et al. (1991) found that, with DOM derived from soils containing relatively low organic matter contents, there was no measurable binding of the herbicides bromacil, metribuzin, or alachlor. The nonresidual herbicides diquat and paraquat were adsorbed by DOM, although this was not preferred to sorption on the soil and therefore the presence of DOM in these soils would not be a mechanism to significantly increase the mobility of these two herbicides. Lee et al. (1990) showed that the presence of DOM decreased the sorption of non-ionic pesticides onto clays. Flores-Cespedes et al. (2002) demonstrated that DOM reduced imidacloprid adsorption to the soil and increased its mobility by about 20%. The nature of the DOM has also been demonstrated to have a marked effect on the sorption and mobility of pesticides. Thomsen et al. (2002) found significant differences in kDOM for esfenvalerate and eight distinct humic materials. Haberhauer et al. (2002) showed that, in a low SOM content sandy soil, the presence of humic acid in the added water slightly increased the mobility of MCPA, while the presence of fulvic acid decreased mobility. However, as both humic and fulvic acids are readily found in naturally occurring organic matter, they might effectively cancel each other out. This appears to be an explanation for the results of the study by Spark and Swift (2002) who found that the naturally occurring DOM in five soils of medium-to-high SOM content had little or no effect on the sorption/transport characteristics of four diverse pesticides, namely, atrazine, 2,4-D, isoproturon, and paraquat. Ben-Hur et al. (2003) studied the behavior of atrazine in several soils containing different concentrations of SOM. They found that the greater the SOM concentration, the greater the affinity of atrazine to the soil phase. The resulting atrazine–DOM complexes decreased the mobility of atrazine.

56

Nanthi S. Bolan et al.

Testing carbamate pesticides, aldicarb, carbofuran, and carbaryl, Fang et al. (1998) found that DOM collected from different locations in the drainage sequence from forest to stream had different binding abilities, with the large molecular weight fractions (>10,000 Da) generally having greater binding capacity. Overall, they found that carbofuran had the greatest binding strength and carbaryl the weakest and concluded that the potential mobility of these carbamate pesticides was significantly enhanced by the presence of DOM in the water. In contrast, napropamide and anilazine mostly formed complexes with DOM in the range of 500– 10,000 Da (Klaus et al., 2000; Williams et al., 1999, 2000). Further to this, Gao et al. (1998) found that DOM had a significant effect on the adsorption–desorption of atrazine and bifeno; they were bound to soluble DOM in the pore water, which resulted in a decreased kd. Such “solubilization-effect” may be important in enhancing desorption and facilitating the transport of pesticides. Dissolved organic matter may exert profound influence on the transformation of certain pesticides. For example, chlorpyrifos exhibited a 32% decrease in hydrolysis rate in the presence of 34.5 mg L 1 DOM, while atrazine, simazine, diazinon, and methylparathion were unaffected (Noblet et al., 1996). In real environmental situations, the presence of DOM enhances the photodegradation of carbaryl and carbofuran (De Bertrand and Barcelo´, 1991). Similarly, Gerecke et al. (2001) demonstrated DOMmediated phototransformation of isoproturon and diuron in sunlit surface waters, while Sakkas et al. (2002) observed a similar result for chlorothalonil. Conversely, Bachman and Patterson (1999) found that increasing concentrations of DOM decreased photolysis of carbofuran and attributed this to the binding of the pesticides to DOM. 6.4.2. Exogenous DOM The addition of organic amendments to the soil introduces another level of complexity and usually alters normal herbicide behavior. In soils amended with sewage sludge 2 years previously, the presence of greater concentrations of DOM doubled the amount of napropamide herbicide in the breakthrough leachate (although < 1.5% of applied herbicide), but the overall greater concentrations of SOM halved the depth at which the center of mass of herbicide was found (Nelson et al., 1998). DOM derived from sewage sludge varied in its sorption ability of atrazine depending on the concentration of DOM and on the way the sludge was treated (Celis et al., 1998; Ling et al., 2005). Generally, low concentrations of DOM increased atrazine sorption on soil but was reduced by higher concentrations. Furthermore, hydrophobic fractions promoted sorption, while the hydrophilic fraction decreased atrazine sorption. Briceno et al. (2008) found that DOM from cow manure decreased the sorption of atrazine. Recent studies indicate reduced sorption and increased mobility of prometryn and diuron in

Dissolved Organic Matter

57

the presence of DOM from various organic amendments (Cox et al., 2007; Jiang et al., 2009; Thevenot et al., 2009). Exogenous DOM from the liquid waste formed associations with 2,4-D but not with simazine. But solid waste generally had less effect on the soil half-life than when the exogenous carbon was in liquid form (Cox et al., 2001). However, the effects of the liquid amendment were disparate, increasing the half-life of 2,4-D but decreasing that of simazine. The enhanced half-life of the former was attributed to the strong association of the more polar 2,4-D with the increased concentrations of DOM in the liquid waste. The association of atrazine and prometryn with DOM extracted from municipal wastewater, swine-derived lagoon wastewater, and HA on sorption by silt loam and sandy loam soils was examined by Seol and Lee (2000). The association of pesticide with DOM normalized to organic matter ranged from 30 to 1000 L kg 1. DOM up to 150 mg C L 1 did not significantly suppress sorption by soils of either atrazine or prometryn. However, in batch experiments with two silt loam soils amended with DOM from poultry, swine and cow effluents and a humic acid, Huang and Lee (2001) found that the hydrophobic insecticide, chlorpyrifos, had a strong affinity for the added DOM, resulting in decreased sorption coefficient for soils and enhancing the potential for DOM-enhanced transport of the pesticide. They noticed that the characteristics of the effluent-derived DOM were not constant—changing with residence time in the waste stabilization ponds. Quantifying the interactions between pyrene and DOM fractions, pyrene partition coefficients (kDOM) of the smaller DOM fractions varied between 4.1 1.03 and 6.8 103 L kg 1, while partition coefficients of the largest fraction of DOM, HA, and FA were 1.5 104, 1.7 105, and 1.1 104 L kg 1, respectively (Herbert et al., 1993). Pyrene partitioning data of the largest DOM fraction suggest that the presence of colloidal organic matter suspended in the soil solution may have a large influence on the transport of non-ionic organic compounds. DOM from several sources was examined as sorbent and potential carrier for hydrophobic polycyclic aromatic hydrocarbons (PAHs) in soil (Raber and Ko¨gel-Knabner, 1997). Partition coefficients (log kDOM) of PAH compounds were 4.8–4.9 for DOM from soil, 4.5–4.7 from compost, and 4.3–4.4 from sewage sludge. Leachates from waste disposal site did not sorb PAHs. The DOM from compost contained a large percentage of organic molecules > 14,000 Da (32–46%), whereas DOM from waste disposal leachate contained only 7–10%, thereby binding less PAHs. Interactions of DOM with a number of herbicides were examined by Pennington et al. (1991). The amount of paraquat bound by DOM ranged from 1.1 to 2.1 mmol g 1 DOM with kd values from 0.050 to 0.187 L kg 1. Diquat was bound at 0.9–1.5 mmol g 1 DOM with kd values from

58

Nanthi S. Bolan et al.

0.044 to 0.143 L kg 1. Bromacil, metribuzin, and alachlor did not bind to the extracts tested. DOM did not increase paraquat solubility in the presence of soil. Binding of these herbicides to DOM in soils used in this study would not be a significant mechanism for increased mobility and groundwater contamination potential. Partitioning studies in interstitial (pore) water of sediments obtained from three Minnesota lakes showed that DOM-bound dieldrin was generally > 30% when pore water DOM concentrations were >20 mg L 1. Pore water DOM partition coefficients (kDOM) were similar among the sediments (Kosian et al., 1995). In summary, the interactions of DOM with pesticides are influenced by many different factors including (i) source, concentration, pH, age, polarity, and molecular configuration of the DOM; (ii) soil properties; (iii) the chemical and physical properties of the pesticide; and (iv) interactions between pesticide–soil, pesticide–DOM, and DOM–soil. This often yields inconsistent results with regard to sorption and pesticide transport.

6.5. Gaseous emission and atmospheric pollution Occurrence of readily decomposable organic matter such as DOM in soil is critical in controlling gaseous emissions from soils through its influence on denitrification of N and methylation reactions of metal(loid)s such as As, Pb, Hg, and Se (Beauchamp et al., 1989; Michalzik et al., 2007; Payne, 1981; Reddy et al., 1982; Robertson and Tiedje, 1984). In the case of N, permanent pastures and forest soils develop surface layers rich in organic material with potential for denitrification of N added through biological N fixation, fertilizer and manure application, and deposition of dung and urine during grazing (Ryden, 1986; Zhang et al., 2009). The presence of ample C substrate can also result in rapid O2 consumption and possible O2 depletion, which may then indirectly enhance the potential for denitrification and methylation (Dodla et al., 2008; Firestone, 1982). Decomposition of plant litter and animal feces and root exudates from the perennial plant cover maintains moderate-to-high concentrations of available carbon including DOM in grazed pasture and forest soils for denitrification (Carran et al., 1995; Nishina et al., 2009; Royer et al., 2007). For example, increased rates of denitrification from urea fertilizers compared with ammonium-based fertilizers (e.g., DAP) have often been attributed to the direct supply of DOM by the urea fertilizer and the solubilization of soil carbon resulting from an increase in soil pH caused by ammonification of urea fertilizer (Barton et al., 1999). Similarly, application of farm effluents such as dairy and piggery effluents has been shown to increase the soil C availability, thereby enhancing the N2O emissions (Bhandral et al., 2007). Other researchers have also attributed increased N2O emissions to the C concentration of animal slurries (Barton et al.,

Dissolved Organic Matter

59

1999; Chadwick et al., 2000) and DOM concentration in plant residues with different C:N ratios (Huang et al., 2004). However, the presence of LMW DOM fractions, not the concentration of DOM, may be an important factor controlling N2O release in upland and peat soils after a major disturbance such as clear-cutting and site preparation (Saari et al., 2009). The accessibility of C supply to microorganisms also has a strong regulatory effect on denitrification, especially in field conditions. The rate could be limited by the diffusion of organic compounds in some soils, which is likely to be faster in the case of DOM when compared with other more stable C sources such as lignin and cellulose in soils (Myrold and Tiedje, 1985). Organic C and DOM concentration decreases with depth in most mineral soils mainly due to adsorption at the soil surface. Thus, although leaching of NO3–N into lower horizons is a common phenomenon in most agricultural soils, the availability of organic C is usually one of the main factors limiting denitrification activity in subsoils (Brye et al., 2001; Luo et al., 1996; Parkin and Meisinger, 1989). The availability of C has also been reported to influence the proportion of N2O and N2 produced. Limited supply of C is likely to cause partial denitrification, resulting in the release of intermediate gases such as NO and N2O. Thus, it is a general content that increasing C availability decreases the ratio of N2O:N2 (Arah and Smith, 1990; Dendooven et al., 1998; Smith and Tiedje, 1979). Indeed, Bhandral et al. (2007) and Vallejo et al. (2006) noticed a decrease in N2O:N2 with increasing concentration of DOM in various sources of effluents. Volatilization of metal(loid)s occurs through microbial conversion to their respective metallic, hydride, or methylated forms. These forms have low boiling points and/or high vapor pressure and are therefore susceptible to volatilization. Methylation is considered to be the major process of volatilization of As, Hg, and Se in soils and sediments, resulting in the release of poisonous methyl gas (Cernansky et al., 2009). Although methylation of metal(loid)s occurs through both chemical (abiotic) and biological processes, biological methylation (biomethylation) is considered to be the dominant process in soils and aquatic environments. For example, biological methylation is effective in forming volatile compounds of As such as alkylarsines, which could easily be lost to the atmosphere (Cernansky et al., 2009; Wood, 1974). At present there is substantial evidence for the biomethylation of Se in soils and aquatic systems (Losi and Frankenberger, 1997). Under anaerobic conditions Hg2þ ions can be biologically methylated to form either monomethyl or dimethyl mercury (Bisogni and Lawrence, 1975). Methylated mercury species are highly toxic and are more biologically mobile than the other forms. A number of studies have shown that the addition of organic matter-rich soil amendments enhances the reduction of metal(loid)s such as Se, thereby enhancing their gaseous emission (Cifuentes et al., 1996; Higgins et al.,

60

Nanthi S. Bolan et al.

1998; Losi et al., 1994; Michalzik et al., 2007). For example, Zhang and Frankenberger (1999) observed that the application of both insoluble (casein) and soluble (casamino acids) organic amendments to Se contaminated soil enhanced dimethyl selenide (DMSe) production and removal of Se from soil through volatilization. Similarly, Michalzik et al. (2007) and Holloway et al. (2009) observed a positive relationship between DOM and methylation of Hg and organo-metal(loid) compounds, respectively. However, complexation of metal(loid)s such as Hg with DOM decreases the concentration of free metal(loid) ions, thereby decreasing the rate of net methylation (Miskimmin et al., 1992).

7. Summary and Research Needs DOM is considered the most mobile and reactive component of the SOM pool despite its occurrence in only small quantities in soils. As such, it regulates major physical and biogeochemical processes not only in soils but also in aquatic ecosystems. This includes processes such as transformation and transport of essential nutrients (e.g., N, P, and S) and organic pesticides (e.g., non-ionic organic herbicides and insecticides), complexation of environmentally important heavy metal(loid)s (e.g., Cu, Pb, Hg, and Cd), sorption of metal(loid)s and organic compounds on clays and metal oxides, and mobility and transport of environmental contaminants. Although DOM is largely a by-product of microbial decomposition, it also serves as an energy substrate for the microbial processing of organic compounds, including indigenous organic matter. Because of these complex functions, DOM can be viewed to serve a dual function in the environment—as a link and as a bottleneck for various ecological processes. For example, as a link, DOM plays a vital role in mobilizing nutrients and heavy metal(loid)s to groundwater and surface water; as a bottleneck, it minimizes the bioavailability of certain inorganic (e.g., heavy metal(loid)s) and organic pesticides to terrestrial and aquatic biota. The nature (i.e., primarily the biochemical composition) and magnitude of DOM may serve as a sensitive indicator of the extent and shift in biogeochemical processes such as those impacted by land use and management practices (e.g., harvesting, conversion of forest to arable land, liming, and fertilization), biosolid (e.g., sewage sludge and livestock manure) application, moisture regime, landscape scale, and acid rain. On a global scale, DOM can also play a vital role in the degradation of endogenous and exogenous organic matter affecting C sequestration and N and P transformation and mobility in soils. Thus, DOM can mediate transformation of organic C into CO2 and CH4 as well as biological reduction of nitrates into N2 and N2O—greenhouse gases that interplay

Dissolved Organic Matter

61

in the global climate change. Thus, from its role in soil formation (e.g., podzolization) and aggregation (i.e., DOM acting as binding agent for microaggregates), mobilization and transport of nutrients, heavy metal (loid)s, and organic pesticides, catalysis of organic matter, and oxidation and reduction of nitrates, it is quite evident that DOM can affect biogeochemical processes from micro- to macroscale. Because DOM dynamics transcends continents, oceans, and landscapes, its overall influence on the cycling and fate of nutrients, organic pesticides, and metal(loid)s in the hydrosphere and biosphere should be probed using multidisciplinary approach. Given the current knowledge of DOM dynamics in soils, the following research areas could be pursued:

7.1. Macroscale (landscape to global) Role of DOM on the nature and extent of C sequestration in soils as in agricultural farms, forest lands, and wetlands (i.e., implication on global climate change). Impact of DOM on the nature and extent of methylation of certain metal(loid)s (Hg, Pb, As, and Se) and reduction of N compound to NO and N2O (i.e., implications on greenhouse gases) Effect of DOM on sorption/desorption of metal(loid)s and organic pesticides and their transport to soil strata, streams, and lakes Quantification of DOM conversion to CO2 and CH4 in large bodies of water such as oceans as source of greenhouse gases

7.2. Microscale (water bodies and soil profile) Extent of sorption/desorption of DOM in various soil horizons, transport in vadose zone, etc., as influenced by soil type, surface charge, metal oxides, and organic matter content Extent of mobility/transport of metal(loid)s and organic pesticides as influenced by DOM dynamics in the rhizosphere Nature and extent of DOC, DON, and DOP mobility as influenced by the quality and source of DOM

7.3. Molecular scale (carbon fractions, organic acids, and microorganisms) Nature and extent of soil mineral–DOM–microbe interactions as influenced by environmental factors such as land use and acid rain, and edaphic factors such as soil type and clay mineralogy Nature of DOM nanoparticles and colloids in nutrients, metal(loid)s, and organic pesticide interactions in terrestrial and aquatic ecosystems

62

Nanthi S. Bolan et al.

Nature of microbial communities as affected by DOM dynamics and their role on the degradation of DOM Improved analytical methods to characterize specific compounds or functional groups in DOM and reliable predictability of the fluxes of DOM using isotopic techniques Mechanistic understanding of the interactions between DOM and metal oxide surfaces in soils as a means to predict DOM transport and cotransport of contaminants associated with DOM and metal oxides

ACKNOWLEDGMENTS The senior author thanks CRC CARE for providing funding (No. 2-3-09-07/08) to undertake research on landfill site remediation; part of the review was derived from this project.

REFERENCES Adriano, D. C. (2001). Trace Elements in Terrestrial Environments; Biogeochemistry, Bioavailability and Risks of Metals. 2nd edn. Springer, New York, NY. Aguilera, P., Briceno, G., Candia, M., Mora, M. D., Demanet, R., and Palma, G. (2009). Chemosphere 77, 785–790. Aira, M., Sampedro, L., Monroy, F., and Dominguez, J. (2008). Soil Biol. Biochem. 40, 2511–2516. Alberts, J. J., Takacs, M., and Pattanayek, M. (2001). Acta Hydroch. Hydrob. 28, 428–435. Albrechtova, J., Seidl, Z., Aitkenhead-Peterson, J., Lhota´kova´, Z., Rock, B. N., Alexander, J. E., Malenovsky´, Z., and McDowell, W. H. (2008). Sci. Total Environ. 404, 424–432. Alves, M., Gonzalez, B. C. G., De Carvalho, G. R., Castenheira, J. M., Pereira, S. M. C., and Vasconcelos, L. A. T. (1993). Water Res. 27, 1333–1338. Al-Wabel, M. A., Heil, D. M., Westfall, D. G., and Barbarick, K. A. (2002). J. Environ. Qual. 31, 1157–1165. Amon, R. M. W., Fitznar, H. P., and Benner, R. (2001). Limnol. Oceanogr. 46, 287–297. Anderson, S., Valeur, I., and Nilsson, I. (1994). Environ. Int. 20, 81–88. Andersson, S., and Nilsson, S. I. (2001). Soil Biol. Biochem. 33, 1181–1191. Annabi, M., Houot, S., Francou, F., Poitrenaud, M., and Le Bissonnais, Y. (2007). Soil Sci. Soc. Am. J. 71, 413–423. Antoniadis, V., and Alloway, B. J. (2002). Environ. Pollut. 117, 515–521. Antoniadis, V., Tsadilas, C. D., and Ashworth, D. J. (2007). Chemosphere 68, 489–494. Apte, S. C., Batley, G. E., Bowles, K. C., Brown, P. L., Creighton, N., Hales, L. T., Hyne, R. V., Julli, M., Markich, S. I., Pablo, F., Rogers, N. J., Stauber, J. L., and Wilde, K. (2005). Environ. Chem. 2, 320–330. Arah, J. R. M., and Smith, K. A. (1990). In “Soils and the Greenhouse Effect” (A. F. Bouwman, Ed.), pp. 475–480. Wiley, Chichester. Arnold, W. R., Cotsifas, J. S., Ogle, R. S., DePalma, S. G. S., and Smith, D. S. (2010). Environ. Toxicol. Chem. 29, 311–319. Ashworth, D. J., and Alloway, B. J. (2008). Commun. Soil Sci. Plant Anal. 39, 538–550. Ashworth, D. J., and Alloway, B. J. (2004). Environ. Pollut. 127, 137–144.

Dissolved Organic Matter

63

Bachman, J., and Patterson, H. (1999). Environ. Sci. Technol. 33, 874–881. Baldock, J. A. (2002). In “Interactions of Soil Particles and Microorganisms” (P. M. Huang, et al., Eds.), Wiley, New York. Ballard, T. M. (1971). Soil Sci. Soc. Am. Proc. 35, 145–147. Barriuso, E., Baer, U., and Calvet, R. (1992). J. Environ. Qual. 21, 359–367. Barton, L., McLay, C. D. A., Schipper, L. A., and Smith, C. T. (1999). Aust. J. Soil Res. 37, 1073–1093. Baskaran, S., Bolan, N. S., Rahman, A., and Tillman, R. W. (1996). Aust. J. Soil Res. 34, 609–622. Bauer, M., and Blodau, C. (2006). Sci. Total Environ. 354, 179–190. Beauchamp, E. G., Trevors, J. T., and Paul, J. W. (1989). Adv. Soil Sci. 10, 113–142. Ben-Hur, M., Letey, J., Farmer, W. I., Williams, C. F., and Nelson, S. D. (2003). Soil Sci. Soc. Am. J. 67, 1140–1146. Benner, R., and Biddanda, B. (1998). Limnol. Oceanogr. 43, 1373–1378. Bertoncini, E. I., D’Orazio, V., Senesi, N., and Mattiazzo, M. E. (2005). Anal. Bioanal. Chem. 381, 1281–1288. Bhatt, M. P., and Gardner, K. H. (2009). Environ. Geol. 58, 867–876. Bhandral, R., Bolan, N. S., Saggar, S., and Hedley, M. J. (2007). Nutr. Cycl. Agroecosys. 79, 193–208. Bisogni, J. J. Jr., and Lawrence, A. W. (1975). J. Water Pollut. Control Fed. 47, 135–152. Blaser, P. (1994). In “Humic Substances in the Global Environment and Implications on Human Health” (N. Senesi and T. M. Miano, Eds.), pp. 943–960. Elsevier, Amsterdam. Boddy, E., Hill, P. W., Farrar, J., and Jones, D. L. (2007). Soil Biol. Biochem. 39, 827–835. Bolan, N. S., Baskaran, S., and Thiagarajan, S. (1996). Commun. Soil Sci. Plant Anal. 27, 2723–2737. Bolan, N. S., Adriano, D., and Curtin, D. (2003). Adv. Agron. 78, 215–272. Borch, N. H., and Kirchman, D. L. (1997). Mar. Chem. 57, 85–95. Bowen, S. R., Gregorich, E. G., and Hopkins, D. W. (2009). Biol. Fertil. Soils 45, 733–742. Boyd, T. J., Wolgast, D. M., Rivera-Duarte, I., Holm-Hansen, O., Hewes, C. D., Zirino, A., and Chadwick, D. B. (2005). Microb. Ecol. 49, 353–366. Bremner, J. M., and Tabatabai, M. A. (1971). In “Instrumental Methods for Analysis of Soils and Plant Tissues” (M. L. Walsh, Ed.). Soil Sci. Soc. Am., Madison. Bricen˜o, M., Escudey, M., Galindo, G., Borchardt, D., and Chang, A. (2004). Commun. Soil Sci. Plant Anal. 35, 1323–1337. Briceno, G., Demanet, R., Mora, M. D., and Palma, G. (2008). J. Environ. Qual. 37, 1519–1526. Brown, T. T., Koenig, R. T., Huggins, D. R., Harsh, J. B., and Rossi, R. E. (2008). Soil Sci. Soc. Am. J. 72, 634–640. Brye, K. R., Norman, J. M., Bundy, L. G., and Gower, S. T. (2001). J. Environ. Qual. 30, 58–70. Bueno, C. S., and Ladha, J. K. (2009). Biol. Fertil. Soils 45, 499–509. Buffle, J., Deladoey, P., Zumstein, J., and Haerdi, W. (1982). Schweiz. Z. Hydrol. 44, 327–362. Carmona, D. M., Cano, A´.F., and Arocena, J. M. (2008). Resumen workshop Macla 10, 115–117. Carran, R. A., Theobald, P. W., and Evans, J. P. (1995). Aust. J. Soil Res. 33, 341–352. Celerier, J., Rodier, C., Favetta, P., Lemee, L., and Ambles, A. (2009). Eur. J. Soil Sci. 60, 33–43. Celik, I. (2005). Soil Tillage Res. 83, 270–277. Celis, R., Cornejo, J., Hermosin, M. C., and Koskinen, W. C. (1998). Soil Sci. Soc. Am. J. 62, 165–171.

64

Nanthi S. Bolan et al.

Cernansky, S., Kolencik, M., Sevc, J., Urik, M., and Hiller, E. (2009). Bioresour. Technol. 100, 1037–1040. Chadwick, D. R., Pain, B. F., and Brookman, S. K. E. (2000). J. Environ. Qual. 29, 277– 287. Chan, K. Y., Heenan, D. P., and Roberts, W. P. (1992). Aust. J. Soil Res. 30, 71–83. Chantigny, M. H. (2003). Geoderma 113, 357–380. Chantigny, M. H., Angers, D. A., and Beauchamp, C. J. (1999). Soil Sci. Soc. Am. J. 63, 1214–1221. Chefetz, B., Hadar, Y., and Chen, Y. (1998). Acta Hydroch. Hydrob. 26, 172–179. Chen, H. Q., Hou, R. X., Gong, Y. S., Li, H. W., Fan, M. S., and Kuzyakov, Y. (2009). Soil Tillage Res. 106, 85–94. Chen, Y., Senesi, N., and Schnitzer, M. (1978a). Geoderma 20, 87–104. Chen, Y., Senesi, N., and Schnitzer, M. (1978b). Soil Sci. 29, 350–359. Cheng, K. Y., and Wong, J. W. C. (2006). Chemosphere 62, 1907–1916. Chepkwony, C. K., Haynes, R. J., Swift, R. S., and Harrison, R. (2001). Plant Soil 234, 83–90. Chiou, C. T., Malcolm, R. L., Brinton, T. I., and Kile, D. E. (1986). Environ. Sci. Technol. 20, 502–508. Chirenje, T., and Ma, L. Q. (1999). J. Environ. Qual. 28, 760–766. Chirenje, T., Rivero, C., and Ma, L. Q. (2002). Geoderma 108, 31–47. Chiti, T., Neubert, R. E. M., Janssens, I. A., Certini, G., Yuste, C. J., and Sirignano, C. (2009). Geoderma 149, 137–142. Christensen, J. B., Tipping, E., Kinniburgh, D. G., Gron, C., and Christensen, T. H. (1998). Environ. Sci. Technol. 32, 3346–3355. Ciavatta, C. M., Govi, M., Antisari, L. V., and Sequi, P. (1991). Commun. Soil Sci. Plant Anal. 22, 795–800. Cifuentes, F. R., Lindemann, W. C., and Barton, L. L. (1996). Soil Sci. 161, 233–241. Cilenti, A., Provenzano, M. R., and Senesi, N. (2005). Environ. Chem. Lett. 3, 53–56. Clark, J. M., Chapman, P. J., Heathwaite, A. L., and Adamson, J. K. (2006). Environ. Sci. Technol. 40, 1776–1783. Condron, L. M., Davis, M. R., Newman, R. H., and Cornforth, I. S. (1996). Biol. Fertil. Soils 21, 37–42. Cooper, R., Thoss, V., and Watson, H. (2007). Hydrol Processes. 21, 622–633. Cox, L., Cecchi, A., Celis, R., Hermosı´n, M. C., Koskinen, W. C., and Cornejo, J. (2001). Soil Sci. Soc. Am. J. 65, 1688–1695. Cox, L., Velarde, P., Cabrera, A., Hermosı´n, M. C., and Cornejo, J. (2007). Eur. J. Soil Sci. 58, 714–721. Cronan, C. S., and Aiken, G. R. (1985). Geochim. Cosmochim. Acta 49, 1697–1705. Currie, W. S., Aber, J. D., McDowell, W. H., Boone, R. D., and Magill, A. H. (1996). Biogeochemistry 35, 471–505. D’Amore, D. V., Fellman, J. B., Edwards, R. T., and Hood, E. (2010). Ecohydrology 3, 249–261. Dafner, E., and Wangersky, P. J. (2002). J. Environ. Monitor. 4, 48–54. Dai, K., David, M. B., and Vance, G. F. (1996). Biogeochemistry 35, 339–365. Dalal, R. (1977). Adv. Agron. 29, 83–117. David, M. B., Vance, G. F., Rissing, J. M., and Stevenson, F. J. (1989). J. Environ. Qual. 18, 212–217. De Bertrand, N., and Barcelo´, D. (1991). Anal. Chim. Acta 254, 235–244. De-Campos, A. B., Mamedov, A. I., and Huang, C. H. (2009). Soil Sci. Soc. Am. J. 73, 550–559. Delprat, L., Chassin, P., Lineres, M., and Jambert, C. (1997). Eur. J. Agron. 7, 201–210.

Dissolved Organic Matter

65

Dendooven, L., Bonhomme, E., Merckx, R., and Vlassak, K. (1998). Biol. Fertil. Soils 27, 5–8. Denef, K., Six, J., Merckx, R., and Paustian, K. (2002). Plant Soil 246, 185–200. Deurer, M., Grinev, D., Young, I., Clothier, B. E., and Muller, K. (2009). Eur. J. Soil Sci. 60, 945–955. Dillon, P. J., Evans, H. E., and Scholer, P. J. (1988). Biogeochemistry 5, 201–220. Dilly, O., Bloem, J., Vos, A., and Munch, J. C. (2004). Appl. Environ. Microb. 70, 468–474. Dodla, S. K., Wang, J. J., DeLaune, R. D., and Cook, R. L. (2008). Sci. Total Environ. 407, 471–480. Donald, R. G., Anderson, D. W., and Stewart, J. W. B. (1993). Soil Sci. Soc. Am. J. 57, 1611–1618. Dosskey, M. G., and Bertsch, P. M. (1997). Soil Sci. Soc. Am. J. 61, 921–927. Dou, F., Wright, A. L., and Hons, F. M. (2008). Commun. Soil Sci. Plant Anal. 39, 667–679. Effler, S. W., and Matthews, D. A. (2008). Lake Reservoir Manage. 24, 122–138. Elmholt, S., Schjonning, P., Munkholm, L. J., and Debosz, K. (2008). Geoderma 144, 455–467. Elovitz, M. S., and Fish, W. (1995). Environ. Sci. Technol. 29, 1933–1943. Engelhaupt, E., and Bianchi, T. S. (2001). Limnol. Oceanogr. 46, 917–926. Enowashu, E., Poll, C., Lamersdorf, N., and Kandeler, E. (2009). Appl. Soil Ecol. 43, 11–21. Escudey, M., Galindo, G., Bricen˜o, M., and Chang, A. (2004). J. Chil. Chem. Soc. 49, 5–9. Fang, F., Kanan, S., Patterson, H. H., and Cronan, C. S. (1998). Anal. Chim. Acta 373, 139–151. Faust, B. C., and Zepp, R. G. (1993). Environ. Sci. Technol. 27, 2517–2522. Favilli, F., Egli, M., Cherubini, P., Sartori, G., and Haeberli, W. (2008). Geoderma 145, 355–369. Fernandez, J. M., Polo, A., Senesi, N., and Plaza, C. (2007). Chemosphere 69, 630–635. Firestone, M. K. (1982). In “Nitrogen in Agricultural Soils” (F. J. Stevenson, Ed.), pp. 289–326. ASA-CSSA-SSA, Madison, Wisconsin, U.S.A. Flores-Cespedes, F., Gonzales-Pradas, E., Fernandez-Perez, M., Vilafranca-Sanchez, M., Socias-Viciana, M., and Urena-Amate, M. D. (2002). J. Environ. Qual. 31, 880–888. Frank, D. A., and Groffman, P. M. (2009). Ecology 90, 1512–1519. Friese, K., Herzsprung, P., and Witter, B. (2002). Acta Hydroch. Hydrob. 30, 141–148. Froberg, M., Berggren, D., Bergkvist, B., Bryant, C., and Knicker, H. (2003). Geoderma 113, 311–322. Froberg, M., Kleja, D. B., and Hagedorn, F. (2007). Eur. J. Soil Sci. 58, 108–114. Ganeshamurthy, A. N., and Nielsen, N. E. (1990). Soil Biol. Biochem. 22, 1163–1165. Gao, H., and Zepp, R. G. (1998). Environ. Sci. Technol. 32, 2940–2946. Gao, J. P., Maguhn, J., Spitzauer, P., and Kettrup, A. (1998). Water Res. 32, 2089–2094. Garcia-Gil, J. C., Plaza, C., Fernandez, J. M., Senesi, N., and Polo, A. (2008). Geoderma 146, 363–369. Garcia-Gil, J. C., Plaza, C., Senesi, N., Brunetti, G., and Polo, A. (2007). CLEAN 35, 480–487. Gasco´, G., and Lobo, M. C. (2007). Waste Manage. 27, 1494–1500. Gerecke, A. C., Canonica, S., Muller, S. R., Scharer, M., and Schwarzenbach, R. P. (2001). Environ. Sci. Technol. 35, 3915–3923. Ghani, A., Dexter, M., Carran, R. A., and Theobald, P. W. (2007). Eur. J. Soil Sci. 58, 832–843. Ghiorse, W. C., and Wilson, J. T. (1988). Adv. Appl. Microbiol. 33, 107–172. Ghosh, A. K., and Bhattacharyya, P. (2004). Environ. Geol. 45, 1010–1016. Gibson, H. S., Worrall, F., Burt, T. P., and Adamson, J. K. (2009). Hydrol. Processes 23, 1901–1911.

66

Nanthi S. Bolan et al.

Gigliotti, G., Kaiser, K., Guggenberger, G., and Haumaier, L. (2002). Biol. Fertil. Soils 36, 321–329. Giroto, J. A., Guardani, R., Teixeira, A. C. S. C., and Nascimento, C. A. O. (2006). Chem. Eng. Process. 45, 523–532. Gissel-Nielsen, G., Gupta, U. C., Lamand, M., and Westernmarck, T. (1984). Adv. Agron. 37, 297–459. Golchin, A., Baldock, J. A., and Oades, J. M. (1997). In “Soil Processes and the Carbon Cycle” (R. Lal, J. Kimble, R. Follett, and B. Stewart, Eds.), pp. 245–266. CRC Press, Boca Raton, FL. Goldstone, J. V., Pullin, M. J., Bertilsson, S., and Voelker, B. M. (2002). Environ. Sci. Technol. 36, 364–372. Gonet, S. S., Debska, B., Zaujec, A., and Banach-Szott, M. (2008). Ekologia (Bratislava) 27, 351–366. Goyne, K. W., Jun, H. J., Anderson, S. H., and Motavalli, P. R. (2008). J. Environ. Qual. 37, 154–163. Grandy, A., Sinsabaugh, R., Neff, J., Stursova, M., and Zak, D. (2008). Biogeochemistry 91, 37–49. Gregorich, E. G., Liang, B. C., Drury, C. F., Mackenzie, A. F., and McGill, W. B. (2000). Soil Biol. Biochem. 32, 581–587. Gu, B., Schmitt, J., Chen, Z., Liang, L., and McCarthy, J. F. (1994). Environ. Sci. Technol. 28, 38–46. Gu, B., Schmitt, J., Chen, Z., Liang, L., and McCarthy, J. F. (1995). Geochim. Cosmochim. Acta. 59, 219–229. Guggenberger, G. (1994). Environ. Int. 20, 31–41. Guggenberger, G., and Kaiser, K. (2003). Geoderma 113, 293–310. Guggenberger, G., and Zech, W. (1993). Geoderma 59, 109–129. Guggenberger, G., Christensen, B. T., and Zech, W. (1994a). Eur. J. Soil Sci. 45, 449–458. Guggenberger, G., Zech, W., and Schulten, H. R. (1994b). Org. Geochem. 21, 51–66. Gullberg, K. R., Goedkoop, W., and Johnson, R. K. (1997). Limnol. Oceanogr. 42, 452–460. Guo, M., Labreveux, M., and Song, W. (2009). Waste Manage. 29, 2151–2159. Haberhauer, G., Temmel, B., and Gerzabek, M. H. (2002). Chemosphere 46, 495–499. Hagedorn, F., and Machwitz, M. (2007). Soil Biol. Biochem. 39, 1759–1769. Hagedorn, F., vanHees, P. A. W., Handa, I. T., and Ha¨ttenschwiler, S. (2008). Global Biogeochem. Cycles 22, GB2004. Han, N., and Thompson, M. L. (1999). J. Environ. Qual. 28, 939–944. Hanc, A., Tlustos, P., Szakova, J., and Habart, J. (2009). Waste Manage. 29, 2282–2288. Harrison, J. A., Caraco, N., and Sietzinger, S. P. (2005). Global Biogeochem. Cycles 19, GB4S04. Harter, R. D. (1991). SSSA. In “Micronutrients in Agriculture, Vol. 4” (J. J. Mortvedt, F. R. Cox, L. M. Shuman, and R. M. WeIch, Eds.), 2nd edn., pp. 59–88. Soil Sci. Soc. Am, Madison, WI. Haynes, R. J. (2000). Soil Biol. Biochem. 32, 211–219. Haynes, R. J., and Swift, R. S. (1989). J. Soil Sci. 40, 773–781. Haynes, R. J., Murtaza, G., and Naidu, R. (2009). Adv. Agron. 104, 165–267. Hedges, J. I. (1987). Nature 330, 205–206. Hedges, J. I., and Oades, J. M. (1997). Org. Geochem. 27, 319–361. Hentschel, K., Borken, W., and Matzner, E. (2007). Plant Soil 300, 21–34. Her, N., Amy, G., McKnight, D., Sohn, J., and Yoon, Y. M. (2003). Water Res. 37, 4295–4303. Herbel, M. J., Suarez, D. L., Goldberg, S., and Gao, S. (2007). Soil Sci. Soc. Am. J. 71, 927–939. Herbert, B. E., Bertsch, P. M., and Novak, J. M. (1993). Environ. Sci. Technol. 27, 398–403.

Dissolved Organic Matter

67

Herencia, J. F., Ruiz, J. C., Melero, S., Galavis, P. A. G., and Maqueda, C. (2008). J. Agr. Sci. 146, 677–687. Hernandez, D., Plaza, C., Senesi, N., and Polo, A. (2007). Eur. J. Soil Sci. 58, 900–908. Hernes, P. J., Robinson, A. C., and Aufdenkampe, A. K. (2007). Geophys. Res. Lett. 34, 10–14. Herre, A., Siebe, C., and Kaupenjohann, M. (2004). Water Sci. Technol. 50, 277–284. Higgins, T. E., Halloran, A. R., Dobbins, M. E., and Pittignano, A. J. (1998). J. Air Waste Manage. 48, 1100–1106. Hildebrand, E. E., and Schack-Kirchner, H. (2000). Nutr. Cycl. Agroecosys. 56, 69–78. Holloway, J. M., Goldhaber, M. B., Scow, K. M., and Drenovsky, R. E. (2009). Chem. Geol. 267, 85–95. Huang, P. M., Wang, M. K., and Chiu, C. Y. (2005a). Pedobiologia 49, 609–635. Huang, Q. Y., Liang, W., and Cai, P. (2005b). Colloids Surf. B 45, 209–214. Huang, X., and Lee, L. S. (2001). J. Environ. Qual. 30, 1258–1265. Huang, Y., Zou, J. W., Zheng, X. H., Wang, Y. S., and Xu, X. K. (2004). Soil Biol. Biochem. 36, 973–981. Impellitteri, C. A., Lu, Y. F., Saxe, J. K., Allen, H. E., and Peijnenburg, W. J. G. M. (2002). Environ. Int. 28, 401–410. Inamdar, S., Rupp, J., and Mitchell, M. (2009). Hydrol. Processes 23, 1912–1926. Ishikawa, T., Trisliana, K., Yurenfrie, Y., Ardianor, A., and Gumiri, S. (2006). Limnology 7, 143–146. Jackson, B. P., Seaman, J. C., and Bertsch, P. M. (2006). Chemosphere 65, 2028–2034. Jaffrain, J., Gerard, F., Meyer, M., and Ranger, J. (2007). Soil Sci. Soc. Am. J. 71, 1851–1858. Jandl, G., Schulten, H., and Leinweber, P. (2002). J. Plant Nutr. Soil Sci. 165, 133–139. Jandl, R., and Sletten, R. S. (1999). J. Plant Nutr. Soil Sci. 162, 623–629. Jansen, B., Nierop, K. G. J., and Verstraten, J. M. (2005). Eur. J. Soil Sci. 56, 537–550. Jardine, P. M., Fendorf, S. E., Mayes, M. A., Larsen, I. L., Brooks, S. C., and Bailey, W. B. (1999). Environ. Sci. Technol. 33, 2939–2944. Jarvis, S. C., Stockdale, E. A., Shepherd, M. A., and Powlson, D. S. (1996). Adv. Agron. 57, 187–235. Jiang, L., Huang, J., Liang, L., Zheng, P. Y., and Yang, H. (2009). J. Agr. Food Chem. 56, 11933–11940. Jin, X. C., Wang, S. R., Zhao, H. C., Zhou, X. N., and Wu, F. C. (2008). Environ. Geol. 56, 391–398. Jinbo, Z., Changchun, S., and Wenyan, Y. (2006). Soil Sci. Soc. Am. J. 70, 660–667. Johnson, K. D., Scatena, F. N., Johnson, A. H., and Pan, Y. D. (2009). Geoderma 148, 346–356. Jones, D. L., Shannon, D., Murphy, D. V., and Farrar, J. F. (2004). Soil Biol. Biochem. 36, 749–756. Jones, R. I. (1992). Hydrobiol. 229, 73–92. Jones, V., Collins, M. J., Penkman, K. E. H., Jaffe´, R., and Wolff, G. A. (2005). Org. Geochem. 36, 1099–1107. Kaiser, K. (2001). Eur. J. Soil Sci. 52, 489–493. Kaiser, K., and Guggenberger, G. (2000). Org. Geochem. 31, 711–725. Kaiser, K., and Guggenberger, G. (2005a). Geoderma 127, 177–187. Kaiser, K., and Guggenberger, G. (2005b). Biogeochemistry 72, 337–364. Kaiser, K., and Zech, W. (1997). Soil Sci. Soc. Am. J. 61, 64–69. Kaiser, K., Guggenberger, G., and Zech, W. (1996). Geoderma 74, 281–303. Kaiser, K., Guggenberger, G., Haumaier, L., and Zech, W. (2001). Biogeochemistry 55, 103–143. Kaiser, K., Guggenberger, G., Haumaier, L., and Zech, W. (2002). Org. Geochem. 33, 307–318.

68

Nanthi S. Bolan et al.

Kalbitz, K., and Kaiser, K. (2003). Geoderma 113, 177–178. Kalbitz, K., and Wennrich, R. (1998). Sci. Total Environ. 209, 27–39. Kalbitz, K., Solinger, S., Park, J. H., Michalzik, B., and Matzner, E. (2000). Soil Sci. 165, 277–304. Kalbitz, K., Schwesig, D., Schmerwitz, J., Kaiser, K., Haumaier, L., Glaser, B., Ellerbrock, R., and Leinweber, P. (2003). Soil Biol. Biochem. 35, 1129–1142. Kalbitz, K., Meyer, A., Yang, R., and Gerstberger, P. (2007). Biogeochemistry 86, 301–318. Karlik, B., and Szpakowska, B. (2001). Pol. J. Environ. Stud. 10, 85–88. Kaschl, A., Romheld, V., and Chen, Y. (2002). Israel J. Chem. 42, 89–98. Kawahigashi, M., Sumida, H., and Yamamoto, K. (2003). Geoderma 113, 381–396. Keiber, D. J., McDaniel, J., and Mooper, K. (1989). Nature 341, 637–639. Khalid, M., Soleman, N., and Jones, D. L. (2007). Soil Biol. Biochem. 39, 378–381. Khokhotva, O., and Waara, S. (2010). J. Hazard. Mater. 173, 689–696. Khomutova, T. E., Shirshova, L. T., Tinz, S., Rolland, W., and Richter, J. (2000). Plant Soil 219, 13–19. Klaus, U., Pfeifer, T., and Spiteller, M. (2000). Environ. Sci. Technol. 34, 3514–3520. Knotek-Smith, H. M., Heather, M., Crawford, D. L., Moller, G., and Henson, R. A. (2006). J. Ind. Microbiol. Biot. 33, 897–913. Koprivnjak, J. F., and Moore, T. R. (1992). Arct. Alp. Res. 24, 204–210. Kosian, P. A., Hoke, R. A., Ankley, G. T., and VanderMeiden, F. M. (1995). Environ. Toxicol. Chem. 14, 445–450. Kothawala, D. N., Moore, T. R., and Hendershot, W. H. (2009). Soil Sci. Soc. Am. J. 73, 1831–1842. Krachler, R., Jirsa, F., and Ayromlou, S. (2005). Biogeosciences Discussions 2, 537–549. Kreutzweiser, D. P., Hazlett, P. W., and Gunn, J. M. (2008). Environ. Rev. 16, 157–179. Kumpiene, J., Lagerkvist, A., and Maurice, C. (2008). Waste Manage. 28, 215–225. Laik, R., Kumar, K., Das, D. K., and Chaturvedi, O. P. (2009). Appl. Soil Ecol. 42, 71–78. Lajtha, K., Crow, S. E., Yano, Y., Kaushal, S. S., Sulzman, E., Sollins, P., and Spears, J. D. H. (2005). Biogeochemistry 76, 261–281. Lang, S. Q., Lilley, M. D., and Hedges, J. I. (2007). Mar. Chem. 103, 318–326. Laudon, H., Hedtjarn, J., Schelker, J., Bishop, K., Sorensen, R., and Agren, A. (2009). Ambio 38, 381–386. Leavitt, P. R., Findlay, D. L., Hall, R. I., and Smol, J. P. (1999). Limnol. Oceanogr. 44, 757– 773. Lee, D. Y., and Farmer, W. J. (1989). J. Environ. Qual. 18, 468–474. Lee, D. Y., Farmer, W. J., and Aochi, Y. (1990). J. Environ. Qual. 19, 567–573. Leenheer, J. A. (1981). Environ. Sci. Technol. 15, 578–587. Lepisto¨, A., Kortelainen, P., and Mattsson, T. (2008). Global Biogeochem. Cycles 22, GB3029. Leytem, A. B., Mikkelsen, R. L., and Gilliam, J. W. (2002). Soil Sci. 167, 652–658. Li, K., Xing, B., and Torello, W. A. (2005). Environ. Pollut. 134, 187–194. Ling, W. T., Xu, J. M., and Gao, Y. Z. (2005). Sci. China Ser. C Life Sci. 48, 57–66. Linn, D. M., and Doran, J. W. (1984). Soil Sci. Soc. Am. J. 48, 1267–1272. Liu, A. G., Ma, B. L., and Bomke, A. A. (2005). Soil Sci. Soc. Am. J. 69, 2041–2048. Liu, J. X., Peng, C. H., Apps, M. J., Dang, Q. L., Banfield, E., and Kurz, W. (2002). For. Ecol. Manage. 169, 103–114. Losi, M. E., Amrhein, C., and Frankenberger, W. T. (1994). Environ. Toxicol. Chem. 13, 1727–1735. Losi, M., and Frankenberger, W. Jr. (1997). Soil Sci. 162, 692. Ludwig, B., Heil, B., Flessa, H., and Beese, F. (2000). Acta Hydroch. Hydrob. 28, 77–82. Luo, J., White, R. E., Ball, P. R., and Tillman, R. W. (1996). Soil Biol. Biochem. 28, 409–417.

Dissolved Organic Matter

69

Luo, L., Zhang, S. Z., Ma, Y. B., Christie, P., and Huang, H. L. (2008). Environ. Sci. Technol. 42, 2414–2419. Lydersen, E., Lofgren, S., and Arnesen, R. T. (2002). Crit. Rev. Env. Sci. Tech. 32, 73–295. Lytle, C. R., and Perdue, E. M. (1981). Environ. Sci. Technol. 15, 224–228. Madhun, Y. A., Young, J. L., and Freed, V. H. (1986). J. Environ. Qual. 15, 64–68. Maestre, S. E., Mora, J., Hernandis, V., and Todoli, J. L. (2003). Anal. Chem. 75, 111–117. Magid, J., Kjaergaard, C., Gorissen, A., and Kuikman, P. J. (1999). Soil Biol. Biochem. 31, 595–602. Magill, A. H., and Aber, J. D. (2000). Soil Biol. Biochem. 32, 603–613. Marin-Benito, J. M., Sanchez-Martin, M. J., Andrades, M. S., Perez-Clavijo, M., and Rodriguez-cruz, M. S. (2009). J. Agric. Food Chem. 57, 9634–9642. Marschner, B., and Bredow, A. (2002). Soil Biol. Biochem. 34, 459–466. Marschner, B., and Kalbitz, K. (2003). Geoderma 113, 211–236. Martin, A. J., and Goldblatt, R. (2007). Environ. Toxicol. Chem. 26, 2594–2603. Martinez-Mena, A., Lopez, J., Almagro, A., Boix-Fayos, C., and Albaladejo, J. (2008). Soil Tillage Res. 99, 119–129. Martin-Olmedo, P., and Rees, R. M. (1999). Biol. Fertil. Soils 29, 386–393. Mattsson, T., Kortelainen, P., Laubel, A., Evans, D., Pujo-Pay, M., Raike, A., and Conan, P. (2009). Sci. Total Environ. 407, 1967–1976. Maurice, P. A., Cabaniss, S. E., Drummond, J., and Ito, E. (2002). Chem. Geol. 187, 59–77. Mayer, L. M., and Xing, B. S. (2001). Soil Sci. Soc. Am. J. 65, 250–258. Mayland, H. F., and Mayland, H. F. (1994). In “Selenium in the Environment” (W. T. Frankenberger Jr. and S. Benson, Eds.), pp. 29–45. Marcel-Dekker, Inc, New York, NY. McBride, M. B. (1994). Environmental Chemistry of Soils. Oxford University Press, New York, 406. McBride, M. B. (1998). J. Environ. Qual. 27, 578–584. McBride, M. B., Richards, B. K., Steenhuis, T., and Spiers, G. (1999). Soil Sci. 164, 613–623. McDonald, A. J., Webster, C. P., Poulton, P. R., Wilmer, W. S., and Goulding, K. W. T. (2001). “Book of Abstracts, 11th Nitrogen Workshop, Reims, France, September 9–12”, pp. 321–322. McDowell, W. H. (2003). Geoderma 113, 179–186. McDowell, R. W. (2005). Aust. J. Soil Res. 43, 853–860. McDowell, R. W., and Koopmans, G. F. (2006). Soil Biol. Biochem. 38, 61–70. McDowell, W. H., and Likens, G. E. (1988). Ecol. Monogr. 58, 177–195. McDowell, R. W., and Sharpley, A. N. (2003). J. Environ. Qual. 32, 207–214. McDowell, R. W., and Stewart, I. (2005). Chem. Ecol. 21, 211–226. McDowell, W. H., and Wood, T. (1984). Soil Sci. 137, 23–32. McDowell, R. W., Biggs, B. J. F., Sharpley, A. N., and Nguyen, L. (2004). Chem. Ecol. 20, 1–40. McDowell, W. H., Currie, W. S., Aber, J. D., and Yano, Y. (1998). Water Air Soil Poll. 105, 175–182. McGill, W. H., Cannon, K. R., Robertson, J. A., and Cook, F. (1986). Can. J. Soil Sci. 66, 1–19. McKnight, D., Bencala, K. E., Zellweger, G. W., Aiken, G. R., Feder, G. L., and Thorn, K. A. (1992). Environ. Sci. Technol. 26, 1388–1396. Means, J. C., and Wijayratne, R. D. (1982). Science 215, 968–970. Meier, M., Namjesnik-Dejanovic, K., Maurice, P. A., Chin, Y. P., and Aiken, G. R. (1999). Chem. Geol. 157, 275–284. Menneer, J. C., McLay, C. D. A., and Lee, R. (2001). Aust. J. Soil Res. 39, 877–891. Mesquita, M. E., and Carranca, C. (2005). Environ. Technol. 26, 1065–1072.

70

Nanthi S. Bolan et al.

Miano, T. M., and Senesi, N. (1992). Sci. Total Environ. 117/118, 41–51. Michalzik, B., and Matzner, E. (1999). Eur. J. Soil Sci. 50, 579–590. Michalzik, B., Ilgen, G., Hertel, F., Hantsch, S., and Bilitewski, B. (2007). Waste Manage. 27, 497–509. Michalzik, B., Kalbitz, K., Park, J. H., Solinger, S., and Matzner, E. (2001). Biogeochemistry 52, 173–205. Michelsen, A., Andersson, M., Jensen, M., Kjoller, A., and Gashew, M. (2004). Soil Biol. Biochem. 36, 1707–1717. Minor, E. C., Dalzell, B. L., Brent, L., Stubbins, A., and Mopper, K. (2007). Aquat. Sci. 69, 440–455. Miskimmin, B. M., Rudd, J. W. M., and Kelly, C. A. (1992). Can. J. Fish. Aquat. Sci. 49, 17–22. Mitchell, A. M., and Baldwin, D. S. (2005). In “Organic Phosphorus in the Environment” (B. L. Turner, E. Frossard, and D. S. Baldwin, Eds.), pp. 309–323. CAB International, Wallingford, Oxon, UK. Mo¨ller, A., Kaiser, K., and Guggenberger, G. (2005). J. Plant Nutr. Soil Sci. 168, 649–659. Monbet, P., McKelvie, I. D., and Worsfold, P. J. (2009). Geochim. Cosmochim. Acta 73, 41027–41038. Moore, T. R. (1985). Soil Sci. Soc. Am. J. 49, 1590–1592. Moore, T. R., and Dalva, M. (2001). Soil Sci. 166, 38–47. Moore, T. R., and Matos, L. (1999). Can. J. Soil Sci. 79, 321–324. Moore, T. R., Desouza, W., and Koprivnjak, J. F. (1992). Soil Sci. 154, 120–129. Moran, M. A., Sheldon, W. M., and Zepp, R. G. (2000). Limnol. Oceanogr. 45, 1254–1264. Mostofa, K. M. G., Yoshioka, T., Konohira, E., and Tanoue, E. (2007). Geochem. J. 41, 323–331. Mu¨ller, K., Magesan, G. N., and Bolan, N. S. (2007). Agric. Ecosyst. Environ. 120, 93–116. Muller, M., Alewell, C., and Hagedorn, F. (2009). Soil Biol. Biochem. 41, 1066–1074. Murphy, D. V., Macdonald, A. J., Stockdale, E. A., Goulding, K. W. T., Fortune, S., Gaunt, J. L., Poulton, P. R., Wakefield, J. A., Webster, C. P., and Wilmer, W. S. (2000). Biol. Fertil. Soils 30, 374–387. Myrold, D. D., and Tiedje, J. M. (1985). Soil Biol. Biochem. 17, 819–822. Nambu, K., and Yonebayashi, K. (2000). Soil Sci. Plant Nutr. 46, 559–569. Navarro, S., Vela, N., Gime´nez, M. J., and Navarro, G. (2004). Sci. Total Environ. 329, 87–97. Neff, J. C., Hobbie, S. E., and Vitousek, P. M. (2000). Biogeochemistry 51, 283–302. Neff, J. C., and Asner, G. P. (2001). Ecosystems 4, 29–48. Nelson, P. N., Baldock, J. A., and Oades, J. M. (1993). Biogeochemistry 19, 27–50. Nelson, P. N., Dictor, M. C., and Soulas, G. (1994). Soil Biol. Biochem. 26, 1549–1555. Nelson, S. D., Letey, J., Farmer, W. J., Williams, C. F., and Ben-Hur, M. (1998). J. Environ. Qual. 27, 1194–1200. Nilsson, J. I. (1985). Ecol. Bull. (Stockholm) 37, 120–137. Nilsson, S. I., Andersson, S., Valeur, I., Persson, T., Bergholm, J., and Wire´n, A. (2001). Forest Ecol. Manag. 146, 55–73. Nishina, K., Takenaka, C., and Ishizuka, S. (2009). Soil Sci. Plant Nutr. 55, 203–214. Noblet, J. A., Smith, L. A., and Suffet, I. H. (1996). J. Agr. Food Chem. 44, 3685–3693. Norton, S. A., Coolidge, K., Amirbahman, A., Bouchard, R., Kopacek, J., and Reinhardt, R. (2008). Sci. Total Environ. 404, 276–283. Oades, J. M. (1984). Plant Soil 76, 319–337. Obernosterer, I., Kraay, G., Ranitz, E. D., and Herndl, G. J. (1999). Aquat. Microb. Ecol. 20, 147–156. Osler, G. H. R., and Sommerkorn, M. (2007). Ecology 88, 1611–1621.

Dissolved Organic Matter

71

Oste, L. A., Temminghoff, E. J. M., and Van Riemsdijk, W. H. (2002). Environ. Sci. Technol. 36, 208–214. Ouyang, D. S., MacKenzie, A. F., and Fan, M. X. (1999). Nutr. Cycl. Agroecosys. 53, 237– 247. Palmer, S. M., Driscoll, C. T., and Johnson, C. E. (2004). Biogeochemistry 68, 51–70. Parfitt, R. L., Mackay, A. D., Ross, D. J., and Budding, P. J. (2009). J. Agr. Res. 52, 69–80. Park, J. H., and Matzner, E. (2003). Biogeochemistry 66, 265–286. Parkin, T. B., and Meisinger, J. J. (1989). J. Environ. Qual. 18, 12–16. Paterson, E., Gebbing, T., Abel, C., Sim, A., and Telfer, G. (2007). New Phytol. 173, 600– 610. Paul, J. W., and Beauchamp, E. G. (1989). Biol. Fert. Soils. 7, 303–309. Paul, K. I., Polglase, P. J., Nyakuengama, J. G., and Khanna, P. K. (2002). Forest Ecol. Manag. 168, 241–257. Payne, W. J. (1981). In “Denitrification”, Wiley, New York, pp. 214. Peckenham, J. M., Nadeau, J. A., Amirbahman, A., and Behr, R. S. (2008). Waste Manage. Res. 26, 163–172. Pennington, K. L., Harper, S. S., and Koskinen, W. C. (1991). Weed Sci. 39, 667–672. Phillips, R. P., Erlitz, Y., Bier, R., and Bernhardt, E. S. (2008). Funct. Ecol. 22, 990–999. Plaza, C., and Senesi, N. (2009). In “Biophysico-Chemical Processes Involving Natural Nonliving Organic Matter in Environmental System. IUPAC Series on BiophysicoChemical Processes in Environmental Systems, Vol. 2” (N. Senesi, P. M. Huang, and B. Xing, Eds.), pp. 147–181. Wiley, Hoboken, NJ. Plaza, C., Senesi, N., Polo, A., Brunetti, G., Garcia-Gil, J. C., and D’Orazio, V. (2003). Soil Till Res. 74, 179–190. Plaza, C., Garcia-Jil, J. C., Polo, A., Senesi, N., and Brunetti, G. (2005a). J. Environ. Qual. 34, 1131–1137. Plaza, C., Garcia-Gil, J. C., Polo, A., and Senesi, N. (2005b). Chemosphere 61, 711–716. Plaza, C., Senesi, N., Polo, A., and Brunetti, G. (2005c). Sci. Technol. Sci. Technol. 39, 7141– 7146. Plaza, C., Senesi, N., Brunetti, G., and Mondelli, D. (2007). Bioresour. Technol. 98, 1964– 1971. Polubesova, T., Chen, Y., Navon, R., and Chefetz, B. (2008). Environ. Sci. Technol. 42, 4797–4803. Prietzel, J., Weick, C., Korintenberg, J., Seybold, G., Thumerer, T., and Treml, B. (2001). Plant Soil 230, 287–305. Prokushkin, A. S., Gavrilenko, I. V., Abaimov, A. P., Prokushkin, S. G., and Samusenko, A. V. (2006). Mitig. Adapt. Strateg. Glob. Change 11, 223–240. Provenzano, M. R., Cilenti, A., Gigliotti, G., and Senesi, N. (2008). CLEAN 36, 748–753. Provenzano, M. R., Cilenti, A., Gigliotti, G., Erriques, F., and Senesi, N. (2006). Compost. Sci. Util. 14, 191–200. Provin, T. L., Wright, A. L., Hons, F. M., Zuberer, D. A., and White, R. H. (2008). Bioresour. Technol. 99, 2672–2679. Pullin, M. J., Progess, C. A., and Maurice, P. A. (2004). Geochim. Cosmochim. Acta 68, 3643– 3656. Qualls, R. G. (2005). Environ. Sci. Technol. 39, 1616–1622. Qualls, R. G., and Haines, B. L. (1991). Soil Sci. Soc. Am. J. 55, 1112–1123. Qualls, R. G., and Haines, B. L. (1992). Soil Sci. Soc. Am. J. 56, 578–586. Qualls, R. G., Haines, B. L., and Swank, W. T. (1991). Ecology 72, 254–266. Raber, B., and Ko¨gel-Knabner, I. (1997). Eur. J. Soil Sci. 48, 443–455. Reddy, K. R., Rao, P. S. C., and Jessup, R. E. (1982). Soil Sci. Soc. Am. J. 46, 62–68. Remington, S. M., Strahm, B. D., Neu, V., Richey, J. E., and da Cunha, H. B. (2007). Soil Sci. 172, 279–291.

72

Nanthi S. Bolan et al.

Rieger, L., Langergraber, G., Thomann, M., Fleischmann, N., and Siegrist, H. (2004). Water Sci. Tech. 50, 143–152. Rixen, T., Baum, A., Pohlmann, T., Balzer, W., Samiaji, J., and Jose, C. (2008). Biogeochemistry 90, 129–140. Robertson, G. P., and Tiedje, J. M. (1984). Soil Sci. Soc. Am. J. 48, 383–389. Rodriguez-Zaraqoza, S., Gonzalez-Ruiz, T., Gonzalez-Lozano, E., Lozada-Rojas, A., Mayzlish-Gati, E., and Steinberger, Y. (2008). Eur. J. Soil Biol. 44, 373–380. Rosenstock, B., Zwisler, W., and Simon, M. (2005). Microb. Ecol. 50, 90–101. Royer, I., Angers, D. A., Chantigny, M. H., Simard, R. R., and Cluis, D. (2007). J. Environ. Qual. 36, 855–863. Ruark, M. D., Brouder, S. M., and Turco, R. F. (2009). J. Environ. Qual. 38, 1205–1215. Ryden, J. C. (1986). In “Nitrogen Fluxes in Intensive Grassland Systems” (H. G. van Meer, J. C. Ryden, and G. C. Ennik, Eds.), pp. 59–73. Martinus Nijhoff Publishers, Dordrecht. Saadi, I., Borisover, M., Armon, R., and Laor, Y. (2006). Chemosphere 63, 530–539. Saari, P., Saarnio, S., Kukkonen, J. V. K., Akkanen, J., Heinonen, J., Saari, V., and Alm, J. (2009). Biogeochemistry 94, 217–231. Sachse, A., Babenzien, D., Ginzel, G., Gelbrecht, J., and Steinberg, C. E. W. (2001). Biogeochemistry 54, 279–296. Saggar, S., Hedley, M. J., and Phimsarn, S. (1998). In “Sulfur in the Environment” (D. G. Maynard, Ed.), pp. 45–94. Marcel Dekker, New York. Sakkas, V. A., Lambropoulou, D. A., and Albanis, T. A. (2002). Chemosphere 48, 939–945. Sanderman, J., Baldock, J. A., and Amundson, R. (2008). Biogeochemistry 89, 181–198. Schaecke, W., Tannenberg, H., and Shilling, G. (2002). J. Plant Nutr. Soil Sci. 165, 609–617. Schindler, D. W., Bayley, S. E., Curtis, P. J., Parker, B. R., Stainton, M. P., and Kelly, C. A. (1992). Hydrobiologia 229, 1–21. Schulze, K., Borken, W., Muhr, J., and Matzner, E. (2009). Eur. J. Soil Sci. 60, 567–577. Schulze, W. (2004). Biogeosciences Discussions 1, 825–853. Senesi, N. (1990a). Adv. Soil Sci. 14, 77–130. Senesi, N. (1990b). Anal. Chim. Acta 232, 77–106. Senesi, N. (1990c). Anal. Chim. Acta 232, 51–75. Senesi, N., and Loffredo, E. (1999). In “Soil Physical Chemistry” (D. L. Sparks, Ed.), 2nd edn., pp. 239–370. CRC Press, Boca Raton, Ch. 6. Senesi, N., and Plaza, C. (2007). CLEAN 35, 26–41. Senesi, N., and Sposito, G. (1984). Soil Sci. Soc. Am. J. 48, 1247–1253. Senesi, N., Bocian, D. F., and Sposito, G. (1985a). Soil Sci. Soc. Am. J. 49, 119–125. Senesi, N., Bocian, D. F., and Sposito, G. (1985b). Soil Sci. Soc. Am. J. 49, 114–119. Senesi, T. M., Miano, M. R., Provenzano, G., and Brunetti, S. (1989). Sci. Total Environ. 81/82, 143–156. Senesi, N., Chen, Y., and Schnitzer, M. (1977a). Soil Biol. Biochem. 9, 397–403. Senesi, N., Chen, Y., and Schnitzer, M. (1977b). Fuel 56, 171–176. Seol, Y., and Lee, L. S. (2000). Soil Sci. Soc. Am. J. 64, 1976–1983. Seol, Y., and Lee, L. S. (2001). J. Environ. Qual. 30, 1644–1652. Serrano, L. (1994). Aust. J. Mar. Fresh Res. 45, 1555–1564. Shao, J., Li, Y., Wei, C., and Xie, D. (2009). Chinese Geographical Science 19, 241–248. Sharp, J. H., Beuregard, A. Y., Burdige, D., Cauwet, G., Curless, S. E., Lauck, R., Nagel, K., Ogawa, H., Parker, A. E., Primm, O., Pujo-Pay, A., Savidge, W. B., Seitzinger, S., Spyres, G., and Styles, R. (2004). Mar. Chem. 84, 181–193. Shen, Y., Chen, L., Yang, J., Wang, M., and Nan, Z. (2008). In “Global Issues Paddock Action. Proceedings of the 14th Australian Agronomy Conference. September 2008, Adelaide South Australia” (M. J. Unkovich, Ed.). # Australian Society of Agronomy www.agronomy.org.au.

Dissolved Organic Matter

73

Shiller, A. M., Duan, S. W., van Erp, P., and Bianchi, T. S. (2006). Limnol. Oceanogr. 51, 1716–1728. Shuman, L. M. (1991). In “Micronutrients in Agriculture” (J. J. Morvedt, et al., Eds.), 2nd edn. SSSA Book Ser. 4, pp. 113–144. Soil Science Society of America, Madison, WI. Siemens, J., Haas, M., and Kaupenjohann, M. (2003). Geoderma 113, 253–272. Siira-Pietikainen, A., and Haimi, J. (2009). Forest Ecol. Manag. 258, 332–338. Sinsabaugh, R. L., Hill, B. H., and Shah, J. J. F. (2009). Nature 462, 795–796. Sirulnik, A. G., Allen, E. B., Meixner, T., and Allen, M. F. (2007). Soil Biol. Biochem. 39, 24–32. Smith, M. S., and Tiedje, J. M. (1979). Soil Biol. Biochem. 11, 261–267. Solinger, S., Kalbitz, K., and Matzner, E. (2001). Biogeochemistry 55, 327–349. Sollins, P., Homann, P., and Caldwell, B. A. (1996). Geoderma 74, 65–105. Spark, K. M., and Swift, R. S. (2002). Sci. Total Environ. 298, 147–161. Spencer, R. G. M., Pellerin, B. A., Bergamaschi, B. A., Downing, B. D., Kraus, T. E. C., Smart, D. R., Dahgren, R. A., and Hernes, P. J. (2007). Hydrol. Processes 21, 3181–3189. Sposito, G., Senesi, N., and Holtzclaw, K. M. (1988). Soil Sci. Soc. Am. J. 52, 632–636. Steenwerth, K., and Belina, K. M. (2008). Appl. Soil Ecol. 40, 359–369. Stevenson, F. J. (1994). Humus Chemistry: Genesis, Composition, Reactions. 2nd edn., John Wiley & Sons, New York. Stevenson, F. J. (1985). Cycles of soil: Carbon, nitrogen, phosphorus, sulfur and micronutrients. John Wiley & Sons, New York. Stewart, A. J., and Wetzel, R. G. (1981). Limnol. Oceanogr. 26, 590–597. Stockdale, E. A., Gaunt, J. L., and Vos, J. (1997). Eur. J. Agron. 7, 145–159. Stottlemeyer, R., and Toczydlowski, D. (2006). Biogeochemistry 77, 409–440. Strobel, B. W., Hansen, H. C. B., Borggaard, O. K., Andersen, M. K., and RaulundRasmussen, K. (2001). Biogeochemistry 56, 1–26. Strobel, B. W., Bernhoft, I., and Borggaard, O. K. (1999). Plant Soil 212, 115–121. Stumpe, B., and Marschner, B. (2010). Environ. Pollut. 158, 148–154. Stutter, M. I., Lumsdon, D. G., and Thoss, V. (2007). Eur. J. Soil Sci. 58, 646–657. Suominen, K., Kitunen, V., and Smolander, A. (2003). Eur. J. Soil Sci. 54, 287. Suseela, K., Sivaparvathi, M., and Nandy, S. C. (1987). Leather Science 34, 149–156. Swift, M., Heal, O. W., and Anderson, J. M. (1979). Studies in Ecology 5, 372. Szajdak, L., Brandyk, T., and Szatylowicz, J. (2007). Agron. Res. 5, 165–174. Tate, C. M., and Meyer, J. L. (1983). Ecology 64, 25–32. Thevenot, M., Dousset, S., Hertkorn, N., Schmitt-Kopplin, P., and Andreux, F. (2009). Sci. Total Environ. 407, 4297–4302. Thomsen, M., Dobel, S., Lassen, P., Carlsen, L., Mogensen, B. B., and Hansen, P. E. (2002). Chemosphere 49, 1317–1325. Thurman, E. M. (1985). Organic Geochemistry of Natural Waters. Martinus Nijhoff/Dr W Junk, Dordrecht, The Netherlands. Tipping, E. (1981). Chem. Geol. 33, 81–89. Titeux, H., Brahy, V., and Delvaux, B. (2002). Geoderma 107, 93–107. Traversa, A., D’Orazio, V., and Senesi, N. (2008). Forest Ecol. Manag. 256, 2018–2028. Tue-Ngeun, O., Sandford, R. C., Jakmunee, J., Grudpan, K., McKelvie, I. D., and Worsfold, P. J. (2005). Anal. Chim. Acta 554, 17–24. Undurraga, P. D., Zagal, E. V., Sepu´lveda, G. W., and Valderrama, N. V. (2009). Chil. J. Agric. Res. 69, 445–454. Uselman, S. M., Qualls, R. G., and Lilienfein, J. (2007). Soil Sci. Soc. Am. J. 71, 1555–1563. Ussiri, D. A. N., and Johnson, C. E. (2004). Soil Sci. Soc. Am. J. 68, 253–262. Valeur, I., Andersson, S., and Nilsson, S. L. (2000). Biogeochemistry 50, 1–20. Vallejo, A., Skiba, U. M., Garcia-Torres, L., Arce, A., Lopez-Fernandez, S., and SanchezMartin, L. (2006). Soil Biol. Biochem. 38, 2782–2793.

74

Nanthi S. Bolan et al.

van Hees, A. W., Andersson, A. M. T., and Lundstro¨m, U. S. (1996). Chemosphere 33, 1951–1966. van Kessel, C., Clough, T., and van Groenigen, J. W. (2009). J. Environ. Qual. 38, 393–401. Vandenbruwane, J., De Neve, S., Qualls, R. G., Sleutel, S., and Hofman, G. (2007). Geoderma 139, 144–153. Vestin, J. L. K., Norstrom, S. H., Bylund, D., and Lundstrom, U. S. (2008). Geoderma 144, 271–278. Vink, S., Ford, P. W., Bormans, M., Kelly, C., and Turley, C. (2007). Biogeochemistry 84, 247–264. Vinther, F. P., Hansen, E. M., and Eriksen, J. (2006). Biol. Fertil. Soils 43, 12–19. Voelker, B. M., Morel, F. M. M., and Sulzberger, B. (1997). Environ. Sci. Technol. 31, 1004–1011. Vogt, K. A., Moore, E. E., Vogt, D. J., Redlin, M. J., and Edmonds, R. L. (1983). Can. J. For. Res. 13, 429–437. Volk, C., Wood, L., Johnson, B., Robinson, J., Zhu, H. W., and Kaplan, L. (2002). J. Environ. Monitor. 4, 43–47. Volk, C. J., Volk, C. B., and Kaplan, L. A. (1997). Limnol. Oceonogr. 42, 39–44. Vulkan, R., Zhao, F. J., Barbosa-Jefferson, V., Preston, S., Paton, G. I., Tipping, E., and McGrath, S. P. (2000). Environ. Sci. Technol. 34, 5115–5121. Wachendorf, C., Taube, F., and Wachendorf, M. (2005). Nutr. Cycl. Agroecosys. 73, 89–100. Waldrop, M. P., and Firestone, M. K. (2004). Oecologia 138, 275–284. Wallstedt, T., Borg, H., Meili, M., and Morth, C. M. (2008). Sci. Total Environ. 407, 405–417. Ward, S. E., Bardgett, R. D., McNamara, N. P., Adamson, J. K., and Ostle, N. J. (2007). Ecosystems 10, 1069–1083. Wetzel, R. G., Hatcher, P. G., and Bianchi, T. S. (1995). Limnol. Oceanogr. 40, 1369–1380. Williams, B. L., and Edwards, A. C. (1993). Chem. Ecol. 8, 203–215. Williams, C. F., Farmer, W. J., Letey, J., and Nelson, S. D. (1999). J. Environ. Qual. 28, 1757–1760. Williams, C. F., Agassi, M., Letey, J., Farmer, W. J., Nelson, S. D., and Ben-Hur, M. (2000). Soil Sci. Soc. Am. J. 64, 590–594. Wong, J. W. C., Li, K. L., Zhou, L. X., and Selvam, A. (2007). Geoderma 137, 310–317. Wood, J. (1974). Science 15, 1049–1052. Worrall, F., and Burt, T. P. (2007). J. Hydrol. 346, 81–92. Worrall, F., Parker, A., Rae, J. E., and Johnson, A. C. (1999). J. Environ. Qual. 28, 595–604. Worrall, F., Burt, T., and Shedden, R. (2003). Biogeochemistry 64, 165–178. Xiao-Gang, L., Li, F., Zed, R., Zhan, Z., and Singh, B. (2007). Geoderma 139, 98–105. Xiao, C. Q., Ma, L. Q., and Sarigumba, T. (1999). J. Environ. Qual. 28, 321–333. Yallop, A. R., and Clutterbuck, B. (2009). Sci. Total Environ. 407, 3803–3813. Yamashita, Y., and Tanoue, E. (2004). Org. Geochem. 35, 679–692. Yongkui, Y., Li, L., and Dingyong, W. (2008). J. Environ. Sci. 20, 1097–1102. Yoshimura, K., Ogawa, T., and Hama, T. (2009). Mar. Chem. 114, 11–18. Young, T. A., Heidler, J., Matos-Perez, C. R., Sapkota, A., Toler, T., Gibson, K. E., Schwab, K. J., and Halden, R. U. (2008). Environ. Sci. Technol. 42, 3335–3340. Yurova, A., Sirin, A., Buffam, I., Bishop, K., and Laudon, H. (2008). Water Resour. Res. 44, W07411. Zablotowicz, R. M., Accinelli, C., Krutz, L. J., and Reddy, K. N. (2009). J. Agr. Food Chem. 57, 4867–4871. Zech, W., Guggenberger, G., and Schulten, H. R. (1994). Sci. Total Environ. 152, 49–62. Zech, W., Guggenberger, G., Haumaier, L., Po¨hhacker, R., Scha¨fer, D., Amelung, W., Miltner, A., Kaiser, K., and Ziegler, F. (1996). In “Humic Substances in Terrestrial Ecosystems” (A. Piccolo, Ed.), pp. 101–170. Elsevier, Amsterdam.

Dissolved Organic Matter

75

Zhang, Y. Q., and Frankenberger, W. T. Jr. (1999). J. Environ. Qual. 28, 1321–1326. Zhang, B., and Horn, R. (2001). Geoderma 99, 123–145. Zhang, J. B., Song, C. C., and Wang, S. M. (2007). Soil Tillage Res. 96, 350–360. Zhang, J. B., Cai, Z. C., Cheng, Y., and Zhu, T. B. (2009). Soil Biol. Biochem. 41, 2551– 2557. Zhao, L. Y. L., Schulin, R., and Nowack, B. (2007a). Environ. Sci. Technol. 41, 2770–2775. Zhao, L. Y. L., Schulin, R., Weng, L. P., and Nowack, B. (2007b). Geochim. Cosmochim. Acta 71, 3407–3418. Zhaohai, S., Li, M., Qiming, X., Yijun, Y., Hui, L., and Hongxia, Y. (2008). J. Environ. Sci. 20, 1075–1081. Zhou, Q., Maurice, P. A., and Cabaniss, S. E. (2001). Geochim. Cosmochim. Acta 65, 803– 812. Ziegler, F. (1991). Environ. Sci. Pollut. Res. 3, 276–277. Zmora-Nahuma, S., Markovitcha, O., Tarchitzkyb, J., and Chena, Y. (2005). Biochem. 37, 2109–2116. Zsolnay, A. (1996). In “Humic Substances in Terrestrial Ecosystems” (A. Piccolo, Ed.), pp. 171–223. Elsevier, Amsterdam. Zsolnay, A. (2003). Geoderma 113, 187–210. Zsolnay, A., and Steindl, H. (1991). Soil Biol. Biochem. 23, 1077–1082.

C H A P T E R

T W O

Genomic Selection in Plant Breeding: Knowledge and Prospects Aaron J. Lorenz,* Shiaoman Chao,† Franco G. Asoro,‡ Elliot L. Heffner,§ Takeshi Hayashi,} Hiroyoshi Iwata,# Kevin P. Smith,k Mark E. Sorrells,§ and Jean-Luc Jannink* Contents 1. Introduction 2. Important Population and Trait Characteristics 3. Single Nucleotide Polymorphism Marker Discovery and Genotyping 4. Statistical Methods 4.1. Random regression best linear unbiased prediction 4.2. Least absolute shrinkage and selection operator 4.3. Reproducing kernel Hilbert spaces and support vector machine regression 4.4. Partial least squares regression and principle component regression 4.5. Bayesian methods 4.6. Statistical methods summary 5. GS Prediction Accuracies 5.1. Evaluating GEBV accuracy through CV 5.2. Reported accuracies 5.3. Marker density, marker type, and training population size 5.4. Effect of training population size and marker number on GEBV accuracy in barley, oat, and wheat data sets 6. Impact of Statistical Model on GEBV Accuracy 7. Modeling Epistasis and Dominance 8. GS in the Presence of Strong Subpopulation Structure

78 80 82 84 85 86 87 89 90 92 94 94 95 97 100 103 107 109

* R.W. Holley Center for Agriculture and Health, USDA-ARS, Ithaca, New York, USA { Biosciences Research Laboratory, USDA-ARS, Fargo, North Dakota, USA { Department of Agronomy, Iowa State University, Ames, Iowa, USA } Department of Plant Breeding and Genetics, Cornell University, Ithaca, New York, USA } Data Mining and Grid Research Team, National Agricultural Research Center, Tsukuba, Ibaraki, Japan # Department of Agricultural and Environmental Biology, Graduate School of Agriculture & Life Sciences, University of Tokyo, Bunkyo, Tokyo, Japan k Department of Agronomy and Plant Genetics, University of Minnesota, St. Paul, Minnesota, USA Advances in Agronomy, Volume 110 ISSN 0065-2113, DOI: 10.1016/B978-0-12-385531-2.00002-5

#

2011 Elsevier Inc. All rights reserved.

77

78

Aaron J. Lorenz et al.

9. Long-Term Selection 10. Summary and Conclusions References

111 114 116

Abstract “Genomic selection,” the ability to select for even complex, quantitative traits based on marker data alone, has arisen from the conjunction of new highthroughput marker technologies and new statistical methods needed to analyze the data. This review surveys what is known about these technologies, with sections on population and quantitative genetic background, DNA marker development, statistical methods, reported accuracies of genomic selection (GS) predictions, prediction of nonadditive genetic effects, prediction in the presence of subpopulation structure, and impacts of GS on long-term gain. GS works by estimating the effects of many loci spread across the genome. Marker and observation numbers therefore need to scale with the genetic map length in Morgans and with the effective population size of the population under GS. For typical crops, the requirements range from at least 200 to at most 10,000 markers and observations. With that baseline, GS can greatly accelerate the breeding cycle while also using marker information to maintain genetic diversity and potentially prolong gain beyond what is possible with phenotypic selection. With the costs of marker technologies continuing to decline and the statistical methods becoming more routine, the results reviewed here suggest that GS will play a large role in the plant breeding of the future. Our summary and interpretation should prove useful to breeders as they assess the value of GS in the context of their populations and resources.

1. Introduction “Genomic selection” comprises methods that use genotypic data across the whole genome to predict any trait with an accuracy sufficient to allow selection on that prediction alone. Imagine: chip a seed, extract its DNA, and discard it or select it as a parent of the next generation. The potential acceleration of the breeding cycle and the increase in selection intensity enabled are breathtaking (Heffner et al., 2010; Schaeffer, 2006). A 4-year breeding cycle, including 3 years of field testing, can be reduced to only the 4 months required to grow and cross a plant. Thousands of selection candidates can be evaluated without ever taking them out to the field. Ongoing field trials are still very much a part of a breeding program using genomic selection (GS), but the purpose, and therefore the practice, shifts: phenotypes are no longer used to select but to train a prediction model (Heffner et al., 2009). The statistical methods used by GS are relatively new to the plant-breeding community. The integration of GS into practical breeding schemes has hardly begun. But it is already clear that

Genomic Selection in Plant Breeding: Knowledge and Prospects

79

GS will dramatically redirect resources and activities in plant-breeding programs and affect the ability to deliver crop improvement. Since the 1980s, plant geneticists and breeders have worked to include molecular marker technology into breeding programs (Stuber et al., 1982; Tanksley et al., 1989). The technologies first led to quantitative trait locus (QTL) identification (Kearsey and Farquhar, 1998; Lander and Botstein, 1989; Paterson et al., 1988), and then to marker-assisted introgression (Hospital and Charcosset, 1997) and selection (Bernardo and Charcosset, 2006; Hospital et al., 1997; Lande and Thompson, 1990; Moreau et al., 1998; Xu and Crouch, 2008). These methods have been quite useful for the manipulation of large-effect alleles with known association to a marker (e.g., Zhong et al., 2006) but not for quantitative traits that have still required extensive field testing (e.g., Moreau et al., 2004). The methods of marker-assisted selection (MAS) or marker-assisted recurrent selection (MARS) assume that the user knows which alleles are favorable, and what their average effects on the phenotype are (Bernardo and Charcosset, 2006; Charmet et al., 1999; Hospital et al., 2000). This assumption is viable for major-gene traits but not for quantitative traits that are influenced by many loci of small effect and the environment. In locus identification and effect estimation for such traits, much uncertainty will remain (Beavis, 1994; Melchinger et al., 1998; Scho¨n et al., 2004). To deal with quantitative traits, new statistical approaches that could account for this uncertainty were needed to generate the best predictions possible. In particular, the difficulty with locus identification entailed that the effects for all marker loci be simultaneously estimated (Meuwissen et al., 2001). A further benefit of this simultaneous estimation was that even effects that might be too small to be declared “significant” might be captured by markers (Meuwissen et al., 2001). In presenting these ideas, Meuwissen et al. (2001) named the methods “genomic selection,” a name that has been maintained within the animal breeding community. In the plant-breeding community, both “genomic selection” (Heffner et al., 2009) and “genome-wide selection” (Bernardo and Yu, 2007) are in use. Because of the older origin of “genomic selection” and because we find the term descriptive of the use of wholegenome data for selection, we will use the term “genomic selection,” or GS. The essential steps of GS are as follows: In a population under selection, some lines, called the training set, have both genotypic and phenotypic data. Statistical analysis of the training set simultaneously estimates allele effects at all loci. Other lines, the selection candidates, are genotyped allowing their genomic estimated breeding values (GEBVs) to be calculated. Selection proceeds on the basis of these GEBVs. The simplicity of this statement masks the magnitude of the shift it represents. As long as the phenotype is the selection criterion, the line is the unit of evaluation. When a prediction based on allele effects is the selection criterion, the allele becomes the unit of evaluation. Alleles are therefore also the unit that needs to be replicated

80

Aaron J. Lorenz et al.

within and across environments. But that replication can occur regardless of the specific lines carrying the alleles such that lines themselves no longer need to be replicated. An analogous point has been made in the context of QTL mapping (Knapp and Bridges, 1990). In the breeding context, removing the need for line replication opens the possibility of dramatically expanding the number of lines pushed through the pipeline of a breeding program, and in turn of increasing selection intensity. In simulation and empirical studies, GEBVs based solely on individuals’ genotype have been remarkably accurate (Habier et al., 2007; Legarra et al., 2008; Lorenzana and Bernardo, 2009; Meuwissen et al., 2001; VanRaden et al., 2009; Zhong et al., 2009). These accuracies depend on the characteristics of the population that is under selection. Consequently, the first section of this review focuses on defining the relevant aspects of linkage disequilibrium (LD), effective population size, and effective QTL number. Second, a primer on marker discovery and genotyping platforms that a GS practitioner may find herself using is given. GS is still in its infancy, and several statistical methods to estimate marker effects and calculate GEBVs are being developed and compared. We describe a cross-section of these methods, seeking to highlight their commonalities. Numerous factors contribute to the success of GS. In particular, we provide discussions of the general impacts of training population size and marker density and type on prediction accuracies; factors that influence the success of particular statistical methods; the value of modeling epistatic and dominant modes of gene action; how subpopulation structure influences GS; and the impacts of GS on long-term selection. Each discussion outlines what is known through theory as well as findings in both in silico and empirical research. Because few empirical results on GS applied to crops have been obtained so far, this review also includes a brief analysis of data sets gathered by the authors. The review ends with summary conclusions and discussion of their implications for the implementation of GS in plant breeding.

2. Important Population and Trait Characteristics In QTL mapping research, an early critical issue to resolve is the choice of population in which segregating causal loci are to be identified, because marker and QTL polymorphism in that population will determine the probability of success. In GS, the assumption is that model development and selection will take place directly in the breeding population of interest. To avoid confusion, we define “population” here as the full set of individuals in a breeding program who are being evaluated for their performance in a particular target environment, and who may be selected and crossed

Genomic Selection in Plant Breeding: Knowledge and Prospects

81

with one another to initiate a future cycle of selection. In our definition, “population” denotes a broader set of individuals than the progeny from a single biparental cross, which we will call a “family,” whether the progeny is inbred or not. Unless otherwise noted, therefore, a “population” in this review consists of multiple families that may have different numbers of progeny and need not originate from any particular mating design. The most important characteristic of the population is its effective size. An obvious measure of population size is its census: how many individuals it contains. But populations with the same census size can behave quite differently in terms of their rates of inbreeding depending on the number of individuals contributing gametes to the following generation (Falconer and Mackay, 1996). For a population of a given rate of inbreeding, the effective size is equal to the census size of a randomly mating (“ideal”) population that would have that same rate. The rate of inbreeding is equivalent to the magnitude of random genetic drift, that is, the shift of allele frequencies that occurs due to the sampling of alleles that are contributed to progeny. Drift is an important cause in generating LD, the nonindependence of alleles at different loci. This nonindependence allows marker alleles to predict the allelic state of nearby QTL, enabling marker genotypes to predict the phenotype. Alleles are independent if the frequency of their combination is equal to the product of their separate frequencies. Hence, drift generates LD because it is unlikely for combinations to be sampled at that frequency by chance. At equilibrium, driftgenerating LD is balanced by recombination, causing it to decay, such that nearby loci are expected to be in higher LD than faraway loci. The lower the effective population size, the more rapidly the drift operates and the stronger LD will be between loci. The first consequence of this relationship between effective population size and LD is that marker density needs to scale with effective population size. That is, for the average LD between adjacent markers to be equal in two populations, the number of markers per centimorgan divided by the effective population size should be the same for each population (Meuwissen, 2009). The second consequence of effective population size comes from analytical formulae giving the expected accuracy of GS as a function of training population size, trait heritability, and the number of marker effects to estimate (Daetwyler et al., 2008; Hayes et al., 2009c). A crucial parameter in these formulae is the ratio between the training population size and the number of marker effects to estimate. As just noted, the number of marker effects to estimate should be proportional to the effective population size. Therefore, to maintain a constant ratio of the training population size to the number of effects estimated, the training and effective population sizes need to scale together (Meuwissen, 2009). For crops, elite breeding programs generally have fairly small effective population sizes. For example, from data in Hamblin et al. (2010) we can infer that

82

Aaron J. Lorenz et al.

for elite barley, effective population size is not greater than 50. Relative to other species such as dairy cattle, or humans, this small size suggests that cereal breeders will be able to make accurate predictions with relatively small training populations. The effective QTL number gives a standardized measure of the degree to which a trait is polygenic. Consider a trait that is influenced by a great many loci. If only three loci have a major impact on the trait, while the rest have a very minor impact, then, to a first approximation, the trait can be modeled as trigenic. If, on the contrary, many loci have equal and small impact on the trait, the trait should be modeled as polygenic. Lande (1981) showed that the effective QTL number can be expressed as the ratio of the square of the sum of QTL variances to the sum of the square of QTL variances. When QTL variances follow a geometric series, that is, they are proportional to 1, a, a2, a3, . . . with a 1, the effective QTL number is then equal to (1 þ a)/(1 a) (Lande and Thompson, 1990). The effective QTL number can be estimated empirically on the basis of its response to divergent selection (Falconer and Mackay, 1996; Lande, 1981; Zeng, 1992) using a method that relies on the fact that selection limits will be more divergent for traits with high than low effective QTL number. Other methods use QTL analysis experiments to estimate, in a false discovery rate framework (Benjamini and Yekutieli, 2005), the proportion of QTL tests for which the null hypothesis is false (e.g., Blanc et al., 2006). From there, the number of loci affecting the trait can be estimated (Chamberlain et al., 2007). Not surprisingly, the performance of GS improves with decreasing effective QTL number (Hayes et al., 2009c; Zhong et al., 2009).

3. Single Nucleotide Polymorphism Marker Discovery and Genotyping The ability to perform GS requires routine genotyping at a high number of loci. Single nucleotide polymorphisms (SNPs) differentiate individuals based on variations detected at the level of a single nucleotide base in the genome. SNPs have become the marker of choice for crop genetics and breeding applications because of their high abundance in genomes, and the availability of a wide array of genotyping platforms with various multiplex capabilities for SNP analysis (Rafalski, 2002). Recent breakthroughs in nextgeneration sequencing (NGS) technologies enabled millions of sequence reads to be generated from a single run at a more affordable cost. The resulting large amount of data provided sequence depth adequate for de novo sequence assembly, which has made the SNP discovery on a large scale a feasible task, particularly for species without completed genome sequences. Successful results on large-scale discovery of SNPs based on

Genomic Selection in Plant Breeding: Knowledge and Prospects

83

NGS methods have been reported in several plant species, including both diploid (Barbazuk et al., 2007; Cheung et al., 2006; Gore et al., 2009; Hyten et al., 2010; Weber et al., 2007) and polyploid (Trick et al., 2009) species, and more are on the way. Details on the mechanisms of current NGS platforms, the available bioinformatics tools for sequence data analysis and SNP identification, and their impact on future genetics studies and crop improvement can be found in several recent review articles (Deschamps and Campbell, 2010; Mardis, 2008; Shendure and Ji, 2008; Varshney et al., 2009). Methods to identify SNPs from gene-enriched and complexity reduced portions of the genome have been developed with the de novo sequencing approach (Gore et al., 2009; Hyten et al., 2010; Maughan et al., 2009; Van Orsouw et al., 2007). Fu et al. (2010) demonstrated the use of sequence capture techniques to enrich nonrepetitive genomic representations to discover SNPs in targeted gene regions, and to recover novel sequences from nonreference alleles. In addition to de novo sequencing, sources of SNP markers and strategies for SNP identification previously described (Ganal et al., 2009) include mining existing expressed sequence tags (ESTs) in public databases (Close et al., 2009), resequencing the PCR amplicons derived from genes of interest with known sequences, and comparing sequenced genomes. In order to achieve a high success rate in genotyping assays, validation of candidate SNPs will be necessary. The conversion rate from discovered SNPs to working assays depends on many factors such as the levels of sequencing error, sequence composition near the targeted SNPs, and the genotyping assay systems used. Validations can be done by resequencing the same samples using a different sequencing method, such as the conventional Sanger’s method (Barbazuk et al., 2007; Hyten et al., 2010), resequencing different samples not included in the discovery panel (Trick et al., 2009), or by genotyping assays (Chao et al., 2009; Close et al., 2009). The particular germplasm and the number of populations used for the SNP discovery panel will affect the levels of SNP polymorphism and allele frequency distribution detected in independent populations; this is known as ascertainment bias (Clark et al., 2005). An example from a recent barley study illustrated that the SNPs discovered and selected based on a small set of cultivars were not able to reveal the same amount of genetic diversity among landraces previously found highly diverse on the basis of simple sequence repeat (SSR) data (Moragues et al., 2010). In general, the principles underlying the SNP assays involve generating allele-specific products in biochemical reactions and identifying the products using detection procedures (Chen and Sullivan, 2003; Syvanen, 2001). Currently, a wide variety of SNP genotyping systems that use different chemistries and detection systems to assay SNPs are commercially available. Moreover, the number of SNP genotypes scored per reaction for each sample ranges from one to over one million on different platforms

84

Aaron J. Lorenz et al.

(Perkel, 2008; Ragoussis, 2006; Syvanen, 2005). The choice of a suitable genotyping platform, thus, will depend on the user’s criteria, including the flexibility of the systems, desired data throughput, the applications, and per sample cost (Chen and Sullivan, 2003). The technological improvements in sequencing and SNP genotyping have resulted in the generation of a wealth of information. As the cost continues to drop for whole-genome sequencing (Podolak, 2010), genetic mapping and genotyping by resequencing are now possible for plants with complete genomic sequences (Huang et al., 2009). For the majority of the crops, however, it is more feasible to develop genome-wide SNP markers using the NGS techniques. Such marker resources will be highly valuable for dissecting the genetic architecture of complex agronomic traits and facilitating GS. The development of highly parallel SNP assays, such as Illumina’s GoldenGate assay with 1536-plex platform (Fan et al., 2003), enabled the genome-wide studies previously not feasible for economically important crops. Using these techniques, SNP-based high-density genetic maps are now available in several crop plants such as soybean (Hyten et al., 2008), maize (McMullen et al., 2009), barley (Close et al., 2009, Szu¨cs et al., 2009), and wheat (Luo et al., 2009). Thus, genotyping lines for use in GS using SNP and direct resequencing with next-generation methods appear likely to be dominant methods for the foreseeable future.

4. Statistical Methods GS emerged out of a desire to exploit high-density parallel genotyping technologies for prediction of genetic value for complex traits (Meuwissen et al., 2001). The authors noted the arbitrariness in MAS of setting marker effects either to zero or to their full value depending on whether their significance is just below or above some predetermined threshold. It is for this reason that marker effects are often greatly overestimated in typical QTL mapping and MAS studies (Lande and Thompson, 1990), that is, the Beavis effect (Xu, 2003). Meuwissen et al. (2001) desired to minimize biased marker effects and hence avoided marker selection during effect estimation and subsequent calculation of genetic values. A consequence of that decision was that more predictor effects, p, needed to be estimated than the number, n, of observations, the so-called “large p, small n” problem. In this case, not enough degrees of freedom exist for estimating all predictor effects simultaneously using least squares (i.e., singular X0 X matrix in the ordinary least squares estimator). Even if sufficient degrees of freedom were available, a high degree of multicollinearity among markers would likely be present, producing an overfitted model. An overfitted model can exaggerate minor fluctuations in the data and will generally have poor predictive ability.

Genomic Selection in Plant Breeding: Knowledge and Prospects

85

To overcome these problems, a variety of statistical models have been proposed for GS. These models can largely be grouped into shrinkage models, variable selection models, kernel methods, and dimension reduction methods. A cross-section of these methods is described. Following previous authors (Moser et al., 2009), the basic model will be denoted as yi ¼ gðxi Þ þ ei

ð1Þ

where yi is an observed phenotype of individual i (i ¼ 1 . . . n) and xi is a 1 p vector of SNP genotypes on individual i, g(xi) is a function relating genotypes to phenotypes, and ei is a residual term. The GEBV is generally equal to g(xi). Further similarities among GS models can be seen by recognizing that they all seek to minimize a certain cost function. In least squares analysis, the well-known cost function is simply the sum of squared residuals, Sei2.

4.1. Random regression best linear unbiased prediction Random regression best linear unbiased prediction (RR-BLUP), also known as ridge regression, was first proposed for MAS by Whittaker et al. (2000) in the context of biparental crosses. gðxi Þ ¼

p X

xik bk

ð2Þ

k¼1

where xik indicates the score for SNP k in individual i, bk is the effect associated with marker k, and the genetic value is the sum of p marker effects. The normal least squares estimators are modified so that b is estimated using ^ ¼ ðX0 X þ lIÞ1 X0 y b

ð3Þ

where X is an incidence matrix relating markers to individuals, I is an identity matrix, and y is a vector of estimated breeding values that may be phenotypes or may come from other analyses. The difference between the RR-BLUP estimator of b and the ordinary least squares estimator is the presence of the lI term, which is introduced to make X0 X nonsingular and reduce collinearity between predictors. Whittaker et al. (2000) chose a l value by testing a range of values and selecting the value that minimized model error. Another way to estimate an optimal l is to assume that marker effects are randomly drawn from a common normal distribution centered on zero, and solve the mixed linear model equations of Henderson (1975). In this case, l is equal to var(e)/var

86

Aaron J. Lorenz et al.

(b), where var(e) is the residual variance and var(b) is the common marker effect variance (Piepho, 2009). Since a small var(b) relative to var(e) will cause marker effects to be more strongly shrunken toward zero, this equivalence shows that large l has that effect too. The effect of l can also be seen in the RR-BLUP cost function: 1X 2 l 0 e þ bb 2 i i 2

ð4Þ

where b is a vector of the b. The second term of this function is called a constraint, penalty, or regularizer. Because of it, a large l creates a greater cost for large b, causing them to shrink more. The fact that all marker effects are identically distributed means that all effects are equally shrunken toward zero (this assumption does not mean that all effects are equal). Meuwissen et al. (2001) used this approach and termed it simply BLUP. In the livestock breeding community, the term BLUP has been overwhelmingly used for animal breeding values predicted using the standard animal model including an additive relationship matrix calculated from pedigree data (Lynch and Walsh, 1998). Gianola et al. (2003) outline the advantages of estimating dense marker effects in a mixed models framework compared to standard RR-BLUP. A major advantage is the flexibility of mixed models in modeling covariance between effects. An important aid in understanding and using RR-BLUP came from the recognition that it is statistically equivalent to fitting a mixed model analysis with lines (rather than markers) as random effects that covary according to a relationship matrix calculated using the marker data (Goddard, 2009; Habier et al., 2007; Hayes et al., 2009c). This equivalence can be seen by putting Eq. (3), which expresses the line effect, in matrix notation, g(xi) ¼ Xb. The covariance of the line effect is then var[g(xi)] ¼ XX0 sb2. Thus, if the line relationship matrix is calculated as XX0 sb2, the mixed model analysis will produce the same predictions for line effects as RRBLUP, irrespective of the trait genetic architecture or the marker density and distribution. When this parameterization is used, it has been referred to as G-BLUP (genomic relationship matrix; Luan et al., 2009) or RA-BLUP (realized additive relationship matrix; Zhong et al., 2009). Piepho (2009) went further and showed the connection between RR-BLUP, genetic covariance, and spatial models and concepts used in geostatistics.

4.2. Least absolute shrinkage and selection operator The relationship between least absolute shrinkage and selection operator (LASSO) and RR-BLUP can best be seen through the LASSO cost function (Bishop, 2006; Tibshirani, 1996):

Genomic Selection in Plant Breeding: Knowledge and Prospects

1X 2 lX e þ jbk j 2 i i 2 k

87

ð5Þ

As in RR-BLUP, there is a penalty for large marker effects (or regression coefficients). In fact, the cost functions of RR-BLUP and LASSO can be generalized to 1X 2 lX e þ jbk jq 2 i i 2 k

ð6Þ

where q ¼ 1 for LASSO, q ¼ 2 for RR-BLUP, and, in general, for 1 q 2, the method is called elastic net regression. Because of the different formulation of the regularizer, LASSO makes coefficients shrink more strongly than RR-BLUP, and some of the coefficients are driven to zero, leading to a sparse model. Thus, LASSO also performs variable selection (hence the “selection operator” in its name). A challenge for LASSO is the choice of l, which influences the size of the variable subset being selected. As l approaches zero, the solution converges on the ordinary least squares solution, while large values of l strongly reduce the absolute value of regression coefficients (de los Campos et al., 2009b). To optimize l, Usai et al. (2009) proposed using a least angle regression (LARS) algorithm that included a cross-validation (CV) step. This avoids the computationally expensive method of quadratic program as originally proposed by Tibshirani (1996). Park and Casella (2008) presented a Bayesian version of LASSO (BL). In the Bayesian view, l sets the prior distribution on the regression coefficients. In the non-Bayesian LASSO, the solution admits up to n 1 nonzero regression coefficients. This is a problem because in the case of dense marker data, there is no reason why the number of individuals in the training set should constrain the number of markers with nonzero effects (de los Campos et al., 2009b). The BL is able to relax this constraint, perhaps producing a more accurate model. de los Campos et al. (2009b) used this model in the context of GS and proposed an alternative version of BL to reduce the impact of the prior. They proposed that the prior on l be formulated using a Beta distribution, allowing vague prior preferences over a wide range of l values.

4.3. Reproducing kernel Hilbert spaces and support vector machine regression Gianola et al. (2006) was the first to apply reproducing kernel Hilbert spaces (RKHS) regression for GS by combining a classical additive genetic model with a kernel function. A kernel function converts predictor variables to a

88

Aaron J. Lorenz et al.

set of distances among observations to produce an n n positive definite matrix to be used in a linear model (Scho¨lkopf et al., 2004). Input space and feature space are terms often encountered in this literature. In the context of GS, the input data are the marker score data and hence, the input space is the multidimensional space in which the location of each individual is determined by their marker scores. The input space is converted to the feature space by applying a kernel function to the input data. RKHS can be represented in matrix notation as gðxÞ ¼ Kh a

ð7Þ

where Kh is the matrix of kernel entries quantifying the distances of individuals to one another, much like the additive relationship matrix quantifying pedigree relationships, and therefore genetic distances, between individuals. The difference here is that kernel functions include one or more smoothing parameters (h) to influence the relationship between distances in the feature space and distances in input space. The n 1 vector a can be construed as individual effects within the feature space that can be estimated using standard mixed-model equations (Gianola and de los Campos, 2008). CV or bootstrap methods can be applied to determine the optimal value of h, allowing great flexibility in models between populations and traits, for instance. RKHS methods can be interpreted within the standard quantitative genetics model framework (de los Campos et al., 2009a) Gianola et al. (2006) argued that standard parametric methods for quantitative genetic analysis are not well suited to model the complex interactions existing in dense marker data for the purposes of predicting phenotype from genotype. Because kernel methods contain a great deal of flexibility and no assumptions of linearity, they may be superior in their ability to capture nonadditive effects of all kinds simultaneously. In fact, Gianola et al. (2006) used a model that included a standard, parametric term for additive genetics, but also included a RKHS term to capture nonadditive effects. RKHS was superior to an RR-BLUP model for predicting simulated traits with a large nonadditive component (Gianola et al., 2006). Also, RKHS performed relatively well for predicting mortality in broilers, a trait of low heritability and categorical nature, making it a challenging trait for standard parametric models. Support vector regression (SVR) is also a kernel method whose cost function can be so-called “e-intensive,” in which case only residuals greater than e are penalized. This cost function can be written as C

X

L ðe Þ i e i

1 þ bT b 2

ð8Þ

Genomic Selection in Plant Breeding: Knowledge and Prospects

89

where C is a constant to weight the contribution of error to the cost function and Le is an “e-intensive loss function,” meaning that errors are not penalized unless they exceed e: 0; forjei j < e Le ðei Þ ¼ e; otherwise e j ij The practitioner sets C and e. Normally, the total sum of residuals, whether large or small, penalizes regression models; here, only models minimizing “large” (>e) residuals are favored, while all other residuals are acceptable. This procedure is designed to develop models that sacrifice many small errors for the exclusion of large errors. The solution to this problem involves Lagrange optimization, which produces a set of coefficients to be multiplied by the sum of all possible inner products between training samples and the observation being predicted. Here, the inner product between training observations i and j is defined as xi 0 xj , where x is a vector of the observation marker scores. To allow nonlinearity, the inner product term can be replaced by a kernel function, effectively transporting the SVR optimization from the input space to a feature space. A more detailed tutorial on SVR was presented by Smola and Scho¨lkopf (2004). Also, a very approachable summary can be found in Thissen et al. (2004). Maenhout et al. (2008) used SVR and a Gaussian kernel (a commonly used kernel function) to predict general (GCA) and specific (SCA) combining abilities of parental inbred lines from SSR and amplified fragment length polymorphism (AFLP) markers. To predict the hybrid performance of specific parent combinations, Maenhout et al. (2007) created a kernel by combining the genetic distance metrics Jaccard similarity measure and Modified Roger’s distance. Their rationale was that hybrid performance is a function of the genetic distance between parents and combining the two metrics allowed optimal use of both the AFLP (dominant marker) and SSR (codominant marker) data. These studies illustrate the flexibility with which kernel and SVR methods can be used for line performance prediction.

4.4. Partial least squares regression and principle component regression Partial least squares (PLS) regression (Wold et al., 1985) and principal component (PC) regression (Coxe et al., 1986) are well-known dimension reduction methods in chemometrics that are useful when the researcher is faced with many variables whose relationships are ill-understood, and the object is merely to construct a good predictive model (Tobias, 1997).

90

Aaron J. Lorenz et al.

In both methods, latent variables are extracted as linear combinations of the predictors and are used for response prediction so that gðxi Þ ¼

w X

til bl

ð9Þ

l¼1

where til is a latent variable extracted from the original p predictors and bl is the effect associated with that variable; that is, til ¼ xiul, and ul is a p 1 vector of weights in the linear combination for latent variable l. Usually w, the number of latent variables, is much smaller than p, and the tls are orthogonal, eliminating the problem of multicollinearity. This procedure covers various techniques, depending on which source of variation is considered most crucial. In PC regression, the latent variables are chosen to explain as much of the variation in the original predictors as possible. This approach yields informative vectors in the predictor space that may not be associated with variation in the responses. In PLS, the latent variables are chosen so that the relationship between the latent variables and the response is as strong as possible. To avoid reduced model accuracy due to too few or too many latent variables (Solberg et al., 2009), the number of latent variables can be determined through CV or related techniques. In both PCR and PLS, g(xi) can be reformulated as a direct function of the predictors without using latent variables: the effect associated with marker k, as in Eq. (3), is bk

¼

w X

bl ulk

ð10Þ

l¼1

4.5. Bayesian methods The assumption made by RR-BLUP that genetic effects are evenly spread across the genome was not satisfactory and Meuwissen et al. (2001) sought to relax it using Bayesian analysis. They first devised an analysis called BayesA where each marker effect k is drawn from a normal distribution with its own variance: N(0, var(bk)). This allows each marker to be shrunken toward zero to a different degree. The variance parameters are in turn sampled from a scaled inverted w2 distribution. This model is also known as Bayesian shrinkage regression (Xu, 2003). They then modified this analysis to one called BayesB, where a probability is given that a marker has no effect at all. This model would better reflect the underlying genetic architecture if genetic variance was present at few loci and absent at many loci (Meuwissen et al., 2001). Recalling our basic model above, this can be represented as

Genomic Selection in Plant Breeding: Knowledge and Prospects

gðxi Þ ¼

p X

xik bk gk

91

ð11Þ

k¼1

where gk is an indicator variable specifying the presence of marker k in the prediction model and the other variables are as before. In the model, bk is assumed to follow the normal distribution with mean zero and finite variance. The prior distribution for the variance of bk follows a mixture distribution: varðbk Þ ¼ 0; with probability p varðbk Þ w2 ðv; SÞ; with probability ð1 pÞ BayesB reduces to BayesA by setting p ¼ 0. Meuwissen et al. (2001) based on their knowledge of the genetic architecture produced by the simulation, set p ¼ 0.95. Reasonable values of this parameter, however, would be unknown when dealing with real organisms and traits. To overcome this problem, an analysis called BayesCp estimates p itself, with the prior distribution for p considered uniform between 0 and 1. In addition, BayesCp assumes that the prior variance for the effects of all markers for which gk ¼ 1 is equal. That is, the effect bk is zero when gk ¼ 0 ^ 2b ) when gk ¼ 1. In turn, the method estimates s ^2b jointly or bk N(0, s over all nonzero markers (Kizilkaya et al., 2009). Grouping markers in this ^ 2b (Gianola way gives the data added weight over the prior in estimating s et al., 2009). A method that predates Meuwissen et al.’s (2001) development of BayesB but that makes much the same prior assumptions, is stochastic search variable selection (SSVS; George and McCulloch, 1993; Verbyla et al., 2009). In SSVS, the mixture distribution for the variance or marker effects is set as ^ 2b =c; varðbk Þ ¼ s ^ 2b ; varðbk Þ ¼ s

with probability p with probability ð1 pÞ

^2b /c is “small” and s where c is chosen so that s ^2b is “large” (e.g., c ¼ 100). The fact that bk is never set to zero simplifies the statistical analysis (Verbyla et al., 2009), making it computationally more efficient. Verbyla et al. (2009) showed the accuracy of SSVS to be identical to that of BayesB, but required only 0.2% of the computing time. A version of SSVS was also used by Iwata and Jannink (in review), who termed their model BayesB0 and found that it performed better than standard BayesB for traits of high heritability, but was inferior for traits of lower heritability.

92

Aaron J. Lorenz et al.

A problem common to all these Bayesian models is that parameter estimates cannot be obtained analytically. It is possible, however, to develop so-called Markov chain Monte Carlo (MCMC) algorithms that sample parameter values from their posterior distributions. Those distributions can then be approximated by sampling repeatedly and calculating appropriate summary statistics such as the mean or median of the distributions. Unfortunately, this solution is very computationally expensive. Non-MCMC methods have therefore been sought to estimate these summary statistics. Xu (2007) proposed a non-MCMC method to fit BayesA by iteratively approaching the posterior mode of SNP effect variances. With these variance estimates, BLUP of the SNP effects themselves could be obtained using standard mixed linear model methods. Yi and Banerjee (2009) reported more rapid computation using an expectation-maximization (EM) algorithm to achieve the same result. Hayashi and Iwata (2010) extended this EM algorithm to BayesB-type models by incorporating variables that weighted SNP effects according to their probabilities for inclusion in or exclusion from the model. Finally, Meuwissen et al. (2009) analytically derived the expectation of the effect of a single SNP, conditional on the data, assuming a BayesB-type prior. They then fitted the BayesB model by calculating this expectation for each SNP while fixing the effects at the other SNP and then iterating until a stable solution was obtained. Compared to MCMC methods, the computing time taken by nonMCMC methods is about 20-fold (Xu, 2007), 50-fold (Hayashi and Iwata, 2010), or several 100-fold (Meuwissen et al., 2009; Yi and Banerjee, 2009) lower. All non-MCMC methods, however, result in approximations of the posterior distributions that reduce their accuracy slightly (by 1–3%; Hayashi and Iwata, 2010; Meuwissen et al., 2009) relative to full MCMC methods. Their much-reduced computational time, however, opens new possibilities for optimizing prior distribution parameters, for example, by using CV techniques that require repeated model fitting (Meuwissen et al., 2009; Xu, 2007). A complete evaluation of these possibilities has not been performed.

4.6. Statistical methods summary Figure 1 illustrates differences in distributional assumptions of a few models presented above. Under the RR-BLUP model, marker effects are samples from a normal distribution with fixed variance. By shrinking all marker effects to the same degree and including all markers in the model, the use of RR-BLUP implies that the practitioner believes the trait to be controlled by many loci with small effects. In contrast, BayesB makes the assumption that most loci have no effect on the trait and thus most markers are left out of the prediction model. These markers are represented by the point mass at var(b) ¼ 0 in Fig. 1. The markers included in the model have effects

93

Genomic Selection in Plant Breeding: Knowledge and Prospects

70 60 Ridge regression BayesB

50

Density

BayesC␲ 40 30 20 10 0 0.00

0.01

0.02 0.03 Var(β)

0.04

0.05

Figure 1 Graphical representation of the priors for the variance of marker effects, b, for ridge regression, BayesB, and BayesCp. Circles represent point mass probabilities; their height on the y-axis has no meaning but is for clarity. Ridge regression uses a single nonzero point mass for the variance of b (circle with black border). BayesB uses a mixture of a point mass at zero of a fixed size and a continuous scale inverted w2 distribution. BayesCp uses a mixture of a point mass at a zero and a nonzero point mass. The nonzero value is estimated (horizontal arrows), as is the mixture probability (allowing the size of the zero and nonzero point mass circles to fluctuate inversely to each other).

sampled from distributions with different variances. BayesB, therefore, effectively presumes that the trait is controlled by relatively few loci that vary in effect size. In the case of BayesCp, the proportion of markers included in the model is estimated from the data, which is indicated by the different point mass sizes in Fig. 1. Marker effects included in the model are sampled from the same distribution whose variance is estimated from the data. This gives BayesCp more flexibility to model oligogenic to polygenic traits. We stress these model priors and their implications to emphasize that the models assume different genetic architectures. To the extent that genetic architectures differ, therefore, there is no single best model for all traits and populations. Where genetic architectures consist of many loci of small effect (Buckler et al., 2009), RR-BLUP and models exploiting global genetic relationship information work well. Where large-effect QTLs explain much genetic variation (Anderson et al., 2001; Munkvold et al., 2009), models such as BayesB should be favored. Furthermore, these example models

94

Aaron J. Lorenz et al.

assume additivity and might not perform well if nonadditive genetic effects were important. Evidence of prevalent nonadditive effects exists in some populations (Dudley and Johnson, 2009); RKHS and SVR may hold an advantage in these situations.

5. GS Prediction Accuracies The prediction accuracy of the GEBVs is evaluated by the correlation between the GEBVs and empirically estimated breeding values, r(GEBV: EBV), where the EBV can be obtained in a number of ways, most simply, as a phenotypic mean. This correlation provides an estimate of selection accuracy and thus directly relates GEBV prediction accuracy to selection response (Falconer and Mackay, 1996). Other statistics such as mean-square error (MSE) are used occasionally (Verbyla et al., 2009). GS accuracy is defined as the correlation between GEBV and the true breeding value (TBV), that is, r(GEBV:TBV). Since we can only measure r(GEBV:EBV), this measure needs to be converted to an estimate of r(GEBV:TBV). To do so, it is assumed that r(GEBV:EBV) ¼ r(GEBV:TBV) r(EBV:TBV). This assumption is correct if the only component common between the GEBV and the EBV is the TBV itself. In other words, the assumption holds if GEBV ¼ TBV þ e1 and EBV ¼ TBV þ e2, where e1 and e2 are uncorrelated residuals. The assumption could be violated if the training and validation data were collected in the same environment. In that case, genotype by environment (G E) interaction would generate a common component of error in both GEBV and EBV, biasing their correlation upward. Thus, training and validation data should be collected in different environments to ensure sound estimates of GEBV prediction accuracy. The correction, r(EBV:TBV), accounts for the fact that the EBV in the validation set is not free of error. When the EBVs are phenotypes, r(EBV:TBV) is equal to the square root of heritability (h) within the validation set (Falconer and Mackay, 1996).

5.1. Evaluating GEBV accuracy through CV All GS studies on empirical data use CV. CV entails splitting the data into a training set and validation set (the machine learning nomenclature used here is prevalent in the GS literature; CV literature normally refers to these data subsets as “estimation set” and “test set,” respectively). The ratio of observations in each set varies, but often a fivefold CV is used, that is, the data set is randomly divided into five sets, with four sets being combined to form the training set and the remaining set designated as the validation set. Each subset of the data is used as the validation set once. Alternatively,

Genomic Selection in Plant Breeding: Knowledge and Prospects

95

observations are sometimes divided into training and validation sets systematically. VanRaden et al. (2009) designated bulls born between 1952 and 1998 as the training set and bulls born between 1999 and 2002 as the validation set so that no sons were present in the training set, as per a real breeding program. In two GS studies on mice (Lee et al., 2008; Legarra et al., 2008), observations were split either across families or within families. In any case, a model is developed using the training data and GEBVs are calculated using genotypes of the validation set.

5.2. Reported accuracies Simulation and empirical studies have consistently shown GS methods to be superior in terms of accuracy to traditional MAS and selection on pedigree information alone. Using simulation and considering intrafamily selection only, Bernardo and Yu (2007) showed that GS provided 18–43% more genetic gain per cycle than MARS, presumably a result of capturing more genetic variance by modeling all marker effects. Depending on the trait heritability and genetic architecture, phenotypic selection is occasionally expected to outperform GS in terms of gain per cycle. In terms of genetic gain per unit time, however, phenotypic selection stands little chance, as the time for a cycle of GS is expected to be one-third or less than that of phenotypic selection for many crops (Heffner et al., 2010; Lorenzana and Bernardo, 2009). This decrease in cycle time will be especially large for perennial species needing several seasons of evaluation and species with long sexual cycles, such as oil palm (Wong and Bernardo, 2008). Meuwissen et al. (2001) and Habier et al. (2007) evaluated the accuracies of RR-BLUP and BayesB using very similar simulation approaches assuming additive gene action and a heritability of 0.5. A combination of their simulation settings and assumptions led to an expected effective QTL number of 6 for Meuwissen et al. (2001) versus 13 for Habier et al. (2007). We believe both of these numbers are low relative to a realistic modeling of polygenic traits. Training population sizes were 2200 versus 1000 for the two studies, respectively. GS accuracies obtained in the two studies were respectively 0.73 and 0.64 for RR-BLUP and 0.85 and 0.69 for BayesB. The greater overall accuracy and greater difference between RR-BLUP and BayesB in Meuwissen et al. (2001) can probably be attributed to the larger variances generated by individual QTL in that study as well as the difference in training population sizes. Zhong et al. (2009) took a different approach to simulation: rather than generating marker data from an ideal population in mutation-recombination-drift equilibrium, they took actual marker data from a diverse set of 42 lines of two-row barley. This approach retains the effects on LD of the more complex and realistic demographic history of a crop. Of the markers available, 1040 were retained as evenly distributed over the genetic map.

96

Aaron J. Lorenz et al.

Of these markers, 80 were picked at random to serve as “QTL” and were removed from the marker data. Additive gene action was assumed and the trait heritability was 0.4. Training populations were simulated by randomly mating the founders to generate 500 lines. In this case, the accuracies for RR-BLUP and BayesB were 0.62 and 0.61, respectively. The primary concept of GS, namely not discriminating among markers used for prediction based on function and significance thresholds, has even reached medical genetics. Wray et al. (2007) considered using wholegenome marker data to predict the risk of complex diseases controlled by 1000 loci each of very small effect. Using simulation, they found that risk could be predicted with an accuracy of greater than 0.75 using wholegenome data and a case-control sample of 10,000 individuals for training. However, some assumptions that are presently unrealistic were made, such as causal SNPs were always included among the markers. Also, the authors ignored LD between simulated SNPs and ignored gene-by-environment interaction effects. Lorenzana and Bernardo (2009) is the only peer-reviewed study that focused on empirical GS accuracy in crops, and therefore, we will review this article in detail wherever their results are pertinent in the following sections. The objective of Lorenzana and Bernardo (2009) was to assess the impact of statistical model, population size, and marker number on the accuracy of genotypic value predictions in biparental mapping populations. Internal and publicly available data on various species and populations were used, including four maize populations, one Arabidopsis population, and two barley populations. Traits ranged from grain yield and indexes of agronomic traits to amylase activity and amino acid content. Marker effects were estimated several ways: multiple linear regression (MLR) using relaxed significance levels, RR-BLUP, and an empirical Bayesian model (e-Bayes). Predictions from the e-Bayes model were made using marker main effects only and both marker main effects and marker–marker interaction effects. GS via RR-BLUP outperformed MLR in almost every instance, sometimes producing twice the prediction accuracy of MLR. RR-BLUP accuracies varied widely, being between 0.26 (stalk lodging in maize) and 0.94 (dry matter % in Arabidopsis). For all but 2 of the 40 trait-population combinations analyzed, GEBV accuracy was greater than 1/2h. This threshold is important because, if we assume that GS can reduce breeding cycle time to half that of phenotypic selection, an accuracy greater than 1/2h will ensure greater gain per unit time by GS than by phenotypic selection (Heffner et al., 2010; Lorenzana and Bernardo, 2009). Empirical GS prediction accuracies have been reported for livestock by several studies since the seminal paper of Meuwissen et al. (2001). VanRaden et al. (2009) conducted the largest single study of this kind using dairy cattle. The training population consisted of some 3576 Holstein bulls, with breeding values measured by progeny testing and genotyped

Genomic Selection in Plant Breeding: Knowledge and Prospects

97

with 38,416 SNP with a minor allele frequency greater than 0.05. Breeding values used for the training population were as estimated in August 2003. These data were used to predict the breeding values of 1759 bulls with validation from progeny testing occurring up to April 2008. They achieved accuracies of 0.44–0.79 for traits ranging in heritability from 0.04 to 0.50. A review of studies from three other dairy cattle GS experiments showed similar results (Hayes et al., 2009a) in that GS methods predicted breeding values better than pedigree information alone, but less well than was expected based on simulations. The range of GS accuracies reported by Hayes et al. (2009a) was 0.37–0.74. Similar results were found among all empirical GS studies in cattle reviewed (Habier et al., 2010; Luan et al., 2009; Moser et al., 2009; Su et al., 2010; Verbyla et al., 2009). A wide range of accuracies are observed, ranging from 0.10 to 0.85 for traits widely varying in heritability and genetic architecture. An important observation is that GEBVs are consistently more accurate than breeding values estimated from pedigrees. The above results suggest that GS models hold excellent potential for predicting breeding values for real, complex traits.

5.3. Marker density, marker type, and training population size The great advantage of GS over traditional MAS is for complex traits. Consequently, dense genome-wide marker coverage is needed to maximize the number of QTL in LD with at least one marker, and large training populations are needed to accurately estimate marker effects. As discussed above, parameters such as effective population size and QTL number strongly influence marker densities and TP sizes required for acceptable accuracy. Indeed, simulations similar to those of Meuwissen et al. (2001) have shown that marker density needs to scale with effective population size (Solberg et al., 2008). Nonetheless, some take-home messages and strategies for maximizing GS program efficiency can be gleaned from the literature. Until very low marker densities were reached, marker number had very little, if any, effect on prediction accuracies within families from various plant species (Lorenzana and Bernardo, 2009). Likewise, GEBV accuracy of several traits in cattle, including net merit, was hardly affected when as many as 75% of the original markers were masked (Luan et al., 2009; VanRaden et al., 2009). In both the plant and cattle studies (Lorenzana and Bernardo, 2009; Luan et al., 2009; VanRaden et al., 2009), TP size had a more important effect on accuracy. VanRaden et al. (2009) showed that the relationship between accuracy and TP size was quite linear up to the maximum size available, while Lorenzana and Bernardo (2009) found that increasing TP size sometimes increased accuracy by as much as 20%. Indeed, it appears that among the empirical studies reviewed, TP size impacted accuracy more than marker number. As expected, there are diminishing returns for both marker number and TP size (Meuwissen

98

Aaron J. Lorenz et al.

et al., 2001; Solberg et al., 2008). For example, Lorenzana and Bernardo (2009) observed that in an Arabidopsis family, 0.10 was gained when TP size was increased from 48 to 96, 0.07 was gained between TP sizes of 96 and 192, and only 0.05 was gained between 192 and 332. Adequate marker density and TP size depend on QTL number and trait heritability. The impact of TP size, as derived from theory, is illustrated in Fig. 2. The figure assumes the genome size of barley and the historic effective population size of 50, which approximates that of modern barley in North America (Hamblin et al., 2010). Consequently, about 1000 independent effects need to be estimated. When TP size is sufficiently large, even traits of extremely low heritability can be predicted quite accurately. Using simulation, Hayes et al. (2009c) showed that in order to achieve equivalent accuracies between traits controlled by 10 and 1000 loci, a fivefold greater marker density is required for the latter trait. Calus and Veerkamp (2007) used the average r2 between adjacent markers as a measure of marker density relative to decay of LD. They found that for a high heritability trait, average adjacent marker r2 of 0.15 was sufficient, but for a low heritability trait, increasing the r2 to 0.20 improved prediction

Genomic selection accuracy

1

0.8

0.6

0.4

0.2

0 0

0.2

0.4 0.6 Trait heritability

0.8

1

Training population size: 20K

5K

1000

200

Figure 2 Predicted accuracy of genomic selection based on (Hayes et al., 2009c) assuming the characteristic of elite barley: effective population size of 50 and genetic map length of 1100 cM. Note that this predicted accuracy assumes that all QTL are in strong LD with one SNP and does not account for accuracy contributed by modeling genetic relatedness.

Genomic Selection in Plant Breeding: Knowledge and Prospects

99

accuracy. Heritability dramatically affects TP sizes required for successful GS, especially at h2 less than 0.40 (Hayes et al., 2009c). Consider a population with an effective population size of 1000 and a desired accuracy of 0.70. In this case, it is predicted that TP size needs to be 9000 if h2 ¼ 0.20, whereas a TP size of less than 3000 is needed if h2 ¼ 0.50. Solberg et al. (2008) used the simulation conditions of Meuwissen et al. (2001) to evaluate the effect of SSR-like multiallelic markers versus SNPlike biallelic markers. They found that the SNP markers required a density of two to three times that of the SSR marker. Also, assembling pairs of adjacent markers into haplotype blocks tended to decrease accuracy relative to using the SNPs separately. Lorenzana and Bernardo (2009) speculated that calculating GEBV accuracies from haplotype effects would have not been beneficial. In cases of high marker density and large TP sizes, however, grouping SNPs into haplotypes may be advantageous (Goddard and Hayes, 2007), and thus, its utility warrants further study. Clearly, the benefit of increasing marker density derives from positioning markers in greater LD with QTL. But a less obvious effect is the decreasing accuracy with which marker effects are estimated as marker density rises. This is because of increasing collinearity between markers when TP sizes are too small (Muir, 2007). Therefore, any positive effect on accuracy from increasing marker numbers can be constrained by TP size, requiring that TP size scale with marker numbers to successfully capture the additional information provided by increased marker density (Meuwissen, 2009; Muir, 2007). This scaling requirement is greatest when a method such as RR-BLUP is used, and would be reduced if a variable selection method such as BayesB was used instead (Fernando et al., 2007; Muir, 2007). It may be possible to have both high SNP density and lower genotyping cost by genotyping only the TP and selection candidate parents at high density, genotyping the selection candidates at low density, and imputing markers not scored on the selection candidates using parental marker information. Using simulation, Habier et al. (2009) evaluated the loss of accuracy using this approach. A model was trained using individuals phenotyped and genotyped with markers spaced every 1 cM. Selection candidates were genotyped with markers spaced every 10 or 20 cM. The authors found that minimal accuracy was lost when markers were scored at a density of one per 10 cM and the remaining markers were imputed using highdensity marker information on both the TP and selected parents. Accuracy losses were reduced further if markers among the low-density set were selected for their effect on the trait of interest using a BayesB algorithm. It would also be desirable to maximize the efficiency of TPs through their design, that is, minimizing phenotyping expenses by evaluating as few individuals as possible while maintaining accuracies as high as those attained with large TPs. This resembles “selective phenotyping” used in the context of QTL mapping ( Jannink, 2005; Jin et al., 2004). Three possible objectives

100

Aaron J. Lorenz et al.

of TP design include maximizing marker variance, reducing collinearity between markers, and uniformly sampling the genetic diversity of the breeding program. Near-infrared reflectance spectroscopy (NIRS) is analogous to GS as an application of multivariate statistics to model development and prediction. Procedures for NIRS calibration, calibration sample set selection, and calibration updating have been researched and refined for decades, and therefore should provide a good starting point for researching the optimization of TP design and GS model updating (reviewed in Jannink et al., 2010).

5.4. Effect of training population size and marker number on GEBV accuracy in barley, oat, and wheat data sets To add to the existing empirical GS-related studies in plants, we analyzed barley (Hordeum vulgare), oat (Avena sativa), and wheat (Triticum aestivum) data sets that were obtained as part of ongoing projects. The traits that were analyzed were plant height and grain yield. The barley data set was a subset from the Barley Coordinated Agricultural Project (Barley CAP; www. thehordeumtoolbox.org) database. All 506 lines were 6-row spring barley and were entered into the Barley CAP by the University of Minnesota (UM), North Dakota State University (NDSU), or Busch Agricultural Resources Inc. (BARI) breeding programs. The oat data set was obtained from cooperative testing nurseries—the Uniform Oat Performance Nursery (1994–2007) and Quaker Uniform Oat Nursery (1997–2005). The phenotypic data were downloaded from http://wheat.pw.usda.gov/GG2/ uopnquery.shtml. The oat data set consisted of 421 lines evaluated for plant height and 436 lines evaluated for grain yield. Testing was conducted across the oat-growing regions of the United States and Canada. Mostly, different varieties were tested in different years, but some overlap in varieties existed from year to year, allowing the recovery of interblock information for the calculation of line BLUPs. The wheat data set consisted of 374 soft white winter elite breeding lines from the Cornell University Wheat Breeding Program. Plant height and grain yield were evaluated in four environments: two locations in 2008 and two different locations in 2009. All locations were near Ithaca, NY. SNPs among the barley lines were scored using two Illumina GoldenGate assays (Close et al., 2009). Among the lines used for this study, 1180 SNPs were defined as polymorphic (MAF > 0.028) and used for analysis. Oat and wheat lines were scored for Diversity Array Technology (DArT) markers ( Jaccoud et al., 2001; www.diversityarrays.com). The oat and wheat marker sets consisted of 1005 and 1554 loci, respectively. A training set was obtained by random sampling; the remaining lines formed the corresponding validation set. A prediction model was built using RR-BLUP:

Genomic Selection in Plant Breeding: Knowledge and Prospects

101

yT ¼ Xb þ Zu þ e where yT is a vector of phenotypes (line means, adjusted means, or BLUPs) training lines, b is a vector of fixed effects, u is a vector of random effects (marker effects in this case), and e is a vector or residuals. X and Z are incidence matrices relating the observations (yT) to the fixed effects and marker effects, respectively. Genotypic value predictions of the validation set (gV) were obtained by multiplying u estimated in the above model by Z of the validation set (ZV). Since variation among observations in the validation set includes variation due to the fixed environmental effect when the data set was unbalanced as it was for barley and oat, Xb was estimated in the validation set (XVbV) and subtracted from yV. Therefore, the correlations reported are between gV and (yV XVbV) so that variation in the variables being correlated is due to random effects only. TP size was set to 100, 200, or 300 lines. Training sets for each TP size were randomly sampled 100 times. Marker numbers were 300, 600, or 900. Markers were also randomly sampled. While varying marker numbers, training population size was fixed at 300. To sample the variation in both marker subsets and training populations, random sampling of markers was nested within a loop for random sampling of training lines. Training populations were sampled 30 times. For each training population, marker subsets were sampled 30 times, producing 900 samples for each marker number. Correlations between predicted and observed values for species and trait are plotted against training population size and marker numbers in Figs. 3 and 4. Before discussing the estimates of accuracy, we note that correlations were not adjusted for the error in the validation set, so the values reported here are biased downward by a factor approximately equal to the heritability in the validation set. The adjustment was not performed for two reasons: (1) data sets for two of the three species (barley and oat) were highly unbalanced, possibly producing an inaccurate h2 estimation and (2) adjusting the correlation by h would result in upwardly biased estimates of accuracy because training and validation data were collected in the same environments. Because of these conditions, accuracies should not be compared between species and traits, but only between levels of TP size and marker number. First, when TP size was 300 and all polymorphic markers were used, r (GEBV:EBV) reached values as high as 0.75 and was greater than 0.40 for five of the six population–trait combinations (Fig. 3). As discussed above, the true accuracy to predict gain from GS would be r(GEBV:TBV). The values of r(GEBV:EBV) reported and biased down relative to that true accuracy by random error, but may be biased up G E deviations in common between the GEBV and EBV. Thus, the direction and precise

cor(GEBV, Pheno)

0.8

0.6

0.4

0.2

100

200 Training pop size

Barley GrYld Barley PH

Oat GrYld Oat PH

300

Wheat GrYld Wheat PH

Figure 3 Correlations between GEBVs and observed phenotypes of the validation set plotted against training population size. Each point represents the mean of 100 replications, where a replication consists of a random sample of lines to form the training population. Traits analyzed were grain yield (GrYld) and plant height (PH). All polymorphic markers were included in the model.

cor(GEBV, Pheno)

0.8

0.6

0.4

0.2

300

600 Marker number

Barley GrYld Barley PH

Oat GrYld Oat PH

900 Wheat GrYld Wheat PH

Figure 4 Correlations between GEBVs and observed phenotypes of the validation set plotted against marker number. Each point represents the mean of 900 replications, where a replication consists of a random sample of 300 lines to form the training population as well as a random sample of markers. Traits analyzed were grain yield (GrYld) and plant height (PH).

Genomic Selection in Plant Breeding: Knowledge and Prospects

103

extent of bias is unknown. Correlations increased with TP size in all cases. The average increase from 100 to 300 TP lines was 8 points (range: 6–12). A small sign of diminishing returns was seen as was reported by Lorenzana and Bernardo (2009). The average increase in accuracy when increasing TP size from 100 to 200 was 5 points (range: 4–7), while that observed by increasing TP size from 200 to 300 was 3 points (range: 2–5). However, the lack of response from increasing marker numbers across all six species-trait combinations was remarkable. It appears that only 300 SNP markers were necessary to capture the genetic variation for these traits in these populations using RR-BLUP. One explanation is that TP sizes of only 300 are not large enough to take advantage of more markers (Muir, 2007). Another possibility is that LD is so high within these populations that 300 markers are sufficient to capture genetic variation caused by most or all QTL. The fact that these are all self-pollinating species that probably have low effective population sizes caused by domestication and modern breeding make the latter explanation more likely. Indeed, Hamblin et al. (2010) showed that there are extensive levels of LD in domesticated barley, especially within the UM, NDSU, and BARI germplasm.

6. Impact of Statistical Model on GEBV Accuracy As discussed in the methods section, RR-BLUP modeling marker effects as random are equivalent to a mixed model analysis wherein observations have random effects that covary according to a relationship matrix calculated using marker data (Habier et al., 2007; Hayes et al., 2009c). Realizing this, Habier et al. (2007) proposed to partition accuracy into a component due to tight marker-QTL LD and a component due to the modeling of genetic relationship. The former component decays slowly over generations of mating between the training and selection populations because the LD it depends on is reduced only by rare recombination events. In contrast, the latter decays rapidly as the genetic relationship between the training and selection populations declines. The relative importance of these two components is not equal across all GS methods. Habier et al. (2007) showed that because RR-BLUP fitted many more markers into the model than BayesB, it modeled genetic relationships more accurately, and a large fraction of its accuracy was due to that component. Conversely, BayesB was able to place more weight on individual markers in LD with QTL so that accuracy due to marker-QTL LD was stronger for BayesB than RR-BLUP (Habier et al., 2007). Consequently, the accuracies of BayesB models persisted longer than those of RR-BLUP models (Habier et al., 2007; Zhong et al., 2009). In simulation contexts, it is straightforward to run through multiple generations to test the decay of accuracy that that causes. Empirically,

104

Aaron J. Lorenz et al.

Habier et al. (2010) proposed partitioning a data set into training and validation populations such that these populations varied in their degree of relatedness. They measured relatedness between populations according to the maximum additive genetic relationship between any cross-population pair of individuals. With the empirical data available, however, it proved difficult to generate pairs of populations that differed widely in their relationship (Habier et al., 2010). Nevertheless, the analysis confirmed that accuracy due to LD was greater for BayesB than RR-BLUP. A similar kind of differential partitioning of training and validation populations can be reconstructed from two studies on a mouse population synthesized from eight inbred mouse lines (Lee et al., 2008; Legarra et al., 2008). This population could be split into training and validation halves in two ways, either across families (different families ending up in the different halves) or within families (different individuals within families ending up in the different halves). Legarra et al. (2008) used RR-BLUP, while Lee et al. (2008) used an analysis with properties similar to BayesB. When training and validation populations were composed of different families, capturing marker—QTL LD was more useful for prediction because the families were weakly related. In contrast, when training and validation populations contained members of the same families, capturing genetic relatedness was also useful. Though the two studies did not analyze the same traits, there were traits of similar heritability. Confirming theory on differences between the methods, prediction across families was more accurate for the BayesBlike analysis (Lee et al., 2008) than for RR-BLUP (Legarra et al., 2008). Conversely, prediction within families was more accurate for RR-BLUP than for the BayesB-like analysis. The relevance of this issue extends beyond the question of the speed at which prediction accuracy decays. Relative to traditional selection using pedigree-based on BLUP, GS is hypothesized to maintain greater genetic diversity for the following reason (Daetwyler et al., 2007). Assuming that no progeny testing is performed, extra information from relatives flows to the selection candidates through their parents. Thus, the extra information for any pair of full sibs is identical, and in general, extra information for any pair of relatives is correlated. From this correlation comes greater similarity of the predicted value of relatives, leading to a greater chance of co-selection of relatives and, finally, to a more rapid decline in genetic diversity in selected populations. Another way of expressing this problem is that information from relatives does not contribute to predicting the value of the alleles each progeny received from its parents through random Mendelian segregation (Daetwyler et al., 2007). This inheritance, however, is unique to each individual and should be an important determinant of its selective value. In principle, GS gains a handle on the Mendelian segregation term because each progeny’s marker genotype tells us what it received. How well GS can take advantage of that information to maintain diversity depends on the

Genomic Selection in Plant Breeding: Knowledge and Prospects

105

balance of components of its prediction accuracy: in cases where the marker—QTL LD component is important, GS should maintain greater diversity than traditional BLUP selection. Conversely, in cases where the component due to genetic relationship is important, GS may do no better than traditional BLUP selection. Parameters that increase the former component include a larger training population size, higher marker density, a genetic architecture with a lower effective QTL number, and Bayesian analysis methods rather than RR-BLUP ( Jannink et al., 2010). More generally, a common finding among empirical studies on GS is that prediction models that assume many QTL evenly distributed over the genome (e.g., RR-BLUP) perform as well as methods that assume fewer QTL of varying effect (e.g., BayesB) (Hayes et al., 2009a,b; Lorenzana and Bernardo, 2009; Luan et al., 2009; Moser et al., 2009; VanRaden et al., 2009; Verbyla et al., 2009). For example, Su et al. (2010) used two Bayesian models, one in which marker effects were uniformly shrunken (like RRBLUP) and one that allowed variable shrinkage (like BayesB). The authors found that the uniform shrinkage model performed better for all traits (fertility, protein, udder health), except for fat percentage, where a mixture distribution performed equally well. In plants, Lorenzana and Bernardo (2009) showed that RR-BLUP, involving “the convenient but unrealistic assumption of equal marker variances,” provided comparable or higher prediction accuracies than a Bayesian model. Several explanations of these observations are possible. The most obvious interpretation is that the true genetic architecture of complex traits is closer to the infinitesimal model, “an infinite number of loci, all with infinitesimally small effect,” than the model of limited numbers (i.e., dozens) of QTL of varying effect. Alternatively, there may be relatively few loci at which variants have a large effect on the phenotype, but these variants are at low frequency and hence contribute little genetic variation at the population level. Loci with several low frequency, large-effect variants, could generate substantial genetic variance and lead to high trait heritability; LD between the causal variants and markers would generally be low, however, because experiments usually include mostly markers with high minor allele frequencies that cannot be in high LD with rare QTL alleles. The LD component of accuracy would therefore be constrained. This genetic architecture generates high heritability and resemblance between relatives, but low association between QTL and markers: it would lead RRBLUP to be more effective than BayesB. This is one of the genetic architectures invoked in “the case of the missing heritability” (Manolio et al., 2009), which refers to instances of human association study where very little variation is explained by associated markers, even for traits with high heritability to which substantial effort at association has been applied (e.g., height studied in panels of 30,000 individuals). This architecture has also been found for flowering time in maize (Buckler et al., 2009), though

106

Aaron J. Lorenz et al.

how common it may be in general and particularly among inbred crops with much smaller effective population sizes is an open question. This “common QTL/rare variant” architecture cannot, of course, explain the success of RR-BLUP in biparental crosses where every variant segregates at an expected frequency of 0.5. The high levels of LD present in biparental crosses, however, may allow the effect of any QTL to be captured by multiple markers spread across a chromosome. That is, the extensive LD makes it appear as though genetic effects are uniformly distributed as RRBLUP presupposes. The fact that RR-BLUP performs well across such dramatically different contexts, however, suggests that it is a robust starting point from which to explore different models. An instructive deviation from the superiority of RR-BLUP comes from dairy cattle and the fat percentage of milk, where prediction accuracies are greater for models assuming fewer QTL of varying effect (Hayes et al., 2009b; VanRaden et al., 2009; Verbyla et al., 2009). In all the three cases, this was because a mutation known to account for a large fraction of the genetic variation for fat percentage (DGAT1) was segregating in the populations under study. The models allowing marker effect distributions to vary were successful in capturing the large effect of markers at the DGAT1 locus, whereas the BLUP method severely shrinks the DGAT1 effect back to zero. An interaction between model and trait architecture is also found by varying the number of simulated QTL and their effect size (Zhong et al., 2009). These authors suggest constructing a hybrid model by adding a polygenic effect to a Bayesian shrinkage model. As is made fairly obvious by the aforementioned findings, the most appropriate statistical method depends on the genetic architecture of the trait analyzed. From the current empirical literature, it appears that most traits will be most accurately predicted by simply calculating BLUPs of markers and summing the values, but exceptions certainly exist and should be anticipated. Another factor affecting the relative performance of different model types is marker number. BayesB, which is good at exploiting markerQTL LD, tends to benefit more from greater marker density than RRBLUP (Meuwissen, 2009; Meuwissen et al., 2001). Moreover, Solberg et al. (2009) showed that the addition of markers to PLS and PCR regression models only increased accuracy by 6–7%, whereas additional markers increased accuracy of BayesB predictions by 17%. Continued advancement in high-throughput, high-density marker platforms and genotyping by sequencing will allow for dense genome-wide marker coverage for all crop species. However, in cases where the GS model does not include a marker that is in LD with each QTL, a pedigree-based polygenic effect can be included in the mixed model to account for the residual genetic variance not captured by the markers (Haley and Visscher, 1998; Muir, 2007). Through selection index theory, both a pedigree-based polygenic effect (i.e., the BLUP-EBV) and the GEBV can be used to

Genomic Selection in Plant Breeding: Knowledge and Prospects

107

calculate a single GEBV (Goddard and Hayes, 2007) where each term’s contribution is weighted by their reliability. This approach has been shown to marginally increase GEBV accuracy in mice (de los Campos et al., 2009b; Lee et al., 2008) and wheat (de los Campos et al., 2009b) and has been used in several dairy cattle breeding programs. The inclusion of a pedigree-based polygenic effect, however, has also been shown to decrease GEBV accuracy (Legarra et al., 2008). This was likely due to collinearity between the marker and pedigree terms, which can arise when markers adequately capture the polygenic effect (Legarra et al., 2008). Therefore, in the short term, inclusion of a pedigree-based polygenic effect may be useful for crops with insufficient marker coverage, but, in the long-term, it is unlikely that this will significantly increase GEBV accuracy, as low-cost genome-wide marker coverage and genotyping-by-sequencing will be readily available (reviewed in Heffner et al., 2009).

7. Modeling Epistasis and Dominance The primary objective of GS is to predict breeding values of individuals without phenotyping in order to expeditiously improve the genetics of the breeding population. However, it will sometimes be desirable to predict genotypic as opposed to breeding values, for example, to augment phenotypic values with estimated genotypic values in variety trials to provide a better prediction of variety performance in future environments. In such cases, accurate prediction of dominance and epistatic effects would be advantageous if these effects make an important contribution to the genotypic value. Only studies harnessing empirical data are illuminating for this topic, since the genetic model used in simulations would entirely dictate any benefit gained in modeling these effects. Lee et al. (2008) and Legarra et al. (2008) evaluated the effect of different models on genotypic value prediction accuracy using the mouse data set of Valdar et al. (2006). Lee et al. (2008) used an analysis assuming few QTL and either included only additive effects (A model) or both additive and dominance effects (AD model) on blood traits and coat color. The AD model outperformed the A model for all the three traits in CV. Variance of dominance deviations was an important component of the genotypic variance; likewise, coat color responded better to the AD model than the other traits. In contrast, Legarra et al. (2008) found no advantage to modeling dominance. Given that Legarra et al. (2008) analyzed different traits than Lee et al. (2008) and used a different model (RR-BLUP), the cause of this contrast is not known. The magnitude of epistatic genetic variance has long been a major question for plant breeders (Anderson and Kempthorne, 1954). While

108

Aaron J. Lorenz et al.

complex interactions between genes and gene products at the molecular level are ubiquitous (Phillips, 2008), the presence and magnitude of epistasis in the form of allelic interaction affecting complex phenotypes greatly differs between studies. Most of this variation in findings stems from whether epistasis was quantified by variance component estimation or generation means comparisons in biometrical studies (Holland, 2001). When investigators calculate epistatic effects in the context of QTL mapping studies, epistasis is more prevalent in studies that reduce genetic background noise, such as those that use near-isogenic line (NIL) populations, as opposed to RIL or F2 populations (Holland, 2001). Using nested association mapping on lines covering the genetic diversity of maize, Buckler et al. (2009) found very little evidence for epistasis for flowering time, which the authors found surprising because of the dependence of flowering time regulation on interactive pathways. Blanc et al. (2006) found significant marker by genetic background interactions in a QTL mapping study using connected populations. Lorenzana and Bernardo (2009) and Dudley and Johnson (2009) both looked into the importance of modeling epistasis for predicting genotypic performance but came to opposing conclusions. Both studies involved the analysis of traits from intermated recombinant inbred line populations (biparental). Lorenzana and Bernardo (2009) found that including marker interactions in the prediction model substantially reduced prediction accuracy; Dudley and Johnson (2009) found that including marker interactions substantially increased prediction accuracy. Unfortunately, these two studies used two different models: Dudley and Johnson (2009) first identified significant markers and marker interactions and then used PLS for prediction; Lorenzana and Bernardo (2009) used a genome-wide Bayesian method that simultaneously fitted all marker and interaction effects. Like the dominance case above, it cannot be known whether the disagreement between these studies stems from the different models used, traits, or populations. Lorenzana and Bernardo (2009) pointed out that including epistatic effects in prediction models will only improve accuracy if two conditions are met: epistasis is present and it can be modeled accurately. Since prediction accuracy was actually lower when epistatic effects were included in predictions, clearly epistasis was poorly modeled with the population sizes in this study. Because they used an intermated RIL population, Dudley and Johnson (2009) declared uncertainty with respect to the pertinence of their findings to real breeding programs. This is because nonintermated populations would contain larger linkage blocks and thus larger blocks of genes over which interactions will more likely average out. It should be noted that both studies used biparental families. As we envision it, GS will utilize genotype and phenotype information from across different biparental populations as well as more complex populations for model development. In this scenario, it is probable that epistasis will contribute to marker effects as observed by Blanc et al. (2006).

Genomic Selection in Plant Breeding: Knowledge and Prospects

109

8. GS in the Presence of Strong Subpopulation Structure In genome-wide association studies (GWAS), subpopulation structure is known to potentially cause spurious long-distance or even unlinked associations between marker alleles and the phenotype. In GWAS, therefore, the concern is with avoiding false discoveries due to structure. In GS, the concern shifts to being able to maintain predictive ability despite a structured training data set. Rare spurious associations will not be an important cause of loss of predictive ability. Rather, consistency of LD across subpopulations, both in strength of correlation and phase across many (all) loci will be critical. If LD is not consistent, allelic effects estimated in one subpopulation will not be predictive for another subpopulation. However, if subpopulations with differing LD relationships are combined to train a model, marker alleles may not be strongly predictive in either subpopulation. Theory addressing persistence of LD across subpopulations shows that it depends on the recombination rate between loci and the number of generations since the subpopulations diverged (Sved, 2009). In this analysis, LD is measured as the correlation of allelic states across loci, denoted by r. With low migration between subpopulations, within subpopulation LD remains the well-known function of the product of effective population size and recombination rate, E(r2) ¼ 1/(1þ4Nec) (Sved, 1971), where Ne is the effective subpopulation size and r2 also expresses the variance of r across subpopulations. Also, given the low migration, any LD between subpopulations is a remnant of ancestral LD that existed prior to subpopulation divergence and that decays at a rate that is a function of the square of the recombination rate. At equilibrium (after many generations under the low migration condition), no between subpopulation LD should remain. Low migration then mimics very large global effective population size, regardless of the subpopulation effective size. The covariance across subpopulations of r is independent of the effective subpopulation size, whereas the correlation of r is not (Sved, 2009). If effective population size is assumed constant, the correlation across subpopulations of r depends only on the time since subpopulation divergence, which can then be inferred (de Roos et al., 2008). For both animal and plant domesticated species, however, effective population sizes have probably decreased with intensifying breeding efforts. This condition will inflate the variance of r, decreasing its correlation across subpopulations, resulting in inflated estimates of divergence time. Early studies looking at the covariance or correlation across subpopulations of allele correlations were in livestock genetics (Andreescu et al., 2007; de Roos et al., 2008; Goddard et al., 2006). These studies, using between 6K

110

Aaron J. Lorenz et al.

and 40K markers, found results consistent with known genetic divergence between breeds: for divergent breeds (e.g., milk versus beef breeds), a high correlation of r existed only between markers that are closer than 50 kb, whereas in related breeds (e.g., milk breeds from different countries), high correlations extend beyond 100 kb. Andreescu et al. (2007) most clearly show that the covariance of r corresponds to the relatedness of subpopulations as determined by allele frequencies: dendrograms constructed from allele frequency or covariance of r distance measures had the same topologies. An interesting point illustrated by these dendrograms was that allele frequency and covariance of r distance measures are nonlinearly related: short distances are shorter in the covariance of r metric than in the allele frequency metric, while long distances are longer. This nonlinearity manifests in the dendrograms as shorter branch lengths between terminal leaves and longer internal branch lengths. The difference may come from the systematic way in which recombination decreases the covariance of r as compared to the stochastic nature of drift. In crops, the correlation of r has, to our knowledge, only been evaluated in barley (Hamblin et al., 2010). Map resolution in that study was insufficient to measure the correlation of r at distances under 1 cM. Except for the most closely related populations (e.g., six-row barley from North Dakota versus Minnesota in the USA), the correlation of r had decayed substantially already at that distance, suggesting that marker densities greater than the 3000 available at that time would be needed to train multipopulation GS models. Assuming that marker alleles can be tagged as to the subpopulation from which they originated, two possibilities are available, that is, to estimate the main effects across subpopulations or the nested effect within subpopulations (Iba´nez-Escriche et al., 2009; Odegard et al., 2009). Conditions that improve the relative performance of main effect estimates are low subpopulation divergence and high marker density (Iba´nez-Escriche et al., 2009): both of those conditions lead to high covariance of r and therefore consistent effects across subpopulations. The main and nested effect models have been compared by simulation at rather high marker density and training data set size (5–20 markers per effective size per Morgan and 10–40 training individuals per effective size per Morgan. For comparison, in elite North American barley, the low figures would entail about 2500 segregating markers and a training data set size of 5000 lines). Under these conditions, the main effect model was favored (Iba´nez-Escriche et al., 2009). Simulations with similar marker and training data set numbers compared training data sets constituted from one or two subpopulations: the latter predicted individuals from a third subpopulation best (Toosi et al., 2010). The authors reasoned that, when marker density is high enough, the LD relevant in multi-subpopulation training data sets existed prior to subpopulation divergence. This LD will be most consistent across subpopulations. This reasoning also explains why adding individuals from a second subpopulation

Genomic Selection in Plant Breeding: Knowledge and Prospects

111

to an existing training data set improves accuracy in the second subpopulation without greatly reducing accuracy in the initial subpopulation: the addition focuses the prediction model on LD that is consistent across the two subpopulations (de Roos et al., 2009). Empirically, Hayes et al. (2009b) estimated marker effects in TPs consisting of Holstein cows only, Jersey cows only, or both breeds combined. As expected, estimating marker effects within one breed and predicting performance of the other breed resulted in low accuracies. This is likely due to changes in marker-QTL phase across breeds. In addition, even if phase is maintained, the different genetic background may change the effect of a causal allele. Accuracies were nearly as great or greater when using a combined-breed versus within-breed training population. This effect was greater for Bayesian than that for RR-BLUP GS models presumably because the former exploits marker-QTL LD to a greater degree, improving its ability to take advantage of the ancestral LD common between the breeds. Because the Jersey population was smaller, it benefited more from the larger training population size than did the Holstein population. Hayes et al. (2009b) concluded that for breeds where few observations are available to form a training population, it would be advantageous to combine breeds to form a multibreed training population. These issues have received little explicit attention in crops, though studies of GS accuracy have often been performed across subpopulations by default (e.g., Iwata and Jannink, 2010, submitted for publication; Zhong et al., 2009). The general conclusions will likely be relevant in plant breeding, though they depend on the existence of adequate marker density to capture ancestral LD.

9. Long-Term Selection Improving gain in the long-term necessarily requires a tradeoff with short-term gain: were it not for this tradeoff, the topic of long-term selection would not merit particular attention. The tradeoff is also often explicit, as in quantitative genetic models that maximize immediate predicted gain subject to a constraint on the rate of inbreeding (Meuwissen, 1997). We break down this study along two axes. First, approaches can select individuals or groups. In individual selection, candidates are characterized in terms of their value for both short- and long-term gains, and those with the best combination are selected. In group selection, groups of candidates are characterized, again for short- and long-term values, and the group with the best combination is selected as a whole. This latter approach allows for the possibility that the future value of an individual depends on what other individuals belong to its group. The second axis involves the method used for study: analytical prediction, deterministic

112

Aaron J. Lorenz et al.

simulation using numerical approaches to optimization, and stochastic simulation. This area of study first gained attention in 1994 when Gibson (1994) showed that selection under complete knowledge of a single QTL, while increasing short-term gain, can decrease long-term gain. Goddard (2009) analytically developed an index for selection of individuals directly on QTL alleles (or on markers in complete LD with the QTL). In this index, weights for each QTL depend on the favorable QTL allele effect and the QTL allele frequency. The index can maximize an objective function of these parameters and of the standardized selection differential summed over multiple selection cycles. This cumulative selection differential represents, in some sense, the duration of selection. Assuming a large cumulative selection differential, the weights depend on initial QTL allele frequencies (Goddard, 2009), such that favorable alleles at low frequency should be weighted more heavily. Though the component accounting for the favorable QTL effect on the objective function should remain constant (Goddard, 2009), the overall weights will change as allele frequencies change. The method assumes that QTLs are independent (not in LD), and the ability to determine an appropriate objective function of course depends on being able to estimate the QTL effects accurately. Further work in this direction should address how to relax these assumptions. We have simulated GS applying or not increased weight on low-frequency favorable alleles (J.-L. Jannink, unpublished). Our study confirmed that weighted GS improves long-term performance, in part by reducing the number of favorable alleles lost (Fig. 5). In several articles, Dekkers and colleagues have described and implemented optimal control theory to maximize gain or profit over a defined number of breeding cycles when selecting on one or two marked QTL and on an estimated polygenic effect (Chakraborty et al., 2002; Dekkers and van Arendonk, 1998; Dekkers et al., 2002). Optimal control theory is a numerical algorithm to optimize a multistage process in which the outcome of each stage depends only on the state of the system in the previous stage. Each stage is also affected by a number of control parameters whose optimal values need to be determined. In the cases discussed here, these parameters were the fraction of individuals of each QTL genotype to be selected (Dekkers and van Arendonk, 1998). This approach allowed the assumption of linkage equilibrium between marked QTL and polygenes to be relaxed. Studies using the approach revealed the importance of maintaining constant selection intensity on the QTL alleles, a result that is likely to transfer to GS. Accurate estimation of QTL effects is still required. Furthermore, at least with the parameterizations used, extending the analysis to more than a few QTL would be difficult. Finally, several group selection approaches have been proposed. These approaches have generally assumed that QTL effects were known. Servin et al. (2004) developed methods for pyramiding genes optimally. Assuming a

Mean genotypic value

0.5

0.4

0.3

0.2

0.1

0.0 0

2

4

6

8

10

8

10

Phenotypic breeding cycle

Mean lost favorable alleles

25

20

15

10

5

0 0

2

4

6

Phenotypic breeding cycle Phenotypic selection Unweighted genomic selection Weighted genomic selection

Figure 5 Simulated response to multiple cycles of phenotypic and genomic selection using ridge regression. Initial populations were based on SNP marker data from an elite barley breeding program. Simulations assumed 100 segregating QTL and heritability of 0.5 on an additive genetic model. One genomic selection breeding cycle took half the time of a phenotypic breeding cycle. Weighted genomic selection increased the weight of favorable alleles in the selection criterion as a function of their frequency among selection candidates, using w ¼ p 0.5, where w is the weight and p is the favorable allele frequency. Genotypic value was scaled to the maximal possible attainable genotypic value. The number of lost favorable alleles is out of the 100 QTL initially segregating. Curves shown are averages across 100 simulations.

114

Aaron J. Lorenz et al.

larger number of QTL, this approach becomes computationally unfeasible. Hospital et al. (2000) proposed recurrent selection where individuals carrying rare favorable alleles might be selected strictly to retain those alleles within the population. Li et al. (2008), working with 100 QTL, identified groups to be selected using a genetic algorithm to maximize objective functions that accounted for QTL allele frequencies or the population inbreeding rate. All methods reduced short-term gain in favor of greater long-term gain, though clear winners could be identified for which the tradeoff was relatively small. To the best of our knowledge, group selection that allows for realistic uncertainty in all QTL parameters has not been studied. Zhong and Jannink (2007) used a Bayesian framework to integrate uncertainty into the selection of pairs of individuals to cross, a potential first step toward group selection. To date, the strongest theoretical framework for group selection comes from classical selection based on pedigrees to reduce inbreeding rates (Meuwissen, 1997). Expanding theory to genomic selection wherein marker genotypes are known and diversity can be assessed on the sub-chromosomal scale is needed.

10. Summary and Conclusions Currently, the lion’s share of research on GS has been performed in livestock breeding, where effective population size, extent of LD, breeding objectives, experimental design, and other characteristics of populations and breeding programs are quite different from those of crop species. Nevertheless, a great number of findings within this literature are very illuminating for GS in crops and should be studied and built upon by crop geneticists and breeders. The application of powerful, relatively new statistical methods to the problem of high dimensional marker data for GS has been nearly as important to the development of GS as the creation of high-density marker platforms and greater computing power. The methods can be classified by what type of genetic architecture they try to capture. Somewhat surprisingly, RR-BLUP, which makes the ostensibly unrealistic assumption that genetic effects are uniformly spread across the genome, often performs as well as more sophisticated models. Exceptions do exist, though, and there is abundant evidence that BayesB is superior for traits with strong QTL effects. Additionally, since BayesB better identifies markers in strong LD with QTL than RR-BLUP, it maintains accuracy for more generations. Finally, the question of whether or not to model epistasis remains open. If epistasis is important for a particular trait in a particular population, the kernel methods and machine-learning techniques such as SVM may be preferable. It is important for the practitioner to consider such issues or

Genomic Selection in Plant Breeding: Knowledge and Prospects

115

test methods on a relevant data set before a method for GEBV calculation is chosen. Although increasing marker density, training population size, and trait heritability are obvious ways to improve GEBV accuracy; these options add cost to the program. Implementing algorithms for marker imputation and training population design holds the potential for essentially free additional accuracy, leading to greater overall GS efficiency. The current drops in genotyping costs, while phenotyping costs remain constant or increase, suggest that efforts to understand how to choose which lines to phenotype on the basis of their genotype, that is, how to design training populations, will be rewarding. Combining training populations from different populations is another way to boost accuracy when individual populations lack sufficient size and assuming that the marker densities required are available. With respect to maximizing long-term selection, we discussed several promising approaches that strive to retain favorable, low-frequency alleles while minimizing loss of short-term gain. Both simulation and empirical results for GS have been quite impressive. Empirical results of GS accuracy in crops, however, are not yet available from the public sector, except in the form of CV within families (Lorenzana and Bernardo, 2009). Further empirical studies of the effects of statistical models, marker density, TP size and composition, and different selection criteria on the effectiveness of GS in breeding populations are urgently needed. In addition, while the CV approach can be instructive, an important caveat should be mentioned. In CV, the training and validation sets belong to the same population. But in GS, the selection candidates will rarely belong to the same population as the training set and may well be several generations removed from it. Recombination during meiosis between generations erodes the association between marker and QTL, systematically reducing accuracy. The effect of selection on allele frequencies and the Bulmer effect can also have detrimental effects on accuracy (Muir, 2007). In order to realistically evaluate GS for crops, studies designed for this purpose should be performed. Clearly, exciting times are ahead of us as public breeding programs launch GS efforts. This review compiles several immediately useful results for breeders wanting to maximize gains through GS. Knowledge of breeding program parameters (effective population size, extent of LD, and trait heritability) allows marker density and training population size to be determined using analytical formulae (Daetwyler et al., 2008; Hayes et al., 2009c). The greatest impact of GS on gain per unit time will come from shortening the breeding cycle (Heffner et al., 2010). Therefore, redesigning crossing and population development schemes to incorporate GS as early as possible will likely be the most effective. Consequently, phenotyping resources will need to be shifted from early generation evaluation for selection to evaluation for model training. Modeling epistasis, while not necessary for parent

116

Aaron J. Lorenz et al.

selection, may be important for determining lines to advance for variety evaluation and release. The importance of epistasis will need to be assessed for each trait. A major paradigm in plant breeding since the availability of molecular marker technology is that mapping and characterizing the genetic loci that control a trait will lead to improved breeding. Often, one of the rationales for cloning of QTL is to develop the “perfect marker” for MAS, perhaps based on a functional polymorphism. In contrast, an advantage of GS is precisely its black box approach to exploiting genotyping technology to expedite genetic progress. This is an advantage in our view because it does not rely on a “breeding by design” engineering approach to cultivar development requiring knowledge of biological function before the creation of phenotypes. Breeders can therefore use GS without the large upfront cost of obtaining that knowledge. In addition, GS can maintain the creative nature of phenotypic selection which couples random mutation and recombination to sometimes arrive at solutions outside the engineer’s scope (Coors, 2006).

REFERENCES Anderson, V. L., and Kempthorne, O. (1954). A model for the study of quantitative inheritance. Genetics 39, 883–898. Anderson, J. A., Stack, R. W., Liu, S., Waldron, B. L., Fjeld, A. D., Coyne, C., Moreno-Sevilla, B., Fetch, J. M., Song, Q. J., Cregan, P. B., and Frohberg, R. C. (2001). DNA markers for Fusarium head blight resistance QTLs in two wheat populations. Theor. Appl. Genet. 102, 1164–1168. Andreescu, C., Avendano, S., Brown, S. R., Hassen, A., Lamont, S. J., and Dekkers, J. C. M. (2007). Linkage disequilibrium in related breeding lines of chickens. Genetics 177, 2161–2169. Barbazuk, W. B., Emrich, S. J., Chen, H. D., Li, L., and Schnable, P. S. (2007). SNP discovery via 454 transcriptome sequencing. Plant J. 51, 910–918. Beavis, W. D. (1994). The power and deceit of QTL experiments: Lessons from comparative QTL studies. In “Proceedings of the 49th Annual Corn and Sorghum Research Conference” (D. B. Wilkinson, Ed.), pp. 250–265. American Seed Trade Association, Washington, DC. Benjamini, Y., and Yekutieli, D. (2005). Quantitative trait loci analysis using the false discovery rate. Genetics 171, 783–790. Bernardo, R., and Charcosset, A. (2006). Usefulness of gene information in marker-assisted recurrent selection: A simulation appraisal. Crop Sci. 46, 614–621. Bernardo, R., and Yu, J. (2007). Prospects for genome-wide selection for quantitative traits in maize. Crop Sci. 47, 1082–1090. Bishop, C. M. (2006). Pattern Recognition and Machine Learning. Springer, New York, USA. Blanc, G., Charcosset, A., Mangin, B., Gallais, A., and Moreau, L. (2006). Connected populations for detecting quantitative trait loci and testing for epistasis: An application in maize. Theor. Appl. Genet. 113, 206–224.

Genomic Selection in Plant Breeding: Knowledge and Prospects

117

Buckler, E. S., Holland, J. B., Bradbury, P. J., Acharya, C. B., Brown, P. J., Browne, C., Ersoz, E., Flint-Garcia, S., Garcia, A., Glaubitz, J. C., Goodman, M. M., Harjes, C., et al. (2009). The genetic architecture of maize flowering time. Science 325, 714–718. Calus, M. P., and Veerkamp, R. F. (2007). Accuracy of breeding values when using and ignoring the polygenic effect in genomic breeding value estimation with a marker density of one SNP per cM. J. Anim. Breed. Genet. 124, 362–368. Chakraborty, R., Moreau, L., and Dekkers, J. C. M. (2002). A method to optimize selection on multiple identified quantitative trait loci. Genet. Sel. Evol. 34, 145–170. Chamberlain, A. J., McPartlan, H. C., and Goddard, M. E. (2007). The number of loci that affect milk production traits in dairy cattle. Genetics 177, 1117–1123. Chao, S., Zhang, W., Akhunov, E., Sherman, J., Ma, Y., Luo, M.-C., and Dubcovsky, J. (2009). Analysis of gene-derived SNP marker polymorphism in US wheat (Triticum aestivum L.) cultivars. Mol. Breed. 23, 23–33. Charmet, G., Robert, N., Perretant, M. R., Gay, G., Sourdille, P., Groos, C., Bernard, S., and Bernard, M. (1999). Marker-assisted recurrent selection for cumulating additive and interactive QTLs in recombinant inbred lines. Theor. Appl. Genet. 99, 1143–1148. Chen, X., and Sullivan, P. F. (2003). Single nucleotide polymorphism genotyping: Biochemistry, protocol, cost and throughput. Pharmacogenomics J. 3, 77–96. Cheung, F., Haas, B. J., Goldberg, S. M. D., May, G. D., Xiao, Y., and Town, C. D. (2006). Sequencing Medicago truncatula expressed sequenced tags using 454 Life Sciences technology. BMC Genomics 7, 272. Clark, A. G., Hubisz, M. J., Bustamante, C. D., Williamson, S. H., and Nielsen, R. (2005). Ascertainment bias in studies of human genome-wide polymorphism. Genome Res. 15, 1496–1502. Close, T., Bhat, P., Lonardi, S., Wu, Y., Rostoks, N., Ramsay, L., Druka, A., Stein, N., Svensson, J., Wanamaker, S., Bozdag, S., Roose, M., et al. (2009). Development and implementation of high-throughput SNP genotyping in barley. BMC Genomics 10, 582. Coors, J. G. (2006). Who are plant breeders, what do they do, and why? In “Plant Breeding: Arnel R Hallauer International Symposium” (K. R. Lamkey and M. Lee, Eds.), pp. 51–60. Blackwell Publishing, Oxford, UK. Coxe, K. L., Johnson, N. L., and Kotz, S. (1986). Principal components regression analysis. In “Encyclopedia of Statistical Science”, Vol. 7, pp. 181–184. Wiley, New York. Daetwyler, H. D., Villanueva, B., Bijma, P., and Woolliams, J. A. (2007). Inbreeding in genome-wide selection. J. Anim. Breed. Genet. 124, 369–376. Daetwyler, H. D., Villanueva, B., and Woolliams, J. A. (2008). Accuracy of predicting the genetic risk of disease using a genome-wide approach. PLoS ONE 3, e3395. de los Campos, G., Gianola, D., and Rosa, G. J. (2009a). Reproducing kernel Hilbert spaces regression: A general framework for genetic evaluation. J. Anim. Sci. 87, 1883–1887. de los Campos, G., Naya, H., Gianola, D., Crossa, J., Legarra, A., Manfredi, E., Weigel, K., and Cotes, J. M. (2009b). Predicting quantitative traits with regression models for dense molecular markers and pedigree. Genetics 182, 375–385. de Roos, A. P. W., Hayes, B. J., Spelman, R. J., and Goddard, M. E. (2008). Linkage disequilibrium and persistence of phase in Holstein-Friesian, Jersey and Angus Cattle. Genetics 179, 1503–1512. de Roos, A. P. W., Hayes, B. J., and Goddard, M. E. (2009). Reliability of genomic predictions across multiple populations. Genetics 183, 1545–1553. Dekkers, J. C. M., and Van Arendonk, J. A. M. (1998). Optimizing selection for quantitative traits with information on an identified locus in outbred populations. Genet. Res. 71, 257–275. Dekkers, J. C. M., Chakraborty, R., and Moreau, L. (2002). Optimal selection on two quantitative trait loci with linkage. Genet. Sel. Evol. 34, 171–192. Deschamps, S., and Campbell, M. A. (2010). Utilization of next-generation sequencing platforms in plant genomics and genetic variant discovery. Mol. Breed. 25, 553–570.

118

Aaron J. Lorenz et al.

Dudley, J. W., and Johnson, G. R. (2009). Epistatic models improve prediction of performance in corn. Crop Sci. 49, 763–770. Falconer, D. S., and Mackay, T. F. C. (1996). Introduction to Quantitative Genetics. fourth ed. Longman Science and Technology, Harlow, UK. Fan, J.-B., Oliphant, A., Shen, R., Kermani, B. G., Garcia, F., Gunderson, K. L., Hansen, M., Steemers, F., Butler, S. L., Deloukas, P., Galver, L., Hunt, S., et al. (2003). Highly parallel SNP genotyping. Cold Spring Harb. Symp. Quant. Biol. 68, 69–78. Fernando, R. L., Habier, D., Stricker, C., Dekkers, J. C. M., and Totir, L. R. (2007). Genomic selection. Acta Agric. Scand. A 57, 192–195. Fu, Y., Springer, N. M., Gerhardt, D. J., Ying, K., Yeh, C.-T., Wu, W., SwansonWagner, R., D’Ascenzo, M., Millard, T., Freeberg, L., Aoyama, N., Kitzman, J., et al. (2010). Repeat subtraction-mediated sequence capture from a complex genome. Plant J. 62(5), 898–909. Ganal, M. W., Altmann, T., and Roeder, M. S. (2009). SNP identification in crop plants. Curr. Opin. Plant Biol. 12, 211–217. George, E. I., and McCulloch, R. E. (1993). Variable selection via Gibbs sampling. J. Am. Stat. Assoc. 88, 881–889. Gianola, D., and de los Campos, G. (2008). Inferring genetic values for quantitative traits non-parametrically. Genet. Res. 90, 525–540. Gianola, D., Perez-Enciso, M., and Toro, M. A. (2003). On marker-assisted prediction of genetic value: Beyond the ridge. Genetics 163, 365–374. Gianola, D., Fernando, R. L., and Stella, A. (2006). Genomic-assisted prediction of genetic value with semiparametric procedures. Genetics 173, 1761–1776. Gianola, D., de los Campos, G., Hill, W. G., Manfredi, E., and Fernando, R. (2009). Additive genetic variability and the Bayesian alphabet. Genetics 183, 347–363. Gibson, J. P. (1994). Short-term gain at the expense of long-term response with selection of identified loci. Proc. 5th World Congr. Genet. Appl. Livest. Prod. 21, 201–204. Goddard, M. (2009). Genomic selection: Prediction of accuracy and maximisation of long term response. Genetica 136, 245–257. Goddard, M. E., and Hayes, B. J. (2007). Genomic selection. J. Anim. Breed. Genet. 124, 323–330. Goddard, M. E., Hayes, B. J., McPartlan, H., and Chamberlain, A. J. (2006). Can the same genetic markers be used in multiple breeds? In “Proceedings of the 8th World Congress on Genetics Applied to Livestock Production, Belo Horizonte, Brazil,” CD-ROM Commun. No. 22-16. Gore, M. A., Wright, M. H., Ersoz, E. S., Bouffard, P., Szekeres, E. S., Jarvie, P., Hurwitz, B. L., Narechania, A., Harkins, T. T., Grills, G. S., Ware, D. H., and Buckler, E. S. (2009). Large-scale discovery of gene-enriched SNPs. Plant Genome 2, 121–133. Habier, D., Fernando, R. L., and Dekkers, J. C. M. (2007). The impact of genetic relationship information on genome-assisted breeding values. Genetics 177, 2389–2397. Habier, D., Fernando, R. L., and Dekkers, J. C. M. (2009). Genomic selection using lowdensity marker panels. Genetics 182, 343–353. Habier, D., Tetens, J., Seefried, F.-R., Lichtner, P., and Thaller, G. (2010). The impact of genetic relationship information on genomic breeding values in German Holstein cattle. Genet. Sel. Evol. 42, 5. Haley, C. S., and Visscher, P. M. (1998). Strategies to utilize marker-quantitative trait loci associations. J. Dairy Sci. 81, 85–97. Hamblin, M. T., Close, T. J., Bhat, P. R., Chao, S., Kling, J. G., Abraham, K. J., Blake, T., Brooks, W. S., Cooper, B., Griffey, C. A., Hayes, P. M., Hole, D. J., et al. (2010). population structure and linkage disequilibrium in U.S. barley germplasm: Implications for association mapping. Crop Sci. 50, 556–566.

Genomic Selection in Plant Breeding: Knowledge and Prospects

119

Hayashi, T., and Iwata, H. (2010). EM algorithm for Bayesian estimation of genomic breeding values. BMC Genet. 11, 3. Hayes, B. J., Bowman, P. J., Chamberlain, A. J., and Goddard, M. E. (2009a). Invited review: Genomic selection in dairy cattle: Progress and challenges. J. Dairy Sci. 92, 433–443. Hayes, B. J., Bowman, P. J., Chamberlain, A. J., Verbyla, K. L., and Goddard, M. E. (2009b). Accuracy of genomic breeding values in multi-breed dairy cattle populations. Genet. Sel. Evol. 41, 51. Hayes, B. J., Visscher, P. M., and Goddard, M. E. (2009c). Increased accuracy of artificial selection by using the realized relationship matrix. Genet. Res. 91, 47–60. Heffner, E. L., Sorrells, M. E., and Jannink, J.-L. (2009). Genomic selection for crop improvement. Crop Sci. 49, 1–12. Heffner, E. L., Lorenz, A. J., Jannink, J.-L., and Sorrells, M. (2010). Plant breeding with genomic selection: Potential gain per unit time and cost. Crop Sci. 50, 1681–1690. Henderson, C. R. (1975). Best linear unbiased estimation and pre-diction under a selection model. Biometrics 31, 423–447. Holland, J. B. (2001). Epistasis and plant breeding. Plant Breed. Rev. 21, 27–92. Hospital, F., and Charcosset, A. (1997). Marker-assisted introgression of quantitative trait loci. Genetics 147, 1469–1485. Hospital, F., Moreau, L., Lacoudre, F., Charcosset, A., and Gallais, A. (1997). More on the efficiency of marker-assisted selection. Theor. Appl. Genet. 95, 1181–1189. Hospital, F., Goldringer, I., and Openshaw, S. (2000). Efficient marker-based recurrent selection for multiple quantitative trait loci. Genet. Res. 75, 357–368. Huang, X., Feng, Q., Qian, Q., Zhao, Q., Wang, L., Wang, A., Guan, J., Fan, D., Weng, Q., Huang, T., Dong, G., Sang, T., et al. (2009). High-throughput genotyping by whole-genome resequencing. Genome Res. 19, 1068–1076. Hyten, D. L., Song, Q., Choi, I.-Y., Yoon, M.-S., Specht, J. E., Matukumalli, L. K., Nelson, R. L., Shoemaker, R. C., Young, N. D., and Cregan, P. B. (2008). Highthroughput genotyping with the GoldenGate assay in the complex genome of soybean. Theor. Appl. Genet. 116, 945–952. Hyten, D. L., Cannon, S. B., Song, Q., Weeks, N., Fickus, E. W., Shoemaker, R. C., Specht, J. E., Farmer, A. D., May, G. D., and Cregan, P. B. (2010). High-throughput SNP discovery through deep resequencing of a reduced representation library to anchor and orient scaffolds in the soybean whole genome sequence. BMC Genomics 11, 38. Iba´nez-Escriche, N., Fernando, R., Toosi, A., and Dekkers, J. (2009). Genomic selection of purebreds for crossbred performance. Genet. Sel. Evol. 41, 12. Iwata, H., and Jannink, J.-L. (2010). Accuracy of genomic selection prediction in barley breeding programs: A simulation study based on the real SNP data of barley breeding lines. Crop Sci. (in review). Jaccoud, D., Peng, K., Feinstein, D., and Kilian, A. (2001). Diversity arrays: A solid state technology for sequence information independent genotyping. Nucleic Acids Res. 29, e25. Jannink, J. L. (2005). Selective phenotyping to accurately map quantitative trait loci. Crop Sci. 45, 901–908. Jannink, J.-L., Lorenz, A. J., and Iwata, H. (2010). Genomic selection in plant breeding: From theory to practice. Brief. Funct. Genomics 9, 166–177. Jin, C., Lan, H., Attie, A. D., Churchill, G. A., Bulutuglo, D., and Yandell, B. S. (2004). Selective phenotyping for increased efficiency in genetic mapping studies. Genetics 168, 2285–2293. Kearsey, M. J., and Farquhar, A. G. L. (1998). QTL analysis in plants; where are we now? Heredity 80, 137–142.

120

Aaron J. Lorenz et al.

Kizilkaya, K., Fernando, R. L., and Garrick, D. J. (2009). Genomic prediction of simulated multi-breed and purebred performance using observed 50K SNP genotypes. J. Anim. Sci.10.2527/jas.2009-2064. Knapp, S. J., and Bridges, W. C. (1990). Using molecular markers to estimate quantitative trait locus parameters: Power and genetic variances for unreplicated and replicated progeny. Genetics 126, 769–777. Lande, R. (1981). The minimum number of genes contributing to quantitative variation between and within populations. Genetics 99, 541–553. Lande, R., and Thompson, R. (1990). Efficiency of marker-assisted selection in the improvement of quantitative traits. Genetics 124, 743–756. Lander, E. S., and Botstein, D. (1989). Mapping Mendelian factors underlying quantitative traits using RFLP maps. Genetics 121, 185–199. Lee, S. H., van der Werf, J. H. J., Hayes, B. J., Goddard, M. E., and Visscher, P. M. (2008). Predicting unobserved phenotypes for complex traits from whole-genome SNP data. PLoS Genet. 4, e1000231. doi:10.1371/journal.pgen.1000231. Legarra, A., Robert-Granie, C., Manfredi, E., and Elsen, J.-M. (2008). Performance of genomic selection in mice. Genetics 180, 611–618. Li, Y., Kadarmideen, H. N., and Dekkers, J. C. M. (2008). Selection on multiple QTL with control of gene diversity and inbreeding for long-term benefit. J. Anim. Breed. Genet. 125, 320–329. Lorenzana, R., and Bernardo, R. (2009). Accuracy of genotypic value predictions for marker-based selection in biparental plant populations. Theor. Appl. Genet. 120, 151–161. Luan, T., Woolliams, J. A., Lien, S., Kent, M., Svendsen, M., and Meuwissen, T. H. E. (2009). The accuracy of genomic selection in Norwegian red cattle assessed by cross validation. Genetics 183(3), 1119–1126. Luo, M. C., Deal, K. R., Akhunov, E. D., Akhunova, A. R., Anderson, O. D., Anderson, J. A., Blake, N., Clegg, M. T., Coleman-Derr, D., Conley, E. J., Crossman, C. C., Dubcovsky, J., et al. (2009). Genome comparisons reveal a dominant mechanism of chromosome number reduction in grasses and accelerated genome evolution in Triticeae. Proc. Natl. Acad. Sci. USA 106, 15780–15785. Lynch, M., and Walsh, B. (1998). Genetics and Analysis of Quantitative Traits. Sinauer Associates, Sunderland, MA. Maenhout, S., De Baets, B., Haesaert, G., and Van Bockstaele, E. (2007). Support vector machine regression for the prediction of maize hybrid performance. Theor. Appl. Genet. 115, 1003–1013. Maenhout, S., De Baets, B., Haesaert, G., and Van Bockstaele, E. (2008). Marker-based screening of maize inbred lines using support vector machine regression. Euphytica 161, 123–131. Manolio, T. A., Collins, F. S., Cox, N. J., Goldstein, D. B., Hindorff, L. A., Hunter, D. J., McCarthy, M. I., Ramos, E. M., Cardon, L. R., Chakravarti, A., Cho, J. H., Guttmacher, A. E., et al. (2009). Finding the missing heritability of complex diseases. Nature 461, 747–753. Mardis, E. R. (2008). The impact of next-generation sequencing technology on genetics. Trends Genet. 24, 133–141. Maughan, P. J., Yourstone, S. M., Jellen, E. N., and Udall, J. A. (2009). SNP discovery via genomic reduction, barcoding, and 454-pyrosequencing in amaranth. Plant Genome 2, 260–270. McMullen, M. D., Kresovich, S., Villeda, H. S., Bradbury, P., Li, H., Sun, Q., Flint-Garcia, S., Thornsberry, J., Acharya, C., Bottoms, C., Brown, P., Browne, C., et al. (2009). Genetic properties of the maize nested association mapping population. Science 325, 737–740.

Genomic Selection in Plant Breeding: Knowledge and Prospects

121

Melchinger, A. E., Utz, H. F., and Schon, C. C. (1998). Quantitative trait locus (QTL) mapping using different testers and independent population samples in maize reveals low power of QTL detection and large bias in estimates of QTL effects. Genetics 149, 383–403. Meuwissen, T. H. E. (1997). Maximizing the response of selection with a predefined rate of inbreeding. J. Anim. Sci. 75, 934–940. Meuwissen, T. H. E. (2009). Accuracy of breeding values of ‘unrelated’ individuals predicted by dense SNP genotyping. Genet. Sel. Evol. 41, 35. Meuwissen, T. H. E., Hayes, B. J., and Goddard, M. E. (2001). Prediction of total genetic value using genome-wide dense marker maps. Genetics 157, 1819–1829. Meuwissen, T. H. E., Solberg, T. R., Shepherd, R., and Woolliams, J. A. (2009). A fast algorithm for BayesB type of prediction of genome-wide estimates of genetic value. Genet. Sel. Evol. 41, 2. Moragues, M., Comadran, J., Waugh, R., Milne, I., Flavell, A., and Russell, J. (2010). Effects of ascertainment bias and marker number on estimations of barley diversity from high-throughput SNP genotype data. Theor. Appl. Genet. 120, 1525–1534. Moreau, L., Charcosset, A., Hospital, F., and Gallais, A. (1998). Marker-assisted selection efficiency in populations of finite size. Genetics 148, 1353–1365. Moreau, L., Charcosset, A., and Gallais, A. (2004). Experimental evaluation of several cycles of marker-assisted selection in maize. Euphytica 137, 111–118. Moser, G., Tier, B., Crump, R. E., Khatkar, M. S., and Raadsma, H. W. (2009). A comparison of five methods to predict genomic breeding values of dairy bulls from genome-wide SNP markers. Genet. Sel. Evol. 41, 56. Muir, W. M. (2007). Comparison of genomic and traditional BLUP-estimated breeding value accuracy and selection response under alternative trait and genomic parameters. J. Anim. Breed. Genet. 124, 342–355. Munkvold, J. D., Tanaka, J., Benscher, D., and Sorrells, M. E. (2009). Mapping quantitative trait loci for preharvest sprouting resistance in white wheat. Theor. Appl. Genet. 119, 1223–1235. Odegard, J., Yazdi, M. H., Sonesson, A. K., and Meuwissen, T. H. (2009). Incorporating desirable genetic characteristics from an inferior into a superior population using genomic selection. Genetics 181(2), 737–745. Park, T., and Casella, G. (2008). The Bayesian Lasso. J. Am. Stat. Assoc. 103, 681–686. Paterson, A. H., Lander, E. S., Hewitt, J. D., Peterson, S., Lincoln, S. E., and Tanksley, S. D. (1988). Resolution of quantitative traits into Mendelian factors using a complete linkage map of restriction fragment length polymorphisms. Nature 335, 721–726. Perkel, J. (2008). SNP genotyping: Six technologies that keyed a revolution. Nat. Methods 5, 447–453. Phillips, P. C. (2008). Epistasis—The essential role of gene interactions in the structure and evolution of genetic systems. Nat. Rev. Genet. 9, 855–867. Piepho, H. P. (2009). Ridge regression and extensions for genomewide selection in maize. Crop Sci. 49, 1165–1176. Podolak, E. (2010). Sequencing’s new race. Biotechniques 48, 105–111. Rafalski, A. (2002). Applications of single nucleotide polymorphisms in crop genetics. Curr. Opin. Plant Biol. 5, 94–100. Ragoussis, J. (2006). Genotyping technologies for all. Drug Discov. Today 3, 115–122. Schaeffer, L. R. (2006). Strategy for applying genome-wide selection in dairy cattle. J. Anim. Breed. Genet. 123, 218–223. Scho¨lkopf, B., Tsuda, K., and Vert, J.-P. (2004). Kernel Methods in Computational Biology. MIT Press, Cambridge, MA. Scho¨n, C. C., Utz, H. F., Groh, S., Truberg, B., Openshaw, S., and Melchinger, A. E. (2004). Quantitative trait locus mapping based on resampling in a vast maize testcross

122

Aaron J. Lorenz et al.

experiment and its relevance to quantitative genetics for complex traits. Genetics 167, 485–498. Servin, B., Martin, O. C., Mezard, M., and Hospital, F. (2004). Toward a theory of marker-assisted gene pyramiding. Genetics 168, 513–523. Shendure, J., and Ji, H. (2008). Next-generation DNA sequencing. Nat. Biotechnol. 26, 1135–1145. Smola, A. J., and Scho¨lkopf, B. (2004). A tutorial on support vector regression. Stat. Comput. 14, 199–222. Solberg, T. R., Sonesson, A. K., Woolliams, J. A., and Meuwissen, T. H. E. (2008). Genomic selection using different marker types and densities. J. Anim. Sci. 86, 2447–2454. Solberg, T. R., Sonesson, A. K., Woolliams, J. A., and Meuwissen, T. H. E. (2009). Reducing dimensionality for prediction of genome-wide breeding values. Genet. Sel. Evol. 41, 29. Stuber, C. W., Goodman, M. M., and Moll, R. H. (1982). Improvement of yield and ear number resulting from selection at allozyme loci in a maize population. Crop Sci. 22, 737–740. Su, G., Guldbrandtsen, B., Gregersen, V. R., and Lund, M. S. (2010). Preliminary investigation on reliability of genomic estimated breeding values in the Danish Holstein population. J. Dairy Sci. 93, 1175–1183. Sved, J. A. (1971). Linkage disequilibrium and homozygosity of chromosome segments in finite populations. Theor. Popul. Biol. 2, 125–141. Sved, J. A. (2009). Correlation measures for linkage disequilibrium within and between populations. Genet. Res. 91, 183–192. Syvanen, A. C. (2001). Accessing genetic variation: Genotyping single nucleotide polymorphisms. Nat. Rev. Genet. 2, 930–942. Syvanen, A. C. (2005). Toward genome-wide SNP genotyping. Nat. Genet. 37, S5–S10. Szu¨cs, P., Blake, V. C., Bhat, P. R., Chao, S., Close, T. J., Cuesta-Marcos, A., Muehlbauer, G. J., Ramsay, L., Waugh, R., and Hayes, P. M. (2009). An Integrated resource for barley linkage map and malting quality QTL alignment. Plant Genome 2, 134–140. Tanksley, S. D., Young, N. D., Paterson, A. H., and Bonierbale, M. W. (1989). RFLP mapping in plant breeding: New tools for an old science. Biotechnology 7, 257–264. Thissen, U., Ustun, B., Melssen, W. J., and Buydens, L. M. C. (2004). Multivariate calibration with least-squares support vector machines. Anal. Chem. 76, 3099–3105. Tibshirani, R. (1996). Regression shrinkage and selection via the lasso. J. R. Stat. Soc. B 58, 267–288. Tobias, R. D. (1997). An Introduction to Partial Least Squares Regression. SAS Institute, Cary, NC. Toosi, A., Fernando, R. L., and Dekkers, J. C. M. (2010). Genomic selection in admixed and crossbred populations. J. Anim. Sci. 88, 32–46. Trick, M., Long, Y., Meng, J., and Bancroft, I. (2009). Single nucleotide polymorphism (SNP) discovery in the polyploid Brassica napus using Solexa transcriptome sequencing. Plant Biotechnol. J. 7, 334–346. Usai, M. G., Goddard, M. E., and Hayes, B. J. (2009). LASSO with cross-validation for genomic selection. Genet. Res. 91, 427–436. Valdar, W., Solberg, L. C., Gauguier, D., Burnett, S., Klenerman, P. O., Cookson, W., Taylor, M. S., Rawlins, J. N. P., Mott, R., and Flint, J. (2006). Genome-wide genetic association of complex traits in heterogeneous stock mice. Nat. Genet. 38, 879–887. van Orsouw, N. J., Hogers, R. C. J., Janssen, A., Yalcin, F., Snoeijers, S., Verstege, E., Schneiders, H., van der Poel, H., van Oeveren, J., Verstegen, H., and van Eijk, M. J. T. (2007). Complexity reduction of polymorphic sequences (CRoPS (TM)): A novel

Genomic Selection in Plant Breeding: Knowledge and Prospects

123

approach for large-scale polymorphism discovery in complex genomes. PLoS ONE 2, e1172. doi:10.1371/journal.pone.0001172. VanRaden, P. M., Van Tassell, C. P., Wiggans, G. R., Sonstegard, T. S., Schnabel, R. D., Taylor, J. F., and Schenkel, F. S. (2009). Invited review: Reliability of genomic predictions for North American Holstein bulls. J. Dairy Sci. 92, 16–24. Varshney, R. K., Nayak, S. N., May, G. D., and Jackson, S. A. (2009). Next-generation sequencing technologies and their implications for crop genetics and breeding. Trends Biotechnol. 27, 522–530. Verbyla, K. L., Hayes, B. J., Bowman, P. J., and Goddard, M. E. (2009). Accuracy of genomic selection using stochastic search variable selection in Australian Holstein Friesian dairy cattle. Genet. Res. 91, 307–311. Weber, A. P. M., Weber, K. L., Carr, K., Wilkerson, C., and Ohlrogge, J. B. (2007). Sampling the Arabidopsis transcriptome with massively parallel pyrosequencing. Plant Physiol. 144, 32–42. Whittaker, J. C., Thompson, R., and Denham, M. C. (2000). Marker-assisted selection using ridge regression. Genet. Res. 75, 249–252. Wold, H., Johnson, N. L., and Kotz, S. (1985). Partial least squares. “Encyclopedia of Statistical Science”, pp. 581–591. Wiley, New YorkVol. 7. Wong, C., and Bernardo, R. (2008). Genomewide selection in oil palm: Increasing selection gain per unit time and cost with small populations. Theor. Appl. Genet. 116, 815–824. Wray, N. R., Goddard, M. E., and Visscher, P. M. (2007). Prediction of individual genetic risk to disease from genome-wide association studies. Genome Res. 17, 1520–1528. Xu, S. (2003). Theoretical basis of the Beavis effect. Genetics 165, 2259–2268. Xu, S. (2007). An empirical Bayes method for estimating epistatic effects of quantitative trait loci. Biometrics 63, 513–521. Xu, Y., and Crouch, J. H. (2008). Marker-assisted selection in plant breeding: From publications to practice. Crop Sci. 48, 391–407. Yi, N., and Banerjee, S. (2009). Hierarchical generalized linear models for multiple quantitative trait locus mapping. Genetics 181, 1101–1113. Zeng, Z. B. (1992). Correcting the bias of WRIGHT’s estimates of the number of genes affecting a quantitative character: A further improved method. Genetics 131, 987–1001. Zhong, S., and Jannink, J.-L. (2007). Using quantitative trait locus results to discriminate among crosses based on their progeny mean and variance. Genetics 177, 567–576. Zhong, S., Toubia-Rahme, H., Steffenson, B. J., and Smith, K. P. (2006). Molecular mapping and marker-assisted selection of genes for septoria speckled leaf blotch resistance in barley. Phytopathology 96, 993–999. Zhong, S., Dekkers, J. C. M., Fernando, R. L., and Jannink, J. L. (2009). Factors affecting accuracy from genomic selection in populations derived from multiple inbred lines: A barley case study. Genetics 182, 355–364.

C H A P T E R

T H R E E

Differences of Some Leguminous and Nonleguminous Crops in Utilization of Soil Phosphorus and Responses to Phosphate Fertilizers Sheng-Xiu Li,* Zhao-Hui Wang,* and Bobby Alton Stewart† Contents 1. Introduction 2. The Difference of P Uptake Amounts of Leguminous and Nonleguminous Crops 3. Leguminous and Nonleguminous Crop Responses to Powdered Rock Phosphates 4. The Relation of Plants’ Root Morphology to Their Capacity of Using Soil Sparingly Soluble P and Responses to P Fertilizers 5. Microorganisms in Rhizosphere Soil and Their Function in Supplying P to Plants 6. Root Exudates (Substances Secreted from Roots) and the Plants’ Capacity to Use Sparingly Soluble P in the Soil and Crop Responses to P Fertilizers 7. Effects of Root Cation Exchange Capacity and Calcium Uptake Amount of Crops on Soil P Absorption and Crop Responses to P Fertilizer 8. The Responses to P Fertilizer Between Leguminous and Cereal Crops with Their Biological Characteristics 8.1. Soil fertility for planting leguminous and cereal crops 8.2. Effect of addition of N on crop uptake of P from soil 8.3. Effect of N fertilization on crop responses to P fertilizer 8.4. The response of leguminous crops to P fertilizer before and after nodulation and N fixation 8.5. Responses of leguminous crops to P fertilizer after inhibition of nodule formation

130 141 146 163 173

176

189 193 195 197 197 199 202

* College of Resources and Environmental Sciences, Northwest Science and Technology University of Agriculture and Forestry, Yangling, Shaanxi, PR China { Dryland Agriculture Institute, West Texas A&M University, Canyon, Texas, USA Advances in Agronomy, Volume 110 ISSN 0065-2113, DOI: 10.1016/B978-0-12-385531-2.00003-7

#

2011 Elsevier Inc. All rights reserved.

125

126

Sheng-Xiu Li et al.

9. Factors Affecting the Responses of Leguminous and Nonleguminous Crops to P Fertilizer 9.1. Relationship between nonleguminous crop responses to P fertilizer and soil-available P 9.2. Relation of the responses of leguminous crops to P fertilizers with soil-available P 9.3. Effect of N fertilizer on leguminous crops’ response to P fertilizer 9.4. Response of leguminous crops to P fertilizer rate 9.5. The availability index for application of P fertilizer to leguminous crops 10. Conclusions 10.1. Importance of P in agriculture 10.2. Demand of P for leguminous and nonleguminous crops 10.3. Responses to RPs between legumes and nonlegumes 10.4. Root morphology 10.5. Soil microorganisms 10.6. Acidification of rhizosphere by roots’ exudates for crops’ responses to P fertilizers 10.7. Effects of root CEC and calcium uptake amount of crops on crop responses to P fertilizer 10.8. Responses to P fertilizer of leguminous crops with N fixation 10.9. Factors affecting crop responses to P fertilizer 10.10. Brief summary Acknowledgments References

203 203 207 209 217 219 222 222 222 223 223 224 224 224 224 225 226 227 227

Abstract As a vital component of a number of macromolecules and an integral part of energy metabolism and major biological processes in photosynthesis, respiration, and membrane transportation, as well as playing a genetic role through ribonucleic acid and energy transfers via adenosine triphosphate, phosphorous is indispensable for all life forms and cannot be substituted by any other element. Being the life-limiting element in natural ecosystems, regular inputs of P fertilizer to replenish the P removed from the soil by crops are one of the characteristics of modern agriculture. The demand for P resources will outstrip supply in the coming decades because the global commercial phosphate reserves may be depleted in another 60–130 years. In addition, rock phosphate (RP) reserves are under the control of a few countries. The P recovery rate is very low and the surpluses of P in soil have produced variable responses of crops to P fertilizers and environmental pollution. Requirements for direct application of RP and improvement of P fertilizer efficiency have led to adoption of specific plant species. Since leguminous crops in general respond better to P fertilizer than cereals, some scientists have proposed the application of P to

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

127

leguminous crops as the first priority. Many hypotheses have been proposed to explain the different responses to P fertilizer between the two types of crops, but most of them have not been substantiated. A series of experiments have been conducted by us on different aspects for more than 40 years, and this chapter reviews the current investigation status and reports our viewpoints based on results obtained mainly from wheat [Triticum aestivum (L.) em. Thell], pea (Pisum sativum L.), maize (Zea mays L.), and soybean [Glycine max (L.) Merr.] One reason for the different response to P fertilizer supposes that legumes require more P than nonlegumes. A long-term experiment in a maize–maize– soybean rotation sequence in which maize and soybean were grown in the same season with almost the same duration of growing period showed that the total P uptake by soybean from unit area was similar to that of maize and in some cases uptake by maize was higher than that by soybean. Results of pot and field experiments conducted by us showed that P uptake amount by leguminous crops was not higher than that by cereal crops, and wheat had a higher capacity to use soil P than do pea and vetch. Without application of N fertilizer, P amounts taken up by legumes were equal to or slightly higher than those of nonlegumes, while cereal crops with N application took up much more P than legumes in most cases either with or without application of P fertilizer. In calcareous soil, the P availability in P-containing minerals is roughly equal to that in RPs. Some agricultural scientists have used RPs as substitutes to test the crops’ ability of using P from soil. Results show that in addition to the physical and chemical properties of RPs and soil pH, some plant species such as rapeseed (Brassica campestris L. and Brassica napus L.), radish (Raphanus sativus L.), and some legumes possess strong abilities in absorbing P from RPs in acid soils, and good responses of rapeseed to the RPs were also found in calcareous soils. However, there were some debatable issues in these studies: comparisons were made not in the same field but in different locations; excessive rates of RPs were used and the available P in the RPs extracted by 2% citric acid was closely correlated with crops’ yield increase; only rapeseed was used for comparison; soils in which yield increase by RPs was several times higher than the control and even higher than superphosphate were particularly unique and had strong responses to calcium carbonate and thus there was no way to separate the effect of P and calcium carbonate contained in RPs; and some results were conflicting. A series of pot and field experiments were conducted by us in two calcareous soils using six crops planted in spring and autumn for both pot and field trials with sufficient N supply. Results showed that crops significantly responded to single superphosphate, while the effect of RPs was seen only at high rates of application. Comprehensive comparisons of the yield increase in absolute amount and in percentage of the control showed that there was almost no difference between legumes and nonlegumes used in our experiments in responses to the RPs. Again, for any crop, the response to RPs was closely related to the citric-acid-soluble P (P2O5) in RPs. Since the P concentration in soil solution is very dilute, the movement of P in soil takes place mainly by diffusion, the diffusion coefficient is very low, and the distance moved is very short, and many scientists hold the view that plant roots

128

Sheng-Xiu Li et al.

play a great role in ensuring sufficient P supply to crops and the different responses are attributed to their root characteristics. Our results indicated that wheat had better developed roots, while pea roots had a higher function in supporting shoots. For a single plant, the total root-absorbing and actively absorbing area of pea was larger than that of wheat, but in one unit volume of soil, root dry weight of wheat was 33% higher than that of pea. The root activities in terms of the 2,3,5-triphenyltetrazolium chloride (TTC) reductive amount and intensity were higher for pea than for wheat. All these root properties could not explain the different responses to P fertilizer between legumes and nonlegumes. Many soil microorganisms are able to transform insoluble forms of P to an accessible soluble form and are regarded as plant growth-promoting microorganisms (PGPM). However, there is no evidence showing their release of P to legumes and nonlegumes, and the very limited knowledge has restrained us from continuing further discussion. Of the hypotheses proposed for explaining the difference of crops in utilizing sparingly soluble P in the soil and their responses to P fertilizer, the most common view is the reduction of rhizosphere pH resulting from the release of protons and organic acids. Our experiments showed that pH in rhizospheric soil was generally one unit lower than that in the bulk soil, but there was no difference in pH between wheat and pea either in rhizospheric or in bulk soil. The available P had the same trend as soil pH. Clearly, the acidification of rhizosphere soil could not differentiate the ability of crops to respond to P fertilizer. The cation exchange capacity (CEC) of plant roots was once considered the basis for crops to exchange cations with those held in soil colloids, and crops with a high root CEC could take up more calcium from soil and thus liberate P bound with calcium for crop use. Our study with wheat and pea showed that the root CEC of pea was several times higher than that of wheat in terms of per kilogram dry root or root weight per pot. However, the P uptake amounts by the two crops did not follow the same pattern as the CEC. In relation with root CEC, crop uptake of the calcium was considered a mechanism for P release, and the ratio of CaO to P2O5 in plant tissue was proposed as an index of the plant’s ability for absorbing P from RPs. However, later researchers have rejected this hypothesis. A series of experiments have shown that the sensitive responses of legumes to P fertilizer are related to their N fixation capacity. These have been evidenced by many facts. It was not always the case that legumes had good responses to P fertilizer; it was true only in a soil deficient in both N and P nutrients. In a soil with sufficient N supply, the difference in P fertilizer response vanished, and the response by cereal crops was even better than by legumes. Addition of N fertilizer to cereal crops significantly increased the P fertilizer effect. In a soil extremely deficient in N and P supply, maize with N addition absorbed almost the same amount of P from the soil as soybean. Without application of N, P uptake by pea was 33% higher than that of wheat, but

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

129

when fertilized with N there was no difference in the amount of P uptake for the two crops. Also, in a soil deficient in N and P supply, application of N fertilizer to cereal crops and no N or a small amount of N to legumes often resulted in much better responses of cereal crops to P fertilizer. This shows that the sensitivity of legume response to P fertilizer is related to its N fixation. Experiments showed that, before pea acquired the ability to fix N, the biomass increase and P uptake amount were much lower than for wheat, but after reaching the stage for fixing N, pea took up much more P than wheat. This reveals that N fixation increased the amount of P uptake of legumes. Application of N fertilizer to pea to depress N fixation and not adding N fertilizer to wheat resulted in almost the same amount of P uptake by the two crops. In contrast, application of N to wheat led to more P uptake than in the case of pea. The same trend was found for maize and soybean. A layer of soil in which soybean had been grown previously had no nodules and such a soil layer was sampled for conducting a pot experiment with maize and soybean. Maize was given two treatments, i.e., with and without P fertilization, whereas soybean was treated with inoculation and without inoculation on both P treatments. Results showed that in both cases with and without P fertilizer, there was almost no difference in P uptake. However, when soybean was inoculated, the P uptake amount and dry matter increase by P addition were much larger than in maize and soybean that was not inoculated. This clearly indicates the importance of N fixation in leguminous crops to respond to P fertilizers. The responses of cereal crops to P fertilizer are mainly determined by the available P in the soil, and this is true also for leguminous crops. Eight plants grown in five soils in a pot experiment showed that organic matter, total N, total P, soil CaCO3 contents, and available N in the soil are not related to the P fertilizer effect, but the P available as determined either by the Olsen (available P extracted by 0.5 mol l 1 NaHCO3 solution) or the Machigin method (a Russian method of extracting available P by 1% (NH4)2CO3 solution) is closely correlated with the crop response to P fertilizer. Field experiments confirmed these results. Application of N fertilizer is beneficial to P fertilization for cereal crops but detrimental to leguminous crops especially when soil mineral N is abundant. The negative effect of N fertilization on the response of leguminous crops to P fertilizer is mainly caused by the reduction of nodule formation and inhibition of root length, and thereby the elimination of the superiority of leguminous crops, leading to low productivity. Since yields of leguminous crops are lower than cereal crops, the P rate for pea can be reduced at least by 20% as compared to wheat. A series of field experiments have further shown that responses to P fertilizer by legume crops were determined by the soil-available P, as was also the case for cereal crops. In a soil having low available P, P fertilization significantly increased pea yield, and the absolute amount of the increase was almost the same as for wheat, but the percentage of increase was higher than that of wheat due to the low yield in the pea control treatment. At a medium level of soilavailable P, pea still had good response to P fertilizer, while wheat had almost

130

Sheng-Xiu Li et al.

no response without application of N fertilizer. Wheat became more and more responsive to P fertilizer with N rate increase, but pea became less responsive. This shows that N was the major constraint limiting wheat response to P fertilizer, and addition of N fertilizer could improve its response to P fertilizer. In a soil with high available P, both pea and wheat had no response to P fertilizer no matter whether N was added or not and no matter what the rate of added N was. Such typical results show that, although leguminous crops need a smaller rate of P fertilizer, their response to soil-available P level is almost the same as for cereal crops. The availability index for the application of P fertilizer to cereal crops could be applicable for leguminous crops as well. To sum up, the different responses of legumes and nonlegumes to P fertilizer are caused neither by the difference of P uptake amount nor by the root characteristics such as root biomass, root surface area, root activity, root exudates, root CEC, and CaO/P2O5 ratio in plant tissues, but by the N fixation of leguminous crops. In a soil deficient in both N and P, the application of P fertilizer alone, the leguminous crops’ responses to P fertilizer have not been restricted by N limitation, the P fertilizer can fully play its role and therefore the effect of P fertilizer to leguminous crops is much better than that to cereal crops. However, when N supply is sufficient, it is another story for nonleguminous crops. With sufficient N supplies, cereal crops have better responses to P fertilizer than leguminous crops. Responses to P fertilizers are determined by the available soil P levels for both legumes and nonlegumes.

1. Introduction Among the 17 elements needed by plants, phosphorus (P) is present in all plant and animal tissues, and deserves special attention. As an element in the periodic table, P cannot be substituted by any other element. It is not only a vital component of a number of macromolecules such as nucleic acids, phospholipids, and sugar phosphates in plants, but also an integral part of the energy metabolism and major biological processes, including photosynthesis, respiration, and membrane transportation (Raghothama and Karthikeyan, 2005). Its genetic role in ribonucleic acid and its function in energy transfers via adenosine triphosphate are indispensable for all forms of life (Ozanne, 1980). As a basic and essential macronutrient, P is absolutely necessary for all lives and plays an extremely important role in plant growth and development and therefore it is vital for agricultural production and for producing the food we eat (Steen, 1998). Without P in the environment, no living organism can exist. In contrast to its indispensability, P is also a limiting nutrient of plant growth in many ecosystems. In any ecosystem, plants obtain P through root uptake from soil. Although in the soil–plant–animal system over 90% of the multiple forms of P exist commonly in soil, less than 10% enters the

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

131

plant–animal life cycle. Within the soil, P is considered in three fractions: the organic P fraction, the insoluble inorganic fraction, and a small, very variable, and soluble fraction that can be absorbed by plants (Ozanne, 1980). However, different from other elements that plants need, P is not readily released from organic residues or recycled in rainfall. Soil accumulates a large amount of the organic phosphate (Po) in organic matter that may hold up to 50–80% of the total P in soil (Dalal, 1977), but generally accounts for around 50% of total insoluble soil P with high organic matter content, such as those in no-tillage management systems (Bayer et al., 2001; Gyaneshwar et al., 2002). The amount and proportion of Po vary greatly in different ecosystems (Otani and Ae, 1999). In natural ecosystems, it occupies 10–90% of the total P, while in the agro ecosystems it accounts for 30–80% (Tiessen, 2008). Long-term experiments have shown that Po is relatively stable, and the variation between years is small, even if a large amount of phosphate fertilizer is applied. Beck and Sanchez (1994) found that the P from phosphatic fertilizers mainly accumulate in the inorganic part, and the inorganic P supplies almost 96% P that plants absorb, whereas the contribution of Po to crop yield and P uptake can be ignored. Sharpley and Smith (1985) reported similar results. Xu et al. (2006) summarized 18 long-term experiments in China and revealed that the changes in Po was much slower compared to inorganic P. Sharpley (1985) estimated that the annually mineralized rate of Po reached 20–74 kg ha 1 and application of inorganic P had no influence on its amount. A large proportion of the Po is represented by inositol phosphates and lesser amounts of other phosphate esters as phospholipids (Richardson, 2001; Richardson et al., 2005; Wakelin et al., 2004). As the most dominant species of Po in soil, the inositol hexakis phosphates are also known as phytate (Turner et al., 2002). Phytate is synthesized in terrestrial ecosystems by plants and highly accumulated in soil, but poorly utilized by plants (Iyamuremye and Dick, 1996; Mudge et al., 2003). The insoluble inorganic fraction (Pi) occupies a large portion and is difficult for plants to use because the prevalence of Al, Fe, and Ca in soil links P to highly insoluble compounds. Although being readily available for plant root uptake, the soluble part in soil solution is too small to meet plant needs, and the natural concentration of orthophosphate in soil solution is often lower than that required for normal plant growth. Because of the slow rate of the diffusion of inorganic P in the soil (Barber, 1962, 1966, 1974), the limitation of the availability in the rhizosphere further intensifies the situation of P deficiency. Influenced by its low availability, P is usually the life-limiting element in natural ecosystems, and in nondisturbed soils P is found naturally in low concentrations (Brady and Weil, 2002). The supply of available P from soils is often deficient for the growth of agricultural plants, particularly when soil is transformed into agricultural management (Steen, 1998). Widespread deficiencies of P in agricultural soils impair

132

Sheng-Xiu Li et al.

agricultural productivity and jeopardize food security (Runge-Metzger, 1995). Unless the soil contains adequate P or it is supplied to the soil from external sources, plant growth will be limited. It has been realized that P deficiency and its availability depend on soil characteristics and contents of labile P fraction. According to Hinsinger (2001), about 5.7 billion ha of land worldwide contain very little available P for sustaining optimal crop production, and according to Batjes (1997) P availability to plant roots is limited in two-thirds of the cultivated soils in the world. Some estimates have shown that about 42.7–47.6% of soils in China have low levels (<10 mg P kg 1) of available P, as determined by the Olsen method (Xie and Tan, 2001; Yang et al., 2001). The deficiency of P has become a serious constraint to crop production in West Africa (Bationo et al., 1986), where the use of P fertilizer is limited by relatively high cost, unavailability, and other logistics. Phosphorus deficiency has caused several problems for crops. For example, deficiency of P in maize has produced negative effects on both yield and quality, affected the morphological and physiological characteristics of roots, reduced its capacity for water and nutrient absorption, decreased the leaf area index, rates of leaf occurrence, expansion, and duration (Fletcher et al., 2006, 2008a,b), reduced biomass accumulation, especially during the first stages of crop development (Colomb et al., 2000; Hajabbasi and Schumacher, 1994; Pellerin et al., 2000), and diminished the final number of grains per cob (Fletcher et al., 2008a). Owing to its deficiency, the P-supplying capacity of the soil determines crop production to a large extent. For sustained crop production, P must be provided as an external input (IFDC, 1998). In modern agriculture, this is usually accommodated by costly applications of phosphate fertilizers. Phosphate (P) fertilizer refers in this chapter to acidulated P fertilizer, especially single calcium superphosphate, which was the first P fertilizer product manufactured in the world as P fertilizer and has been widely used since then. More recently, additional concentrated P fertilizers have become common. Generally, the deficiency of P can be overcome or corrected by applying these phosphate fertilizers, although the efficiency of plants for acquiring P fertilizer is low because up to 80% of the applied P fertilizer can be immobilized rapidly into forms other than orthophosphate (Holford, 1997). Plants require P to grow, and food production needs an adequate supply of P on agricultural fields to sustain crop yields. Therefore, application of P through edaphic fertilization has been carried out in different ways. Historically, crop production relied on natural levels of soil P and the addition of locally available organic matter such as manure and human excreta (Ma˚rald, 1998). Chinese have used silkworm excreta and human excreta (“night soil”) as fertilizers since the first century AD (Li and Wang, 2010), and the Japanese have done it from the twelth century (Matsui, 1997). In Europe, soil degradation and recurring famines during the seventeenth and

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

133

eighteenth centuries created the need to supplement animal and human excreta with other sources of phosphorus (Ma˚rald, 1998). In the early nineteenth century, for instance, England imported large quantities of bones from other European countries. In addition to the application of P from new sources, improved agricultural techniques enabled European agriculture to recover from the famines of the eighteenth century (Ma˚rald, 1998). These improvements included crop rotation, improved handling of manure, and, in particular, the introduction of new crops such as clover that could fix nitrogen from the atmosphere. To keep up with increased food demand for rapid population growth in the twentieth century, guano and later rock phosphate (RP) were applied extensively to food crops (Brink, 1977; Smil, 2000a). Modern agriculture is dependent on regular inputs of mineral P in phosphate fertilizers derived from RP that was formed 10–15 million years ago (White, 2000) to replenish the P removed from the soil by the growing and harvesting of crops. For high crop yields, P must be applied, and it has been applied for quite a long time in many countries. It is acknowledged that approximately 90% of society’s global demand and use of P from RP is for food production (including fertilizers, feed, and food additives) (European Fertilizer Manufacturers Association, 2000; Rosmarin, 2004; Smil, 2000a). Since the end of World War II, global extraction of RP has tripled to meet industrial agriculture’s demand for NPK fertilizers (UNEP, 2005), and the inputs of P fertilizers during 1950–1980 have increased threefold (from 10 to 30 kg ha 1) in many European countries (Granstedt, 2000; Tunney et al., 2003). Currently, around 148 million metric tons of phosphate rock is produced per year for P fertilizers (Gunther, 2005; Smil, 2000a,b), and P sourced from mined phosphate rock as phosphate fertilizers accounts for around 15 million metric tons in terms of P per year (Gumbo, 2005; Gumbo and Savenije, 2001; Jasinski, 2006; Rosmarin, 2004). Therefore, the rational fertilization of P constitutes one of the most important measures to improve crop yield, and the use of mineral P fertilizers is the most effective and essential way to improve the P supply for crop production and to minimize yield loss in the soil ( Johnston, 1994; Lo¨bermann et al., 2007). The future availability of P resources is hard to forecast due to the uncertainties in both the size of the global resources of RP and future P demand. The accuracy of any reserve estimate depends on a large number of factors, the most important of which is the amount of information the estimator has at his disposal. Owing to the lack of reasonably accurate data and the differences in grades concerned, the total world reserve estimated earlier was greatly different. For example, in 1938, the total world reserves were estimated to be 15.9 billion metric tons by Barr (1960) but 46.7 billion metric tons by Jacob (1953). In 1971, the British Sulphur Corporation estimated world reserves of all grades to be 130 billion metric tons,

134

Sheng-Xiu Li et al.

equivalent to about 19.5 billion metric tons of P (British Sulphur Corporation, 1971). The Institute of Ecology (1971) stated that the known reserves of RP might be exhausted in 90–130 years. Concerning the extent of the resource base, current estimates vary widely. Although the estimates are burdened by uncertainties, reserves of RP are clearly limited. According to Klindworth (2000), markets for fertilizer products are global in scope due to the geographical concentration of the natural resources. This is certainly the case for P. It is predicted that the demand for P may increase by 50–100% by 2050 with increased global demand for food and changing diets (Steen, 1998), and the demand for P resources will outstrip supply in the coming decades. However, like oil, phosphate rock is a nonrenewable resource, and P for fertilizer is extracted from the limited reserves of RP and there is no alternative to depletion in the long term. While P demand is projected to increase, the expected global peak in P extraction is predicted to occur around 2030. Different projections or studies suggest that at current rates of extraction the global commercial phosphate reserves will be depleted or exhausted in another 50–100 years (Runge-Metzger, 1995; Steen, 1998) or 60–130 years (Steen, 1998; Wagner, 2003). This implies that the known reserves will last approximately 50–100 years (Gunther, 2005; Smil, 2000a; Steen, 1998). The remaining potential reserves are of lower quality or more costly to extract. Because natural forces dominate the mobilization of elemental P (Klee and Graedel, 2004), the anthropogenic use of the resource is burdened with high losses, especially in agriculture (Baccini and Brunner, 1991). In addition, RP reserves are concentrated in a few regions, and now the world’s remaining phosphate rock reserves are under the control of only a handful of countries. United States, China, West Sahara (especially Morocco), and Russia are supplying RP for about 72% of the P fertilizer produced (FAO, 2004), with West Sahara and China having the largest reserves (FAO, 2004). In contrast, Western Europe and India are totally dependent on imports ( Jasinski, 2006; Rosmarin, 2004). For this reason, phosphate rock supply is subject to international political influence. Morocco has a near monopoly on Western Sahara’s reserves. The United States, historically the world’s largest producer, consumer, importer, and exporter of phosphate rock and phosphate fertilizers, has approximately 25 years of domestic reserves left ( Jasinski, 2007, 2008; Stewart et al., 2005) or less than 30 years ( Jasinski, 2006; Rosmarin, 2004), and U.S. companies import significant quantities of phosphate rock from Morocco to supply their phosphate fertilizer factories ( Jasinski, 2008). This is geopolitically sensitive, as Morocco currently occupies Western Sahara and controls its phosphate rock reserves. The Western Sahara Resource Watch (WSRW) claims that “extracting and trading with phosphates from Western Sahara are contrary to international law” (WSRW, 2007).

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

135

Moreover, increasing market integration of the agricultural sector and urbanization have separated the geographical link between production and consumption processes. Large amounts of P drawn from agricultural soils end up in the sewer systems of cities. In this setting, P recycling from wastewater can have the double benefit of mitigating the depletion of resource stocks and reducing P emissions into receiving waters (Schmid Neset et al., 2008). However, the recycling is never complete and thus the ultimate depletion of phosphate reserves cannot be avoided. The modern agricultural system is characterized by a requirement of annual applications of P-rich fertilizer and harvesting crops prior to their decay phase, and transporting them to food manufacturers and consumers all over the world. Thus, it is unlike the natural biochemical cycle that recycles P back to the soil in situ via dead plant matter. Phosphate rock and phosphate fertilizers are both commodities on the international market, and international data are available for mining, fertilizer production, and application. However, after fertilizer application, there are little accurate data available for use in a global analysis of P. This is particularly true for organic P sources that are often not quantified in investigations of P flows in the food production and consumption process, as researchers are presently more interested in losses to water bodies causing eutrophication. Calculations by Smil (2000a,b) suggest the total P content in annual global agricultural harvests is approximately 12 million tons (Mt) P, of which 7 Mt is processed for feed and food and fiber, while 40% of the remaining 5 Mt P of crop residues is returned to the land. The continuous application of P fertilizers, particularly in many cases when P in manure and fertilizers has been applied in excess of recommendations for good practice (Sharpley, 1999; Tamminga, 2003), or in some instances the recommendations themselves have exceeded the minimum necessary for optimum crop production (Tunney et al., 2003), has resulted in P inputs to agricultural soils exceeding the P uptake by crops. This has led to considerable annual surpluses (input – output) and substantial accumulations of P in agricultural soils. In the Netherlands, France, and Germany, recent national P surpluses average as high as 25–30 kg ha 1 (Smil, 2000a; Tunney et al., 2003), while in Sweden, Norway, and the UK, the P surpluses in intensive livestock farms are about 8–20 kg ha 1 (Ule´n et al., 2007). In China, excess P fertilizer application has significantly increased the P accumulation in soils ( Ju et al., 2007), similar to what happened in Europe 10 years ago (Barberis et al., 1995; De Smet et al., 1996; Djodjic et al., 2005). Blake et al. (2000) compared the P balance for three European long-term field experiments and reported that the soil type significantly influenced the P dynamics at each location. As a consequence of the surplus P balances, soil P levels have increased over the past half century in many countries. The recovery of P fertilizer applied to soil is low. Johnston and Poulton (1992) showed that only about 13% of the accumulated P reserve from past

136

Sheng-Xiu Li et al.

applications of fertilizers and manures was present as Olsen-P at the Rothamsted experimental farm, and other relative studies have shown that each 100 kg P ha 1 P surplus increased Olsen-P by only 2–6 mg P kg 1 (Aulakh et al., 2007; Shepherd and Withers, 1999). The surplus amount is surprisingly high. In Ireland, for example, the increase has been fourfold (Tunney et al., 2003) and in Finland two- to threefold (Saarela et al., 2004; Uusitalo et al., 2007). The surpluses of P in soil have two consequences. The first is that crops have variable responses to P fertilizers. For example, a number of short- and long-term experiments have been conducted in Denmark (Rubæk and Sibbesen, 2000), Norway (Riley, 2007), and Canada (McKenzie et al., 2001, 2004, 2005). Many of these trials have shown highly variable yield responses to P fertilization. In Denmark, the effect of P fertilization was negligible (0–10%), and in most cases it was statistically nonsignificant (Rubæk and Sibbesen, 2000). In Canada, the average yield increase was 7%, with values ranging from 12% to þ33%, but the effect was nonsignificant for most trials (McKenzie et al., 2001, 2005). In China, with the purpose of achieving high yields, farmers have used higher than the recommended levels of fertilizers in many areas. The total P fertilizer (as P2O5) usage has increased 2.1 times from 1990 (6.0 Mt) to 2005 (12.7 Mt) (CAY, 1991, 2006). Yet crop yield only increased by about 7% during the same period (CAY, 1991, 2006). In long-term experiments conducted in Norway for over 20 years on a loam soil, yields increased by 10–45% depending on the crop (Riley, 2007). Environmental pollution by P is the other consequence. The P contained in topsoil is a key factor controlling P transport from agricultural fields into the environment (Heckrath et al., 1995; Turtola and Yli-Halla, 1998). The phosphorus saturation in soils becomes more and more significant after long-term fertilizer or manure application (Beauchemin and Simard, 1999; Brett and Mallarino, 2006). From the agronomic point of view, the amount of P lost annually from soils is generally low, but from the perspective of water quality, very small annual inputs and changes in P concentrations may change the trophic status of a water body (Hart et al., 2004). Eutrophication contributes to a number of water quality problems, including drastic hangers in bottom vegetation, excessive growth of phytoplankton and algae, reduction of water clarity, changes in fish populations, and, in severe cases, oxygen deficiency in bottom waters leading to the death of benthic animals and fish. The Baltic Sea, as one of the world’s largest bodies of brackish water, is ecologically unique. Since the nineteenth century, its state has changed from oligotrophic to eutrophic (Valkama et al., 2009). Now, protection of the environment from surface water eutrophication is an urgent need for rational fertilization of P, and the other need at the same time is to improve agricultural practices so as to reduce diffuse source pollution and adverse environmental effects as well as to bring about

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

137

more efficient nutrient utilization (Valkama et al., 2009). Although both N and P contribute to the eutrophication, many scientists consider that phosphorus control is critical to mitigate eutrophication (Daniel et al., 1998; Nair et al., 2004; Robert and Hans, 2008; Stephen, 2008). At present, the developing and developed countries have markedly different problems concerning P. The former still have low P concentrations in their P-deficient agricultural soils (IFA, 2005), and continue to increase the use of P fertilizers for improving food supplies for a growing population and shifts in diet. An increasing proportion of calories from meat, eggs, and milk requires more P fertilizer per unit of food energy. By comparison, developed countries are coping with the consequences of excessive P use and environmental pollution, and are restricting the consumption of P fertilizer due to saturated agricultural soils, stable population, food consumption patterns, and environmental pollution (Weikard and Seyhan, 2009) Application of P fertilizer produced from RP has other disadvantages. Phosphate anions can be immobilized through precipitation with cations such as Ca2þ, Mg2þ, Fe3þ, and Al3þ, and therefore most of the soluble P in fertilizers applied to soils is rapidly converted into less soluble forms or sparingly soluble compounds (Batten, 1992; Davies et al., 2002; Hinsinger, 2001) that cannot be easily utilized by plants, and, with time, such compounds become unavailable to plants (Lin et al., 1997; Wang et al., 1997; Zhang et al., 1994). For example, in calcareous soils, the efficiency of P from applied fertilizers is inhibited due to the fast reaction of fixation between the applied P and the high calcium contents in the soil (Hedley and McLaughlin, 2005; Sharpley, 2000). That is, P in the fertilizer interacts with Ca, forming calcium phosphates that have low solubility and thus reduce its availability to crops (Afif et al., 1993; Hedley and McLaughlin, 2005; McBeath et al., 2006; Sa´nchez, 2007; Wang et al., 2005). The strategy of ameliorating the low P fertility production constraint in acidic soils with corrective applications of P is limited economically and environmentally due to the high amounts of P fertilizer required and the high P-fixing capacity of these soils (Hinsinger, 2001). As a result, the recovery of applied P is notoriously low. It has been estimated that in China, the recovery of phosphate fertilizer in the year applied is only 15%, and it is still less than 25% even when the residual effect is included (Li, 1999). Since the input amount greatly exceeds that taken up by plants, and a large amount of P accumulates in soil, studying its changes in soil and adopting adequate techniques to raise its use efficiency have become of great concern. In recent years, numerous agricultural scientists, including those specialized in soil science and plant nutrition, have concentrated on seeking new ways for reducing P fixation in soil and improving the recovery of P fertilizer in addition to transforming P that cannot be used by plants to become available for direct use by plants.

138

Sheng-Xiu Li et al.

Despite its nonrenewability, P has some important features that distinguish it from other nonrenewable resources. It differs from most metals because it is nonsubstitutable in its main use and it differs from fossil fuels because it is not used up by consumption but can be recovered to some extent (Weikard and Seyhan, 2009). Given the nonsubstitutability and the nonrenewability of RP, efficient P management is of strategic importance, and the development of technologies for fully using P fertilizer and the recycling of P is an inevitable consequence of agricultural production (Sharpley, 1985; Weikard and Seyhan, 2009). Careful long-term stock and flow management could slow down resource depletion (Seyhan, 2006). The prevalence of low available P in soil, coupled with transformation of applied P to unavailable P for plant uptake, has inspired interest in developing alternative technologies aimed at increasing P utilization efficiency (Buckman and Brady, 1990). Two trends become more and more important for improvement of P nutrition to plants. The first one is to use the soil P that is difficult to solubilize, and the other is to directly use RP. Soil contains a large amount of P, but most of it cannot be used by plants. Liberation of the sparingly soluble P in soil for plant use has been attempted for a long time and is still the focus of many scientists. Cultivation of plants is the core of agriculture and many measures are being used for promoting plant growth. Considering the properties of agricultural production, there are two ways, and only two ways, that can be used to manage P nutrition. One is the improvement of conditions for vigorous plant growth; the other is to utilize plants that have a high ability to use either the sparingly soluble P in soils or P added as rock phosphate P, or have strong responses to P fertilizers. Improvement of plant growth conditions is the most common method widely used in agriculture for cultivating vigorously growing plants. These plants can effectively use soil nutrients, including soil P, rock phosphate P, and fertilizer P applied, and thereby can make the P resource fully play its role. These include improvement in managing water, nutrients, air, and heat in soil, as well as the use of P fertilization practices such as band placement, combination of P fertilizer with N or K fertilizer, deep application, application of P fertilizer to soil deficient in P supply, and others (Kababo and Gilkes. 1988; Stecker et al., 2001; Steen, 1998). Application of P fertilizer in combination with organic materials is believed to improve the recovery of phosphate fertilizers (Tisdale et al., 1985). Utilization of specific plant species under nutrient-stressed conditions is the cheapest way for improvement of plant P nutrition as well as for full utilization of P fertilizers, and this has received considerable attention. Although a viewpoint exists that all crops have similar abilities for absorbing P from water or acid-soluble phosphate fertilizers (Li, 1966a,b), agricultural practices and scientific research findings have shown that different crops

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

139

respond differently to P fertilizers. It has been found that the P-efficient crops have two major mechanisms for improving their growth under P-deficient condition: increasing P acquisition (root morphology, root exudation, and P uptake mechanisms), and enhancing P utilization (internal mechanisms associated with conservable use of absorbed P at the cellular level) (Bates and Lynch, 2001; Raghothama, 1999; Vance et al., 2003). Of these, the P acquisition ability has received great notes. It is well recognized that plant species differ in their ability for absorption, transfer, utilization, accumulation, and redistribution of mineral elements (Clark, 1983; Li et al. (2003c) and so do the genotype varieties within the same species (Clark and Duncan, 1991; Ramirez and Lizaso, 2006; Yan et al., 1992a,b). Evidences indicate that plants can solubilize P from RP (Bolan et al., 1990; Hinsinger, 1998; Liu et al., 2004), and some species such as buckwheat [Fagopyrum tataricum (L.) Gaertn.], oil seed crops, and legumes are particularly efficient in using P from RP (Hinsinger and Gilkes, 1995; Zoysa et al., 1998a). This is also true for crop genotype varieties that differ in their ability to use P from RP in their rhizosphere (Flach et al., 1987). Shi et al. (2004) reported significant differences among rapeseed varieties for the depletion of P and K in the rhizosphere and a wider depletion range in the rhizosphere of the varieties tolerant to low nutrient supply. To adapt to the mineral stress environment and to raise P fertilizer or RP efficiency, selection, adoption, and cultivation of crop species that can grow in soils with low levels of available P and that can efficiently utilize soil P compounds of low solubility (sparingly soluble P), rock phosphate P, or the P fertilizer applied could be a promising approach for effective improvement of crop growth and sustainable crop production (Horst et al., 2001) as well as for saving the phosphate mine resources. Release of P-efficient genotypes in both high- and low-input farming systems would reduce the production costs associated with P fertilizer applications, minimize environmental pollution, and contribute to the maintenance of P resources globally (Cakmak, 2002; Vance et al., 2003). Of the crops that have different responses to P fertilizers, leguminous and nonleguminous crops have attracted great attention. On fields where application of P fertilizers has yielded good result, leguminous crops or leguminous green-manure crops generally respond much better than nonleguminous crops, particularly cereal crops. Application of P fertilizer can result in a relatively high and stable effect on yield increase of leguminous crops (Chen et al., 1963a; Guo, 1981; He, 2004; Huang et al., 1964; Jiang, 1997; Li, 2005; Liu, 1964, 1980; Liu et al., 1963, 1981; Lu et al., 1998; Shi et al., 1962; Wang, 2004; Ye, 2002; Zang et al., 1965; Zhao and Tong, 1964), while a less pronounced effect is seen on nonleguminous crops. Due to this difference, leguminous crops are regarded as “P-like crops” (Peng, 1990; Peng and Pei, 1980) or “crops needing more-P” (Chen et al., 1963b; Extension Station of Agricultural Technology in Bijie District, Guizhou

140

Sheng-Xiu Li et al.

Province, 1964; Guo, 1981). The good responses of leguminous crops to P fertilizer have been confirmed in our experiments. In the last 40 years and earlier, we have conducted over 200 field experiments for studying the effect of application of P fertilizers on wheat and maize and have chosen several fields to compare their effect with pea and soybean. Results show that in most of cases, pea responds much better than wheat, and soybean than maize. Based on these facts, Sun and Wang (1980) proposed that P fertilizers should be applied to leguminous crops as the first priority. Some scientists even expressed the view that “if P fertilizer was not applied to leguminous crops but to nonleguminous crops, the effect would not be obvious, and it would be impossible to have good results when P fertilizer was not applied to leguminous but to other crops or to fallow field” (Chen et al., 1963a,b). What causes the different responses to P fertilizers between leguminous and nonleguminous crops? Many hypotheses have been put forward (Chilikef, 1962; Li, 1959, 1964, 1966a,b, 1986; Li and Hu, 1956; McLean et al., 1956; Wallace, 1962). However, most of the viewpoints were not substantiated by past research results, and some conflict in the results of recent research have also appeared. In this review, we have used some leguminous and nonleguminous crops to make comparisons in order to reveal why leguminous crops generally have better responses to P fertilizers. For seeking the mechanisms of the different responses to P fertilizers existing between leguminous and nonleguminous crops, we used acidulated P fertilizers, particularly calcium superphosphate, and conducted investigations from different aspects, including P uptake amount, responses to RP, root morphology, root exudates (protons, low weight molecular organic acids, amino acids) and root cation exchange capacity (CEC), and N supplying capacity, and attempted to find out if there are any differences in the ability of the two kinds of crops for utilizing the sparingly soluble P in soil or RP. In most cases, we used our results of more than 40 years as evidence to explain some of these phenomena. For most studies, soybean and pea were used as leguminous crops and wheat and maize as nonleguminous crops. This is because these leguminous and nonleguminous crops are grown over a wide range of climatic and edaphic conditions, and widely cultivated in China (Tong et al., 2003), especially in the dry areas. These crops play a major role in agricultural production and food security. For instance, as the most important food crop, wheat is planted on more land than any other crop, and in yield occupies the second position in crops worldwide (FAO, 2008; Li, 2008). Known as “feed king, high-yielding king, and raw material-processing king,” maize is extremely important for both animal feed and human food (Hamdia et al., 2004), and its production is now higher than wheat worldwide according to FAOSTAT (FAO, 2008). In China, maize accounted for 26.9% of the total grain yield in 2006, and is now higher than wheat but only lower than rice, and China

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

141

ranks second in maize production in the world according to a recent report (National Statistics Bureau, P.R. China, 2006a,b). Soybean is an important crop as a source of protein, oil, and micronutrients for humans and animals (Yan et al., 2009). It is grown in a wide range of climatic and edaphic conditions. Substantial differences have been found among soybean genotypes grown in P-deficient soils and their abilities to respond to P fertilizers (Ding and Li, 1997, 1998; Tong and Yan, 1999). Substantial differences among genotypes were also found when they were grown in an acid red soil (Tong and Yan, 1999; Tong et al., 2000; Zhao et al., 2004). Therefore, use of soybean as an experimental crop is significant. Pea is a traditional crop widely planted in China and is still used in rotation systems for improving soil fertility. Pea and wheat grow during the same time of year (both can be sown at autumn and grown till summer next year), compared to soybean and maize that also grow during the same time period (sown at summer or spring and grown till autumn). For this reason, the crops can be compared under the same conditions without influence by other factors.

2. The Difference of P Uptake Amounts of Leguminous and Nonleguminous Crops The different responses of crops to P fertilizers have been attributed to the P uptake amount by some agricultural researchers. Some scientists hold the view that leguminous crops take up much more P than nonleguminous crops and this has been considered the reason for leguminous crops to have better responses to P fertilizers than other crops (Chen et al., 1963b; Extension Station of Agricultural Technology in Bijie District, Guizhou Province, 1964; Guo, 1981; He, 2004). Ding et al. (2006) conducted a pot experiment to evaluate the abilities of 20 genotypes of soybean for absorbing P from soil with and without P fertilizers. They concluded that P uptake amount was one of the most important indices for evaluating the ability of a cultivar to absorb P under deficient conditions. The P uptake amount is the total dry matter times P concentration at a specific supply of P and is defined as the amount of P absorbed by plants to produce a particular weight. If the P concentration in dry matter is consistent or without great changes, the dry matter amount will be the decisive factor to determine the P uptake amount, and, therefore, a high dry matter production will yield a high amount of P uptake. Due to such relations between the P uptake amount and crop yields, it is important to use the amount of P uptake in evaluating crop response to P fertilizers or the ability of crops to take up P from soil. However, for the evaluation of the amount of P uptake, many factors have to be considered. It is well known that a number of factors affect the

142

Sheng-Xiu Li et al.

amount of P uptake by crops, including environmental differences and internal characteristics of plants themselves. Even for the same crop, due to differences in location and growing season, the amounts of P uptake are greatly different. As a major factor, the length of plant-growing period determines to a large extent the crop productivity potential and thus the P uptake amount. For this reason, a reliable conclusion can be obtained only if both leguminous and nonleguminous crops grow on the same field during the same season with almost the same duration of growing time so that the influence of soil fertility and climatic factors can be avoided. Based on these considerations, we compared some experimental results carried out under same conditions by others, in addition to also conducting similar experiments ourselves. Zhu et al. (2009) conducted a long-term experiment in the lower reaches of the Liao River Plain, northeast China, to study nutrient distribution in crop plants in a maize–maize–soybean rotation sequence. The results showed that, on average, the concentration or content of P was 5.35 g kg 1 in soybean seeds and 0.63 g kg 1 in the straw. In contrast, the concentration of P was 2.20 g kg 1 in maize seeds and 0.48 g kg 1 in the residue. The grain harvest index (the ratio of seed weight to the total weight of the above-ground biomass) was 0.51 for soybean and 0.44 for maize, while the P harvest index (the P uptake amount in seeds divided by the total uptake P in the aboveground biomass) was 0.90 for soybean and 0.82 for maize. These data show that most of the P was concentrated in the seeds. Li (2004a) surveyed a large amount of data of P uptake amounts for various varieties of maize in China and concluded that the P amount removed by 100 kg maize varied from 0.17 to 0.52 kg with an average of 0.37 kg. Compared to the average, the P uptake amount in the rotation sequence was much lower. This might have been caused by the soil that (contained 10.1 mg kg 1 Olsen-P) was regarded as one with medium level of deficiency in P supply (Li et al., 1979a). However, even for the maize taking up much less P than usual, the total P uptake from a unit area was similar to, and in some cases higher than, soybean since the yield of soybean was only about one-third of maize. We conducted several field experiments to compare the effect of P fertilizers applied to various crops. Data were obtained to determine the relationship between P uptake amounts and crop responses to P fertilizers. Results of the experiments were strikingly similar and we will use three of the field trials (Experiments 1, 2, and 3) to demonstrate the results. Experiment 1 was carried out in 1962 at Experimental Station 1 of the Northwestern College of Agriculture (now the Northwest Science and Technology University of Agriculture and Forestry). The soil used was developed on the loess parent material and the land had uniformly been cropped with grass for 2 years before the field trial started to consume soil nutrients so that the soil would be in a uniform low-fertility condition for the experiment. The cultivated soil layer (0–20 cm) contained 10.3 g kg 1

143

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

Table 1

Some summer-sown crops responses to phosphate fertilizer Seed yield (kg ha 1)

Crop

Maize Millet Buckwheat Soybean White sweet clover (forage)

P uptake (kg ha 1)

Increase by Increase by Without P With P P addition Without P With P P addition

2550 1748 1607 1178 4250

2795 1953 1782 1448 5090

245 205 175 270 840

8.34 8.34 7.35 5.85 8.60

9.56 9.30 8.12 7.16 10.09

1.22 0.96 0.77 1.31 1.49

Modified from Li et al. (1995).

organic matter, 0.78 g kg 1 total N, and 0.23 g kg 1 total P. The plantavailable N was 27 mg kg 1 nitrate-N, determined by the nitrification power method using a 5:1 water/soil ratio (w/w) and incubated at 25 C for 1 week. The available P (Olsen-P) was 5.8 mg kg 1. Based on these values, the soil was unfertile and deficient in both available N and P. Five summersown crops, i.e., maize, (foxtail) millet [Setaria italica (L.) Beauv.] or (Panucum italicum L or P. miliaceum L.), buckwheat, soybean, and white sweet clover (Melilotus albus Desr.), were row-planted at locally used population rates, and four replications were implemented. Each crop had two treatments: control (without application of fertilizer) and P fertilization (application of 30 kg P ha 1 using superphosphate as P source). As shown in Table 1, on such an unfertile soil with no application of N fertilizer, P addition significantly increased crop yield for both leguminous and nonleguminous crops but the increase was not as high as expected. Compared to nonleguminous crops, the increase in leguminous crop yield was higher, and so was the increased amounts of P uptake resulting from P fertilization. However, for the total P uptake amount, it is another story: without application of P fertilizers, the total P uptake by white sweet clover was not significantly higher than that of maize, millet, and buckwheat, and that of soybean was much lower than the three nonleguminous crops. With added P, however, the total P uptake amount by white sweet clover was higher than for the other crops and soybean still had the lowest amount of P uptake. The results did not confirm the hypothesis that P uptake amount was the reason for crops responding to P fertilizers. Experiment 2 was conducted in 1973 and 1974 on three manural soils of village fields around the College, and the soils were also developed on the loess parent materials with basic characteristics such as organic matter, total P, and total N (data not shown) similar to Experiment 1, but the amounts of available N and P for the three soils being different (Table 2). Two autumn-

Table 2 Responses of pea and wheat to P fertilizer and P uptake from soil (kg ha 1)

Soil no.

Pea 75001 75002 75008 Wheat 75001 75002 75008

Available Na (mg kg 1)

Olsen-P (mg ka 1)

35 28 41 35 28 41

Seed yield (kg ha 1)

P uptake (kg ha 1)

Without P

With P

Increase by P

Without P

With P

Increase by P

2.2 10.9 12.2

653 1370 1671

1463 1988 1893

810 618 222

2.55 5.40 6.60

5.70 8.10 7.50

3.15 2.70 0.90

2.2 10.9 12.2

1176 2907 3200

1884 3282 3626

708 375 426

3.45 8.55 9.30

5.55 9.45 10.50

2.10 0.90 1.20

Modified from Li et al. (1995). a Determined by nitrification power method (Begerburcki, 1961).

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

145

sown crops, wheat and pea, were used to investigate whether the results obtained in summer-sown crops would be in agreement with autumn-sown crops. The same treatments, control and P fertilization, were adopted. Again, application of P fertilizers significantly increased crop yield for both pea and wheat for the three soils, as well as the amounts of P uptake (Table 2). Pea responded to the P fertilizer much better than wheat and absorbed more P from the fertilizer in two of the three soils. Soil 75008 contained higher amounts of available N and P than the other two soils. According to the indices used for guiding P application based on Olsen-P determinations (Li and Xiao, 1992), values between 8 and 16 mg kg 1 are still in the range for recommending P fertilization when N supply is adequate. The high soilavailable N supply increased the added P effect for wheat but also promoted stem and leaf growth of pea but inhibited seed formation. This may be the reason why pea and wheat had different responses to the P fertilizer in this soil (see later sections for more detailed discussions). Although the P concentration in pea was considerably higher than in wheat, its yield was much lower than that of wheat. As a result, the total P uptake amount was lower for pea than wheat for both the control and added P treatments. The two experiments discussed above were carried out only with control and P fertilizer treatments without application of N fertilizers. Since leguminous crops have the ability to fix N from the atmosphere, whereas cereal crops have no such ability, this design will lead to different N levels between legumes and cereals. In almost all soils, N deficiency limits crop production as well as crop responses to P fertilizers. Due to soil N limitation, crops often have no response to P fertilizers even in a P-deficient soil because N deficiency is the major constraint. For this reason, we hypothesized that the good response of leguminous crops might be related to their N fixation ability. According to our experimental results, pea and hairy vetch (Vicia villosa Roth) fixed about 90 kg N ha 1 during their growing period (Li and Li, 1992). For leguminous and cereal crops to have the same N supplying level during their growing period, Experiment 3 was conducted from 1985 to 1986 in a manural soil at Experimental Station 1 in the Northwestern Agricultural University (Northwestern College of Agriculture until name was changed in 1984). Crops included wheat, pea, and hairy vetch. For wheat, 112.5 kg N ha 1 was applied, while for hairy vetch and pea 22.5 kg N ha 1 was applied. In addition, a control without the application of N fertilizer was arranged for pea. Application of such a small amount of N to the leguminous crops was based on these considerations: (1) to supplement N to leguminous crops so that both wheat and the two leguminous crops would have the same amount of N supply (however, the results showed that it was not necessary to add N to leguminous crops); and (2) to stimulate early growth of leguminous crops. In our areas, nodulation for leguminous crops sown in autumn begins around 2 weeks after planting, while the biological N fixation takes place

146

Sheng-Xiu Li et al.

much later, generally in March the following year since low temperature in winter is a limiting constraint for crop growth and also for nodulation and N fixation. During this period, the leguminous crop often grew poorly due to deficiency of N supply. Application of a small amount of N at planting called “starter N” is beneficial to improve early growth and yield of leguminous crops in most cases, as shown in other areas (Afza et al., 1987; Osborne and Riedell, 2006; Starling et al., 1998), whereas a high amount of starter N fertilization has not only had a detrimental effect on infection, nodulation, and N fixation capacity of rhizobium bacteria (Beard and Hoover, 1971; Koutroubas et al., 1998) but was also found harmful for the biological N fixation process. A small amount of N is enough to promote leguminous crop growth. Afza et al. (1987) reported that lower N rates (<40 kg N ha 1) did not inhibit N fixation. Based on each N rate (125.5 kg N ha 1 for wheat, 22.5 kg N ha 1 for hairy vetch, and 0 and 22.5 N ha 1 for pea), we conducted a study with six rates of added P (0, 19.5, 39, 58.5, 78, and 97.5 kg P ha 1) to the soil with three replications. There were 24 treatments as shown in Table 3. Urea and calcium superphosphate were used as the N and P sources. Both N and P fertilizers were applied as basal dressing before sowing the crop. Wheat and pea were harvested at maturity, and hairy vetch at the fully flowering period. Results (Table 3) showed that, with such rates of N fertilizer applied, the P effect on wheat yield increase was much higher than for pea and hairy vetch, and wheat absorbed the highest amount of P, followed by hairy vetch, while pea took the lowest amount. Using the average of six P levels to make comparison, wheat took up 10.4 kg P ha 1, which is 38% higher than for hairy vetch and 57% higher than for pea. Calculated by the difference method, hairy vetch absorbed the largest amount of P from the P fertilizer, whereas wheat and pea took up almost the same amount. From the results above, it can be concluded that the P uptake amount by leguminous crops was not higher than by cereal crops: without supply of N fertilizer, the P uptake amount was almost equal; while with the application of N fertilizer, wheat took up more P than leguminous crops. Wheat took up more P from the soil than either pea or hairy vetch. This shows that wheat has a higher capacity to use soil P than pea and vetch. As a result, the magnitude of P uptake amount is not the major cause for crop responses to P fertilizer.

3. Leguminous and Nonleguminous Crop Responses to Powdered Rock Phosphates The ability of crops to absorb sparingly soluble P from soil can be evaluated by their ability to take up P from ground RPs. It is reported that, in calcareous soil, the P availability for crop uptake in P-containing minerals

147

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

Table 3 Responses of wheat, pea and hairy vetch to P rate and P uptake amount (field experiment conducted from 1985 to 1986) Yield (kg ha 1)

P rate (kg ha 1) Shoota Seed

Yield increase (%) Shoot

Seed

Plant P (%)

Wheat (with application of 112.5 kg N ha 1) 0 5103 2040 0.140 19.5 7118 3254 40 60 0.136 39.0 7757 3566 52 75 0.118 58.5 8375 3791 64 86 0.125 78.0 9441 4229 85 107 0.130 97.5 9711 4491 93 120 0.136 1 Pea (with addition of 22.5 kg N ha ) 0 2418 1109 0.147 19.5 2985 1385 23 25 0.166 39.0 3207 1464 33 32 0.178 58.5 4089 1709 69 54 0.177 78.0 4947 2175 105 96 0.179 97.5 4997 2067 107 87 0.182 Pea (without addition of N) 0 2618 1224 0.147 19.5 3099 1545 18 26 0.145 39.0 3600 1626 38 33 0.148 58.5 4886 2301 87 88 0.169 78.0 5330 2558 104 109 0.168 97.5 5223 2378 100 94 0.184 1 Hairy vetch (with addition of 22.5 kg N ha ) 0 1611 0.179 19.5 2142 33 0.191 39.0 2856 77 0.224 58.5 3729 132 0.273 78.0 3494 117 0.298 97.5 3954 145 0.275

Plant P uptake N uptake N (%) (kg ha 1) (kg ha 1)

1.44 1.26 1.25 1.20 1.22 1.15

7.2 9.8 9.2 10.5 12.3 13.2

73.5 89.7 96.9 100.5 115.2 111.8

2.28 2.08 2.30 2.18 2.28 2.17

3.6 5.0 5.7 7.2 8.9 9.2

55.2 62.1 73.8 89.1 112.8 108.5

2.21 2.23 2.16 2.27 2.17 2.37

3.9 4.5 5.4 8.3 9.0 9.6

57.9 69.2 77.7 110.9 115.7 123.8

2.91 2.89 2.95 3.10 3.11 2.98

2.9 4.1 6.5 10.5 10.4 10.8

47.0 63.0 84.3 115.7 108.6 117.9

Modified from Li and Li (1992). a Shoot was all aboveground biomass including the grain.

is roughly equal to that in RPs (Li, 1966a,b). For this reason and also for finding a way to directly apply RPs, some agricultural scientists have used RPs to test the crops’ ability to utilize sparingly soluble soil P. As already mentioned, the supply of RPs is short and controlled by only a few countries. Also, even in the countries with abundant RP resources, highgrade RP is still limited. This is particularly true in China. China is rich in RP resources, but high-grade RP occupies only a small portion, whereas

148

Sheng-Xiu Li et al.

the low- and middle-grade RPs exist in large quantities, about 80% of the total. Segregation of high-grade RPs from the low- and middle-grade rocks for fertilizer production consumes energy and labor, and in some cases it may be difficult to carry out such a process. For RPs unsuitable for processing into P fertilizers, direct application has been identified as a promising alternative (Boiland, 1993; Khasawneh and Doll, 1984; Li et al., 1992a; Wei et al., 2001). This saves energy and allows the P resources to be fully utilized for agricultural production. The effect of direct application of RPs is determined by many factors. The first is the RP characteristics. As raw materials for phosphate fertilizer industry, RPs generally contain no or little water-soluble P (Khasawneh and Doll, 1978), and therefore some weak organic acids such as formic acid, lactic acid, tartaric acid, 2% citric acid, and neutral citrate ammonium (Li, 1966a,b) are used as chemical-extracting agents rather than water to determine the soluble P from RPs as indices of their availability. Of these, 2% citric acid and 2% neutral citrate ammonium are particularly noted. The soluble P extracted by such weak acids is called plant-available P, different from the water-soluble P. It has been found that the availability of RPs for plant uptake is well correlated with the amount of soluble P extracted by 2% citric acid. However, the characteristics and amounts of the accompanying minerals with RPs influence the determination of the citric-acid-soluble P (Lei and Wang, 2005; Meng, 1989; Shi et al., 1982) and therefore the availability of RPs ( Jiang, 1988; Wang et al., 2006; Zhu et al., 1991). Calcium carbonate is such an accompanying mineral that can consume and react with citric acid and make the determination results lower. In contrast, 2% citrate ammonium is not affected by ammonium carbonate. For solving this problem, both 2% citric acid and 2% citrate ammonium are used together to determine the plant-available P. When the citrate ammonium P is less than 5–6% and the citric-acid-soluble P less than 10% of the total P, the availability of RPs to plants is generally low; while when it is larger than 10% for the former and 20% for the latter, the RPs have at least medium availability (Li, 1966a,b; Li et al., 1992a; Three Institutes, 1977). The physical and chemical properties of RPs determine the RP availability to plants to a large extent. Seen from the chemical components, RPs are mainly composed of fluorapatite. Due to substitution, some anions such as CO32, OH, Cl, as well as cations such as Naþ, are removed from the RP crystalline structure. Influenced by the accompanying minerals, some other elements are mixed in the RPs such as Si, Fe, Al, and S. Therefore, the composition of RPs is complicated. The substitution of CO32 for PO33 in the RP crystalline structure results in an increase in its availability by making the crystalline structure loose. However, owing to the existence of CO3 2 either in the crystalline structure by substitution or in the accompanying minerals that mix with the RPs or occluded on the surface of the RPs, it is difficult to separate the latter by physical or chemical

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

149

methods. Thus, in the past, scientists have failed to use the ratio of CO3 2 (sometimes CO2 was used instead of CO3 2) to P2O5 to indicate the degree of RP availability. According to their shape and crystal degree, together with the available P as main criterion, RPs can be divided into three types. Type 1 is the low plant-available P type. In general, it is well crystallized, with apatite (such as fluorapatite, hydroxylapatite, and chlorapatite) as a dominant component; its structure is dense and its specific surface area is around 1 m2 g 1. In most cases, it contains a very small amount of the citricacid-soluble P (<3%). Using 2% citrate ammonium for extraction, Li et al. (1992a) identified the Shimen RP and the Jinpin RP from China as belonging to this type. The Shimen RP, produced in Hunan Province, contains 32.5% total P but only 1.7% citrate-ammonium-extracted P, whereas the Jinpin RP from Jiangsu Province contains 36% total P with only 1.2% citrate-ammonium-extracted P. As a result, the availability is low and it is difficult to be directly used by crops. In this type, the mineral composition and chemical reaction of RPs vary greatly ( Jiang et al., 1988; Li, 1966a,b; Three Institutes, (1977). Almost all elements in apatite can be partly substituted by isomorphous substitution under geological environments, and the degree of PO43 substituted by CO32 in apatite is the key in determining the availability of carbonate apatite (Shi et al., 1982). In contrast, type 2 is strong for supplying P to crops and is suitable for direct application. It is not crystallized but is apedal or colloidal in shape. The citrate-ammonium-extracted P is 15–20% of the total, much higher than type 1, and the sparingly soluble P is easily changed into the plantavailable form. The specific surface area varies from 10 to 20 m2 g 1. RPs produced in Morocco as well as Kaiyang and Kunyang in China are typical examples of such type. The Morocco RP contains 32% total P and 11% citrate-ammonium-extracted P, with the ratio of the latter to the total being as high as 34%; thus it can supply a large amount of P that can be used by crops. The Kaiyang RP produced in Guizhou Province contains 36% total P and 8.4% citrate-ammonium-extracted P, with a ratio of 23.4% of the latter to the total. The Kunyang RP produced in Yunnan Province contains 38% total P and 8% citrate-ammonium-extracted P, with 21% as the ratio of the latter to the total. Also, the Yulin RP produced in Guangxi Province, Fengdai RP in Anhui Province, Emei RP in Sichuan Province, and Helanshan RP in Ningxia also belong to this type (Li et al., 1992a). With characteristics between type 1 and type 2, type 3 is a medium type without a clear crystalline structure. Its specific surface area varies from 5 to 10 m2 g 1, and the 2% citrate-ammonium-extracted P is around 5% or less. The Xiangyang, Lanxi, and Shifang RPs from China fall under this type. The Xiangyang RP in Hubei Province contains 32.4% total P and 5.2% citrate-ammonium-extracted P, with 16% as the ratio of the latter to the former. The Lanxi RP in Zhejiang Province contains 24% total P and 3.3% citrate-ammonium-extracted P, with 14% as the ratio of the latter to the

150

Sheng-Xiu Li et al.

total. The Shifang RP in Sichuan Province is also regarded as this type (Three Institutes, 1977). Although the availability of the RPs varies greatly with the type, location, impurity, forming conditions, grade, and the extracted P either by citric acid or by citrate ammonium, the grade (total P content) of RP is the essential quality of the raw material for manufacturing P fertilizers. However, the high content of total P does not necessarily mean high availability to plants. In some cases, the high grade of RP contains high citric-acidextracted P, but in some other cases it is not so because of the type as illustrated above (Three Institutes, 1977). Soil solution pH and P-supplying capacity considerably affect the availability P in RPs. When the soil pH is lower than 5.5, RPs may be significantly affected, and direct application of RPs may produce good results (Li et al., 1992a). Baenes and Kamprath (1975) found that when the soil pH was above 5.2, the effect of RP was greatly reduced. Stamford et al. (2007) pointed out that direct application of RPs as a source of P was only feasible under acidic soil conditions. In the past, most of the good results for the direct use of RPs were obtained in south China with acidic soils (Li, 1964, 1966a,b; Li and Hu, 1956; Li et al., 1952, 1953) and some recent reports have further confirmed this conclusion (Meng et al., 2007; Wu et al., 2002). The effect of RPs in other soils has been doubtful or uncertain. Akande et al. (1998) stated that a direct application of RPs for short-duration crops has not shown good promise in near-neutral soils. Sun (2001) points out that the effect of RPs is also determined by P and Ca concentrations; any changes of these factors will lead to changes of the RP effect on crops. In northern China, the calcareous soil having high pH and Ca limits the solubility of RP and thus leads to unstable or no effect by its direct use (Li et al., 1992a; Wei et al., 2001). The extent of direct use of RPs has increased at least 3–4 times during the twentieth century, but it is still not widely used on a large scale, especially in northern China. A general conclusion is that, in calcareous soils, apatite is the major form of soil P, and the RP applied to soil has little opportunity to be changed into the available form. For this reason, it is considered unsuitable to apply RP to calcareous soil even if the soil is deficient in available P except for some special crops (Sun, 2001). Among the factors influencing RP effect, plant species have attracted great attention. For quite a long time, scientists have tried to seek crops having a high ability to utilize the RPs. Since 1952, numerous papers on crop responses to RPs have been published in China (Li, 1959, 1964, 1966a, b, 1986; Li and Hu, 1956; Li et al., 1952, 1953; Rock Phosphate Research Group, Institute of Soil Science, Chinese Academy of Sciences, 1966) and other countries (Agilars and van Diest, 1981; Maloth and Prased, 1979; Murdock et al., 1960). As common conclusions, the conditions for effective application of RPs include acidic soil, high amount of P extracted by either citric acid or citrate ammonium in RPs, and crops having good responses to

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

151

RPs, in addition to how finely they are ground. Among the experiments, those carried out by Li et al. (1952, 1953), Li and Hu (1956), Li (1966a), and the Rock Phosphate Research Group, Institute of Soil Science, Chinese Academy of Sciences (1966), are typical. They conducted field and pot experiments from 1951 to 1964. The primary experiments were carried out in a red soil (pH 4.5) from 1951 to 1954 in Ganjiashan, Jiangxi Province, using Haizhou Jinpin RP containing 36–38% total P2O5 (no citric acid or citrate-ammonium-extracted P amount reported). The experiments included three or four treatments. The three treatments included no application of P fertilizer (control), application of 2250 kg ha 1 of Haizhou RP, and application of 600 kg ha 1 single superphosphate (SSP) (containing 20% water-soluble P2O5); for the four treatments, application of 600 kg ha 1 bone meal (containing 22–24% P2O5) was added. The soil was excessively acidic and extremely deficient in nutrients. Application of N, P, and limestone could at least double the yield, while treatments without added P fertilizer produced little or no yield of cereals. In such a soil, the effect of RP was much better than SSP, perhaps due to the calcium carbonate contained in the RP that tended to neutralize the acid. Thereafter, researchers continued to conduct such experiments in red soils with pH between 4.5 and 6.0 in Jinxian of Jiangxi Province, Zhanjiang and Guangzhou of Guangdong Province, and Jinhua of Zhejiang Province using Kunyang RP containing 36% total P2O5 and 6.5% P2O5 extracted by 2% citric acid. In each field experiment, 1500 kg ha 1 Kunyang RP was used for testing and compared to control and 500 kg ha 1 SSP (containing 20% water-soluble P2O5) treatments. In pot experiments, 6 kg soil was added to each pot with two treatments: 7.5 g Kunyang RP per pot and 5 g SSP per pot, respectively. Adequate N fertilizer (ammonium sulfate) was applied to all treatments as a base for both field and pot experiments. In the process of experimentation, 11 crops, i.e., wheat, rice (Oryza sativa L.), buckwheat, millet, peanuts (Arachis hypogaea L.), tomato (Lycopersicon esculentum Mill or Solanum lycopersicum L.), rapeseed, vetch (Vicia Linn.), Chinese vetch or milk vetch (Vicia chinensis Franch), radish, and rice bean [Vigna umbellate (Thumb.) Ohwi et Ohashi], were investigated under both field and pot conditions, whereas studies for other crops only used experimental pots. The summary of results for the crops’ responses to Kunyang RP is presented in Table 4. Since the experiments were continued for many years, and crop yields for different years were greatly different, relative effects in percentage of RP to SSP were used for comparison rather than absolute yields. From the results, the authors concluded that rapeseed and radish, the main green manure crops in red paddy acidic soil of middle China, possessed very strong abilities for absorbing P from RPs, and that application of RPs to these crops was a significant practice. In addition, almost all leguminous crops suitable to grow in acidic soils as major winter green manures or winter crops such as vetch, Chinese vetch, and pea, as well as those having an important position in rotation systems, such as soybean and

152

Sheng-Xiu Li et al.

Table 4 Relative effect of rock phosphate on different crops as compared to single superphosphate (SSP) Responses to rock phosphate (RP)

High effect Good effect

Representative crops

Seed rape, radish, buckwheat Vetch, pea Soybean, rice bean, Chinese vetch (milk vetch) Peanut, (striped) crotalaria, sesbania, (shrub) lespedeza Median effect Maize Potato, sweet potato Sesame No significant effect Millet Wheat, rye, oats Rice

Average equivalent to SSP (%)

80 70–80 70 60–70 50–60 50 40 20–30 15–30 25–50

The RP was produced in Kunyang (containing 16% total P (36% P2O5), 2.2% (6.5% P2O5) acid-soluble P, 33% SiO2, 2.2% MgO, 1.26% Fe2O3, and 1.63% Al2O3. It was the sedimentary rock phosphate, dark gray in color, and the ground particle was less than 0.15 mm) Yunnan Province, and rate applied was 1500 kg ha 1, while the rate of calcium superphosphate containing 20% of water-soluble P2O5 as control was 600 kg ha 1. Modified from Rock Phosphate Research Group, Institute of Soil Science, Chinese Academy of Sciences (1966).

peanuts, had strong capacities to use P from RPs, and that attention should be paid to applying RPs to these crops. The good results for rapeseed to utilize RPs have been confirmed recently by other researchers. Wu et al. (2002) conducted four rigorous experiments in acidic paddy soils (pH from 5.2 to 5.4, organic matter 9.5– 25 g kg 1, available P2O5 6.3–10.8 mg kg 1 determined by Bray-1 method and CEC 9.8–11.3 cmol kg 1) in Lanxi district, Zhejiang Province, to study the effect of Kunyang RP powder (containing 316 g kg 1 total P2O5, 63 g kg 1 citric-acid-soluble P2O5 and 495 g kg 1 CaO) from China and Tunisia Gafsa RP powder (containing 290 g kg 1 total P2O5, 85 g kg 1 citric-acid-soluble P2O5, and 481 g kg 1 CaO) on rice and rapeseed. The same rate of 45 or 50 kg P2O5 was used for the two RPs, and SSP as control. Sulfur 95 (sulfur–bentonite granular fertilizer containing 95% of S) as S source was added to one of the two RPs so that the same amount of S would be available as that in the SSP. Results show that at the same rate of total P2O5, the two powdered RPs had more significant effects on rapeseed than on rice. The rapeseed yield grown on paddy soil derived from quaternary red clay (available P2O5 6.3 mg kg 1) increased by 14–22 times and 3–4 times in alluvial deposit and red sandstone (available P2O5 10.8 mg kg 1).

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

153

In contrast, rice yield grown on the corresponding soils increased by only 1.4–3.8% and 3.5–7.3%, respectively. The effect of ground Gafsa RP was equal to 86–105% of SSP, while that of the ground Kunyang RP was equal to 47–65% of SSP. With the application of RPs, S fertilizer increased rapeseed yield by 8.4–19.5% and rice by 1.6%. Meng et al. (2007) made a further study in two paddy soils developed on the quaternary red clay with rapeseed cultivar Shuang 72 as the test crop. The field was located in Lanxi district, Zhejiang Province. The soil pH was 5.14 and 5.26, organic matter 27.3 and 26.5 g kg 1, and available P 6.3 and 5.6 mg kg 1 (determined by Bray-1 method). On the basis of application of 120 kg N ha 1, six treatments were adopted: without application of P and S fertilizer; application of 45 kg P2O5 ha 1 of Kunyang RP and Gafsa RP, respectively; application of 45 kg P2O5 ha 1 þ 40 kg elemental S for the two RPs, respectively, and application of 45 kg P2O5 ha 1 of SSP. Results showed that Gafsa RP increased rapeseed by 22–23 folds, Kunyang RP by 14 folds, and SSP by 21 folds compared to control. In addition, Kunyang RP, Gafsa RP, and SSP increased P uptake by 7, 16, and 14 folds, respectively. Although the addition of S had no significant effect on rapeseed yield as compared to the treatments without S, addition of S increased rapeseed P uptake by 11% from Kunyang RP and 69% from Gafsa RP. In a calcareous soil (pH 8.5), Sun (2001) also found significant effects of RPs on the increase of rapeseed yield. The soil used for the experiment was very poor, containing only 6.4 g kg 1 organic matter, 0.43 g kg 1 total N, 1.52 g kg 1 total P, and 2.5 mg kg 1 Olsen-P. The powdered RPs were produced from Jinpin, Caiyang, and Jingxiang in China and Morocco, together with SSP as the control. The rates were 1125 kg ha 1 for RPs and 225 kg ha 1 for SSP. Fertilizers were applied at the transplantation period of rapeseed by banding with 58 kg N ha 1 of ammonium bicarbonate. At the same time, demonstrations were carried out on three soils of different texture and pH (7–8.8) and used two RPs, from Morocco and Jinpin (from China), respectively. Results (Table 5) showed that the application of RPs greatly increased rapeseed yield even with a low grade of RP such as Jinpin RP, and the degree of yield increase by RPs was even significantly higher than that by SSP. The demonstrative results (Table 6) also showed the excellent effects of RPs on rapeseed yield in addition to soil textures that exerted an effect on RP: higher effect of the RP in slight clay soil while lower in the sand loam soil. From the results obtained on acidic and calcareous soils, we found some issues in the crop comparisons that needed further clarification. The first is that different crops were not planted in the field on the same soil with homogeneous fertility but on different soils at different locations even though on the same soil type. Soils respond differently even if similarities exist at the locations. Different soils have different characteristics and fertilities, which greatly affect crop growth and response to RPs, resulting in

154

Sheng-Xiu Li et al.

Table 5 Properties of powdered rock phosphates used for experiments and rapeseed responses to rock phosphates Properties of PRP (%)

Type of PRP

Total Water P2O5

Jinpin 1.75 Caiyang 14.30 Jingxiang 6.70 Morocco 8.11 SSP Control

34.2 29.2 34.5 36.2

2% citric acid extracted P2O5

0.51 3.29 4.65 8.46 8.9

Rapeseed responses to PRP Yield increase in kilogram Yield Increase by a kilogram (kg ha 1) (%) of RP or SSP

1312 2365 2624 3302 2917 962

36 146 173 243 203

0.31 1.25 1.48 2.06 8.69

The rate applied was 1125 kg ha 1 for each of RPs and that of SSP (single superphosphate) was 225 kg ha 1. Modified from Sun (2001).

different effects even for the same crop. For this reason, the comparisons were made of the percentage yield increase resulting from application of RPs to that of SSP which was set to 100. However, whether the percentage comparison can lead to a correct evaluation of the ability of a crop to respond to RPs is not clear. The second is the rate of RPs used in the experiments. Until now, in most of the trials, the rate of RPs used by the researchers has been too large. For example, Li et al. (1952, 1953) and Li and Hu (1956) used 1500 kg ha 1 and even 2250 kg ha 1 RP, and Sun (2001) used 1125 kg ha 1. With such a high rate of RP, the P extracted by 2% citric acid was considerably large and played the major role rather than the amount of total P. Based on rate of RPs applied per hectare and the 2% citric-acid-extracted P2O5 in RPs as reported by Sun (2001), we calculated the amounts of the citric-acidextracted P2O5 applied for a hectare. For example, the Jinpin RP contained 0.51% citric-acid-extracted P2O5, and the rate applied was 1125 kg ha 1; so the citric-acid-extracted P2O5 was 5.5 kg ha 1. Calculated in this way, it was 37 kg P2O5 ha 1 for Caiyang RP, 52 kg P2O5 ha 1 for Jinxiang RP, and 95 kg P2O5 ha 1 for Morocco RP, compared to 20 kg water-soluble P2O5 ha 1 for SSP. With the exception of the Jinpin RP, the citric-acidextracted P2O5 in other RPs was much higher than the water-soluble P in SSP per hectare. Using the calculated results and the increased rapeseed yield (kg ha 1) provided in his paper, we drew Fig. 1. As shown in this figure, rapeseed response to RPs closely correlated with the citric-acidextracted P2O5 (P < 0.01) but not with the total P2O5 in RPs (P > 0.05). This means that the total phosphorus did not play significant functions in the

Table 6

Effects of Morocco and Jinpin RP on rape yield in three soils with different textures Rapeseed yield (kg ha 1) from rock phosphates

Soil properties Soil texture

pH

OM (g kg 1)

Total N (g kg 1)

Total P (g kg 1)

Available P (mg kg 1)

Morocco

Jinpin

SSP

Control

Sandy loam Sandy loam Slight clay

7.70 8.70 6.75

9.42 9.24 9.96

0.575 0.578 0.703

0.161 0.173 0.163

7.02 6.60 4.88

1178 2100 1508

1072 1512 1140

1686 2255 1628

859 1100 713

SSP, single superphosphate. Modified from Sun (2001).

Sheng-Xiu Li et al.

4000 3500 3000 2500 2000 1500 1000 500 0

Relation of rape yield to available P amount

Relation of rape yield to available P content (%) in RP

y = 23.827x + 1198 2 R = 0.9417 50

0

100

Available P amount (kg

150

ha−1)

Rape yield (kg ha−1)

Rape yield (kg ha−1)

156

4000 3000 2000

y = 273.01x + 1189.7 R 2 = 0.9474

1000 0

0

y = 3.8955x + 937.35 R 2 = 0.4802

0

100

200

300

400

P amount (kg ha−1) in RP

4

6

8

10

Relation of rape yield to P content (%) Rape yield (kg ha−1)

Rape yield (kg ha−1)

Relation of rape yield to P amount in RP 3500 3000 2500 2000 1500 1000 500 0

2

Available content (%) in RP

in RP

500

3500 3000 2500 2000 1500 1000 500 0

y = 43.864x + 936.58 R 2 = 0.4811

0

10

20

30

40

P content (%) in RP

Figure 1 The relation of rapeseed responses to available P and total P in RPs Drawing after Sun (2001).

increase of rapeseed yield. This will cause wastage of large amounts of RP resources, and hence direct application of RPs may not be a good way for full utilization of P resources in such a case. Third, some researchers such as Sun (2001), Wu et al. (2002), and Meng et al. (2007) only used rapeseed as the test crop, so the results cannot be extended to other crops, particularly leguminous and cereal crops, for comparison. Numerous reports have shown that citric acid can be exuded from plant roots to liberate sparingly soluble P from soil, and therefore the citric-acid-soluble P has been regarded as P that can be used by plants (Bolan et al., 1994; Drouillon and Merckx, 2003; Fox and Comerford, 1990; Geelhhoed et al., 1999; Hinsinger, 2001; Jones, 1998; Jones and Darrah 1994; Kraffczyk et al., 1984; Lambers et al., 2002; Lo´pez-Bucio et al., 2000; Shahandeh et al., 2005; Staunton and Leprince, 1996; Stro˝m et al., 2002). The importance of the citric-acid-extracted P to plant growth is clearly illustrated in Fig. 1. If this is the case, not only rapeseed but other crops should also respond to RPs. However, such information was not provided. Fourth, the soils used by Li et al. (1952, 1953), Li and Hu (1956), Sun (2001), Wu et al. (2002), and Meng et al. (2007) were particularly unique. In these soils, the crop yield was very low in the control treatment and the effect of RPs was very high, several times higher than the control and even higher than the SSP treatment. This is unthinkable. In the strongly acidic soils such as those used by Li et al. (1952, 1953) and Li and Hu (1956), the soils were so acidic that plant growth was inhibited and had a strong

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

157

response to calcium carbonate for neutralizing the acid, and the RPs contained sufficient calcium carbonate that might have partially neutralized the soils. Whether the good results obtained by the addition of RPs was caused by P or by calcium carbonate remains unknown. Last but not least is that some conflicting results have been reported. The experiments carried out by the Beijing Agricultural University and Experimental Farm of the Agriculture Ministry (1966) are typical examples. Their field trials were conducted using an albic bleached soil with pH 6.3 in Heilongjiang Province, and included two major treatments: control (without application of fertilizer) and RP addition (900 kg ha 1 Moccoro RP). Results showed that addition of RP increased maize yield from 3057 ha 1 (control) to 5693 kg ha 1, an 82% increase, and soybean yield from 2250 to 3300 kg ha 1, a 47% increase. Application of RP increased green manure such as rapeseed from 3203 kg (control) to 21,475 kg ha 1 or 6.7-fold, while buckwheat from 2736 to 4231 kg ha 1 or 55%, and soybean as green manure from 14,432 to 20,100 kg ha 1, a 39% increase. Obviously, crop response to RP did not follow the same pattern as shown in Table 4. For further evidence to reveal the differences of crop responses to RPs, we conducted a series of pot and field experiments during the period from 1972 to 1976. The soil used in the pot experiment was calcareous with pH 8.2, organic matter 12.4 g kg 1, total N 0.11 g kg 1, and Olsen-P 3.8 mg kg 1. Each pot contained 10 kg air-dried soil, and eight replications were adopted for each treatment. The field experiment was arranged in a soil with pH 8.3, organic matter 8.76 g kg 1, total N 0.76 g kg 1, and Olsen-P 4.3 mg kg 1. Four replications were adopted. The two soils were classified as seriously P-deficient (Li and Xiao, 1992). Previous studies indicated that any crop grown at any time in the two soils had a good response to P fertilizer. Six crops were planted in both pot and field trials in two seasons: spring and autumn. Maize and soybean were sown in spring, while wheat, pea, hairy vetch, and rapeseed in autumn. Three RPs from different locations were used in the studies. One was Morocco RP containing 32% total P2O5 and 8.89% citric-acid-soluble P2O5 in addition to 51.6% CaO, 6.4% CO2, 3.4% F, 0.2% Fe2O3, 0.14% Al2O3, and 1.59% SiO2. The second was Kunyang RP containing 38.1% total P2O5 and 7.96% citric-acid-soluble P2O5, and the third was Hejiayan RP from Shaanxi Province with 24.4% total P2O5 and 1.8% citric-acidsoluble P2O5 in addition to some impurities including 47.3% CaO, 13.25% CO2, 0.7% Fe2O3, 1.05% Al2O3, and 3.43% SiO2. Since the soils used for the experiments were deficient in N supply, sufficient N was applied to meet the crop demand so that the crop responses to P fertilizer would not be limited by N deficiency. In the pot experiment, 0.15 g N to each kilogram soil and in field trials 150 kg N ha 1 were applied to cereal crops and rapeseed, but no N was applied to leguminous crops (pea,

158

Sheng-Xiu Li et al.

hairy vetch, and soybean). Application of SSP was also included as a treatment for comparison. In this case, a crop that had no response to either SSP or RPs might indicate that the soil had sufficient P for crop use but could not show that the RP had no effect. However, crops having good responses to SSP but no response to RPs would definitely show that RPs had no effect. The responses of crops planted in the same soils and under such conditions could be used to evaluate the crops abilities to use P from RPs. Results (Table 7) showed that, in the two calcareous soils, crops had significant responses to SSP. Application of SSP remarkably increased yields of all crops used. Compared to SSP, the effect of RPs was different depending on the rates applied. At low rates, RPs had almost no effect on crop yield for either leguminous or nonleguminous crops, whereas at high rates RPs also significantly raised crop yield though not as high as SSP. Considering the yield increase in absolute amount (kg ha 1) for the autumn-sown crops, wheat seed yield increased more than pea and rapeseed, but less than hairy vetch that was harvested as shoot dry matter. For spring-sown crops, maize yield increased more than soybean. The pot experiments followed the same pattern as the field experiments. In pot and field experiments, rapeseed, pea, and soybean were lower in seed yield than wheat and maize, and therefore their yield increases by SSP or RP addition were also lower. Since different crops have different productive potential and the yield increases by P fertilizer addition have a certain range, comparison of the yield increase cannot reflect their productive potentials. If the increase in seed yield or shoot dry matter (hairy vetch) was expressed as percentage of the control, the higher percentage by SSP or RP addition was for low-yield crops, rather than for high-yield crops. In the pot experiment, autumn-sown crops, i.e., pea and rapeseed, were higher than wheat and hairy vetch, and for spring-sown crops soybean was higher than maize. The same trend was also found in the field experiment with the exception of pea, which was equal to wheat in the percentage. Obviously, the yield increase percentage is related not only to the crop response (yield increase) but also to the control yield. If the control yield is low, a small increase in yield can lead to a high increase in percentage. In contrast, when the control yield is high, a significant increase in yield may lead to only a low increase in percentage. Due to such discrepancy, comparisons with the yield increase in percentage of control were also problematic. Based on these issues, we believe both of them should be considered in addition to the relative effect of RPs to SSP as used by other authors. The latter is to compare the ratio or percentage of yield increase by RPs to that by SSP (the yield increase being set to 100%). Results showed that for application of 1200 kg ha 1 of Morroco RP, the ratios varied from 54% to 63.8%, and for 1200 kg ha 1 of Kunyang RP, the variations were from 50% to 57.7% – both being in the same range of a type as set up by the Rock Phosphate Research Group, Institute of Soil Science, Chinese Academy of Sciences

Table 7 Some crops responses to rock phosphates in calcareous soils (1972–1976)

Pot experiment Yield (g pot 1) Control (without fertilizer) SSP 4 g (0.72 g water-soluble P2O5 pot 1) Morocco RP 20 g (6.4 g P2O5 pot 1) Kunyang RP 20 g (7.6 g P2O5 pot 1) Hejiayan RP 20 g (4.9 g P2O5 pot 1) Morocco RP 100 g (32 g P2O5 pot 1) Kunyang RP 100 g (38 g P2O5 pot 1) Hejiayan RP 100 g (24.4 g P2O5) LSD0.05 for P Yield increase in absolute amount (g pot 1) SSP 4 g Morocco RP 100 g (32 g P2O5 pot 1) Kunyang RP 100 g (38 g P2O5 pot 1) Hejiayan RP 100 g (24.4 g P2O5) Yield increase in percentage (%) of control SSP 4 g Morocco RP 100 g (32 g P2O5 pot 1) Kunyang RP 100 g (38 g P2O5 pot 1) Hejiayan RP 100 g (24.4 g P2O5) Yield increase in percentage (%) of SSP SSP 4 g Morocco RP 100 g (32 g P2O5 pot 1)

Wheat

Pea

Hairy vetch*

Rapeseed

Maize

Soybean

32.8 43.4 33.1 32.9 32.4 38.3 37.6 34.8 1.32

17.8 24.9 18.3 18.6 17.6 21.9 21.3 19.1 1.45

34.6 49.8 35.3 34.4 33.5 45.3 44.8 39.2 4.07

13.5 20.9 14.3 14.8 14.6 18.5 18.2 15.8 1.56

50.4 71.7 52.1 51.9 51.4 64.4 63.5 54.2 3.91

10.6 15.9 11.1 10.9 10.5 13.9 12.9 12.1 1.21

10.6 5.5 4.8 2.0

7.1 4.1 3.5 1.3

15.2 10.7 10.2 4.6

7.4 5.0 4.7 2.3

21.3 14.0 13.1 3.8

5.3 3.3 2.3 1.5

32.3 16.8 14.6 6.1

39.9 23.0 19.7 7.3

43.9 30.9 29.5 13.3

54.8 37.0 34.8 17.0

42.2 27.8 26.0 7.5

50.0 31.1 21.7 14.2

100 51.9

100 57.7

100 70.4

100 67.6

100 65.7

100 62.3 (continued)

Table 7

(continued)

1

Kunyang RP 100 g (38 g P2O5 pot ) Hejiayan RP 100 g (24.4 g P2O5) Field experiment Yield (kg ha 1) Control SSP 300 kg ha 1 Morocco RP 300 kg ha 1 Kunyang RP 300 kg ha 1 Hejiayan RP 300 kg ha 1 Morocco RP 1200 kg ha 1 Kunyang RP 1200 kg ha 1 Hejiayan RP 1200 kg ha 1 LSD0.05 for P Yield increase in absolute amount (kg ha 1) SSP 300 kg ha 1 Morocco RP 300 kg ha 1 Kunyang RP 300 kg ha 1 Hejiayan RP 300 kg ha 1 Morocco RP 1200 kg ha 1 Kunyang RP 1200 kg ha 1 Hejiayan RP 1200 kg ha 1 Yield increase in percentage of control SSP 300 kg ha 1 Morocco RP 300 kg ha 1 Kunyang RP 300 kg ha 1

Wheat

Pea

Hairy vetch*

Rapeseed

Maize

Soybean

45.3 18.9

49.3 18.3

67.1 30.3

63.5 31.1

61.5 17.8

43.4 28.3

3578 4779 3612 3598 3582 4226 4178 3724 226

2334 3115 2400 2384 2352 2772 2732 2466 321

13532 18876 13971 13923 13645 16944 16341 14017 1054

1787 2585 1808 1798 1792 2278 2212 1859 105

3698 4712 3759 3745 3700 4323 4215 3783 214

1678 2330 1756 1721 1689 2078 2054 1740 184

1201 34 20 4 648 600 146

781 66 50 18 438 398 132

5344 439 391 113 3412 2809 485

798 21 11 5 491 425 72

1014 61 47 2 625 517 85

652 78 43 11 400 376 62

33.6 1.0 0.5

33.5 2.8 2.1

39.5 3.2 2.9

44.7 1.2 0.6

27.4 1.6 1.3

38.9 4.6 2.6

Hejiayan RP 300 kg ha 1 Morocco RP 1200 kg ha 1 Kunyang RP 1200 kg ha 1 Hejiayan RP 1200 kg ha 1 Yield increase in percentage (%) of SSP SSP 300 kg ha 1 Morocco RP 300 kg ha 1 Kunyang RP 300 kg ha 1 Hejiayan RP 300 kg ha 1 Morocco RP 1200 kg ha 1 Kunyang RP 1200 kg ha 1 Hejiayan RP 1200 kg ha 1

0.1 18.1 16.8 4.1

0.8 18.8 17.1 5.7

0.8 25.2 20.8 3.6

0.3 27.5 23.8 4.0

0.1 16.9 14.0 2.3

0.7 23.8 22.4 0.4

100 2.8 1.7 0.3 54.0 50.0 12.2

100 8.4 6.4 2.3 56.1 51.0 16.9

100 8.2 7.3 2.1 63.8 52.6 9.1

100 2.7 1.4 0.6 61.5 53.3 9.0

100 6.0 4.6 0.2 61.6 51.0 8.4

100 11.9 6.6 1.7 61.3 57.7 9.5

Each pot contained 10 kg soil, and 4 g single calcium superphosphate (containing 18% P2O5) and 20 g rock phosphate was applied for the corresponding treatment. Modified from Li (1972–1976, unpublished data).

162

Sheng-Xiu Li et al.

Hejiayan

Kunyang

Yield increase or available P (kg ha−1)

Available P2O5 700 600 500 400 300 200 100 0

Hejiayan

Kunyang

Yield increase or available P (kg ha−1)

Morocco

Kunyang

Available P2O5

Morocco

Yield increase

Pea

Hejiayan

Kunyang

Available P2O5 700 600 500 400 300 200 100 0

Yield increase

Rapeseed

Hejiayan

700 600 500 400 300 200 100 0

Yield increase

Maize

Available P2O5 700 600 500 400 300 200 100 0

Morocco

Yield increase or available P (kg ha−1)

Wheat

Yield increase or available P (kg ha−1)

700 600 500 400 300 200 100 0

4000 3500 3000 2500 2000 1500 1000 500 0

Morocco Yield increase

Soybean

Hejiayan

Kunyang

Available P2O5 Yield increase or available P (kg ha−1)

Yield increase or available P (kg ha−1)

(1966) (see Table 4). By comprehensive comparisons, we concluded that, as a whole, there was almost no difference between leguminous and nonleguminous crops used in our experiments in responses to the RPs, and the leguminous crops, i.e., pea, soybean, and hairy vetch, had no special ability to use P from the RPs; the same was true for rapeseed, at least in the calcareous soils. Again, our results also showed that, for any crop, the response to RPs was closely related with the citric-acid-soluble P (P2O5) in RPs as shown in Fig. 2. Results obtained in other countries indicate that the availability of RPs is very low, and that a yield increase from direct application is doubtful for any crop. As reported by the Council for Agricultural Science and Technology

Morocco

Yield increase

Hairy vetch

Hejiayan

Kunyang

Available P2O5

Morocco

Yield increase

Figure 2 The relation of crop yield increase (kg/ha) to available P (in term of P2O5) (2% citrate acid-soluble P2O5) (kg ha 1). Drawing after Li (1972–1974, unpublished data).

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

163

(1980), RP found in the United States varied in solubility but generally had very low immediate availability to plants even when finely ground. One 24-year study showed RP to be only one-sixth as effective as acidulated phosphate in a corn–oats–alfalfa rotation in slightly acidic soils. RP was even less effective on a less acidic soil (Webb, 1982, 1984).

4. The Relation of Plants’ Root Morphology to Their Capacity of Using Soil Sparingly Soluble P and Responses to P Fertilizers Phosphorus is absorbed by plant roots in the form of orthophosphate, mainly monovalent dihydrogen phosphate ion H2PO4, although some metaphosphate and pyrophosphate that can be changed into orthophosphate in soil or in plants may also be absorbed (Sun and Wang, 1980). Unlike N and S in nitrate and sulfate, the P atom in phosphate is not reduced in the plant to a lower oxidation state. In soils, P occurs almost exclusively in the form of orthophosphate with total P concentrations usually in the range of 500–800 mg kg 1 dry soil (Mengel and Kirkby, 2001). However, the bulk of such a form of P (more than 90%) is present as insoluble and fixed forms including primary phosphate minerals, humus P, insoluble phosphate of Ca, Fe, and Al, as well as P fixed by hydrous oxides and silicate minerals. Phosphate in soil solution that is completely accessible makes up only a minute fraction of the total soil P. Commonly, the concentration of the soil solution P is around 0.05 mg l 1 (Barber et al., 1963) and is seldom higher than 0.3 mg l 1 (Fried and Shapiro, 1961). Such a fraction cannot meet the plant need for high yields in most of the cases. Plants obtain P from soil by three ways: diffusion, mass flow, and interception (direct absorption of nutrients by roots contacting the soil) (Barber, 1982). Not only is the concentration of P in the soil solution usually very low but also the movement of phosphorus in soil is carried out mostly by diffusion; the diffusion is very slow and the distance moved is very short (Bhat and Nye, 1973; Li and Zou, 1962; Lin et al., 1964; Liu and Cao, 1980; Mengel, 1984; Murdock et al., 1960; Xu and Liu, 1984). The diffusion coefficient of P in the soil solution was found by Olsen et al. (1962) to be less than 10 2 of that in water, whereas that estimated by George and Richardson (2008) was only about 1:1000 of that in water. Barber et al. (1963) reported values for the rate of diffusion of 32P in soil as low as 4 10 11 cm2 s 1, while Lewis and Quirk (1965) gave the mean velocity of phosphate ions in Seddon soils as only 0.04 mm day 1 at an applied P level of 100 mg kg 1 soil. As a result of this very slow movement in soil, P tends to accumulate close to the place where the P fertilizer was applied.

164

Sheng-Xiu Li et al.

Whether or not soluble or insoluble P fertilizers are applied to the soil, they are retained close to the surface. Consequently, the migration of phosphate ions in soil is difficult, and the amount of P uptake amount by diffusion is limited to a small space. Also, due to very dilute P concentration in soil solution (Barber, 1982; Fox et al., 1974; He et al., 1988; Li, 1986; Parfitt, 1978), P originating from mass flow driven by transpiration to plants is only a very small part, less than 1% of the total P required by the plant (Barber, 1982; Li, 1966a,b). Also, the organic P supply to plants by mass flow accounts for only 4% of P plants’ need (George and Richardson, 2008). For this reason, the area of the roots of the plant in contact with soil surface becomes a decisive factor and root systems play important roles in nutrient uptake. Plant roots proliferate in this P-rich surface zone and a small proportion penetrates the subsoil, where the bulk of the soil water, potentially available to plants, is stored. For this reason, root distributions down to the soil profile and root length are important for plants to take up P from soil. The importance of root length in the uptake of P adsorbed by soil has been hypothesized by Barley (1970). Andrews and Newman (1970) showed that reducing root length by pruning reduced P uptake, but did not significantly decrease the uptake of N. Root length is, of course, a major determinant of the absorbing surface area, and data showing this were presented by Christie (1975). There is much evidence that plant roots do differ in the efficiency with which they take up P from the soil. The continuous absorption of water and nutrients from soil solution by roots not only supplies large amounts of nutrients to plants but also leads to the depletion of water and nutrients in the rhizospheric soil zone, and forms gradients of nutrients and water potential between rhizosphere and bulk soil, thereby promoting the continuous progress of mass flow and diffusion. During the growing period, plants continuously form root branches and root hairs, and develop root system and arbuscular mycorrhizal fungi. All these increase the root-absorbing area, increase root spaces, and ensure the uptake of P (Parfitt, 1978). It has been shown that in the process of roots contacting the soil, the direct function between roots and soil played a great part for ensuring sufficient P supply to the crops (Mengel, 1984). Although Wang et al. (1999a) found that the P efficiency of common bean or kidney bean (Phaseolus vulgaris Linn.) was related to the seed size and the total content of P in seeds (a large-sized seed stores higher P and therefore has higher tolerance to low-P stress than the small-sized one), most scientists believe that it is the roots that have made some plants grown in soil deficient in available P to obtain P from soil for normal growth rather than other characteristics (Barber, 1982; Schenk and Barber, 1980; Hisanobori and Tanaka, 1983). Also, roots and aerial parts are contiguous in a plant, and the important roles roots play lie not only in supplying water and nutrients to the aerial parts but also participating in life activities through root metabolites. Amino acids, for instance, in roots can be synthetic precursors of

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

165

hormones and regulate nitrogen transportation between the source and the sink (Aulakh et al., 2001; Liu and Song, 2007; Nozu et al., 2006; Tan et al., 2006). As an interface of plant and soil, plant roots absorb water and nutrients from the soil, and react with some chemical compounds in the soil and change them, by which the growth and development of a plant is regulated (Mou et al., 1996). All these characteristics of P and roots have made the root function noted by many researches. Some special techniques have been used for such studies. For example, Da Silva and Gabelman (1992) selected high-P-efficient inbred lines of maize grown in sand culture system with phosphorous aluminum oxide. Li et al. (2003c) evaluated the tolerant ability of maize to low-P stress by seedling bed culture. Zhang et al. (2005) found that the P-absorbing efficiency at the seedling and elongation stages was the major characteristic determining most of the variance of the maize inbred lines tolerant to low-P stress. In studying the relationship between root characteristics and P absorption, great attention has been paid to the root morphological characteristics. A large number of investigations have shown that the P uptake ability by crops is related to the root morphology. It has been documented that under P-deficient conditions, plants suffering from P deficiency had some specific root properties: the ratio of root to shoot dry weight increased (Tang et al., 2001); root length extended (Alves et al., 2001; Cao et al. 2002); proliferous lateral roots developed (Schenk and Barber, 1979); the number of root hairs increased (Fo¨hse and Jungk, 1983), special root structure such as cluster roots (Pate and Watt, 2001) formed or root architecture changed (Yan, 2005); and the P harvest index declined (the ratio of P in seeds to the total of the above-ground P including the seed) (Gill et al., 2004). Maize genotypes with high efficiency of P uptake are characterized by vigorous growth of root systems, large absorbing areas, strong root activities, and high root affinity to P (Liu et al., 2006). The lengths of the preliminary branch roots and the taproots of maize inbred lines are reduced for varieties with low P efficiency, while those with high P efficiency are less reduced (Mi et al., 2004). Steffens (1984) compared the root status of ryegrass (Lolium multiflorum Lam.) and red or purple clover (Trifolium pratense Linn.) with P uptake amount in field, and found that the P uptake amount was closely related to the root length and fresh root weight. The root length of ryegrass was five times that of red clover, and its roots had more root hairs than red clover, whereas the average root diameter of red clover was two times larger than that of ryegrass. As a result, ryegrass took up much more P from soil than red clover. Itoh and Barber (1983) found that the P uptake amounts of six plant species were closely correlated with the total root lengths, root hairs, and even root hair densities. Pan et al. (2008) used 96 genotypes of soybean in a P-deficient soil to compare the growth and P efficiency. Results showed that the P-efficient genotypes facilitated biomass accumulation, root growth, and P uptake and shoot P utilization, and therefore these genotypes

166

Sheng-Xiu Li et al.

were highly associated with root dry weight, root to shoot ratio, root length, and root surface area. Similar results were observed in other crops such as wheat (Horst et al. 1993; Osborne and Rengel, 2002a,b), maize (Li et al. (2003c)), and rice (Guo et al., 2002). Tong and Yan (1999) reported that the relative value of P uptake (the ratio of P uptake in a deficient condition to P uptake in an adequate supply condition) and the relative value of shoot P content (the ratio of shoot P uptake in a P-deficient condition to that in an adequate P supply condition) were the most useful characteristics for the evaluation of P-efficient crops. Conflicting results also exist; for example, Bole (1977) did not find any correlation between root hair density and P uptake amount for wheat. The effectiveness of roots depends on their absorbing surface, and therefore on the extent of the root systems. Continually, vertical and horizontal extension and branching to form a large root system through the soil provide a basis for absorption of water and minerals from different spaces in the soil. Plant roots would not have the full ability to absorb abundant water and nutrients to meet their need and guarantee the supply to shoots unless they occupy a relatively large space in the soil for growth. Previous results have revealed the effects of root spatial distribution on the growth and yield of rice (Ling and Ling, 1984; Zhu et al., 2001), millet, and sorghum [Sorghum bicolor (L.) Moench and Sorghum vulgare (L.) Pers] (Zaongo et al., 1994). Our study further revealed the function of root distributive space of maize and wheat in plant growth and yield in such a way that restriction of roots in nylon bags considerably reduced root biomass and resulted in a remarkable reduction in yield (Li et al., 2009). For leguminous and nonleguminous crops, some researchers hold the view that the former can take up more P from soil than cereal crops since they have large root systems and strong ability to absorb P from soil (Extension Station of Agricultural Technology in Bijie District, Guizhou Province, 1964; Liu, 1964; Zhao and Tong, 1964). Although such a view was proposed as early as in the 1970s in China, it has been contested until now (He, 2004; Jiang, 1997; Li, 2005; Lu et al., 1998; Peng and Pei, 1980; Sun and Wang, 1980; Wang, 2004; Ye, 2002). For studying the relationship of roots to P fertilizer response between leguminous and nonleguminous crops, we carried out a series of experiments. In a pot experiment, two manural soils, 86001 and 86002, having different N and P supply levels were used. Soil 86001 contained 57.6 mg ka 1 of NaOH–hydrolyzable N (Cornfield, 1960) and 10.6 mg kg 1 of Olsen-P, and soil 86002 contained 41 mg kg 1 of NaOH–hydrolyzable N and 2.5 mg kg 1 of Olsen-P. Mitscherlich-type pots were used, and 10 kg dry soil was added to each pot. Two crops, wheat and pea, were planted with the plant density of 12 and 10, respectively. The high plant densities ensured that there were enough plants to fully absorb P both from soil and the P fertilizer. There were two treatments: without

167

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

application of P fertilizer (control) and with P treatment (application of 0.11 g P kg 1). The plants were harvested five times, and three pots for each treatment were harvested every time to investigate plant roots and P uptake. The results (Table 8) showed that, although the plant density of pea was higher than it was in the field, its root weight was still much lower than that of wheat. Without the application of P fertilizer, the average weight of five harvests of pea was 3.99 g per pot for soil 86001 and 2.18 g per pot for soil 86002, whereas that of wheat was 10.74 and 4.59 g per pot, respectively, for the two soils. After application of the P fertilizer, pea root weight was 3.53 Table 8 Root and shoot dry weight of pea and wheat and their P uptake amount (1986–1987)

Soil no.

Shoot weight P uptake Root weight Sampling (g pot 1) (mg pot 1) (g pot 1) time (d/m) Without P With P Without P With P Without P With P

Pea 86001 7/3 15/4 2/5 20/5 1/6 Mean 86002 7/3 15/4 2/5 20/5 1/6 Mean Wheat 86001 7/3 15/4 2/5 20/5 1/6 Mean 86002 7/3 15/4 2/5 20/5 1/6 Mean

0.71 3.14 5.58 5.58 4.93 3.99 0.65 2.53 2.20 2.83 2.75 2.18

0.73 2.91 5.50 4.49 4.00 3.53 0.91 4.16 5.03 4.75 4.95 3.96

1.03 3.44 12.43 22.36 27.62 13.38 1.00 1.93 5.77 14.78 18.71 8.44

1.18 8.03 19.80 30.57 39.08 19.73 1.16 8.39 26.13 40.74 41.56 23.59

1.79 6.90 19.52 38.21 45.15 22.31 2.01 3.71 8.91 15.50 19.56 9.96

3.62 22.40 22.58 44.63 58.86 32.40 3.67 29.78 38.17 43.01 51.15 33.76

6.34 9.40 15.99 14.40 7.54 10.74 2.58 4.75 7.76 4.86 3.01 4.59

4.55 9.60 15.43 13.13 10.02 10.55 5.76 8.25 10.71 11.66 8.57 8.99

4.78 17.14 32.91 34.02 33.51 24.49 1.77 3.21 11.78 19.07 16.57 10.48

5.49 18.57 39.07 38.61 40.32 26.81 6.21 16.69 30.35 33.26 37.23 34.75

12.71 23.01 34.19 34.62 36.07 27.95 2.89 4.67 12.62 14.63 13.84 9.52

32.58 37.69 49.13 45.15 46.29 42.18 22.58 28.34 31.75 31.83 38.56 30.61

Modified from Li et al. (1995).

168

Sheng-Xiu Li et al.

and 3.96 g per pot, whereas wheat root weight was 10.55 and 8.99 g pot 1, respectively. Due to difference in biomass of the two crops, there is no basis for making a direct comparison, and such a comparison of their dry root weight would not reveal the reality. The ratio of roots to shoots can reflect the relative magnitude of root function to shoots. The root to shoot ratio for pea was 0.30–0.54, while for wheat it was 0.47–0.81, indicating that wheat had better developed roots, while pea roots had a higher function in supporting the shoots. Whether the results obtained from the pot experiment were in agreement with field results needs examination. For getting support from field evidence, we sampled both wheat and pea roots from the same field in Yuanshi Village in Yangling, Shaanxi Province, during the period from April 19 to 21, 2002. At the time of sampling, wheat was at the heading stage and pea at the flowing stage. Both wheat and pea were sown in rows normally and grown under normal conditions. The space between rows was 16 cm for both wheat and pea, and the distance between two plants in the same row averaged 2 cm for wheat and 10 cm for pea. The soil sample was taken as follows: selecting a plant row as the center and measuring 50 cm along the row as length with 5 plants for pea or 25 plants for wheat, harvesting the above-ground parts of the plants, and measuring 8 cm distance from the row to the middle between two rows as width, and then dividing the width into inner (from the row to 4 cm) and outer (from 4 to 8 cm) parts, each being 4 cm wide. The inner soil sample was taken in five layers, 0–5, 5–10, 10–15, 15–20, and 20–30 cm, while the outer in four layers, 0–5, 5–10, 10–15, and 15–20 cm. The soil was sampled layer by layer. Each layer of soil was taken on both sides of the center (the plant row) and recorded in inner or outer zone followed by the layer depth in centimeters in the relevant tables. The samples were placed on a plastic sheet, the soil was carefully smashed into pieces, and roots were carefully taken out by hand. Root weight of each layer was the total of both sides and was recorded as inner or outer depth in centimeters. Two replications were adopted. The total area for sampling plants was 800 cm2 (16 50 cm). Taking the average of the duplication, the above-ground fresh weight of wheat was 6.683 kg m 2 and that of pea 5.040 kg m 2, equivalent to a total dry weight 12,030 kg ha 1 for wheat and 10,692 kg ha 1 for pea. As shown in Table 9, both pea and wheat roots were mainly distributed in the inner 0–4 cm zone of both sides of the plant row and a 0–5 cm deep layer, followed by a 5–10 cm deep layer. Beyond the two layers of the inner zone, few roots were distributed. Compared to the weight and the distribution, wheat roots were much higher in weight (although the ratio of wheat root to pea root was not as high as in the pot) and occupied a larger space in the outer 4–8 cm zone and deeper layer than pea. Again, the root weight could not explain why pea was more sensitive to the application of P fertilizer.

Table 9 Root fresh and dry weight of pea and wheat sampled on both sides around the plant row center from inner zone (0–4 cm from the center) and outer zone (4–8 cm from the center) in different layers with a soil volume of 0.004 m3 (g/100 2 4 5 cm3) or 0.08 m3 (g/100 2 4 10 cm3) Fresh and dry weight (g) of pea

Fresh and dry weight (g) of wheat

Sampling area

Fresh

Dry

Water (%)

Fresh

Dry

Water (%)

Inner 0–5 cm deep Inner 5–10 cm deep Inner 10–15 cm deep Inner 15–20 cm deep Inner 20–30 cm deep Total root weight in inner zone of 0.024 m3 soil Outer 0–5 cm deep Outer 5–10 cm deep Outer 10–15 cm deep Outer 15–20 cm deep Total root weight in outer zone of 0.016 m3 soil Total root weight in 0.04 m3 soil volume

54.034 10.054 3.226 2.472 1.618 71.404 9.808 5.968 1.896 0.238 17.910 89.314

6.400 1.192 0.388 0.304 0.210 8.494 1.036 0.690 0.210 0.026 1.962 10.456

88.16 88.16 87.97 87.97 87.18

59.458 16.466 11.256 4.784 2.248 94.212 11.722 8.238 3.316 2.684 25.960 120.172

10.422 2.886 2.230 0.948 0.444 16.930 2.084 1.458 0.666 0.540 4.748 21.678

82.47 82.47 80.18 80.18 80.18

Data in brackets are means of water content (%). Modified from Li, S. X. (2002, unpublished data).

89.44 89.44 88.97 88.97

82.29 82.29 79.91 79.91

170

Sheng-Xiu Li et al.

Nutrient uptake by plants is not all determined by root weight, but mainly by the root surface area. It was reported that the total P uptake amounts of soybean were highly correlated with the total root areas (Borkert and Barber, 1983; Liu et al., 2006). As mentioned, Steffens (1984) showed that ryegrass root length was five times longer than that of red clover, while the average root diameter of red clover was two times larger than that of ryegrass. In our pot experiments, the root surface areas of wheat and pea were not determined, but we can use the result of ryegrass and red clover for estimation. Obviously, wheat is different from ryegrass and pea from red clover. However, for the same family of crops, their root characteristics have certain similarity. Wheat and ryegrass belong to cereal crops, while red clover and pea belong to the leguminous crops. Like ryegrass, wheat roots are fibrous and dense and have a large number of fine roots. Also, like red clover, pea has a taproot system, it is rough, and it has a colloidal shape at the surface (Steffens, 1984) (Plate 1). Therefore, the root parameters of red clover and ryegrass can be roughly used to estimate those of pea and wheat in a pot experiment. Wheat root weight is as much as two times that of pea, and if the root diameter of pea is two times that of wheat, then the number of roots per unit weight of wheat is at least four times that of pea. Since root hairs and root surface areas are proportional to the number of roots, the root surface area of wheat should be at least four times that of pea. If the root development degree was the dominant factor determining the crops’ uptake amount of P and crop response to P fertilizer, then wheat with a well-developed root system would have taken up more P from soil than pea, and therefore would have declined its sensitive response to P fertilizer. However, this is not true. In fact, in the soil with low P- and N-supply levels, wheat took up less P than pea, while in the soil having high

Plate 1

Wheat roots in left side and pea roots in right side.

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

171

N- and P-supply levels wheat took up more P from the soil. This shows that the root surface area, like root weight, cannot explain the difference between the two crops’ response to P fertilizer. The conclusion from pot experiments was supported by the measurement of field wheat and pea roots. The root samples taken from the field were also used to determine the total absorbing area and actively absorbing area. Results (Table 10) showed that, for a single plant, the total rootabsorbing area and actively absorbing area of pea were much larger than those of wheat. However, considering the root areas in unit volume of soil, the same trends existed as the fresh or dry weight of the root: wheat areas were almost one-third of those of pea. Also, the areas were closely related to root biomass or fresh and dry weights. The correlation coefficient (r) of pea root fresh weight was 0.9505 with total absorbing area and 0.9277 with actively absorbing area, and that of root dry weight was 0.9504 with total absorbing area and 0.9277 with actively absorbing area. The same relation was also found for wheat: the correlation of root fresh weight was 0.9551 with total absorbing areas and 0.9603 with actively absorbing area, and that of the root dry weight was 0.9544 and 0.9598 with the total absorbing and actively absorbing areas. With no exception, the correlations were all significant at the P < 0.05 level (n ¼ 5, r0.05 ¼ 0.878), and some were significant at the P < 0.01 level (n ¼ 5, r0.01 ¼ 0.959). That means that the higher the root weight, the higher the total root-absorbing area or the actively root-absorbing area. As a consequence, root weight and root surfac provides almost the same information. Root function characteristics of the absorbing capacity of nutrients and water are mainly exhibited in two aspects: physical and physiological. In addition to the physical properties of root morphology such as root number, length, and area (total root area, total root-absorbing area, root actively absorbing area, specific absorbing area, and specific actively absorbing area), the root activities in terms of the 2,3,5-triphenyltetrazolium chloride (TTC) reductive amount and reductive intensity can be regarded as the main physiological aspects for indicating the root absorptive capacity (Ding, 1987). TTC is a redox indicator used to differentiate between metabolically active and inactive tissues. The white compound TTC is enzymatically reduced to red 1,3,5-triphenylformazan (TPF) in living tissues due to the activity of various dehydrogenases (enzymes important in oxidation of organic compounds and thus cellular metabolism), while it remains white TTC in areas of necrosis since these enzymes have been either denatured or degraded. Calculations based on the measurements of the total and actively absorbing areas from field pea and wheat roots show that the TTC reductive amount expressed in terms of per plant or per unit volume of soil was higher for pea than for wheat. The TTC reductive amount was 1299 mg TTC h 1 per plant in a 0.04 m3 soil volume for pea, while it was 166 mg TTC h 1 for wheat, the former being 7.8 times the latter. The TTC reductive intensity

Table 10 Root absorbing and actively absorbing areas for a single plant and for all plants of pea and wheat sampled on both sides around the plant row center from inner zone (0–4 cm from the center) and outer zone (4–8 cm from the center) in 10 cm deep layers for each sample with a soil volume of 0.08 m3 (g/100 2 4 10 cm3)

Sampling zone and layer

Absorbing area （m2） Actively absorbing area Total absorbing area (m2) Actively absorbing area per plant distributed in (m2) per plant distributed for all plants sampled in (m2) for all plants 0.008 m3 soil 0.008 m3 soil in 0.008 m3 soil sampled in 0.008 m3 soil

Pea Inner 0–10 cm deep 8.30 Inner 10–20 cm deep 0.50 Inner 20–30 cm deep 0.22 Outer 0–10 cm deep 1.78 Outer 10–20 cm 0.28 Total area in 0.04 m3 soil 11.08 volume Wheat Inner 0–10 cm deep 2.78 Inner 10–20 cm deep 0.10 Inner 20–30 cm deep 0.08 Outer 0–10 cm deep 0.06 Outer 10–20 cm deep 0.08 Total area in 0.04 m3 soil 3.10 volume Modified from Li S. X. (2002, unpublished data).

2.58 0.14 0.06 0.72 0.14 3.64

41.5 2.5 1.1 8.9 1.4 55.4

12.9 0.7 0.3 3.6 0.7 18.2

0.98 0.06 0.06 0.02 0.04 1.16

69.5 2.5 2.0 1.5 2.0 77.5

24.5 1.5 1.5 0.5 1.0 29.0

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

173

was 85.3 mg TTC h 1 g 1 fresh root on average for pea, while it was 33.6 mg TTC h 1 for wheat (Table 11), the former being double the latter. However, since plant density for pea was different from that of wheat, comparison of a plant or per unit weight could not give a correct indication. Compared from the total TTC reductive amount in a soil volume of 0.04 m3, pea was 6494 mg TTC h 1 whereas wheat was 4160 mg TTC h 1. In any case, pea was higher in root activity than wheat. The results were in conflict with root weight and root surface. Many experiments have shown that crop yields are closely related to root biomass (Liu et al., 1994; Wang et al., 1999b). Our previous results also provided such evidences. Evaluated by the shoot biomass, seed yield, and nutrient uptake as criteria, the root biomass or root weights were parallel to shoot growth and, finally, to crop yield (Fig. 3). For this reason, the root biomass weights could comprehensively indicate the root absorptive capacity for the same crop and were much better than root activities. In most cases, root biomass and total root surface area were almost equal in terms of reflecting root absorptive capacity (Song and Li, 2003, 2006). Accompanied by the root weight decrease in unit area, total root-absorbing area was correspondingly reduced (Song and Li, 2003). In contrast, the specific absorbing or specific actively absorbing area of roots in term of area per unit weight might bear no such relations to root biomass, and the root activity expressed in TTC reductive amount per unit volume or area was another situation: it was related to the root biomass to a certain extent and this occurred in most cases, but not in all. Clearly, the root activity may be playing a minor role in reflecting root absorptive capacity compared to root biomass. Although the root biomass and its related root surface areas can indicate the yield and nutrient uptake amount for some crops, the difference between leguminous and nonleguminous crops to respond to P fertilizers cannot be satisfactorily explained when judged either by root biomass or by root-absorbing area.

5. Microorganisms in Rhizosphere Soil and Their Function in Supplying P to Plants Microbial populations are key components of the soil–plant continuum where they are involved in a framework of interactions affecting plant development (Barea et al., 2005; Vassilev et al., 2006a,b; Wakelin et al., 2004). Many soil microorganisms are able to transform insoluble forms of P to an accessible, soluble form, contributing to plant nutrition as PGPM. The P-solubilizing microorganisms (PSM) can dissolve and mineralize P from inorganic and organic pools of total soil P, and may be used as inoculants to increase P availability to plants (Illmer et al., 1995; Kucey et al., 1989; Richardson, 1994, 2001; Whitelaw et al., 1999). Such work may help in

Table 11 Fresh root weight, TTC reductive amount for a single plant and for all plants of pea and wheat sampled on both sides around the plant row center from inner zone (0–4 cm from the center) and outer zone (4–8 cm from the center) in 10 cm deep layer for each sample with a soil volume of 0.08 m3 (g/100 2 4 10 cm3) as well as reductive intensity for a unit weight (g) of fresh root in different zones and layers

Sampling zone and layer

Pea Inner 0–10 Inner 10–20 Inner 20–30 Outer 0–10 Outer 10–20 Total in 0.04 m 3 soil Wheat Inner 0–10 Inner 10–20 Inner 20–30 Outer 0–10 Outer 10–20 Total in 0.04 m3 soil

TTC reductive amount (mg TTC h 1 plant 1）

TTC reductive intensity (mg TTC g 1 FW h 1）

Total TTC reductive amount in 0.008 m3 soil (mg TTC h 1）

64.090 5.704 1.618 15.776 2.122 89.314

894.4 154.8 48.0 197.2 4.4 1298.8

69.8 135.7 148.3 62.5 10.4 (85.3)

4472 774 240 986 22 6494

75.924 16.040 2.248 19.960 6.000 120.172

123.8 16.8 6.2 17.0 2.6 166.4

40.8 26.2 69.0 21.3 10.8 (33.6)

3095 420 155 425 65 4160

Fresh root weight (g) in 0.008 m3 soil volume

The root activity was determined by reduction method and expressed in term of mg TTC g 1 DW h 1 (reduced amount of TTC in microgram per gram fresh weight per hour) (Ding, 1987). Data in brackets are means. Modified from Li S. X. (2002, unpublished data).

175

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

5500 y = 8.7003x − 1254 R 2 = 0.9345

14,500

Wheat yield (kg ha−1)

Shoot weight (kg ha−1)

16,500

12,500 10,500 8500 6500 4500 700

900

1100

1300

Root weight (kg

1500

ha−1)

1700

5000

y = 3.2171x − 513.84 R 2 = 0.8624

4500 4000 3500 3000 2500 2000 1500 500

800

1100

1400

1700

Root weight (kg ha−1)

Figure 3 Relation of wheat shoot and seed weight to root biomass. Drawing after Li et al. (2001, unpublished data)

developing soil microbial community management schemes based on specific functions of P dissolution and mineralization, and selection of microorganisms as potential microbial inoculants (biofertilizers) for possible use of PSM in industrial bioprocessing of RP. Several studies have indicated that the presence of different plant species or genotypes influences the microbial community due to the differential response of these organisms to different root signaling and exudation patterns, especially when plants are under environmental stress (Lynch and Whipps, 1990; Richardson, 1994) such as phosphorus stress (Hinsinger, 2001). Oliveira et al. (2009) isolated, screened, and evaluated the phosphate solubilization activity of microorganisms in maize rhizosphere soil, and found that bacteria produced the greatest dissolution in a medium containing tricalcium phosphate, while the fungal population was the most effective in dissolving P sources of aluminum phytate and lecithin. A greater diversity of P-dissolving microorganisms was observed in the rhizosphere of the P-efficient maize genotypes, suggesting that their P efficiency may be related to the potential to enhance microbial interactions of P-dissolving microorganisms. Narsian et al. (1994) found lower rates of dissolution from P–Al than from P–Ca among eight microorganisms tested in tricalcium and aluminum phosphate. Similar results were also found by Whitelaw et al. (1999) and Barroso and Nahas (2005), probably due to the higher solubility of the calcium phosphates in culture solutions. Some authors have pointed out that the phosphorus liberated from clay minerals along with AlPO4 can increase the level of toxic Al3þ in solution (Illmer et al., 1995), and this could possibly explain the suppression of the P-solubilizing activity of the PSM in the P–Al medium. A low correlation was observed between pH reduction and increase of soluble P in the liquid culture, indicating that the decline in the concentration of insoluble phosphorus might have been influenced by other factors (Barroso and Nahas, 2005; Whitelaw et al., 1999).

176

Sheng-Xiu Li et al.

The ability of soil microorganisms to dissolve various forms of inorganic P is well documented (Goldstein et al., 2003; Kucey et al., 1989; Richardson, 1994; Whitelaw et al., 1999). Some results have shown that the genus Aspergillus was the most efficient in solubilizing phytate (Yadav and Tarafdar, 2003) and the fungus Aspergillus terreus has been cited in the literature as an important phytase-producing fungus (Mitchell et al., 1997). Nacamulli et al. (1997) report that Burkholderia cepacia represents one of the predominant bacterial species among the microorganisms occurring in maize rhizosphere. Several studies have shown that B. cepacia can compete with the indigenous microflora, survive and colonize roots of various maize cultivars (Bevivino et al., 2000; Nacamulli et al., 1997), enhance the yield of several crop plants (Baldani et al., 2001; Rodrigues and Fraga, 1999), and antagonize and repress the major soilborne fungal pathogens of maize (Bevivino et al., 2000). Oliveira et al. (2009) found B. cepacia capable of solubilizing unavailable sources of phosphorus in maize, that fungus A. terreus was able to solubilize P from aluminum phosphate phytate and lecithin, and that Talaromyces rotundus was one of the highest phosphatase producers in the phytate and lecithin media, indicating that this fungus represents an important species for producing phytase and phospholipases. These microorganisms are of interest as having the potential for inoculants to improve the efficiency of P acquisition from the soil and fertilizer sources, especially RPs. Of the organisms around plant roots, attention has been given to the mycorrhiza, especially Arbuscular mycorrhiza. A number of reports have presented results of A. mycorrhiza liberating P from soil or from RP for plant use. Prakash et al. (2004) reported a significant effect of mycorrhizae in dissolution of P from RP. Similarly, Vassilev et al. (2006a,b) considered that the use of microbial dissolution of RP was a cheaper way to use sparingly soluble soil P and to achieve better yield. There is no consistent viewpoint for microorganisms on the release of P to plants between legumes and nonlegumes, and there exist no publications showing their effect on responses of the two type crops to P fertilizer; therefore the very limited knowledge do not permit us to carry on further discussion.

6. Root Exudates (Substances Secreted from Roots) and the Plants’ Capacity to Use Sparingly Soluble P in the Soil and Crop Responses to P Fertilizers Of the hypotheses proposed for explaining the difference of crops in utilizing sparingly soluble P in the soil and their responses to P fertilizers, the most common and important points are the differences in the release of protons and organic acids from the roots (Li, 1966a,b; Xie, 1966).

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

177

Plant roots can secrete different kinds of chemical substances and these substances may influence soil inorganic or organic phosphate availability and endow crops with different abilities for using soil P and therefore the amount of P taken up from soil as well as responses to P fertilizers. The substances related to P availability include protons, organic acids, and acid phosphatase (APase). The former two may be connected with inorganic phosphate availability, while the latter with the organic phosphate availability. Many researchers have focused on these aspects. Under P-deficient conditions, many crops can release protons from roots and acidify the rhizosphere soil. This phenomenon has been reported by many scientists, and has been explained by the hypothesis of cation and anion balance (Coic et al., 1962; Kirkby, 1968). According to this hypothesis, the ratios of anions to cations absorbed by different crops are considerably different if the ions are expressed in terms of moles. Some crops may absorb more anions than cations, while some other may absorb more cations than anions, and a few may take up equal quantities. Such unbalanced uptakes of nutrients can induce pH changes in rhizosphere areas and thereby lead to the dissolution and absorption of RPs. It is said that protons released by ATPase pumps located in the plasma membrane during nutrient uptake solubilize P from RP for meeting plant growth requirements (Gill et al., 1994). We know now that most cereal crops absorb more anions than cations, especially when nourished by nitrate-N. In this case, for the electronic balance between the inside of the root and the outer medium, the surplus anions must be neutralized. How the plants neutralize the anions is not actually known. For crops called alkalinized plants, most scientists believe that roots may exude OH. As a result, pH in the rhizospheric area is increased. Some types of crops take up more cations than anions and the plants secrete protons (Hþ) into the rhizospheric areas, lowering the pH in those areas. These types of plants are called acidified crops. If the dominant cation taken up by the plant is calcium, then P solubility in RPs could be enhanced. As early as 1950, Sokolov (A. B. Coкoлoв, 1950 in Russian, 1959 in Chinese) cited in his book Agrochemistry a result from Domontowich (M. К. Дoмoнтoвич) that strong acidification occurred in lupine rhizosphere soil and that cereal crops sown mixed with lupine could increase the absorption efficiency of P from RPs. The acidification of rhizosphere soil was further proven in white lupin (Lupinus albus L. cv. Amiga) by other researchers (Dinkelaker et al., 1989; Hinsinger and Gilkes, 1995; Neumann and Ro¨mheld, 1999), and such a phenomenon was also found in other crops such as rapeseed (Hoffland, 1992), maize, soybean, sorghum, sudan grass [Sorghum sudenense (Piper) Stapf], wheat, oats (Avena sativa L.), and barley (Hordeum vulgare L.) (Gollany and Schumacher, 1993). The release of protons decreases rhizosphere soil pH. The decline in pH is an induced mechanism for plants’ adaptation to the low-P stress (Polomski and Kuhn,

178

Sheng-Xiu Li et al.

2002) and can increase the solubility of the sparingly soluble P in the rhizosphere soil (Polomski and Kuhn, 2002) and therefore the availability of phosphorus (Agilars and van Diest, 1981; Hisanobori and Tanaka, 1983). Owing to the very short diffusion distance of phosphoric acid, P supply by rhizosphere soil plays a more important role than bulk soil. Different crops have different effects on the rhizospheric soil pH, and thus on the phosphorus availability (Hedley et al., 1983). It has been reported that the plant uptake amount of P has an exponential relation with the soil pH; therefore, with the decrease of pH in certain ranges the uptake amount of P increases (Holfold and Ratrick, 1979; Van Ray and Van Diest, 1979). The supply status of phosphorus from the rhizosphere soil is related to the crops, particularly to the amount of protons (Hþ) they release from the roots. Hedley et al. (1983) showed that rapeseed acidified the soil if deprived of an adequate supply of phosphate. Phosphorus deficiency strongly induced the net release of protons from the roots of tomato (Lycopersicon esculentum L., cv. Moneymaker) and chickpea (Cicer arietinum). Such release of protons by plants probably determines their ability to utilize acid-soluble calcium phosphates in calcareous soils or RP (Neumann and Ro¨mheld, 1999). Gillespie and Pope (1989) found that alfalfa rhizosphere acidification could enhance P uptake by walnut ( Juglans L.) tree seedlings in alfalfa (Medicago sativa L.)/black walnuts ( Juglans nigra L.) intercropping system. Bolland et al. (1999) found that faba bean (horse bean, broad bean) (Vicia faba L.) had higher yields without addition of P and lower yield response to added P than wheat in both greenhouse and field experiments. Nuruzzaman et al. (2005) revealed that faba bean had higher P concentration and content than field pea (Pisum sativum L.), white lupine (L. albus L.), and wheat. Li et al. (1999) observed the interspecific facilitation in faba bean/maize intercropping systems. The total P uptake was increased by 20% when no P was added and 38% when 33 kg P ha 1 was applied in the faba bean/maize intercropping system, as compared to the corresponding monoculture in a field experiment (Li et al., 2003a,b). Li (2004b) found that the capacity of faba bean to mobilize insoluble soil P was much greater than that of maize, and hypothesized that a decrease of rhizosphere pH played an important role in P mobilization. Zhou et al. (2009) observed that, in addition to a large root system, the proton efflux of entire plants and the specific proton efflux of faba bean were much higher than in maize and soybean; and therefore the intensive mobilization ability of soil P by faba bean was not only due to its large root system at early stages but also due to the rate of proton efflux per unit root length. Acidification of rhizosphere soil by roots enhances the effectiveness of RP utilization of some crops such as white lupine (Sas et al., 2002), clover and ryegrass (Hinsinger and Gilkes, 1996), lowland rice (Saleque and Kirk, 1995), and camellia plants (Zoysa et al., 1997, 1998b). Hinsinger and Gilkes (1996) reported that the release of protons by roots of clover and ryegrass

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

179

was the driving force for the root-induced dissolution of RP. Evidences show that some plants can solubilize P from RP by release of Hþ from their roots and acidify the rhizosphere soil (Bolan et al., 1990; Hinsinger, 1998; Hinsinger and Gilkes, 1995; Liu et al., 2004; Zoysa et al., 1998a). There are also some genotype varieties with this characteristic (Flach et al., 1987). Phosphorus utilization efficiency is the amount of biomass produced per unit of P taken up by the plant (Siddiqi and Glass, 1981). It varies widely among the various wheat cultivars depending upon their capability to acidify the rhizosphere and solubilize P from RP. Ramirez et al. (2001) reported that maize lines improved P utilization efficiency from RP by acidifying the rhizosphere. Schenk and Barber (1979) revealed that root acidification was an important factor in P uptake and its utilization efficiency. Flach et al. (1987) and Hoffland et al. (1989) reported dissolution of RP through root acidification. Liu et al. (2004) showed that the roots of the seedlings induced significantly higher acid and solubilized a significantly higher amount of RP in the rhizosphere than in the bulk soil, and proposed that it was likely caused by increased soil acidification and oxlate production in the rhizosphere. Gill et al. (1994) found that P uptake by wheat cultivars was significantly related to the drop in root medium pH presumably due to Hþ efflux from their roots. Rahmatullah et al. (2009) revealed in a hydroponics experiment using RP as P source that seven wheat cultivars showed significantly different growth status and P utilization from RP, and the P uptake and utilization were negatively correlated to the drop in pH of the root medium. The most efficient P cultivar solubilized and utilized more P from RP added by releasing more Hþ than the less efficient ones. Development of P-efficient genotypes with a great ability to grow and yield in P-deficient soil is, therefore, an important goal in plant breeding (Hash et al., 2002; Rengel, 1999; Wissuwa et al., 2002; Yan et al., 1992a,b, 2004). Using a mathematical model, Saleque and Kirk (1995) concluded that solubilization of soil P by acidification accounted for at least 80% of the P taken up for lowland rice. The intensity and amount of proton release are greatly influenced by the uptake of nutrient forms, and N forms greatly affect such characteristics. The amount of the sparingly soluble P taken up by plants depends to certain extent on the environmental conditions, and N forms certainly exert a great influence on the environment. Nitrate-N applied to soil increases the rhizosphere pH (alkalization of rhizosphere) (Nye, 1981), while ammonium-N addition acidifies the rhizosphere. With ammonium-N as the N source, the alkalinized crops may even be changed into acidified crops. Ryegrass can acidify soil and dissolve RPs. Rapeseed, in contrast, can significantly acidify soil with ammonium-N nutrition, while it can still use P from RPs with nitrate supply due to promotion of Ca absorption by nitrate-N even if the soil is slightly acidified (Li et al., 1992a). However, buckwheat has a strong ability to absorb Ca and Mg, and needs less N, but

180

Sheng-Xiu Li et al.

has strong ability to acidify rhizospheric areas. Soybean can strongly acidify soil by use of nitrogen fixed by nodule bacteria that are present in roots. The effect of N forms on the rhizosphere pH has been verified in many crops such as soybean (Riley and Barber, 1971), ryegrass (Gahoonia et al., 1992), sunflower (Helianthus annuus L. and other species) (Ro¨mheld et al., 1984), rice (Kirk et al., 1999), and tea (Amellia spp.) (Zoysa et al., 1998b). Besides releasing protons through the exchanges of cations from plant roots with soil nutrients, many plant roots can secrete low-molecularweight (LMW) acids such as citric acid, malic acid, oxalic acid, malonic acid, amber acid, and tartaric acid (Liu et al., 2009). Citrate is usually considered the most powerful organic anion, next to oxalate (Schlesinger 1997) and malate (Bolan et al. 1994). These acids have been detected in many plants such as pigeon pea [Cajanus cajan (Linn.) Millsp.] (Otani et al., 1996), radish, rapeseed (Zhang et al., 1997), mustard (Brassica hirta Moench, B. juncea L., and other species) (Gerke and Meyer, 1995), barley (Gahoonia et al., 2000), rice (Kirk et al., 1999), grape (Brasica nigra L.), and red or purple clover (Liu et al., 2009). Organic acid concentrations have been reported to be significantly lower in cultivated soils than those under native vegetation (Fox and Comerford, 1990; Hue et al., 1986). Exudation of LMW carboxylates is regarded as an important P acquisition strategy for plants in strongly P-fixing soils. The LMW organic acids could have solubilized part of the sparingly soluble P and improved the soil P availability ( Johnson and Loepertt, 2006; Onthong et al., 1999). Not only pristine forests (Schlesinger 1997) but also species rich in ecosystems are reported to have plants that are capable of exudation. Proteaceae vegetations are known to use exudation to mobilize sparingly soluble phosphate in severely phosphate-impoverished soils (Lambers et al., 2002; Lo´pez-Bucio et al., 2000; Sas et al., 2001). Exudation is also reported to play an important role in wetlands, as they probably live longer in O2-poor conditions (Mitsch and Gosselink, 2000). Crops that secrete LMW organic acids could release soil P continuously during growing period for meeting their requirement (Wang et al., 2007). Different organic acids have different capacity to release soil P (Lu et al., 1998), perhaps due mainly to their chelating effect and different functional groups (Kpomblekou and Tabatabai, 2003). Dinkelaker et al. (1989) showed that NO3-fed white lupine decreased rhizosphere soil pH in a P-deficient calcareous soil due to the excretion of a large amount of citric acid. The exudation of the organic acids depends on plant species or genotype cultivars. Li et al. (2007) observed a much higher amount of malate and citrate exuded into rhizosphere than maize and soybean from a Chinese cultivar of faba bean (V. faba L. cv. No. 5 Lincan), while Nuruzzaman et al. (2005) did not detect a significant amount of carboxylates in the rhizosphere in an Australian cultivar (V. faba L. cv. Ascot). Tang et al. (2007) showed that soybean released very small amounts of organic acids and that P mobilization from sparingly soluble P sources did not correlate

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

181

with the exudation of organic acids or rhizosphere pH. The release of P from soil is a process that ranges from fast to slow (Amer et al., 1955; De Smet et al., 1998; Elrashidi et al., 1975). In the Hedley method (Hedley et al., 1982), the soil P has been classified into seven fractions: H2O-P, NaHCO3-Pi, NaOH-Pi, NaHCO3-Po, NaOH-Po, HCl-P, and residual-P. The most active and most available inorganic P forms for plants are H2O-P and NaHCO3-Pi. The former is a solid inorganic P being balanced with soil solution P, and its amount and changes are closely associated with the transformation of various forms of different P pools (Qin et al., 2007; Xiang et al., 2004), while the latter is a fraction adsorbed in soil particle surface with a relatively strong activity. NaOH-Pi is firmly absorbed on the surface of Fe and Al compounds by chemical force with medium activity. The NaHCO3-Po is mainly the readily mineralized soluble organic P absorbed on the soil surface (Bowman and Cole, 1978; Hedley et al., 1982), whereas the NaOH-Po, i.e., the organic P formed by combination of Fe and Al with humic acid and pheophytin, has certain availability after action by phosphatase. Organic P can be transformed into NaHCO3-Pi after mineralization (Bowman and Cole, 1978). Being the form of P associated with calcium, the HCI-P is a relatively stable form in soil. In calcareous soils, P mainly exits in the apatite form, and in highly weathering soils the occluded phosphorus is also included in the HCI-P type. The nonextracted fraction by reagents is the residual-P that is more stable and relatively insoluble, and thus is difficult to be used by plants (Hedley et al., 1982). Liu et al. (2009) adopted different concentrations of citric and malic acids to study their effect on total P release from soil and the released amount from different P fractions. Results showed that the activation of soil P by the acids was a continuous and dynamic process, and the intensity of activation increases with the acid concentration while its rate decreased with time. The activation curve could be divided into fast, slow, and very slow parts, corresponding to ready-release P, slow-release P, and sparingly soluble P in the soil. The released amount by citric acid was larger than that by malic acid. The two acids could promote the release of inorganic phosphorus fraction (H2O-P and NaHCO3-Pi) and the mineralization of organic P fractions (NaHCO3-Po and NAOH-Po). Of the P fractions, NaOH-Pi was more influenced by organic acids followed by the HCl-P; NaHCO3-Pi and H2O-P could be released more with the rise of the acid concentration while there was no effect on residual-P. In concentrations of 0.5 and 1.0 mmol l 1 of citric acid and 1.0 mmol l 1 of malic acid, the released amount of iron oxide- and aluminum oxide-associated NaOH-Pi was larger than the calcium-associated HCl-P, while the calcium-associated HCl-P was larger than plant-available inorganic P (NaHCO3-Pi > H2O-P). The released amounts of soluble iron and aluminum were related to the acid addition, and significantly correlated with the phosphorus amount released.

182

Sheng-Xiu Li et al.

This suggests that iron oxide- and aluminum oxide-associated inorganic P (NaOH-Pi) should play an important role in phosphorus release. Van der Zee et al. (1987) found that, when crop took up P from soil solution, P absorption at the surface of iron- and aluminum oxides was fast initially and then slowly desorbed. Holfold and Ratrick (1979) held that in acid soil, the cause of soluble P increase was the reduction and dissolution of iron oxides and then the release of P bonded or occluded in the internal site of iron and aluminum oxides. What are the mechanisms that cause P release by the exudation of the organic acids? Different theories have been proposed. As a whole, the effects of carboxylates on enhancing P availability can be roughly grouped into two types: direct and indirect. The former refers to the immediate P release, including the blocking of P adsorption sites or ligand exchange of organic anion with PO43, a process in which an organic ligand exchanges inorganic P on the mineral surface points and makes P release to soil solution (He et al., 1994; Lopez-Hernander et al., 1986; Nagarajar et al., 1968) and also to the oxide dissolution by complexing Al or Fe held in minerals or mobilization of P held in metal–humic substances (Staunton and Leprince, 1996). The indirect P mobilization effects of exudation are far less known compared to the direct effects. In this type, both the stimulation of the microbial activity and a soil pH alteration, mostly a pH decrease, may be involved. When pH is decreased, the hydrolysis of part of the organic P may be possible (Ding et al., 2005). Disagreements exist as to whether citric acid is exuded in its acid form, and dissociates only once it is outside the cell environment, thereby causing acidification (Haynes and Mokolobate, 2001; Staunton and Leprince 1996), or in its conjugate base form citrate3, as cytosol pH is 7.1–7.3 (Hinsinger, 2001; Jones, 1998). In the latter case, the soil pH decrease is not caused by citric acid dissociation, but by the simultaneously excreted counteracting ion Hþ to maintain electrical neutrality. However, Sas et al. (2001) found that organic anion release was only partly accompanied by Hþ extrusion, and suggested Kþ as the alternative accompanying cation, as did Hinsinger (2001). For most organic acid release situations, the counter ions remain to be identified ( Jones, 1998). A pH increase ( Jones and Darrah, 1994; Jones and Kochian, 1996), in contrast, is reported, and fewer possible mechanisms are also proposed, except for the citrate3 equilibrium with citrate2 and Hþ that is not regarded as the only counter ion to adjust the anion—cation balance of the cell membrane. Haynes and Mokolobate (2001) named—next to the complexation of Hþ by citrate—microbial decarboxylation as a proton-consuming process. This effect is sometimes referred to as the self-liming effect in the context of organic residue amendment. The ligand exchange between the organic acid and soil solid phase hydroxyl groups may be another possibility for the pH increase (Yan et al., 1996).

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

183

Chelating dissolution of part of the phosphate combined with Fe and Al is another effect (Earl et al., 1979; Johnson and Loepertt, 2006; Johnson et al., 1996; Kirk et al., 1999). It results from the fact that the complexation of Al and Fe bound to soil humic substances by organic anions renders them more soluble, smaller in size, and more accessible to soil or root phosphatases. This facilitates enzymatic hydrolyzation of organically bound phosphorus that may be associated with these humic substances (Haynes and Mokolobate, 2001; Jones, 1998). Especially in soils containing a large amount of organic matter, this mechanism may be crucial because the organic matter is often associated with Al and Fe (Borie and Zunino, 1983). As long as organic P is bound with Al or Fe, it remains protected. Therefore, the limiting step in the utilization of the large organic P pool is not phosphatase activity but the release of Al and Fe from organic P (Otani and Ae, 1999). However, the importance of the chelating ability of organic acids in comparison with their acidifying ability is still under debate (Subbarao et al., 1997). Although our knowledge is limited about the mechanisms by which carboxylates improve P availability, the exudation yields a competitive advantage to plants in natural ecosystems. The carboxylate exudation as a major plant response occurs not only in the deficiency of P (Stro¨m et al., 2002) but also in a number of well-defined stress situations, including Al toxicity (Heim et al., 1999) and deficiency of Fe (Marschner and Ro¨mheld, 1994), Kþ (Kraffczyk et al., 1984), and O2 (anoxia) ( Jones, 1998). In the case of P deficiency, direct modification of the plant rhizosphere by carboxylate exudation is not the only possible P acquisition strategy, but other mechanisms can be used such as excreting phosphatases to release organically bound P and the provision of extra C (Lo´pez-Bucio et al., 2000) as a booster for microorganisms, which in turn can produce carboxylates as well as phosphatases ( Jones, 1998). Drouillon and Merckx (2003) reported that when both extractants were adjusted to soil pH prior to extraction, the amount of molybdate reactive P extracted by citric acid was larger than that extracted by CaCl2; the total P content of citric acid extracts far exceeded the total P content of CaCl2. However, the ratios of molybdate reactive P to total dissolved P remained constant. On adding citric acid (pH being adjusted to that of CaCl2 for eliminating release of P through acidification) to the CaCl2 extract, a significant pH increase was noted. Simultaneously, molybdate reactive P increased significantly in all but one soil. This suggests that the P availability enhancing properties of citric acid should be not only due to acidification of the plant rhizosphere but also due to its Al and Fe complexing capacity. Of the organic acids exuded from roots, amino acids have received great attention. Amino acids play a major role in plant physiology, and the contents and components of the essential amino acids have been regarded as main indices for evaluation of plant growth states (Huang and Liu, 1989;

184

Sheng-Xiu Li et al.

Mo, 1993). Many factors affect the secretion of amino acids. Nutrient application, particularly N rates and time applied, greatly influences their concentrations and components secreted from roots (Hong and Nie, 2003; Shen and Yan, 2001; Wang et al., 2001; Xu et al., 2003). Environmental factors such as temperature and light during the grain-filling stage influence their contents in seeds (Dai et al., 2006; Shi et al., 1995; Zhou et al., 1997). As an important component of root exudates, amino acids can alter the pH value and redox potential in the rhizosphere and thus influence the nutrient validity (Hong and Nie, 2003; Lu et al., 1998; Tu et al., 2000). Since components and concentrations of root exudations reflect plant growth status (Li et al., 2004; Ming et al., 2000, 2002), it is speculated that the decrease in amino acids in root exudations may be associated with the senescence of plants under nitrogen stress, while the changes in amino acids in root exudations may be related to the adaptability of plants to P stress. Chang et al. (2008) found 17 amino acids in rice root exudates during the grain-filling stage, and revealed that nitrogen stress significantly reduced whereas phosphorus stress significantly increased amino acids in root exudation, due mainly to the increase in concentrations of acidic and neutral amino acids, but reduction in contents of protein and total and essential amino acids. Many studies have been conducted on the adaptability of plants to P deficiency. One of the hot points focuses on the production of enzyme phosphatase from plants. The enzyme phosphatase is mainly produced by the roots of higher plants as well as by numerous microorganisms (Aspergillus, Penicillium, Mucor, Rhizopus, Bacillus, and Pseudomonas). Oliveira et al. (2009) found that T. rotundus was one of the highest phosphatase producers in the phytate and lecithin media. In soil, the organic phosphates occupy a large part of the total. The ultimate process by which organic phosphates are rendered available is by cleavage of inorganic phosphates by means of a phosphatase reaction, and the principle of the reaction is a process of hydrolysis (Mengel and Kirkby, 2001). Phosphatases are phosphate esterases that hydrolyze inositol phosphates, nucleic acids, and phosphoglycerates. It has been reported that in the modified Hedley method for P fractions (Bowman and Cole, 1978; Hedley et al., 1982), the NaOH-Po (organic P extracted by NaOH), which is a soluble organic P adsorbed in soil surface and formed by Fe and Al in combination with humic acid and pheophytin, can be mineralized (Qin et al., 2007). This form of P has certain availability after reaction by phosphatase (Liu et al., 2009). Liu et al. (2004) showed that the seedling roots of maize induced significantly alkaline phosphatase activities in the rhizosphere. Of the phosphatases, the APase has received considerable attention. It plays a great part in transfers of polyphosphate and organic P, and catalyzes phosphatide or phospholipids into inorganic P (Pi) and the corresponding fatty acid. The Pi produced thus can be reused by plants. Also, the APase in

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

185

outer layer cells of root tips and that released to the rhizosphere soil can participate in the decomposition of organic P, and therefore improve the soil P supply to plants. All these are beneficial to plant P nutrition. High activity of the APase in root systems can promote the hydrolysis of organic phosphate and thus improve the P availability in root microecological systems (Fan et al., 2001; Zhang et al., 2007) as well as the P uptake efficiency by plants (Duff et al., 1994). Ding and Li (1997, 1998) found that, in addition to leaf area, dry matter weight, and cumulative nitrogen content, the APase activity correlated significantly with grain and biomass yield of soybean under P-deficient conditions and could be used as one of the parameters for screening the ability of plants tolerant to low P at the seedling stage. It is known that the APase is an induced enzyme, and its activity is affected by P supply to plants (Ae et al., 1991; Wang et al., 1999a; Zhang, 2001). Low P-stressed conditions can induce plants to produce a large amount of the APase and increase its activity. For plants stressed by P deficiency, the activity of APase secreted from roots is significantly increased, and at early stages of P stress, the rise of the activity of APase occurs in roots, especially in root tips. With the increase of the P stress, its activity is significantly increased in plant stems and leaves (Wang et al., 1999a; Xu, 1989). The rise of the secreted amount of the APase and its activity is an adaptable phenomenon, reflecting the plant’s tolerance to P-deficient stresses and it is also a biological index for selection of P-tolerant varieties (Ding and Li, 1997; Wang et al., 1999a). Zhang et al. (2004) found that under P-deficient conditions, APase activities of three genotype varieties of wheat were all raised. APase activity is regarded as one of the adaptable mechanisms for P-deficient soils. However, some scientists consider that APase activity cannot be regarded as a special mechanism for the regulation of P deficiency but only as one phenomenon of plant’s adaptability to P deficiency, since it is not greatly related to plant’s tolerance to P deficiency (Liu et al., 1999; Wang et al., 1999a). Yuan et al. (2010) studied the relation of phosphatase activity to P deficiency with three cultivars of maize by water culture and found that, without addition of P to the solution, the phosphatase activity in root tissue was increased for all three cultivars while that in the intact root was different among the three cultivars (Table 12). Fan et al. (2001) studied the APase activities of intact roots and ground root tissues of maize grown in high and low P solutions and found that the APase activity in ground root tissue decreased from the root tip region to root base, whereas in base and middle root region it was influenced by medium P levels, decreasing with the increase of medium P concentration. In the case of intact root regions, APase activities along root axis from tip to base showed a similar changing trend as it did in ground tissue. However, the APase activities of intact roots were not influenced by the P levels. For these reasons, a conclusion was drawn that the APase secreted from roots and in intact root apoplast would not be

186

Sheng-Xiu Li et al.

Table 12 Effect of P deficiency on the APase activities in root tissue and in intact roots for three maize cultivars

Cultivar

Treatment

APase activities in root tissue (mmol P g FW 1 h1)

Liyu16

þP P þP P þP P

5.44 bc 6.91 a 4.45 cd 5.59 b 4.21 d 4.96 bcd

Jidan28 Kuancheng10

APase activities of intact root (mmol P g FW 1 h1)

0.37 ab 0.40 a 0.28 bc 0.27 bc 0.42 a 0.21 c

Measurement of phosphatase activity in intact root by Gilbert et al.’s (1999) and Guan’s (1983) method (wash plants or seedlings, use tissue to dry water on plants, and place the prepared plants into 100 ml solution containing 0.5 g l 1 p-NPP and buffered by PbAC. After 30 min of incubation, take out the plants, and add 5 ml NaOH solution of a concentration of 0.5 mol l 1. This solution was centrifuged for 20 min with 10,000 rpm at 4 C, and measured by colorimetry under 510 nm. Measurement of phosphatase activity in root tissue also by Gilbert et al.’s (1999) and Guan’s (1983) method (wash and dry roots as described above, cut roots from root crown, weigh the fresh roots’ weight and take 0.5 g root sample, ground roots in ice-bath and transfer to centrifugation tubes, centrifugalize for 20 min with 10,000 rpm at 4 C. The upper clear solution in the tube was used for measuring the activity). Modified from Yuan et al. (2010).

useful in improving maize organic P acquisition efficiency under a P-stressed condition. According to Setanac et al. (1980), root cell walls were high in the APase activity. In intact root systems growing in a sterile nutrient solution, the LMW organic P compounds contained, including lecithin, sodium phytate, and sodium glycerophosphate, were hydrolyzed at higher rates than necessary to meet the plant demand (Tarafdar and Claassen 1988). Similar findings have been reported by Dou and Steffens (1993). These authors found that the APase in soils was stimulated by the addition of inorganic and, especially, organic phosphates. Phosphatase activity and phosphate turnover were much higher in the rhizosphere compared to the bulk soil (Hetal and Sauerbeck, 1984; Tarafdar and Jungk, 1987). Of all root exudates, protons and organic acids are commonly regarded as decisive factors affecting P solubility. It is reported that legumes absorbed more cations than anions through N2 fixation and released more protons than other crops (Raven et al., 1990; Tang et al., 1998). In addition, some N2-fixing legumes can acidify rhizosphere even when nitrate-N is supplied (Marschner and Ro¨mheld, 1983). From these facts, we may deduce that the difference of such substances between leguminous and nonleguminous crops causes the different response to P fertilizers, and that legumes may have a higher ability to use soil P. However, no comparison has been made between legumes and nonlegumes under the same conditions. Also, as discussed above, the release of protons and organic acids in rhizosphere soil has been found not only in leguminous crops but also in nonleguminous

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

187

crops, and there is no evidence as to whether the legumes or nonlegumes released more protons or organic acids than the other. Furthermore, Zhou et al. (2009) visualized the rhizosphere acidification of faba bean, soybean, and maize as well as their intercropping combination in agar-gel film images, and Yuan et al. (2010) repeated this in three maize cultivars. From their images, the changes in acidification for faba bean and soybean in the rhizosphere were almost in the same range as the three maize cultivars, showing that the acidification by leguminous crops was similar as in cereal crops. No matter what kind of acids or how many protons are released, the reduction of pH is the final result. If the acidification of rhizosphere is the cause for leguminous and nonleguminous crops to have different ability in utilization of sparingly soluble P in soil, the lower the pH in rhizosphere soil declined by a crop, the higher the ability of absorption of such P should be. For examining this hypothesis, evaluation of pH decline under the same conditions should provide useful information. Li et al. (1992a) conducted sand culture experiments to study pH changes of some crops and found that buckwheat and soybean significantly acidified the environment, while radish, which had high ability to use P from RPs (see Table 4), significantly increased pH (Table 13). Also, they found that the extreme pH (Mattson, 1948) of some crop roots (determined by electrodialysis of roots to form Hþ-saturated roots and then placing the roots in 1 mol l 1 of potassium solution to measure the pH) was not in agreement with crop responses to RPs. For example, oats, a crop having weak ability to use P from RPs, have almost the same extreme pH as radish, buckwheat, and soybean, which have been regarded as crops having strong ability to use P from RPs (Table 14). Obviously, these determinations have not provided evidence to support the hypothesis proposed. The same results were also obtained by our experiments. During wheat and pea growing period, we collected soil samples in a field trial from both rhizospheric and bulk soil when the soil water contents were suitable for sampling. The roots were dug out with soil mass, and the large part of soil was carefully separated from the roots by slight shaking, and the very small Table 13

Effects of some crops grown in sand medium on solution pH pH in sand solution

Days after sowing

Radish

6 17 36 50

7.6 7.9 8.2

Modified from Li et al. (1992a).

Soybean

7.5 6.3 4.1

Buckwheat

3.7 3.3 5.3

188

Sheng-Xiu Li et al.

Table 14 Root extreme pH of some crops Crop

Root extreme pH

Radish Buckwheat Soybean Barley Oats

3.52 3.59 3.36 4.07 3.62

Modified from Li et al. (1992a).

Table 15 pH and available P in rhizospheric soil and bulk soil of wheat and pea Crop

pH Pea Wheat Available P (Olsen-P) (mg kg 1) Pea (adding 30 kg P ha 1 to soil) Pea (without adding P) Wheat (adding 30 kg P ha 1 to soil) Wheat (without adding P)

Rhizospheric soil

Bulk soil

5.59 6.55

7.57 7.58

5.2 3.2 4.8 3.7

6.6 4.2 6.5 4.4

Modified from Li et al. (1995).

part of soil still sticking on the roots was removed. The former was collected as bulk soil and the latter as rhizospheric soil. Analysis showed that pH in rhizospheric soil was greatly different from the bulk soil for both wheat and pea. The pH was generally one unit lower in the rhizospheric soil than in the bulk soil. However, there was no pH difference between wheat and pea either in rhizospheric soil or in bulk soil. The available P (Olsen-P) had the same trend as soil pH: it was lower in the rhizospheric soil but higher in the bulk soil. Without application of P fertilizer, the Olsen-P in the rhizospheric soil was almost 1 mg kg 1 lower than in bulk soil, while with application of P fertilizer it was about 1.5 mg kg 1 lower. Again, there was no difference of Olsen-P between wheat and pea either in rhizospheric or in bulk soil. T-test for significance showed that irrespective of whether P fertilizer was applied or not and whether it was rhizospheric soil or bulk soil, the utilization of P from both soil and the P fertilizer showed no significant differences at the 0.05 statistical levels. These results show clearly that the acidification of rhizosphere soil cannot differentiate the ability of crop responses to P fertilizer (Table 15).

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

189

7. Effects of Root Cation Exchange Capacity and Calcium Uptake Amount of Crops on Soil P Absorption and Crop Responses to P Fertilizer In 1916, a French chemist Devaux reported that plant roots had CEC produced by pectinase in the cell walls of root hairs (Mehlich and Drake, 1955, see Chinese version translated by Yuan et al., 1959). Later researchers found that 70–90% CEC was produced from the free carbonyl and the remainder from the proteins of cell walls composed of pectin, cellulose, hemicelullose, and from the protoplasm (He and Meng, 1987). The CEC can be measured by titration of the salt (generally using KCl) solution in which the fresh electrically dialyzed roots were placed with NaOH till pH 7. The root CEC was once regarded as the basis for crops to exchange cations with those held in soil colloid. The higher the capacity, the higher the possibility for crops to take up cations from soil. Crops with high CEC not only have high ability to take up more calcium from soil, but can also liberate phosphorus that is bound with calcium so it can be absorbed by them. In addition, crops with high CEC generally have high anion exchange capacity and this may help plants absorb P from soil. Some results show that the root CEC and the bond energy of the root surface to calcium are the main physiological properties for plants to use rock phosphate P (Peng, 1990; Peng and Pei, 1980; Sun and Wang, 1980). Li (1966a,b) and Li et al. (1992a) conducted sand culture experiments and measured the root CEC of nine crops that were planted in sand medium for 2 months. The results showed that the root CEC was related to the crop’s ability to absorb P. When the root CEC was in the range from 9 to 15 cmol kg 1 dry roots (weight) for small grain crops such as rice, wheat, barley, oats, and millet, the P-absorbing ability was weak. However, when roots with CEC from 15 to 20 cmol kg 1 dry roots for maize, 35–60 cmol kg 1 dry roots for leguminous crops (soybean, alfalfa, vetch, lupine, rice bean), 40 cmol kg 1 dry roots or so for buckwheat, and 40–50 cmol kg 1 dry roots for radish, the crops had high ability to use P from RP (Table 16). Furthermore, plants having high bond energy with calcium and high Hþ activity in the root surface also had high capacity to use P from RP, but the mechanism needs further study (Peng, 1990; Peng and Pei, 1980; Sun and Wang, 1980). In order to identify the relationship between plant CEC and the P uptake amount, we determined the root CEC of both wheat and pea of the same age (same growing date) using water culture plants. The results showed that the CEC of pea was 69.4 cmol kg 1 dry root, whereas that of wheat was 13.5 cmol kg 1 dry root. Although the root weight and root total surface areas of pea were smaller than those of wheat, the root CEC

190 Table 16

Sheng-Xiu Li et al.

Root cation exchange capacity (CEC) of some crops

Crop

Root CEC (cmol kg 1 dry root)

Barley Tomato Oats Soybean Ryegrass Radish Lucerne Spinach Buckwheat Cabbage Rice bean

9.5 33.3 12.9 35.2 14.3 36.7 25.5 40.0 29.6 40.8 31.5

Modified from Li et al. (1992a).

was several times higher than that of wheat. Calculated with their root dry weight, the actual root CEC of pea was 2.80 cmol per pot, while that of wheat was 1.64 cmol per pot on average. The former was still almost double the latter (Table 17). However, the P uptake amounts by the two crops were not as striking as those demonstrated in the section on P uptake amount. In relation to root CEC, the uptake of calcium by crops was considered as a mechanism for P release and P uptake. The reason is simple. In neutral to calcareous soils, the concentration of phosphate in soil solution is governed mainly by the formation and dissolution of calcium phosphate, which in turn depends on soil pH and Ca2þ concentration in the soil solution. The lower the Ca/P (in weight) ratios in soil, the higher the solubility of the sparingly soluble P in water (Mengel and Kirkby, 2001). To liberate P from such compounds, a reduction of calcium is required. Provided that a crop can absorb a large amount of Ca from the calcium phosphates, the Ca/P ratio in these P compounds will be reduced, and P will be released. The released P will be taken up by the crop. In this case, even if the soil-available P is insufficient, the crop may still obtain adequate amount of P from the soil. The utilization of Ca by plants to liberate P from RPs is just the same as factories use RPs for the production of P fertilizers. Different crops have different requirements of Ca and P. For expressing the relations of crop uptake amounts of P to Ca, the CaO/P2O5 ratio in plant tissue was traditionally used in the literature. For comparison, in this chapter, we have still used the CaO/P2O5 for discussion, but we have converted the CaO/P2O5 ratio into Ca/P ratio and retained the two ratios in the relevant tables for reference. As early as in 1950, a Russian scientist, Cherekov (F. B. Чиpикoв), found that crops having a ratio of CaO to

191

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

Table 17 Root cation exchange capacity (CEC) of wheat and pea

N treatment

Wheat Without N With N addition Pea Without N With N addition

Soil no.

Root weight (g pot 1)

Root CEC (cmol kg 1 dry root)

Root total CEC (cmol pot 1)

62001 62002 62001 62002

14.4 7.8 17.0 9.1

135 135 135 135

1.944 1.053 2.295 1.229

62001 62002 62001 62002

5.6 2.8 3.7 4.0

694 694 694 694

3.886 1.943 2.568 2.776

Modified from Li et al. (1995).

P2O5 (in weight) higher than 1.3 (or 0.397 in Ca/P ratio) in their tissue could take up P from RP, while those with a ratio lower than 1.3 could not utilize P from RP (Sokolov or Coкoлoв, 1950 in Russian, and 1959 in Chinese). This means that plants strongly needing calcium had a high ability to use P from RP. However, the hypothesis of CaO/P2O5 ratio has not been proven in the past 60 years, and in some cases the results were even in conflict with the ratio. It has been shown that not only different crops had different proportions of CaO and P2O5 demand, but the same crop had different requirement of CaO and P2O5 at different growing stages as well. Tomato is a typical example. It absorbs CaO and P2O5 almost at the same amount (the ratio of CaO to P2O5 being 1.13) at transplantation time, while the amount of CaO taken up was six times higher than P2O5 (the ratio of CaO to P2O5 being 7:1) at the bud formation and flowering stages (Asarov, 1959). This means, the higher the CaO to P2O5 ratio in their tissue, the lower the P uptake from the RP. Also, the CaO/P2O5 ratio in cotton (Gossypium hirsutum L. and other species) was much higher than in sweet clover [Melilotus alba, Melilotus indica (L.) All., Melilotus officinalis Lam., and other species], but its ability to use P from RP was much lower than of sweet clover (Peng, 1990; Peng and Pei, 1980). Li et al. (1992a) determined the ratio of CaO to P2O5 for some crop shoots (Table 18) and found that the CaO/P2O5 ratios could not be used as indices to differentiate the crop’s ability to use P from RPs. For examples, oats and ryegrass, which had been regarded as crops with weak ability to use the sparingly soluble P, had a high CaO/P2O5 ratio in their shoots (around 7 or so), while buckwheat, regarded as a crop with strong ability to absorb the sparingly soluble P, had a low ratio, being only 1.1.

192

Sheng-Xiu Li et al.

Table 18 CaO/P2O5 and Ca/P ratios in some crop tissues P2O5/CaO ratio (original)

P/Ca ratio (converted)

Crop

CaO (%)

P2O5 (%)

P2O5/CaO

Ca (%)

P (%)

P/Ca

Radish Buckwheat Lucerne Barley Oat

1.64 0.77 2.10 1.12 0.76

0.80 0.67 0.87 0.15 0.11

0.49 0.81 0.32 0.14 0.14

1.17 0.55 1.50 0.80 0.54

0.35 0.29 0.38 0.07 0.05

0.30 0.53 0.25 0.09 0.09

Modified from Li et al. (1992a).

However, Li et al. (1992a) also found that, if the ratio of P2O5 to CaO was used to show how much P2O5 would be taken up by the uptake of one unit CaO in the case of the RP being the sole source of P, the crops having strong ability to use P from RP such as radish, buckwheat, and soybean could absorb 0.32–0.81 unit of P2O5 by taking up a unit CaO, while those with low ability only took up less than 0.2 units of P2O5 by taking up a unit CaO in weight. This shows that there exist some relation between P2O5 and CaO uptake (Table 19). Considering this aspect, the root CEC and the ratio of P2O5 to CaO in plant tissue may be associated with the crop’s ability to absorb the sparingly soluble phosphate from the soil. However, whether judging their ability to use P from ground RP or from their CEC, no reason has been given to explain the difference in leguminous and nonleguminous crops’ response to P fertilizers. Tao (1983) related the CaO/P2O5 ratio in plant tissue to the crop’s responses to F fertilizer. In pot experiments, he planted 17 crops, including faba bean, mung bean (green gram, golden gram, mash bean, or green soy) [Vigna rabiata (Linn.) Wilczek], soybean, soybean (in black color), pea, lentil (Lens culinaris Medik), sesame (Sesamum indicum), flax (Linum usitatissimum L.), scarlet runner bean [Phaseolus coccineus (L.)], wheat, oats, buckwheat, qingke barley, proso (broom corn millet or panic grass) (Ipomoea species), (foxtail) millet, maize, and sunflower. Each plant was treated in two ways: with and without application of P fertilizer. The N, P, and Ca amounts taken up by plants were determined, and the Ca/P (CaO/P2O5) and N/P ratios were used to correlate with dry matter increase (%) and P uptake increase (%), respectively. He also correlated the P uptake increase (%) with N uptake increase (%), Ca uptake increase (%) with P uptake increase (%), and Ca uptake increase (%) with N uptake increase (%) (Table 20). All the increases in percentage were calculated from the comparison of P addition with the control. The results showed that the correlation coefficient was 0.4169* (*indicating significant at P < 0.05 level) for Ca/P versus dry matter increase, 0.5327* for N/P versus dry matter increase, 0.4089 for Ca/P

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

Table 19

193

The CaO/P2O5 and Ca/P ratios in shoots of some crops

Crop

CaO/P2O5 (original)

Ca/P (converted)

Radish Barley Buckwheat Oats Rice bean Ryegrass Soybean Tomato Lucerne Spinach

2.1 1.1 1.1 6.9 7.8 7.3 2.4 7.3 3.1 7.4

6.87 3.60 3.60 22.57 22.51 23.88 7.85 23.88 10.14 24.21

Modified from Li et al. (1992a).

versus P uptake increase, and 0.4403* for N/P versus P uptake increase. Also, the coefficient was 0.8690** (**indicating significant at P < 0.01 level) for P versus N in increased %, 0.6956** for Ca versus P in increased %, and 0.7243** for Ca versus N in increased %. From these results, it seems that the CaO/P2O5 ratio can to a certain extent reflect the crop’s response to P fertilizer.

8. The Responses to P Fertilizer Between Leguminous and Cereal Crops with Their Biological Characteristics As mentioned above, all reasons or hypotheses proposed so far (including crop P uptake amount, crop responses to RP, crop root morphology, crop root exudates, crop root CEC, and the CaO/P2O5 in plant tissues) cannot explain the difference in response to P fertilizer between leguminous and cereal crops. Also, the reasoning that leguminous crops have higher ability to use P from the sparingly soluble phosphate than cereal crops is in conflict with the fact that, if the leguminous crops really have high ability to use P from the sparingly soluble phosphate, they would take up more P from soil, and when grown in soil deficient in available P supply, they would show no or less P deficiency than cereal crops. As a direct result, the sensitivity of leguminous crops to P fertilizers should be lower and they should have less response to P fertilizers than cereal crops. The high sensitivity of leguminous crops to P fertilizer just shows their low rather

Table 20

Relation of CaO/P2O5 or Ca/P and N/P ratios in plant tissue to crop responses to P fertilizer Dry matter (g pot 1)

Crop

Without P

With P

CaO/P2O5 Ca/P (original) (converted) N/P

Faba Mung bean Soybean (yellow) Soybean (black) Pea Lentil Sesame Flax Scarlet runner bean Wheat Oat Buckwheat Qingke barley Proso Millet Maize Sunflower

11.3 60.6 90.6 86.5 10.4 3.6 23.8 6.6 57.2 13.6 27.7 39.5 12.8 60.5 52.2 44.5 44.3

13.4 74.8 148.2 158.8 12.8 5.6 24.2 7.1 53.2 14.8 28.6 48.7 15.2 59.7 45.4 66.7 66.7

6.75 7.99 6.71 11.81 6.84 4.74 7.89 3.78 10.57 1.60 3.21 8.69 1.84 3.38 2.36 3.83 10.67

Modified from Tao (1983).

2.06 2.44 2.05 3.61 2.09 1.45 2.41 1.15 3.23 0.49 0.98 2.65 0.56 1.03 0.72 1.17 3.26

5.52 8.62 8.83 11.86 8.16 8.78 3.86 8.69 5.58 7.99 4.81 3.96 8.10 3.04 3.47 4.37 7.85

Yield or nutrient uptake increase (%) by P addition Dry matter

P

N

Ca

18.6 23.4 63.5 83.6 22.8 56.7 1.6 7.0 7.0 8.6 3.2 23.1 18.8 1.4 12.9 71.3 50.1

47.9 73.9 36.4 119.9 34.4 62.7 2.8 2.4 13.4 6.1 20.8 39.5 10.7 5.1 17.9 113.1 34.3

18.7 43.3 90.1 140.8 22.2 64.1 0.3 2.8 9.1 13.2 30.1 11.2 2.4 9.8 4.4 122.3 9.0

41.7 33.3 69.2 81.2 23.5 82.4 17.0 13.3 6.0 59.4 10.4 2.3 43.8 19.5 3.7 93.9 66.9

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

195

than high ability to use P from soil. For such considerations, the difference of leguminous and cereal crops to P fertilizers needs to be examined from other aspects. Considering the biological properties, one important difference is that the leguminous crops have nodules with N fixation bacteria and can fix N from the atmosphere, while the cereal crops have no such property. Due to this characteristic, the leguminous crops can obtain N from the atmosphere and their responses to P fertilizer are not restricted by soil N deficiency in any case, whereas cereal crops are often stressed by soil N limitation and this in turn affects their responses to P fertilizer. Is it the cause for leguminous crops to have higher responses to P fertilizer than the cereal crops? Some reports have shown that application of P fertilizer can promote nodule formation and increase the N fixation ability (Guo, 1981), and that P fertilization can regulate the ratio of N to P in soil for leguminous crops to have vigorous growth, and therefore it is more important to apply P fertilizer to leguminous crops than to other crops. Also, N fixation of leguminous crops can influence the root CEC, thus influencing the P uptake amount (Extension Station of Agricultural Technology in Bijie District, Guizhou Province, 1964). Consequently, leguminous crops with a high N to P ratio in their tissue may have strong ability to absorb P from soil as well as from P fertilizer (Xie, 1966). As mentioned, Tao (1964b, 1983) used the ratio of N to P2O5 in crop tissue to indicate crop’s responses to P fertilizer, and concluded that, when the ratio was higher than 3, crops would have good responses to P fertilizer (Tao, 1964a). However, using the ratio as a criterion, there should be almost no difference between leguminous and cereal crops in response to P fertilizer (Tao, 1983), but this is not the case. To seek for direct evidence supporting N fixation on leguminous crop responses to P fertilizer, we carried out a series of experiments from different aspects.

8.1. Soil fertility for planting leguminous and cereal crops Agricultural practices have shown that it is not always the case that leguminous crops have good responses to P fertilizer, but it is so only in a soil that is deficient in both N and P nutrients. Traditionally, in China, leguminous crops have been used for improving soil structure and building up soil fertility, and are often planted on unfertile lands that are deficient in N supply. China has a long history of use of leguminous crops for such purposes. As early as in the sixth century AD, a book entitled Qi Min Yao Shu (Manual of Important Arts for the People) written by Jia Sixie, recorded that “of the methods for making soil fertile, planting mung bean was most effective, followed by lentils and fenugreek (Trigonella foenum-graecum Linn.). All the crops were seeded in May or June, and plowed to bury them in July or August. Using these crops as green manure for spring

196

Sheng-Xiu Li et al.

cereals, the cereals could harvest 10 dan (a Chinese unit, 1 dan ¼ 50 kg) per mu (A Chinese unit, 1 mu ¼ 1/15 ha), and the effect was as good as application of silkworm excreta or decomposed manures” (You, 1983). Certainly, the record should be much later than the beginning time for leguminous planting. From the record, we can estimate that Chinese have used leguminous crops for fostering and nurturing their lands for at least more than 2500 years. During the long period of the cultivation process, leguminous crops have been used in two ways. Use as green manure and forage crops is one, and rotation systems for producing beans is the other. In either case, the leguminous crops were planted in lands that had almost no productivity for continuous cereal cultivation due mainly to serious deficiency in nutrients, especially N. These lands are generally distributed in remote areas far away from the village where farmers were living. As a cultivated pattern, farmers applied manure to the lands close to the village for saving energy and labor since roads were bad and therefore transfer of manure to the remote sites was difficult in the past. As a result, a special nutrient cycling was formed. The lands around the village were fertilized with manure, while those far away from the village were rotated with leguminous crops for a short time. In this way, P and other nutrients in the remote sites were allowed to migrate to the lands near the village via the transfer of harvest products, while deficiency of this nutrient in the remote sites became more and more serious. At the time of use of leguminous crop rotation, these lands had been cultivated with cereals for quite a long time, and were unsuitable to grow such crops continuously. Such unfertile lands were extremely deficient in both N and P. Cereal crops, due to the limitation of N deficiency, could not utilize P fertilizer, while leguminous crops could. For recovering soil fertility, the growing of cereal crops was stopped, whereas leguminous crops were cultivated for this purpose. In this case, P function and leguminous crop characteristics could be fully utilized. This may be the reason why Chen et al. (1963b, 1987) concluded that “if P fertilizer was not applied to leguminous crops but to nonleguminous crops, the effect would not be obvious, and it would be impossible to have good results when P fertilizer was not applied to leguminous but to cereal crops or to fallow fields.” This conclusion was reached in the 1950s when N fertilizer was in short supply and rate of N applied was extremely low. However, in a soil with sufficient N supply and deficient in P supply, such as fields growing alfalfa for many years and those that only depend on leguminous crop rotation or on application of N fertilizer alone to support a certain harvest, the difference of crop response to P fertilizer between leguminous and cereal crops would be diminished. Furthermore, with application of large amounts of N fertilizers, the situation of the deficiency of N in remote sites has been changed and cereal crops’ response to P fertilizer is much more obvious in these lands at present compared to the 1950s. In some cases, the response of cereal crops to P fertilizer is even better than that of leguminous crops.

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

197

8.2. Effect of addition of N on crop uptake of P from soil If the difference of leguminous and cereal crop responses to P fertilizer is caused by N supply, then addition of N fertilizer to cereal crops might influence the crop’s ability for absorption of sparingly soluble P from soil. For testing such a hypothesis, a soil seriously deficient in available P was selected to conduct a pot experiment. Extracted with Olsen method, the available P of the soil was only 2.4 mg P kg 1. In such a soil extremely deficient in available P, crops taking up the sparingly soluble P from soil would not be affected by soil P supply and their ability could be fully utilized, and thus an objective evaluation of the P uptake ability between leguminous crops and cereal crops could be correctly conducted. Mitscherlich-type pots were used and filled with 8 kg of soil. The experiments were done twice: one used summer-sown crops and the other used autumn-sown crops. For the former, maize and soybean were adopted, and one plant of maize was planted in each pot compared to three plants of soybean. Nitrogen fertilizer (0.2 g kg 1) was applied to maize but not to soybean. The plants were harvested after they grew for 60 days. For the latter study, wheat and pea were used. Twelve plants of wheat and eight plants of pea were cultivated in each pot. For wheat 0.15 g N kg 1 soil was added, while for pea it was 0.02 g N kg 1 soil. The plants were harvested at maturity. The results (Table 21) showed that, under such conditions of different plant density and biomass, the summer-sown crops maize and soybean took up 7.12 and 7.20 mg P per pot, respectively, i.e., almost the same amount of P from the soil. For the autumn-sown crops, without application of N fertilizers, pea took up 44.75 mg P per pot, and wheat took up only 31.70 mg P per pot, the former being higher than the latter by nearly 33%. However, on fertilization with N, wheat and pea showed no difference in P uptake amount: wheat took up 44.75 mg P per pot and pea 44.56 mg P per pot. Compared to plants grown without N fertilizer, the P amount taken up by pea was no different, but that by wheat was greatly increased. No matter whether we consider the above-ground parts or the shoots plus roots, the amount of P uptake was the same for both pea and wheat when N fertilizer was added. This shows that the two crops have no difference in the absorptive ability of P from soil.

8.3. Effect of N fertilization on crop responses to P fertilizer Numerous phenomena imply that the sensitivity of leguminous crop’s response to P fertilizer is related to their N fixation. The N and P nutrition of crops are closely related. For cereal crops, if N supply is not sufficient, it is difficult for P fertilizer to play its role (Chen et al., 1963a; Fertilizer laboratory, Soil and Fertilizer Institute, Shandong Province, 1976; Institute of Plant Protection and Soil Fertilizer, Hebei Province, 1975; Li,

Table 21

Phosphorus uptake (mg P pot 1) by summer- and autumn-sown crops from soil deficient in available P (pot experiments) Ground part (root)

Crop

Weight (g pot 1)

P content (%)

Aboveground part P uptake (mg pot 1)

Summer-sown crop Soybean (without N addition) Maize (with application of 0.2 g N kg 1) Autumn-sown crop Without application of N Pea 2.75 0.19 5.23

Organ

Total Total

Weight (g pot 1)

2.59 3.53

Leaf–stem 9.26 Seed 9.45 Total 18.71 Wheat 3.01 0.10 3.01 Leaf–stem 9.66 Seed 6.58 Total 16.24 1 1 With application of 0.15 g N kg for wheat and 0.01 g N kg for pea Pea 3.14 0.16 5.02 Leaf–stem 14.59 Seed 9.61 Total 24.20 Wheat 4.64 0.13 6.03 Leaf–stem 13.72 Seed 14.77 Total 28.49 Modified from Li et al. (1995).

P content (%)

P uptake (mg pot 1)

0.275 0.204

7.12 7.20

0.09 0.33

8.33 31.19 39.52 5.00 23.69 28.69

0.05 0.36

0.08 0.29 0.03 0.24

11.67 27.87 39.54 4.12 35.45 39.57

Total P uptake (mg pot 1)

44.75

31.70

44.56

45.60

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

199

1962, 1981, 1983; Li et al., 2009; Lu and Jinag, 1962; Soil Science Institute, Liaoning Academy of Agricultural Sciences, 1974; Yuan et al., 1962). In contrast, under the condition of sufficient N supply, the N amount taken up by crops is increased. Accompanied with N increase, P uptake amount is correspondingly increased. Various field experiment results show that no matter whether leguminous crops or cereal crops are used, N and P in plants are increased at a certain rate, and both N and P are highly related. It should be noted that even in a soil deficient in N and P supply, application of N fertilizer to cereal crops and no N or a small amount of N to leguminous crops so that both cereal and leguminous crops have almost the same supply of N level during their growing period (Li and Li, 1992; Li et al., 1992b), the cereal crops had much better responses to P fertilizer and took up a higher amount of P than the leguminous crops (Table 22). This shows from another aspect that the sensitivity of leguminous crops’ response to P fertilizer is related to its N fixation.

8.4. The response of leguminous crops to P fertilizer before and after nodulation and N fixation The phenomena mentioned above only provided indirect evidence. For providing direct evidence, the best way is to use a leguminous crop that has no nodules for fixing N and compare it with cereal crops. However, since it is difficult to find out such a leguminous crop that has been used in agriculture, the study can only be carried out in an indirect way. Determination of leguminous response to P fertilizer before and after nodule formation and N fixation is one way to show the role of N in promoting the role of P. Leguminous crops do not fix N from atmosphere at all times, but only when nodules are formed and N fixation function can be performed (this can be seen from the red color nodules). Before nodules are formed and N fixation function is implemented, they do not have the ability to fix N from the atmosphere. For this reason, comparison of leguminous crops’ response to P fertilizer before and after formation of nodules and N fixation function can give some indications about N function in the promotion of P roles. If the response of leguminous crops is caused by N fixation, then the sensitivity of leguminous response to P fertilizer will be played only after N fixation is performed, and the response to P fertilizer of leguminous crops before N fixation should show no great difference than with cereal crops. Based on such considerations, two autumn-sown crops, wheat and pea, in the pot experiment with soil 86002 (see Table 22) were treated with and without P application either with N addition or without. Results (Table 23) show that, in such a soil deficient in available P (2.5 mg kg 1 of Olsen-P), the P uptake amount for the control (without application of P fertilizer) was approximately equal for both wheat and pea before N fixation (the first harvest), and they had almost

Table 22 The responses of leguminous and nonleguminous crops to phosphate fertilizer with N addition to soils having different Olsen-P levels NþP

N Soil No.

Weight (g pot 1)

P uptake (mg pot 1)

Autumn-sown crop: Wheat 85001 53.82 110.1 85002 23.49 39.6 Autumn-sown crop: Pea 85001 39.86 123.3 85002 24.20 39.6 Summer-sown crop: Soybean 87001 5.86 17.8 Summer-sown crop: Maize 87001 8.67 15.6

Increase amount by P

Increase (%) by P

Weight (g pot 1)

P uptake (mg pot 1)

Weight

P uptake

Weight (g pot 1)

P uptake (mg pot 1)

55.82 50.61

130.7 117.0

2.0 22.1

20.6 77.4

3.7 77.6

18.1 195.5

39.41 42.82

131.1 122.5

0.5 18.6

7.8 82.9

1.3 76.9

6.3 209.3

20.42

57.0

14.6

39.2

249.1

220.2

30.80

58.6

22.1

42.0

254.9

353.0

Soil 85001 was fertile, containing 21.7 mg Olsen-P kg 1, while 85002, an unfertile soil, containing 3.6 mg Olsen-P kg 1. Soil 87001 containing 6.7 mg Olsen-P kg 1. Modified from Li et al. (1995).

201

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

Table 23 Pea response to P fertilizer before and after fixation of N in comparison with wheat under conditions without and with N addition (pot experiment with soil 86002, see Table 8) Plant sampling (d/m)

Pea 7/3 15/4 2/5 20/5 1/6 Wheat 7/3 15/4 2/5 20/5 1/6

P uptake (mg pot 1) without N addition Without P

With P

P uptake increase (%) by P

2.01 3.71 8.91 15.50 19.56

3.67 29.78 38.17 43.05 54.15

2.89 4.67 12.62 14.63 13.84

22.58 28.34 31.75 31.83 38.56

P uptake (mg pot 1) with N addition Without P

With P

P uptake increase (%) by P

83 703 328 178 177

1.62 5.85 13.88 17.54 19.48

4.14 14.98 33.93 47.82 52.79

156 157 144 173 170

681 507 152 118 179

2.71 6.94 13.10 16.59 19.91

24.85 48.60 53.10 54.15 55.11

816 600 305 226 178

Modified from Li et al. (1992d).

the same pattern of P uptake. With application of P fertilizer, the P uptake amount was significantly increased for both wheat and pea. However, before N fixation (the first harvest), the P uptake amount by P input was much lower for pea than for wheat, being 681% increase for wheat while only 83% for pea. In contrast, after N fixation was performed (after the first harvest), P input increased P uptake of pea by 703%, while it decreased that of wheat to 507%. Since then, P uptake in pea increased significantly than in wheat, and at harvest pea absorbed 19.56 and 54.15 mg P per pot, while wheat absorbed only 13.84 and 38.56 mg per pot under conditions without and with P addition, respectively. The results mentioned above were the case without N addition. With N addition, due mainly to reduction in N fixation capacity, pea depended largely on soil N supply, and therefore P input effect was increased before N fixation, whereas it decreased after N fixation. Wheat, however, significantly increased P uptake and absorbed almost the same amount of P as pea at harvest. The P uptake of pea in relation to N fixation of pea could also be seen from the N amount in the entire plant (including that in roots) before and after N fixation. It was found that P addition significantly promoted N fixation of pea, especially while N fertilizer was not added. At the beginning of the N fixation stage, there was a sharp flush in N increase by P addition.

202

Sheng-Xiu Li et al.

For example, with no N addition, the N amount in pea plant was 58 mg per pot before N fixation and 83.5 mg per pot after N fixation in the case without P input. With N increase in plant tissue, the P uptake amount increased from 2.01 to 3.71 mg per pot, and there was no significant flush. However, with P addition, the N amount sharply increased from 83.5 mg per pot before N fixation to 438.8 mg per pot after N fixation; accompanied by N increase in plant tissue, the P uptake amount also flushed from 3.71 to 29.78 mg per pot. The response of pea to P fertilizer after fixation of N strongly reveals the great contribution of N fixation by leguminous crops to P fertilizer response.

8.5. Responses of leguminous crops to P fertilizer after inhibition of nodule formation Application of N fertilizer can inhibit nodule formation and N fixation of leguminous crops (Li et al., 1990). This is called “N depression” to leguminous crops (Agegnehu et al., 2005; Chen and Chen, 2004; Izaurralde et al., 1992; Jensen, 1996; Santalla et al., 2001; Xiao et al., 2004). This provides another way to study the sensitivity of the response of leguminous crops to P fertilizer. Two sub-manural soils seriously deficient in P and N supply were used for pot experiments, and 8 kg of soil was added to each of the Mitscherlich pots. Summer-sown crops (maize and soybean) as well as autumn-sown crops (wheat and pea) were planted. Four treatments were applied: without application of fertilizer (control) as well as with application of 0.2 g N kg 1, 0.05 g P kg 1, and 0.2 g N plus 0.05 g P kg 1, using calcium superphosphate and urea as P and N source, respectively. The purpose for application of the N fertilizer was to depress the fixation of N by leguminous crops so that they could depend on soil N supply just as cereal crops and therefore their responses to P fertilizer could be judged under the same condition of dependence on soil N status. Results show that, taking the average of two soils for the autumn-sown crops, pea took up 78 and 93 mg P per pot from the control (without P addition) and P addition treatments, respectively, while wheat took up only 32 and 38 mg P per pot, respectively. With application of N fertilizer, the nodule formation of pea was almost completely depressed, and the P uptake amount by pea (54 mg P per pot) was not greatly different from wheat (58 mg P per pot). However, with both N and P applications, pea took up 98 mg P per pot, and wheat 136 mg P per pot, the latter being much higher than the former. The same trend was also found for maize and soybean. Sterilizing leguminous seeds and planting leguminous and cereal crops in soil without Rhizobium infection can objectively evaluate whether the leguminous crop response to P fertilizer is caused by N fixation. For this purpose, we sampled the old-cultivated layer of the manural soil to conduct a pot experiment. According to a previous experiment, soybean grown in

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

203

this soil had no nodules. The soil was extremely deficient in available P, and the Olsen-P determined was only 2.1 mg P kg 1. Maize and soybean were planted. For maize, there were two treatments: without application of P fertilizer (control) and addition of P; and soybean was treated with inoculation and without inoculation both with and without application of P fertilizer. For the inoculation treatment, soybean seeds were inoculated prior to planting with commercial peat inoculants. Results showed that, without inoculation, soybean took up 9.3 mg P per pot and maize 10.6 mg P per pot without P addition, whereas with P addition soybean took up 16.4 mg P per pot and maize 15.6 mg P per pot. In both cases with and without application of P fertilizer, there was almost no difference in P uptake for the two crops in P treatments. However, when soybean was inoculated, the P uptake amount increased to 12.6 mg per pot in case without P addition and 24.9 mg P per pot in case with P addition. Inoculation of soybean made its dry matter increase by 40% and P uptake by 36% in the case without P addition, while the corresponding figures were 64% and 52% in the case with P addition. The inoculated soybean increased N content as well as P uptake. As a result, its uptake of P amount was much higher than maize, and also higher than soybean that was not inoculated (Table 24). This clearly indicates the importance of N fixation to leguminous crops responses to P fertilizers.

9. Factors Affecting the Responses of Leguminous and Nonleguminous Crops to P Fertilizer 9.1. Relationship between nonleguminous crop responses to P fertilizer and soil-available P Numerous experiments have found that nonleguminous crops’, especially cereal crops’, response to P fertilizers is not determined by soil organic matter, total N, total P and soil texture, soil CaCO3 contents, or soil pH, but by the soil-available P, which remarkably reflects the soil’s P-supplying capacity and greatly influences the effect of P fertilizer (Li, 1965, 1981, 2007; Li et al., 1978, 1979a); if soil contains abundant available P, crops will be able to obtain P from the soil for their needs, and therefore will have no response or only a small response to P fertilizers. However, in a soil with a low available P, the application of P fertilizer greatly increased the crop yield. Of the procedures for extraction of the available P, Olsen-P (extracted by 0.5 mol l 1 NaHCO3 solution) and Machigin-P (a Russian method with extraction of available P by 1% (NH4)2CO3 solution) methods were once widely used in China for the calcareous soil (Zhou, 1980, 1987). Comparison has been made between the two methods, and results showed

Table 24

Responses of maize and soybean with inoculation and without inoculation to phosphate fertilizer

Dry matter (g pot 1)

N uptake (mg pot 1)

P uptake (mg pot 1)

Without inoculation

Without inoculation

Inoculation

Without inoculation

79.8

141.0

110.3 43.6

Inoculation

Soybean without P addition 4.51 6.32 Soybean with P addition 6.87 11.24 Increase (%) by P addition 52.3 77.8 Maize without P addition 4.73 Maize with P addition 6.60 Increase (%) by P addition 39.5 Modified from Li et al. (1995).

Inoculation

Dry matter increase by inoculation (%)

P increase by inoculation (%)

9.3

12.6

40.1

35.5

179.3

16.4

24.9

63.6

51.8

27.0

76.3

97.6

33.4

10.6

34.6

15.6

4.0

47.2

205

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

that their effectiveness in reflecting soil P-supplying capacity was almost the same though the values obtained by the two methods were different (Li, 1963). Due to its simplicity, the Olson method has been widely used later on in China. Results show that, when the available P in the soil determined by the Machigin method was less than 10 mg kg 1, P fertilizer gave good effect and application of P fertilizer was required, while above 15 mg kg 1 the application of P fertilizer alone had no consistent effect (Chen et al., 1963a; Lin et al., 1964). Similarly, the Olsen-P in soil is also very well correlated with crop yield increase by application of P fertilizers (Table 25) For the rational application of P fertilizer, P availability indices have been worked out for different locations (Table 26). Due to different natural conditions and crop productivity in different areas, the indices are quite different even for the same crop. In addition to available P, soil N-supply levels influence crop responses to P fertilizer to a great extent, being important in the determination of P fertilizer effect. Nitrogen and P are closely related in crop nutrition. The supply of N is necessary to allow crops to use P (Li and Zhao, 1990), while the metabolism of N is inhibited with an inadequate supply of Table 25 The relationship between soil available P and crop yield increase by application of P fertilizer

Crop

Available P (mg kg 1)

Yield (kg ha 1)

Yield Without P With P increase (%) Source

Xinjiang Autonomous Region Irrigated wheat <3 2418 3–5 2519 5–10 4202 10–15 3108 > 15 5250 Gansu Province Spring wheat <5 4282 (irrigated) 5–10 4463 10–15 3956 15–20 5648 > 20 5994 Dryland winter < 5 2474 wheat 5–10 3546 10–15 3264 Ningxia Hui Autonomous Regions Wheat (irrigated) 4–6 5232 9–14 6167 > 14 6525

2928 2910 4413 3242 5355

20.4 15.5 5.0 4.2 2.0

Li (2004a)

5103 4872 5781 5706 6209 3062 3990 3534

19.2 9.2 6.0 1.0 3.6 23.7 12.6 8.3

Jin (1989)

5886 6450 6620

12.5 4.6 1.4

Li(2004a)

Jin (1989)

206

Sheng-Xiu Li et al.

Table 26 P availability indices in some dryland locations (Olsen-P; mg kg 1)

Soil

Extremely low Low

Gansu Province Warping Desert < 2.0 Dark loessial < 6.0 Shaanxi Province Loessial soil < 1.5 Manural loessial < 4.4 Helan Province Drab soil < 2.8 Meadow < 3.3 Shandong Province Drab soil < 2.2 Meadow soil < 3.2

2–5 6–11

Medium

High

Source

6–12 11–15

> 12 > 15

Jin (1989)

1.5–4.1 4.1–9.6 > 9.6 Liu and Zhao (1988) 4.4–8.7 8.7–17.5 > 17.5 Li et al. (1979a) 2.8–6.3 6.3–14 3.3–5.7 5.7–10

> 14 > 10

Li (2004a)

2.2–6.6 6.6–19.5 > 19.5 Li (2004a) 3.2–8.1 8.1–20.4 > 20.4

P. Consequently, crop response to P fertilizer largely depends on the soil N-supplying capacity or N addition. When soil N supply is deficient, even if the available P in the soil is deficient, crops may be limited by the N supply with little or no response to P fertilizer. Organic matter is low in soils worldwide and so is the N-supplying capacity. This problem has long been realized by most scientists in China (Chen et al., 1963a; Gao and Wang, 1963; Guo and Luo, 1986; Li, 1965, 2004c; Li and Wang, 1965; Li and Zou, 1962; Lin et al., 1964; Shanxi Institute of Soil and Fertilizer, the Chinese Academy of Agricultural Sciences, 1963; Zhao, 1962). In as early as the 1960s, some scientists had pointed out the importance of soil N level on crop response to P fertilizers. Li and Zou (1962) suggested that the balance of N to P in soils had a very important impact on the response of P fertilizer. Liu et al. (1963) reported that N and P levels in soil were undoubtedly key factors controlling the effects of both P and N fertilizers and emphasized that the combined application of P and N fertilizers should be made to cereals, whereas application of P fertilizer alone be done for soybean and pea. Li (1965) conducted field and pot experiments in Guanzhong Plain, Shaanxi Province, and concluded that the available P in soil was the major factor affecting P fertilizer efficiency. In the absence of N fertilizer, the level of available N in soil had a great influence on the P fertilizer effect. Lin et al. (1964) claimed that, when soil-available P exceeded 15 mg kg 1 (Machigin method), it was necessary to apply N fertilizer along with the P fertilizer; only in fields with sufficient N but deficient available P was it possible to have good response by application of P fertilizer alone. Tao (1983) used the available P (determined by the

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

207

Machigin method) and the ratio of available N (determined by nitrification incubation method) to available P in soil to guide application of P fertilizer: if the available P was less than 15 mg kg 1 and the ratio less than 1.5, P fertilizer had a significant effect when N fertilizer was applied; if available P was less than 15 mg kg 1 but the ratio greater than 1.5, application of P fertilizer alone increased yield significantly, but application of N fertilizer as well would increase the efficiency of P use; if the available P was greater than 15 mg kg 1 and the ratio less than 1.5, the soil was rich in P but deficient in N, and there was no need to apply P fertilizer. Li et al. (1978, 1979a) found that when the Olsen-P in the soil was less than 8 mg kg 1, P deficiency was the major constraint and crops had good responses to P fertilizer in almost all cases; when the Olsen-P in the soil was between 8 and 16 mg kg 1, the effect of P fertilizer was unstable depending on N addition, and with adequate N supply the P effect was significant but otherwise it would not have had any effect; when the Olsen-P was larger than 16 mg kg 1, the soil was rich in P supply and crops showed no response to P fertilizer either with or without the application of N fertilizer. Such results have been proven in many areas. Due to the importance of N and P relations, research seeking the optimum ratios of N to P fertilizer have been carried out in most regions of the northern territory of China and different guidelines have been established (Tables 27 and 28).

9.2. Relation of the responses of leguminous crops to P fertilizers with soil-available P The discussions above are related to nonleguminous crops, especially cereal crops. Whether the effect of P fertilizer on leguminous crops depends on the soil-available P remains unknown. Some results have provided indication that leguminous crops should have the same requirement as cereal crops in responses to P fertilizer. As shown in Table 8, soil 86001 and 86002 were different in the available P, and both wheat and pea had striking differences in responses to P fertilizer in the two soils. For further Table 27 Optimum ratios of soil N to soil P2O5 for wheat with different fertilities Soil fertility Province

Helan Hebei

Low

Medium

High

Source

1:0.7–1.0 1:0.5–0.6 1:0.3–0.4 Li (2004a) 1:1 1:0.5–0.75 1:0.25–0.5 Liu and Zhou (1992) Liu et al. (1989) Ningxia (irrigated) 1:0.5–0.7 1:0.3–0.5 1:0–0.3 Li (2004a)

208

Sheng-Xiu Li et al.

Table 28 Optimum ratios of N to P2O5 for different crops Crop

Shandong Province Wheat Maize Cotton Soybean and peanut Gansu Province Dryland winter wheat Irrigated winter wheat Irrigated spring wheat Maize Tianjin District Wheat Maize Rice

Ratio

Source

1:0.5 1:0.46 1:0.65 1:1

Li (2004a)

1:0.83 1:0.4 1:0.65 1:0.55

Jin (1989)

1:0.6 1:0.4 1:0.1

Li (2004a)

Table 29 Some soil properties for pot experiment conducted from 1965 to 1966

a

Soil no.

OM (%)

Total N Total P CaCO3 (%) (%) (%)

Available P (mg kg 1) Available N (mg kg 1)a Olsen-P Machigin-P

65001 65002 65003 65004 65005

0.80 1.25 1.50 1.15 1.11

0.07 0.12 0.14 0.08 0.11

25.5 73.0 51.0 44.5 32.3

0.12 0.11 0.16 0.12 0.11

6.35 6.10 5.84 6.00 6.47

1.7 5.4 7.9 13.8 21.7

3.3 8.7 10.3 20.5 29.8

Modified from Li et al. (1988). Determined by nitrification power method (Begerburcki, 1961).

confirmation of the pot experimental results, a series of other pot and field experiments were implemented. A pot experiment was conducted in 1965–1966 using soil sampled from the 0– 25 cm layer of five soils with widely different fertility levels from fields near Yangling, Shaanxi Province (Table 29). The soil organic matter, total N, total P, CaCO3, and available N (determined by nitrification power method; Begerburcki, 1961) were measured. Eight crops (white sweet clover, pea, soybean, mung bean, maize, barley, wheat, and rapeseed) were planted. Two treatments were included, without application of P

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

209

fertilizer (control) and with application of 0.05 g P kg 1 soil using calcium superphosphate produced in Nanjing as P source. For the leguminous crops, N fertilizer was not added, while for other crops 0.15 g N kg 1 soil was applied both for the control and P fertilization treatments. P fertilizer was applied as basal fertilization by mixing with soil before crops were planted and the N fertilizer was top-dressed with irrigation water during the plantgrowing period. Six replicates were adopted for each treatment. White sweet clover, soybean, mung bean, and maize were sown on April 12, 1965 and harvested on September 25 in the same year. The other crops were sown in October 2–3, 1965 and harvested on June 5, 1966. All crops were planted in same sized pots and with same treatments. Results (Table 30) showed that crops’ responses to P fertilizer had no relation to organic matter, total N, total P, CaCO3, or available N in soils. For showing the relations, the yield increase by P addition to pea is demonstrated in Fig. 4. In contrast, all crops’ seeds and shoots significantly increased with soil-available P increase, and the yield increase by P addition was significantly decreased as soil-available P increased, whether or not the soilavailable P was determined by the Olsen method or by the Machikin method (Table 30 and Fig. 5). The same trends were found for all other crops, although the relations of P effects for some crops were not as good as those for pea. This reveals that leguminous crops have the same requirement of soil-available P conditions for response to P fertilizer as nonleguminous crops. For confirmation of the pot experiment result, five field experiments were conducted with pea and wheat from 1974 to 1975 with soils selected on the basis of Olsen-P levels. Two P treatments, without application of P fertilizer and with addition of 35 kg P ha 1, were designed for wheat and three levels of N fertilizer (0, 30, and 60 kg N ha 1 were designed, respectively) for pea without and with P addition to investigate the effect of N rate on pea response to P fertilizer as well. The results showed that, as in the pot experiments, wheat and pea seed yields significantly increased, while the seed and shoot yield increase by P fertilizer both significantly declined (Table 31 and Fig. 6) with the Olsen-P level increase. The results further confirmed that the effect of P fertilizer on pea, as in the case of wheat, depended on soil P levels, and that the soil-available N determined by nitrification power method (Begerburcki, 1961) was not related to the P effect in these experiments.

9.3. Effect of N fertilizer on leguminous crops’ response to P fertilizer The study of the influence of N fertilization on pea’s response to P fertilizer was carried out in both pot and field trials. The pot experiments for testing the effect of N fertilizer on pea and its response to P fertilizer were

210

Sheng-Xiu Li et al.

Table 30 Responses to of different crops to P fertilizer in five soils (pot experiment conducted from 1965 to 1966) Seed yield (g pot 1) Soil no.

Shoot yield (g pot 1)

GYI by P GYI by SYI by P SYI by P P Control added (g pot 1) P (%) Control added (g pot 1) P (%)

Pea 65001 1.5 14.3 65002 4.0 13.5 65003 7.4 14.8 65004 15.7 18.7 65005 16.3 18.0 Mung bean 65001 1.4 15.1 65002 10.8 21.5 65003 13.5 28.4 65004 15.9 18.8 65005 17.2 16.0 Soybean 65001 2.3 12.7 65002 12.6 20.0 65003 13.4 19.4 65004 19.4 25.1 65005 21.2 19.6 White sweet clover 65001 65002 65003 65004 65005 Maize 65001 0.0 26.8 65002 9.3 48.8 65003 18.2 64.8 65004 47.5 65.4 65005 48.4 53.4 Barley 65001 2.1 14.3 65002 7.3 29.9 65003 11.8 27.4 65004 21.3 30.4 65005 24.0 25.4

12.8 9.5 7.4 3.0 1.7

853 238 100 19 10

3.8 16.8 25.1 50.8 53.9

35.7 40.1 50.5 62.9 60.5

31.9 23.3 25.4 12.7 6.6

839 139 101 25 12

13.7 10.7 14.9 2.9 1.2

979 99 110 18 7

1.9 25.5 33.7 42.2 50.5

40.0 56.2 55.6 53.5 50.0

38.1 30.7 21.9 11.3 0.5

2005 120 65 27 1

10.4 7.4 6.0 5.7 1.6

452 59 45 29 8

8.0 40.9 40.6 59.9 61.0

43.6 68.3 60.6 70.1 64.8

35.6 27.4 19.7 10.2 3.8

445 67 49 17 6

2.1 13.8 20.9 30.2 38.5

31.5 38.7 41.0 32.8 38.1

29.4 24.9 20.1 2.6 0.5

1400 180 96 9 1

26.8 39.5 46.6 17.9 5.0

425 256 38 10

3.1 54.3 62.2 125.5 136.4

93.8 140.8 154.8 160.1 151.4

90.7 86.5 92.6 34.6 15.0

2926 159 149 28 11

12.2 22.6 15.6 9.1 1.4

581 310 132 43 6

7.1 22.5 28.9 50.1 50.4

25.7 62.1 64.2 64.1 57.6

18.6 39.6 35.3 14.0 7.2

262 176 122 28 14

211

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

Table 30 (continued) Shoot yield (g pot 1)

Seed yield (g pot 1) Soil no.

GYI by P GYI by SYI by P SYI by P P Control added (g pot 1) P (%) Control added (g pot 1) P (%)

Wheat 65001 2.5 65002 5.1 65003 8.5 65004 21.4 65005 20.2 Rapeseed 65001 0.4 65002 9.5 65003 11.4 65004 12.4 65005 14.2

17.6 29.1 27.2 29.8 22.3

15.1 24.0 18.7 8.4 2.1

604 471 220 39 10

7.8 24.3 35.9 69.7 60.4

50.8 88.9 82.4 89.3 67.8

43.0 64.6 46.5 19.6 7.4

551 266 130 28 12

8.6 16.2 17.8 15.8 15.8

8.2 6.7 6.4 3.4 1.6

2050 71 56 27 11

2.3 23.3 25.3 35.5 48.7

36.8 47.9 48.2 43.8 50.1

34.5 24.6 22.9 8.3 1.4

1500 106 91 23 3

GYI, grain yield increase; SYI, shoot yield increase including grain. Modified from Li et al. (1992c).

conducted in different years using six soils with different fertilities (Table 32), but the design, pea variety, pot size, and cultivation conditions were the same for all experiments and therefore the results could be compared. Results showed that the response of pea to P fertilizer was soildependent: in soils deficient in the available P supply, pea responded significantly to P fertilizer; otherwise the responses were lower. As a whole, N fertilization was detrimental to pea and to its P effect. Seen from the average of six soils from pot experiments, N addition obviously decreased pea yield in both the P-fertilized and non-fertilized cases (Table 33 and Fig. 7). The higher the N rate, the more was the decline in yield. Also, N fertilization decreased the yield increase by P addition. However, for soils 64001 and 64002, application of N fertilizer alone significantly improved pea shoot and seed yield (Table 33). The two soils were sampled in extremely unfertile and remote lands and had been abandoned for cultivation. The good results of N fertilizer to the two soils might be related to the little residual mineral N left in the soils since farming without N fertilization was implemented for the two soils for quite a long time. The field experiments for testing the N effect on pea response to P fertilizer were conducted in combination with tests for P fertilizer effect. The results are shown in Fig. 8 and were already shown in Table 31.

Sheng-Xiu Li et al.

15

Relation of total N to yield increase (g pot−1) by P

10 5 0

0

0.5

1

1.5

2

5 0

0.1

0.05

0

0.15

Relation of total P to yield increase (g pot−1) by P

Relation of CaCO3 to yield increase (g pot−1) by P

5

0.05

0.1

0.15

0.2

Total P (%)

Yield increase (g pot−1)

10

Total N (%)

10

0

15

Organic matter (%)

15

0

Yield increase (g pot−1)

Relation of OM to yield increase (g pot−1) by P

Yield increase (g pot−1)

Yield increase (g pot−1)

Yield increase (g pot−1)

212

15 10 5 0 5.8

6

6.2

6.4

6.6

CaCO3 (%)

Relation of available N to yield increase (g pot−1) by P 15 10 5 0

0

20

40

60

80

Available N (mg kg−1)

Figure 4 Relation of pea response to P fertilizer with some important soil properties (pot experiment conducted from 1965 to 1966). Drawing after Li et al. (1992c).

Again, the N effect on pea yield and pea response to P fertilizer had the same trends as the pot experiments. Addition of N fertilizer significantly decreased pea yield and P effect. An exception was soil 74001. In this unfertile soil, the addition of 30 kg N ha 1 slightly increased pea yield in both cases without and with P fertilization. As a typical example, Table 3 provided useful information on pea responses to different P rates under conditions either without N fertilization or with addition of 22.5 kg N ha 1 in comparison with wheat that was fertilized with 112.5 kg N ha 1. Results show that with sufficient N supply, wheat strongly responded to P fertilizer, and the yields linearly increased with P rate increase. However, pea responded to P fertilizer differently. Without addition of such a small amount of N, pea responded better to P fertilizer than with N addition (Fig. 9). In the N addition case, pea yield

213

20

Yield and yield increase (g pot−1)

Yield and yield increase (g pot−1)

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

15 10 5 0 3.3

8.7

10.3

20.5

20 15 10 5 0

29.8

5.4

1.7

Machkin P (mg kg−1) Yield (g pot−1)

7.9

13.8

21.7

Olese-P (mg kg−1)

Yield increase by P (g pot−1)

Yield (g pot−1)

Yield increase by P (g pot−1)

Shoot and shoot increase (g pot−1)

Shoot and shoot increase (g pot−1)

60 60 40 20

40 20 0

0 3.3

6.7

10.3

20.5

29.8

Machkin-P (mg kg−1) Shoot (g pot−1)

Shoot increase by P (g pot−1)

1.7

5.4

7.9

13.5

21.7

Olsen-P (mg kg−1) Shoot (g pot−1)

Shoot increase by P (g pot−1)

Figure 5 Relation of pea responses to P fertilizer with soil available P determined by Olsen and Machigin methods (pot experiment conducted from 1965 to 1966). Drawing after Li et al. (1988).

decreased and so did the P effect. As the average of the six P rates, pea shoot yield (the total above-ground biomass including the seed) was 4126 kg ha 1 and the seed yield 1936 kg ha 1, while with addition of 22.5 kg N ha 1 it was 3774 kg ha 1for shoot and 1661 kg ha 1 for seed. In addition, with yield decline by N addition, P effect on the yield increase correspondingly declined. Without N fertilization, P increased shoot yield by 1810 kg ha 1 and seed yield by 858 kg ha 1 on average of six P rates, whereas on adiition of N fertilizer the corresponding increase was 1627 kg ha 1for shoot and 651 kg ha 1 for seed. Why has N addition decreased pea yield and its response to P fertilizer? Two reasons are obvious. The first is that N fertilizer impacted nodule formation. An investigation using soil 78001 in the pot experiment (see Table 33) showed that P fertilization significantly increased while N fertilization significantly reduced the number of nodules. The reduction in the number of nodules depended on the N rate: the higher the N rate, the larger the reduction in number (Table 34). Such a phenomenon was also observed in soybean (Li et al., 1990). Leguminous crops depend on nodules to fix N from the atmosphere, and the reduction of nodule numbers by N fertilizer will influence their N fixation capacity. This will in turn eliminate leguminous crops’ superiority and affect their productivity, while N input may be unable to compensate their loss by N fixation. The second is the inhibition effect of N on the root length. Experiments have proven that crop yields are

214

Sheng-Xiu Li et al.

Table 31 Pea and wheat response to P fertilizer and N rate effect on pea response to P fertilizer (field experiment conducted from 1974 to 1975) Seed yield (kg ha 1) Soil no

Olsen-P Available N Without With (mg kg 1) (mg kg 1)a P 35 kg P ha 1

Yield Yield increase (kg ha 1) increase (%) by P by P

Pea response to P fertilizer without application of N fertilizer (modified from Li et al., 1992c) 74001 2.2 35.0 653 1463 810 124 74002 10.9 27.4 1370 1863 493 36 74003 12.2 30.8 1671 1895 224 13 74004 15.4 35.0 2333 2339 6 0 74005 18.4 52.0 3390 3420 30 1 Mean 1883 2236 352 38 Pea response to P fertilizer with addition of 30 kg N ha 1 (modified from Li et al., 1988) 74001 2.2 35.0 668 1508 840 126 74002 10.9 27.4 1238 1788 550 44 74003 12.2 30.8 1542 1715 173 11 74004 15.4 35.0 2108 2220 112 5 74005 18.4 52.0 3180 3221 41 1 Mean 1747 2090 343 37 Pea response to P fertilizer with addition of 60 kg N ha 1 (modified from Li et al., 1988) 74002 10.9 27.4 1119 1556 440 39 74003 12.2 30.8 1465 1603 138 9 74004 15.4 35.0 2085 2166 81 4 74005 18.4 52.0 3120 3155 35 1 Wheat response to P fertilizer without application of N fertilizer (modified from Li et al., 1992c) 74001 2.2 35.0 1176 1884 708 60 74002 10.9 27.4 2607 3082 475 18 74003 12.2 30.8 3200 3556 356 11 74004 15.4 35.0 3182 3251 69 2 74005 18.4 52.0 4146 4215 69 2 Mean 2923 3252 329 18 a

Determined by nitrification power method (Begerburcki, 1961).

closely related to root biomass and its distributing space (Liu et al., 1994; Wang et al., 1999b). For a crop planted in nylon bags with limited soil, the root biomass was significantly reduced and so was the crop yield due to the limitation of root distributed space (Song and Li, 2003). The nutrient input

215

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

Wheat response to P fertilizer

4000 2

3000

R = 0.8778

2000 1000 0

2

R = 0.9124 0

5

10

15

20

Seed and seed increase (kg ha−1)

Seed and seed increase (kg ha−1)

Pea response to P fertilizer without N addition 5000 4000 3000 2000 1000 0

Olsen-P (mg kg−1) Seed yield (g pot−1)

Seed increase by P (g pot−1)

2 R = 0.9593

2

R = 0.9136 0

5

Seed yield (kg ha−1)

10 15 Olsen-P (mg kg−1)

20

Yield increase by P (kg ha−1)

Figure 6 Relation of pea and wheat responses to P fertilizer in relation with soil Olsen-P (field experiments conducted from 1974 to 1975). Drawing after Li et al. (1992c, 1988).

Table 32

Some soil properties for pot experiments conducted in different years

Total Soil no. OM (%) N (%)

Total P (%)

Available N C/N (mg kg 1)*

Olsen-P (mg kg 1)

pH (H2O)

63002 64001 64002 78001 85001 85002

0.145 0.132 0.118 0.120 0.165 0.185

10.0 8.7 9.2 8.6 10.7 9.9

10.4 3.9 4.0 7.0 5.7 24.1

7.9 8.0 8.0 8.1 8.2 8.2

1.12 0.42 0.54 0.80 1.16 1.40

0.086 0.028 0.034 0.054 0.063 0.082

38.2 36.5 22.8 14.0 41.0 57.6

Soils were sampled in Yangling area, Shaanxi Province. There was no great difference in soil texture for the six soils used in the pot experiments. For example, soil 85001 was composed of 12.3% sand, 38.6% course silt, 25.1% fine silt, and 24% clay, and soil 85002, 14.2% sand, 38% course silt, 24.7% fine silt, and 23.1% clay. Modified from Li et al. (1988).

directly affects root growth, and the effects of nutrient input on root growth depend on the kind of nutrient elements and on the supplying capacity of the soil of these nutrients. In general, P promotes root growth (Barber, 1984), especially when soil is deficient in available P and high in available N, while N fertilization reduces the root length. Although in some cases P addition may not cause a significant increase in shoots, its function in increasing root length depends on whether the soil was sufficient or deficient in this element. As a result, P has a special function in increasing the root biomass and extension of root length in any case, while N supply increases root weight but reduces its length. In a soil deficient in P but sufficient in N, the effect of N fertilization on the inhibition of root length is more obvious. For this reason, N input limits the root system to a small space, reduces the root/shoot ratio, and this influences pea uptake of P from a wider area (Barber, 1984), thus affecting its yield and P fertilizer effect.

216

Sheng-Xiu Li et al.

Table 33 Effect of N fertilization on pea responses to P fertilizer (pot experiments conducted in different years) Yield (g pot 1) Soil no.

Control

P (0.05 g kg 1)

Seed yield Without addition of N fertilizer 63002 9.45 13.12 64001 0.11 11.33 64002 0.29 10.92 78001 8.34 12.57 85001 3.45 12.10 85002 14.73 14.71 Mean 6.06 12.45 With addition of 0.03 g N kg 1 63002 10.63 15.67 78001 9.45 14.71 With addition of 0.1 g N kg 1 63002 7.30 9.33 64001 0.83 9.70 64002 0.43 7.72 78001 7.90 10.20 85001 2.17 6.26 85002 10.42 12.70 Mean 4.84 9.32 With addition of 0.2 g N kg 1 78001 5.60 8.48 Shoot dry matter production including seed Without addition of N fertilizer 63002 24.31 48.95 64001 1.94 39.80 64002 2.92 35.47 78001 17.46 45.16 85001 16.90 41.54 85002 32.56 43.09 Mean 16.02 42.34 With addition of 0.03 g N kg 1 63002 25.46 51.74 78001 17.97 46.07 With addition of 0.1 g N kg 1 63002 22.70 43.06 64001 2.56 30.89 64002 2.90 31.82

Yield increase by P (g pot 1)

3.67 11.22 10.63 4.23 8.65 0.02 6.39

Yield increase by P (%)

39 10200 3667 51 251 0.1

5.04 5.25

47 56

2.03 8.87 7.29 3.30 4.09 2.28 4.64

28 1069 1895 42 188 22

2.88

51

24.64 37.86 32.55 27.70 24.64 10.53 26.32

101 1952 1115 159 146 32

26.28 28.10

103 156

20.36 28.33 28.92

90 1107 997

217

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

Table 33 (continued) Yield (g pot 1) Soil no.

P (0.05 g kg 1)

Control

78001 15.40 34.76 85001 10.54 28.14 85002 30.60 35.99 Mean 14.12 34.11 With addition of 0.2 g N kg 1 78001 13.44 30.18

Yield increase by P (g pot 1)

Yield increase by P (%)

19.36 17.60 5.39 19.99

126 167 18

16.74

125

Modified from Li et al. (1988).

Seed yield (g pot−1) without N

Seed yield (g pot−1) with addition of 0.1 g N kg−1

10 5 0

3.9

4

5.7

7

10.4

24.1

Seed and seed increase (g pot−1)

Seed and seed increase (g pot−1)

15

15 10 5 0

Olsen-P (mg kg−1)

−5

3.9

Shoot and shoot increase (g pot−1)

Shoot and Shoot increase (g pot−1)

30 20 10 5.7

7

10.4

10.4

24.1

Shoot yield (g pot−1) with addition of 0.1 g N kg−1

40

4

7

Seed yield (g pot−1) Yield increase by P (g pot−1)

Shoot yield (g pot−1) without N

3.9

5.7

Olsen-P (mg kg−1)

Seed yield (g pot−1) Yield increase by P (g pot−1)

0

4

24.1

Olsen-P (mg kg−1) Shoot yield (g pot−1) Shoot increase by P (g pot−1)

40 30 20 10 0

3.9

4

5.7

7

10.4

24.1

Olsen-P (mg kg−1)

Shoot yield (g pot−1) Shoot increase by P (g pot−1)

Figure 7 Pea responses to P fertilizer in relation with N fertilization (pot experiments conducted in different years). Drawing after Li et al. (1988).

9.4. Response of leguminous crops to P fertilizer rate Dinitrogen is described as the most stable diatomic molecule known. The two atoms of nitrogen in the diatomic molecule are joined by a very stable triple bond. This triple bond requires high energy (945 kJ, or 226 kcal, per mole) to break and therein lies one of the major challenges of fixed

Sheng-Xiu Li et al.

Pea seed yield (kg ha−1) without N addition 4000 3000 2000 1000 0

2.2

10.9

12.2

15.4

Pea seed yield with addition of 30 kg N ha−1 Seed and seed increase (kg ha−1)

Seed and seed increase (kg ha−1)

218

4000 3000 2000 1000 0

18.4

2.2

10.9

Olsen-P (mg kg−1)

12.2

15.4

18.4

Olsen-P (mg kg−1) Seed increase by P (g pot−1)

Seed yield (g pot-1)

Figure 8 Relation of pea responses to P fertilizer to soil Olsen-P under conditions without and with N fertilization (field experiments conducted from 1974 to 1975). Drawing after Li et al. (1992c, 1988). Wheat

Wheat Seed yield (kg ha−1)

Shoot yield (kg ha−1)

10,000 8000 6000 4000 2000 0

0

39 58.5 78 19.5 P rate (kg ha−1)

5000 4000 3000 2000 1000 0

97.5

0

19.5

Pea

5000 4000 3000 2000 1000 0

97.5

3000

6000 Seed yield (kg ha−1)

Shoot yield (kg ha−1)

Pea

39 58.5 78 P rate (kg ha−1)

0

19.5

58.5 78 P rate (kg ha−1)

Shoot without N (kg ha−1)

39

97.5

Shoot with N (kg ha−1)

2500 2000 1500 1000 500 0

0

19.5

39 58.5 78 P rate (kg ha−1)

Seed without N (kg ha−1)

97.5

Seed with N (kg ha−1)

Figure 9 Effects of N fertilization on wheat and pea responses to P rate (field experiment conducted in 1985 to 1986, data from Table 3). Drawing after Li and Li (1992).

dinitrogen chemically or biologically. Dinitrogen fixation is energy expensive because it requires much energy to break the triple bond and also to provide the hydrogen necessary to reduce dinitrogen to two ammonia molecules. Just as the chemical fixation of dinitrogen is energy expensive, so too is its fixation in biological systems. The principal differences lie in the sources of the reductant and energy and the fact that biological N2 fixation takes place at ambient pressures and temperatures. That is quite a feat when

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

219

Table 34 Effect of N fertilization on pea nodulation Treatment

Control and P addition Without application of fertilizer Adding 0.05 g P kg 1 soil Adding 0.1 g P kg 1 soil N addition Adding 0.02 g N kg 1 soil Adding 0.1 g N kg 1 soil Adding 0.2 g N kg 1 soil Low P rate with different N rate Adding 0.05 g P þ 0.02 g N kg 1 soil Adding 0.05 g P þ 0.1 g N kg 1 soil Adding 0.05 g P þ 0.2 g N kg 1 soil High P rate with different N rate Adding 0.1 g P þ 0.02 g N kg 1 soil Adding 0.1 g P þ 0.1 g N kg 1 soil Adding 0.2 g P þ 0.2 g N kg 1 soil

Nodules per plant

18 66 41 18 4 1 57 43 21 53 32 15

Measured from pot experiment conducted in different years, using soil 78001 (see Table 31). Modified from Li et al. (1988).

we consider the rigors of the industrial process. Energy for biological N2 fixation comes from the oxidation of organic carbon sources, such as glucose (Graham, 1998; Keyser and Li, 1992; Zuberer, 1998). For fixation of N from the atmosphere, leguminous plants consume a large amount of energy and photosynthetic products. This may be one of the reasons why leguminous crops produce lower yield than cereal crops in addition to their biological productivity. Considering their low yield, their requirement of P fertilizer rate should be different from cereal crops. From Table 3, we can see clearly that the wheat yield and its yield increase by P addition were continuously increased with P rate increase until the highest rate. In contrast, the highest pea yield and yield increase for both shoot and seed yields were at the P rate of 78 kg P ha 1, and thereafter both decreased. This shows that for obtaining the highest yield of pea, the P rate can be reduced by 20% as compared to wheat.

9.5. The availability index for application of P fertilizer to leguminous crops Three typical experiments were compared in different fields (including soil 74001 as shown in Table 31) representing low, middle, and high P levels in soil to study pea and wheat responses to P fertilizer. Lands were selected

220

Sheng-Xiu Li et al.

mainly on the basis of their Olsen-P values. For eliminating the detrimental effect of mineral N in soil to pea, lands were chosen where the only N applied in recent years was in the form of organic fertilizer. The rate of P was 35 kg ha 1 with SSP as P source, and the N rate was 30 and 60 kg ha 1 with urea as N source. As shown in Table 35, probably due to the low mineral N and high soil buffering capacity of such soils, N fertilization did not significantly decrease or increase pea yield. Since both wheat and pea were in normal growth, the comparisons of N fertilizer effect on their responses to P fertilizer might be more adequate than if pea were not in normal growth. From such typical experiments (Table 35), we can see that, although leguminous crops need a lower rate of P fertilizer application, their response to soil-available P level is almost the same as for cereal crops. In the soil having a low available P, such as that containing 2.2 mg Olsen-P kg 1 soil, both pea and wheat had a striking effect on yield increase by P fertilization. The yield increase in absolute amount was similar for the two crops, but the percentage increase for pea was much higher than that for wheat due to the low yield of the control. In the medium level of soil available P (12.2 mg Olsen P kg 1 soil), pea still had good response to P fertilizer with application of P fertilizer alone, while wheat had almost no response without the application of N fertilizer. However, with N rate increase, wheat became more and more responsive to P fertilizer whereas in the case of pea it became negligible. This showed that N had been the major constraint limiting wheat response to P fertilizer, and addition of N fertilizer could improve its response to P fertilizer. With high availability of P in the soil, both pea and wheat had no responses to P fertilizer irrespective of whether N was added or not and the rate at which N was added. The yield increase in percentage by pea was not caused by P addition but by N addition, which reduced its control yield. From the results obtained, we can conclude that the responses of both leguminous and nonleguminous crops to P fertilizer are mainly determined by soil-available P. If soil-available P is sufficient, there will be no responses for any crop to P fertilizer. The soil P availability index for nonleguminous crops may be also suitable for leguminous crops, although the legumes require a less rate of application of P fertilizer. For input of P fertilizer to nonleguminous crops, N-supply level of the soil and N addition should be taken into account by considering the available P levels in the soil. To sum up, the different responses of leguminous crops and cereal crops to P fertilizer are caused neither by the difference of P uptake amount nor by the root characteristics that might have different ability to use the sparingly soluble phosphate, but by the N fixation of leguminous crops. In a soil deficient in both N and P conditions, and with and with application of P fertilizer alone, the leguminous crop responses to P fertilizer have not been restricted by N limitation, the P fertilizer can fully play its role and therefore the effect of P fertilizer to leguminous crops is much better than

Table 35 Effect of application of phosphate fertilizer on yield increase of pea and wheat Pea Rate of P and N

Yield (kg ha 1)

Wheat Increase (kg ha 1) by P

Increase (%) by P

Grown in soil with Olsen-P 2.2 mg kg 1 (74001) (modified from Li et al., 1992c) Control 653 35 kg P ha 1 1463 810 124.0 668 30 kg N ha 1 (30 kg P þ 30 kg N) ha 1 1508 840 125.7 1 Grown in soil with Olsen-P 12.4 mg kg (modified from Li et al., 1979b) Control 1665 35 kg P ha 1 1893 228 13.7 30 kg N ha 1 1743 1767 89 5.1 (35 kg P þ 35 kg N) ha 1 60 kg N ha 1 1866 1903 75 4.0 (35 kg P þ 60 kg N) ha 1 1 Grown in soil with Olsen-P 22.4 mg kg (modified from Li S. X, unpublished data) Control 2450 35 kg P ha 1 2512 62 2.5 2413 30 kg N ha 1 (35 kg P þ 30 kg N) ha 1 2496 83 3.4 2411 68 kg N ha 1 (35 kg P þ 60 kg N) ha 1 2470 59 2.5

Yield (kg ha 1)

1176 1884 1201 2106 3161 3250 3871 4278 4653 5179 4260 4294 4871 4894 5253 5298

Increase (kg ha 1) by P

Increase (%) by P

708

60.2

905

75.4

89

2.8

407

10.5

523

11.7

34

0.7

23

0.5

45

0.8

222

Sheng-Xiu Li et al.

that to cereal crops. In contrast, when N supply is sufficient, it is another story for nonleguminous crops. Sufficient N supply enables the nonleguminous crops, especially cereal crops to have better responses to P fertilizer than the leguminous crops.

10. Conclusions 10.1. Importance of P in agriculture Phosphorus is a vital component of a number of macromolecules and is an integral part of energy metabolism and major biological processes including photosynthesis, respiration, and membrane transportation. Its genetic role in ribonucleic acid and function in energy transfers via adenosine triphosphate are indispensable. Absolutely necessary for all forms of life, P cannot be substituted by other elements in agricultural production. As the life-limiting element in natural ecosystems, regular inputs of phosphate fertilizer to replenish the P removed from crops are one property of modern agriculture. However, RP is a nonrenewable resource and the global commercial phosphate reserves may be depleted in another 50–130 years. In addition, RP reserves are under the control of only a few countries. The recovery rate of the P fertilizer is very low; the surpluses of P in soil have led to variable responses of crops to P fertilizers and to environmental pollution. For sustainable agriculture, requirements of direct application of RP and improvement of P fertilizer efficiency have led to use of specific plant species. Since legumes have better responses to P fertilizer in general than cereals, some scientists have proposed the application of P to leguminous crops as the first priority. Many hypotheses have been proposed to explain the different responses to P fertilizer between the two kinds of crops but most of the viewpoints are not substantiated. This chapter has reviewed the current investigation status and reported our own viewpoints on the responses of the two type crops to P fertilizer based on studies of more than 40 years.

10.2. Demand of P for leguminous and nonleguminous crops One of the reasons proposed for explaining the difference in responses to P fertilizer is the higher P demand of legumes than nonlegumes. A long-term experiment in a maize–maize–soybean rotation sequence showed the total P uptake by maize from unit area was similar to soybean, and in some cases even higher than soybean. Results of our pot and field experiments have shown that P uptake amount by legumes was not higher than in cereal crops, and wheat had a higher capacity to use soil P than pea and vetch. Without application of N fertilizer, P amounts taken up by legumes were equal to or slightly higher than those of nonlegumes, while with N

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

223

application cereal crops took up much more P than legumes in most cases either with or without P fertilization.

10.3. Responses to RPs between legumes and nonlegumes Since in calcareous soil the P availability in P-contained minerals is roughly equal to that in RPs, some agricultural scientists have used RPs to test the crops’ ability for using sparingly soluble P from soil. Results show that, in addition to the physical and chemical properties of RPs and soil pH, some plant species such as rapeseed, radish, and some leguminous crops possess strong ability to absorb P from RPs in acidic soils; the good responses of rapeseed to the RPs were also found in calcareous soils. However, there exist some debatable issues in the studies: comparison was made not in the same field but in different locations; much higher rate of RPs was used, and P in the RPs extracted by 2% citric acid was closely correlated with the crops’ yield increase; only rapeseed but no other crop was used for comparison; the soils were particularly unique in that rapeseed yield increase by RPs was several times higher than the control, and had strong responses to calcium carbonate and thus the P effect could not be separated from calcium carbonate contained in the RPs; and some conflicting results existed. A series of pot and field experiments were conducted by us in two calcareous soils using six crops planted in spring and autumn for both pot and field trials with sufficient N supply. Results showed that crops had significant responses to SSP, while the effect of RPs was not as high as SSP and occurred only at high rates. By comprehensive comparisons of the yield increase in absolute amount and in percentage of the control, there was almost no difference between legumes and nonlegumes used in our experiments in responses to the RPs. Again, for any crop, the response to RPs was closely related with the citric-acid-extracted soluble P (P2O5) in RPs.

10.4. Root morphology Since the P concentration in soil solution is very dilute, the P movement in soil takes place mainly by diffusion; the diffusion coefficient is very low, and the distance moved from one place to another is very short; plant roots play a great part for ensuring sufficient P supply to crops; and the different responses of crops are attributed to their root characteristics. Our results indicate that wheat has better developed roots, while pea roots have a higher function in supporting the shoots; for a single plant, the total root-absorbing and actively absorbing area of pea was larger than that of wheat, but in unit volume of the soil, wheat root dry weight was 33% larger than for pea. The root activities in terms of the TTC reductive amount and intensity were higher for pea than for wheat. Clearly, the root biomass, root-absorbing

224

Sheng-Xiu Li et al.

areas, and root activity could not explain the different responses to P fertilizers between legumes and nonlegumes.

10.5. Soil microorganisms Many soil microorganisms are able to transform insoluble forms of P into an accessible soluble form, being regarded as PGPM. However, there are no evidences showing their release of P to legumes and nonlegumes, and the very limited knowledge prevented us from carrying forward further discussion.

10.6. Acidification of rhizosphere by roots’ exudates for crops’ responses to P fertilizers Of the hypotheses proposed for explaining the difference of crops in utilizing sparingly soluble P in the soil and their responses to P fertilizer, the most common points are differences of the reduction of the rhizosphere pH through the release of protons and organic acids from roots. Our experiments showed that pH in rhizospheric soil was generally one unit lower than that in the bulk soil, but there was no pH difference between wheat and pea either in rhizospheric or in bulk soil. The available P had the same trend as soil pH. It seems that the acidification of rhizosphere soil could not differentiate the ability of crop responses to P fertilizer.

10.7. Effects of root CEC and calcium uptake amount of crops on crop responses to P fertilizer The CEC of plant roots was once regarded as the basis for crops to exchange cations with those held in soil colloid, and crops with high root CEC could take up more calcium from soil and thus liberate P bound with calcium for crop use. Our study with wheat and pea showed that the root CEC of pea was several times higher than that of wheat in terms of per kilogram dry root or root weight per pot. However, the P uptake amounts by the two crops did not follow the same pattern as root CEC. In relation to root CEC, crop uptake of the calcium amount was considered a mechanism for P release and for crop uptake, and the ratio of CaO to P2O5 in plant tissue was proposed as the index for indicating the plant’s ability to absorb P from RPs. However, later researchers have rejected this hypothesis.

10.8. Responses to P fertilizer of leguminous crops with N fixation A series of experiments have shown that the sensitive responses of legumes to P fertilizer were related to their N fixation capacity. This has been evidenced by many facts.

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

225

It was not always the case that leguminous crops had good responses to P fertilizer but only in a soil deficient in both N and P nutrients; in a soil with sufficient N supply, the different response to P fertilizer of the two type crops vanished and the response of cereal crops to P fertilizer was even better than for leguminous crops. Addition of N fertilizer to cereal crops significantly increased the P fertilizer effect. In a soil extremely deficient in N and P supply, maize with N addition absorbed almost the same amount of P as soybean from the soil. Without application of N, the P uptake amount of pea was 33% higher than that of wheat; when fertilized with N, the P uptake amount showed no difference from that of pea. Also, in a soil deficient in N and P supply, application of N fertilizer to cereal crops and no N or a small amount of N to leguminous crops often resulted in much better responses of cereal crops to P fertilizer than leguminous crops. This shows that the sensitivity of leguminous crops’ response to P fertilizer is related to its N fixation. Our experiments showed that, before pea had the ability to fix N, the biomass increase and P uptake amount by pea were much lower than in wheat, whereas after acquiring the ability to fix N, pea took up much more P than wheat, strongly revealing the great contribution of N fixation by legumes to P fertilizer response. Application of N fertilizer to pea for depressing its N fixation ability and no N addition to wheat have resulted in almost the same P uptake amount for the two crops; with N application, wheat took up much more P than pea. The same trend was also found for maize and soybean. An old-cultivated layer of a soil in which soybean grown previously had no nodules was sampled to conduct a pot experiment for planting maize and soybean. Maize was treated in two ways: without and with application of P fertilizer, while soybean was treated with inoculation and without inoculation on both P treatments. Results showed that in both cases with and without P fertilizer, there was almost no difference in P uptake for the two crops. However, when soybean was inoculated, the P uptake amount and dry matter increase by P addition were much larger than those of maize and also more than soybean that was not inoculated. This clearly indicates the importance of N fixation to leguminous crops responses to P fertilizers.

10.9. Factors affecting crop responses to P fertilizer The responses of nonleguminous crops, especially cereal crops, to P fertilizer are mainly determined by the soil-available P, which remarkably reflects the soil P-supplying capacity and greatly influences the effect of P fertilizer. This is true also for leguminous crops. Eight plants grown in five soils in a pot experiment show that organic matter, total N, total P, soil CaCO3 contents, and soil available N are not related to the P fertilizer effect, but soil-available P, determined by either Olsen or Machigin

226

Sheng-Xiu Li et al.

methods, is closely correlated with crop response to P fertilizer. Field experiments further confirmed the results. Application of N fertilizer is beneficial to P fertilization for cereal crops, but is detrimental to leguminous crops in most cases, especially when soil mineral N is abundant. The negative effect of N fertilization on the response of leguminous crops to P fertilizer is mainly caused by its reduction in nodule formation and inhibition of root length, thereby eliminating the superiority of leguminous crops and leading to low productivity. Since yields of leguminous crops are lower than those of cereal crops, the P rate can be reduced by at least 20% as compared to wheat. Typical experiments show that, although leguminous crops need a lower rate of P fertilizer application, their response to soilavailable P level is almost the same as for cereal crops. In a soil having a low available P, P fertilization significantly increased pea yield, and the absolute increase was almost no different from wheat, but the increased percentage was higher than wheat due to low yield in pea control. At a medium level of soil-available P, pea still had good response to P fertilizer, while wheat had almost no response without application of N fertilizer. However, with N rate increase, wheat became more and more responsive to P fertilizer whereas pea became less responsive. This shows that N was the major constraint limiting wheat response to P fertilizer, and addition of N fertilizer can improve its response to P fertilizer. In a soil with high available P, both pea and wheat had no responses to P fertilizer irrespective of whether N was added or not and irrespective of the rate of N added. From these results, we can conclude that the availability index for the application of P fertilizer to leguminous crops can be used for leguminous crops as well.

10.10. Brief summary To summarize, the difference in response of legumes and nonlegumes to P fertilizer is caused neither by the difference of P uptake amount nor by the root characteristics such as root biomass, root surface area, root activity, root exudates, root CEC, and CaO/P2O5 ratio in plant tissues, but by the N fixation of the leguminous crops. The responses of both leguminous and nonleguminous crops to P fertilizer are mainly determined by the soilavailable P. If the soil-available P is sufficient, there will be no response of the crops to P fertilizer. In addition, the N-supplying levels in the soil greatly influence the P effect. In a soil deficient in both N and with application of P fertilizer alone, the leguminous crops’ responses to P fertilizer have not been restricted by N limitation, the P fertilizer can fully play its role and therefore the effect of P fertilizer to leguminous crops is much better than that to cereal crops. It is another story when N supply is sufficient for nonleguminous crops. Sufficient N supply will make the cereal crops respond better to P fertilizer than leguminous crops.

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

227

ACKNOWLEDGMENTS This chapter contains work that has been carried out for more than 40 years. In the early period (from 1962 to 1981), it was supported by the Shaanxi Provincial Ministry of Agriculture and the Northwestern College of Agriculture. Later, it was continuously supported by the common project (30971866, 30971868 and 30871596), key project (30230230), important project (49890330), and project toward agriculture (30070429) from the National Natural Science Foundation of China (NSFC) and the National Key Basic Research Special Funds (2009CB118604), and the program “IRT0749”of Ministry of Education of China. The authors would like to take the opportunity to express their gratitude to all organizations for their kindness in supporting such projects.

REFERENCES Ae, N., Arihara, J., and Okada, K. (1991). Phosphorus uptake mechanisms of pigeon peagrown in alfisols and vertisols. In “Phosphorus Nutrient of Grain Legume in the Sentiarid Tropic” (C. Johansen, K. K. Lee, and K. L. Sahrawat, Eds.), pp. 91–98. ICRISAT, Patancheru, India. Afif, E., Matar, A., and Torrent, J. (1993). Availability of phosphate applied to calcareous soils of West Asia and North Africa. Soil Sci. Soc. Am. J. 57, 756–760. Afza, R., Hardarson, G., Zapata, F., and Danso, S. K. A. (1987). Effects of delayed soil and foliar N fertilization on yield and N2 fixation of soybean. Plant Soil 97, 361–368. Agegnehu, G., Ghizaw, A., and Sinebo, W. (2005). Yield performance and land-use efficiency of barley and faba bean mixed cropping in Ethiopian highlands. Eur. J. Agron. 25(3), 202–207. Agilars, A., and Van Diest, A. (1981). Rock phosphate mobilization induced by the alkaline uptake pattern of legumes utilization symbiotically fixed nitrogen. Plant Soil 45, 295–300. Akande, M. O., Aduayi, E. A., Olayinka, A., and Sobulo, R. A. (1998). Efficiency of Sokoto rock phosphate as a fertilizer source for maize production in southwestern Nigeria. J. Plant Nutr. 21, 1339–1353. Alves, V. M. C., Parentoni, S. N., Vasconcellos, C. A., Bahia Filho, A. F. C., Pitta, G. V. E., and Schaffert, R. E. (2001). Mechanisms of phosphorus efficiency in maize. In “Plant Nutrition-Food Security and Sustainability of Agroecosystems Through Basic and Applied Research” (W. J. Horst, et al., Eds.), pp. 556–567. Kluwer Academic Publishers, Dordrecht. Amer, F., Buldin, D. R., and Black, C. A. (1955). Characterization of soil phosphorus by anion exchange resin and adsorption and 32P equilibrium. Plant Soil 6, 391–408. Andrews, R. E., and Newman, I. (1970). Root density and competition for nutrients. Oecol. Plant. 5, 319–334. Asarov, X. K. (1959). “Agrochemistry” (A Chinese textbook translated from Russian into Chinese by Beijing Agricultural University for agricultural university students) High Education Press, Beijing, China, vol. 1, pp.197–205. Aulakh, M. S., Wassmann, R., and Bueno, C. (2001). Characterization of root exudates at different growth stages of ten rice (Oryza sativa L.) cultivars. Plant Biol. 3, 139–148. Aulakh, M. S., Garg, A. K., and Kabba, B. S. (2007). Phosphorus accumulation, leaching and residual effects on crop yields from long-term applications in the subtropics. Soil Use Manag. 23, 417–427. Baccini, P., and Brunner, P. H. (1991). The Metabolism of the Anthroposphere. Springer, Berlin.

228

Sheng-Xiu Li et al.

Baenes, J. S., and Kamprath, E. J. (1975). Availability of North Carolina rock phosphate applied to soils. N. C. Agr. Exp. Stn. Tech. Bull. 229. Baldani, V. L. D., Baldani, J. I., and Do¨bereiner, J. (2001). Inoculation of rice plants with the endophytic diazatrophs Herbaspirillum seropedicae and Burkholderia spp. Biol. Fertil. Soils 30, 485–491. Barber, S. A. (1962). A diffusion and mass flow concept of soil nutrient availability. Soil Sci. 93, 39–49. Barber, S. A. (1966). The role of root intercept, mass flow, diffusion in regulating the uptake of ion by plants from soil. Tech. Rep. I.A.E.A. 65, 39–45, 55–89. Barber, S. A. (1974). Influence of the plant root on ion movement in the soil. In “The Plant Root and Its Environment” (E. W. Carson, Ed.), pp. 525–564. University Press of Virginia, Charlottesville. Barber, S. A. (1982). The P uptake amount by plants depending on the relation of soil to roots. Soil. Fertil. 1982, 52–54, (Agriculture in Foreign Countries). Barber, S. A. (1984). Soil Nutrient Bioavailability—A Mechanistic Approach. John Wiley and Sons Inc., New York, pp. 55–85, 90–113 and 201–225. Barber, S. A., Walker, J. M., and Vasey, E. H. (1963). Mechanisms for the movement of plant nutrients from the soil and fertilizers to the plant root. J. Agric. Food Chem. 11, 204–207. Barberis, E., Ajmone Marsan, F., Scalenghe, R., Lammers, A., and Schwertmann, U. (1995). European soils over fertilized with phosphorus. Part I. Basic properties. Fertil. Res. 45, 199–207. Barea, J. M., Pozo, M. J., Azco´n, R., and Azco´n-Aguilar, C. (2005). Microbial co-operation in the rhizosphere. J. Exp. Bot. 56, 1761–1778. Barley, K. P. (1970). The configuration of the root system in relation to nutrient uptake. Adv. Agron. 22, 159–201. Barr, J. A. (1960). Phosphate rock. In “Industrial Minerals and Rocks” (J. L. Gillson, Ed.), third ed.. pp. 649–668. American Institute of Mining and Metallurgical Engineers, New York. Barroso, C. B., and Nahas, E. (2005). The status of soil phosphate fractions and the ability of fungi to dissolve hardly soluble phosphates. Appl. Soil Ecol. 29, 73–83. Bates, T. R., and Lynch, J. P. (2001). Root hairs confer a competitive advantage under low phosphorus availability. Plant Soil 236, 243–250. Bationo, A., Mughogho, S. K., and Mokwunye, A. U. (1986). Agronomic evaluation of phosphate fertilizers in tropical Africa. In “Management of Nitrogen and Phosphorus Fertilizers in Sub-Saharan Africa” (A. U. Mokwunye and P. L. G. Vlek, Eds.), pp. 283– 318. Martinus Nijhoff, Dordrecht, The Netherlands. Batjes, N. H. (1997). A world data set of derived soil properties by FAOUNESCO soil unit for global modeling. Soil Use Manag. 13, 9–16. Batten, G. D. (1992). A review of phosphorus efficiency in wheat. Plant Soil 146, 163–168. Bayer, C., Martin-Neto, L., Mielniczuk, J., and Sangoi, L. (2001). Changes in soil organic matter fractions under subtropical no-till cropping systems. Soil Sci. Soc. Am. J. 65, 1473–1478. Beard, B. H., and Hoover, R. M. (1971). Effect of nitrogen on nodulation and yield of irrigated soybeans. Agron. J. 63, 815–816. Beauchemin, S., and Simard, R. R. (1999). Soil phosphorus saturation degree: Review of some indices and their suitability for P management in Quebec, Canada. Can. J. Soil Sci. 79(4), 615–625. Beck, M. A., and Sanchez, P. A. (1994). Soil phosphorus fraction dynamics during 18 years of cultivation on a typic paleudult. Soil Sci. 34, 1424–1431. Begerburcki, A. B. (1961). Worker’ Handbook for Agricultural Chemistry Laboratory. Agricultural Literature, Journal and Geology Press, Moscow, pp. 111–115.

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

229

Beijing Agricultural University and Experimental Farm of Agriculture Ministry. (1966). Preliminary result of the experiment of the Morocco powered rock phosphate in the bleached soil. Acta Pedol. Sin. 14(1), 80–82. Bevivino, A., Dalmastri, C., Tabacchioni, S., and Chiarini, L. (2000). Efficacy of Burkholderia cepacia MCI 7 in disease suppression and growth promotion of maize. Biol. Fertil. Soils 31, 225–231. Bhat, K. K. S., and Nye, P. H. (1973). Diffusion of phosphate to plant roots in soil. 1. Quantitative autoradiography of the depletion zone. Plant Soil 38, 161–175. Blake, L., Mercik, S., Koerschens, M., Moskal, S., Poulton, P. R., Goulding, K. W. T., Weige, A., and Powlson, D. S. (2000). Phosphorus content in soil, uptake by plants and balance in three European long-term field experiments. Nutr. Cycl. Agroecosyst. 56, 263–275. Boiland, M. D. A. (1993). Summary of research on soil testing for rock phosphate fertilizers in Western Australia. Fertil. Res. 35(1), 83–91. Bolan, N. S., White, R. E., and Hedley, M. J. (1990). A review of the use of phosphate rocks as fertilizers for direct application in Australia and New Zealand. Aust. J. Exp. Agr. 30, 297–313. Bolan, N. S., Naidu, R., Mahimairaja, S., and Baskaran, S. (1994). Influence of low-molecularweight organic acids on the solubilization of phosphates. Biol. Fertil. Soils 18, 311–319. Bole, J. B. (1977). Influence of root hairs in supplying soil phosphorus to wheat. Can. J. Soil Sci. 58, 169–175. Bolland, M. D. A., Siddique, K. H. M., Loss, S. P., and Baker, M. J. (1999). Comparing responses of grain legumes, wheat and canola to application of superphosphate. Nutr. Cycl. Agroecosyst. 53, 157–175. Borie, F., and Zunino, H. (1983). Organic matter P associations as a sink in P-fixation processes in allophonic soils of Chile. Soil Biol. Biochem. 15, 599–603. Borkert, C. M., and Barber, S. A. (1983). Effect of supplying P to a portion of the soybean on root growth and P uptake kinetics. J. Plant Nutr. 6, 895–910. Bowman, R. A., and Cole, C. V. (1978). Transformation of organic phosphorus substrate in soils as evaluated by NaHCO3 extraction. Soil Sci. 125, 49–54. Brady, C. N., and Weil, R. R. (2002). The Nature and Properties of Soil. 13 ed. Prentice Hall, Upper Saddle River, NJ, pp. 592–596. (correct for the publisher). Brett, L. A., and Mallarino, A. P. (2006). Relationships between extractable soil phosphorus and phosphorus saturation after long-term fertilizer or manure application. Soil Sci. Soc. Am. J. 70, 454–463. Brink, J. (1977). World resources of phosphorus. In “Ciba Foundation Symposium, 13–15,” pp. 23–48. British Sulphur Corporation. (1971). A World Survey of Phosphate Deposits. third ed.. British Sulphur Corporation, London, 180pp. Buckman, H. O., and Brady, N. C. (1990). The Nature and Properties of Soils. Macmillan Publishing Co., New York. Cakmak, I. (2002). Plant nutrition research: Priorities to meet human needs for food in sustainable ways. Plant Soil 247, 3–24. Cao, A. Q., Liao, H., and Yan, X. (2002). Root architectural responses to low P availability from common bean in the soil in relation to P efficiency. Acta Pedol. Sin. 39, 276–282. CAY (1991). China Agriculture Yearbook. Chinese Agricultural Press, Beijing. CAY (2006). China Agriculture Yearbook. Chinese Agricultural Press, Beijing. Chang, E. H., Zhang, S. F., Wang, Z. Q., Wang, X. M., and Yang, J. C. (2008). Effect of nitrogen and phosphorus on the amino acids in root exudates and grains of rice during grain filling. Acta Agron. Sin. 34(4), 612–618. Chen, W. X., and Chen, W. F. (2004). Exertion of biological nitrogen fixation in order to reducing the consumption of chemical nitrogenous fertilizer. Rev. China Agric. Sci. Technol. 6(6), 3–6.

230

Sheng-Xiu Li et al.

Chen, S. J., Guo, Y. D., Linag, D. Y., Zhang, M. C., and Chen, Y. H. (1963a). Studies on the effect of P fertilizer in calcareous soil. Sci. Agric. Sin. 1, 34–38. Chen, Y. P., Liao, S., and Wang, J. J. (1963b). Agricultural effect of increasing N by P fertilization. Sci. Agric. Sin.(2), 9–13. Chen, S. J., Liu, L. X., and Yang, Z. (1987). Studies on crop requirement to P rate and recommendation of P fertilization. Soil Fertil.(6). Chilikef, B. (1962). Agrochemistry of Phosphorus and Potassium (A Chinese Version Translated by W.Q. Ying) Shanghai Science and Technology Press, China, pp. 29–263. Christie, E. K. (1975). Physiological responses of semiarid grasses. II. The pattern of root growth in relation to external phosphorus concentration. Aust. J. Agric. Res. 26, 437–446. Clark, R. B. (1983). Plant genotypes differences in the uptake, translocation, accumulation and use of mineral elements required for plant growth. Plant Soil 72, 175–196. Clark, R. B., and Duncan, R. R. (1991). Improvement of plant mineral nutrition through breeding. Field Crop. Res. 27, 219–240. Coic, Y., Lesaint, C., and Le Roux, F. (1962). Effects of ammonium and nitrate nutrition and a change of ammonium and nitrate supply on the metabolism of anions and cations in tomatoes. Ann. Physiol. Ve´g. 4, 117–125. Colomb, B., Kiniry, J. R., and Debaeke, P. (2000). Effect of soil phosphorus on leaf development and senescence dynamics of field-grown maize. Agron. J. 92, 428–435. Cornfield, A. H. (1960). Ammonia released on treating soils with N sodium hydroxide as a possible means of predicting the nitrogen—Supplying power of soils. Nature 187, 260–261. Council for Agricultural Science and Technology. (1980). Organic and conventional farming compared. Report No. 84, Ames, Iowa. Da Silva, A. E., and Gabelman, W. A. (1992). Screening maize inbred lines for tolerance to low-P stress conditions. Plant Soil 146, 181–187. Dai, P. A., Zheng, S. X., Li, X. B., Nie, J., Yi, G. Y., Yuan, D. R., and Huang, K. Y. (2006). Effect of proportion of nitrogen fertilizer for promoting panicle development on nitrogen uptake, grain amino acid contents and grain yield of two-line hybrid rice. Chin. J. Rice Sci. 20, 79–83. Dalal, R. C. (1977). Soil organic phosphorus. Adv. Agron. 29, 83–117. Daniel, T. C., Sharpley, A. N., and Lemunyon, J. L. (1998). Agricultural phosphorus and eutrophication: A symposium overview. J. Environ. Qual. 27(2), 251–257. Davies, T. G. E., Ying, J., Xu, Q., Li, Z. S., Li, J., and Gordon-Weeks, R. (2002). Expression analysis of putative high-affinity phosphate transporters in Chinese winter wheats. Plant Cell Environ. 25, 1325–1339. De Smet, J., Hofman, G., Vanderdeelen, J., Van Meirvenne, M., and Baert, L. (1996). Phosphate enrichment in the sandy loam soils of West-Flanders, Belgium. Fertil. Res. 43, 209–215. De Smet, J., Vanderdeelen, J., and Hofman, G. (1998). Effect of soil properties on the kinetics of phosphate release. Common Soil Sci. Plant Anal. 29, 2135–2147. Ding, Z. Y. (1987). Determination of root activity by TTC reduced method. In “Handbook for Determining Plant Physiological Characteristics” (S. T. Wang, Ed.), pp. 35–36. Shaanxi Science and Technology Publisher, Xi’an, China. Ding, H., and Li, S. X. (1997). Genotypic differences of phosphorus accumulation and utilization efficiency between soybean varieties. J. Chin. Oil Plants 19, 52–54. Ding, H., and Li, S. X. (1998). Genetic differences in tolerance to low phosphorus and response to phosphorus fertilizer for soybean varieties. Plant Nutr. Fertil. Sci. 4, 257–263. Ding, Y. Z., Li, Z. A., and Zou, B. (2005). Low-molecular weight organic acids and their ecological roles in soil. Soils 37(3), 243–250.

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

231

Ding, Y. C., Chen, M. C., Cheng, B., Li, L. J., and Zhang, H. S. (2006). The selection of spring soybean genotypes with high phosphorus efficiency in Northern China. Plant Nutr. Fertil. Sci. 12(4), 597–600. Dinkelaker, B., Ro¨mheld, V., and Marschner, H. (1989). Citric acid excretion and precipitation of calcium citrate in the rhizosphere of white lupin (Lupinus albus L.). Plant Cell Environ. 12, 285–292. Djodjic, F., Bergstro¨m, L., and Grant, C. (2005). Phosphorus management in balanced agricultural systems. Soil Use Manag. 21, 94–101. Dou, H., and Steffens, D. (1993). Recovery of 15N labeled urea as affected by fixation of ammonium by clay minerals. Z. Pflanzenerna¨hr. Bodenk. 158, 351–354. Drouillon, M., and Merckx, R. (2003). The role of citric acid as a phosphorus mobilization mechanism in highly P-fixed soils. Gayana Bot. 60(1), 55–62. Duff, S. M. G., Sarath, G., and Plaxton, W. C. (1994). The role of acid phosphatase in plant phosphorus metabolism. Plant Physiol. 90, 791–800. Earl, K. D., Syers, J. K., and Mclaughlin, J. R. (1979). Origin of the effect of citrate tartrate and acetate on phosphate sorption by soils and synthetic gels. Soil Sci. Soc. Am. J. 43, 674–678. Elrashidi, M. A., van Diest, A., and El-Damaty, A. H. (1975). Phosphate determination in highly calcareous soils by the use of an anion exchange resin. Plant Soil 42, 273–286. European Fertilizer Manufacturers Association. (2000). Phosphorus: Essential Element for Food Production. European Fertilizer Manufacturers Association, Brussels. Extension Station of Agricultural Technology in Bijie District, Guizhou Province. (1964). The effect of crop yield increase by application of P fertilizer in the yellow soil areas of Bijie District, Guizhou Province. Chin. J. Soil Sci.(6), 11–15. Fan, M. S., Xu, B., and Wang, Y. (2001). Acid phosphatase activities of intact roots and ground root tissues of maize grown in high P or low P nutrient solution. Rev. Agric. Sci. Technol. 3(3), 33–36. FAO. (2004). Use of Phosphate Rocks for Sustainable Agriculture. Food and Agriculture Organization of the United Nations, Rome. FAO. (2008). FAOSTAT. http://faostat.fao.org/. Fertilizer Laboratory, Soil and Fertilizer Institute, Shandong Province. (1976). Wise application of P fertilizer for raising crop yield. Soil Fertil.(1), 47. Flach, E. N., Quak, W., and van Diest, A. (1987). A comparison of the rock phosphate mobilizing capacities of various crop species. Trop. Agric. 64, 347–352. Fletcher, A. L., Moot, D. J., and Stone, P. J. (2006). The effect of fertiliser P on crop biomass production, partitioning and quality in ‘Challenger’ sweet corn. Aust. J. Agric. Res. 57, 1213–1219. Fletcher, A. L., Moot, D. J., and Stone, P. J. (2008a). Solar radiation interception and canopy expansion of sweet corn in response to phosphorus. Eur. J. Agron. 29, 80–87. Fletcher, A. L., Moot, D. J., and Stone, P. J. (2008b). Radiation use efficiency and leaf photosynthesis of sweet corn in response to phosphorus in a cool temperate environment. Eur. J. Agron. 29, 88–93. Fo¨hse, D., and Jungk, A. (1983). Influence of phosphate and nitrate supply on root hair formation of rape, spinach and tomato plants. Plant Soil 74, 359–368. Fox, R. L., Nishimoto, J. K., Thompson, J. K., and de la Pena, R. S. (1974). Comparative external P requirements of plants growing in tropic soils. Trans. 10th Int. Cong. Soil Sci. 4, 232–239. Moscow, Russia: Izd. Nauka. Fox, T. R., and Comerford, N. B. (1990). Low-molecular weight organic acids in selected forest soils in the Southeastern USA. Soil Sci. Soc. Am. J. 54, 1139–1144. Fried, M., and Shapiro, R. E. (1961). Soil plant relationship in ion uptake. Ann. Rev. Plant Physiol. 12, 91–112.

232

Sheng-Xiu Li et al.

Gahoonia, T. S., Claassen, N., and Jungk, A. (1992). Mobilization of phosphate in different soils by ryegrass supplied with ammonium or nitrate. Plant Soil 140, 241–248. Gahoonia, T. S., Asmar, F., and Giese, H. (2000). Root-released organic acids and phosphate uptake of two barley cultivars in laboratory and field experiments. Eur. J. Agron. 12, 281–289. Gao, H. J., and Wang, R. T. (1963). The effect of soil fertility on the efficiency of P fertilizer. Chin. J. Soil Sci. 6, 53–54. Geelhhoed, J. S., Van Riemsdijk, W. H., and Findenegg, G. R. (1999). Simulation of the effect of citrate exudation from roots on the plant availability of phosphate sorbed on goethite. Eur. J. Soil Sci. 50, 379–390. George, T. S., and Richardson, A. E. (2008). Potential and limitations to improving crops for enhanced phosphorus utilization. In “The Ecophysiology of Plantphosphorus Interactions” ( P. J. White and J. P. Hammond, Eds.), Plant EcoPhysiology Series vol 7, Springer, Dordrecht, Netherlands. Gerke, J., and Meyer, U. (1995). Phosphate acquisition by mustard on a huic podzol. J. Plant Nutr. 18, 2409–2429. Gilbert, G. A., Knight, J. D., Vance, C. P., and Allan, D. L. (1999). Acid phosphatase activity in phosphorus-deficient white lupin roots. Plant Cell Environ. 22, 801–810. Gill, M., Rahmatullah, A., and Salim, M. (1994). Growth responses of twelve wheat cultivars and their phosphorus utilization efficiency from rock phosphate. J. Agron. Crop Sci. 173, 204–209. Gill, H. S., Singh, A., Sethi, S. K., and Behl, R. K. (2004). Phosphorus uptake and use efficiency in different varieties of bread wheat (Trticum aestivum L.). Arch. Agron. Soil Sci. 50, 563–572. Gillespie, A. R., and Pope, P. E. (1989). Alfalfa N2-fixation enhances the phosphorus uptake of walnut in interplantings. Plant Soil 113, 291–293. Goldstein, A., Lester, T., and Brown, J. (2003). Research on the metabolic engineering of the direct oxidation pathway for extraction of phosphate from ore has generated preliminary evidence for PQQ biosynthesis in Escherichia coli as well as a possible role for the highly conserved region of quinoprotein dehydrogenases. Biochim. Biophys. Acta 1647, 266–271. Gollany, H. T., and Schumacher, T. E. (1993). Combined use of colorimetric and microelectrode methods for evaluating rhizosphere pH. Plant Soil 154, 151–159. Graham, P. H. (1998). Biological dinitrogen fixation: Symbiotic. In “Principles and Applications of Soil Microbiology” (D. M. Sylvia, J. J. Fuhrmann, P. G. Hartel, and D. A. Zuberer, Eds.), pp. 322–345. Prentice-Hall, Inc. Simon & Schuster/A Viacom Company, Upper Saddle River, NJ. Granstedt, A. (2000). Increasing the efficiency of plant nutrient recycling within the agricultural system as a way of reducing the load to the environment-experience from Sweden and Finland. Agric. Ecosys. Environ. 80, 169–185. Guan, S. Y. (1983). Soil Enzymes and Research Method. China Agriculture Press, Beijing, China, pp. 309–313. Gumbo, B., 2005. Short-Cutting the Phosphorus Cycle in Urban Ecosystems. PhD thesis, Delft University of Technology and UNESCO-IHE Institute for Water Education Delft. Gumbo, B., and Savenije, H. H. G. (2001). Inventory of phosphorus fluxes and storage in an urban-shed: Options for local nutrient recycling. In “Internet Dialogue on Ecological Sanitation,” 15 November–20 December 2001, Delft. Gunther, F. (2005). A solution to the heap problem: The doubly balanced agriculture: integration with population. Available: http://www.holon.se/folke/kurs/Distans/Ekofys /Recirk/Eng/balanced.shtml. Guo, G. M. (1981). Application of P fertilizer to pea can increase yield. Shanxi Agric. Sci.(6), 277.

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

233

Guo, H. Y., and Luo, W. C. (1986). Effect of different ratios of N and P fertilizers applied on their recovery to wheat. Xinjiang Agric. Sci.(1), 19–21. Guo, Y. C., Lin, W. X., Shi, Q. M., Liang, Y. Y., Chen, F. Y., He, H. Q., and Liang, K. J. (2002). Screening methodology for rice (Oryza sativa) genotypes with high phosphorus use efficiency at their seedling stage. Chin. J. Appl. Ecol. 13, 1587–1591. Gyaneshwar, P., Kumar, G. N., Parekh, L. J., and Poole, P. S. (2002). Role of soil microorganisms in improving P nutrition of plants. Plant Soil 245, 83–93. Hajabbasi, M. A., and Schumacher, T. E. (1994). Phosphorus effects on root growth and development in two maize genotypes. Plant Soil 158, 39–46. Hamdia, A. B. E., Shaddad, M. A. K., and Doaa, M. M. (2004). Mechanisms of salt tolerance and interactive effects of Azospirillum brasilense inoculation on maize cultivars grown under salt stress conditions. Plant Growth Regul. 44(2), 165–174. Hart, M. R., Quin, B. F., and Nguyen, M. L. (2004). Phosphorus runoff from agricultural land and direct fertilizer effects: A review. J. Environ. Qual. 33, 1954–1972. Hash, C. T., Schaffert, R. E., and Peacock, J. M. (2002). Prospects for using conventional techniques and molecular biological tools to enhance performance of ‘orphan’ crop plants on soils low in available phosphorus. Plant Soil 245, 135–146. Haynes, R. J., and Mokolobate, M. S. (2001). Amelioration of Al toxicity and P deficiency in acid soils by additions of organic residues: A critical review of the phenomenon and the mechanisms involved. Nutr. Cycl. Agroecosyst. 59, 47–63. He, W. S. (2004). Introduction to Plant Nutrition. Ningxia People Publishing House, Yinchuan, China, pp. 176–177. He, N. Z., and Meng, C. F. (1987). Principles of Plant Nutrition. Shanghai Science and Technology Press, Shanghai, China. pp.32–33. He, Z. L., Zhu, Z. X., Yuan, K. N., and Huang, C. Y. (1988). The adsorptive characteristics of soil to P and its relation to P supplying indices. Acta Pedol. Sin. 25(4), 397–404. He, Z. L., Yang, X., Yuan, K. N., and Zhu, Z. X. (1994). Desorption and plant availability of phosphate sorbed by some important minerals. Plant Soil 152, 89–97. Heckrath, G., Brookes, P. C., Poulton, P. R., and Goulding, K. W. T. (1995). Phosphorus leaching from soils containing different phosphorus concentrations in the Broad balk experiments. J. Environ. Qual. 24, 904–910. Hedley, M. J., and McLaughlin, M. J. (2005). Reactions of phosphate fertilizers and byproducts in soils. In “Phosphorus: Agriculture and the Environment” ( J. T. Sims and A. N. Sharpley, Eds.), pp. 181–252. ASA, CSSA, and SSSA, Madison, WI, Agron. Monogr. No. 46. Hedley, M. J., Stewart, J. W. B., and Chauhan, B. C. (1982). Changes in inorganic and organic soil phosphorus fractions induced by cultivation practices and by laboratory incubation. Soil Sci. Soc. Am. J. 46, 970–975. Hedley, M. J., Nye, P. H., and White, R. E. (1983). Plant induced changes in the rhizosphere of rape (Brassica napus var emerald) seedlings. IV. The effect of rhizosphere phosphorus status on the pH, phosphatase activity and depletion of soil phosphorus fractions in the rhizosphere and on the cation–anion balance in the plants. New Phytol. 95, 69–82. Heim, A., Luster, J., Brunner, I., Frey, B., and Frossard, E. (1999). Effect of Aluminum treatment on Norway spruce roots: Aluminum binding forms, element distribution, and release of organic substances. Plant Soil 216, 103–116. Hetal, H., and Sauerbeck, D. (1984). Influence of roots on C and P metabolism in soil. Plant Soil 76, 175–182. Hinsinger, P. (1998). How do plant roots acquire mineral nutrients? Chemical processes involved in the rhizosphere. Adv. Agron. 64, 225–265. Hinsinger, P. (2001). Bioavailability of soil inorganic P in the rhizosphere as affected by root-induced chemical changes: A review. Plant Soil 237, 173–195.

234

Sheng-Xiu Li et al.

Hinsinger, P., and Gilkes, R. J. (1995). Root-induced dissolution of phosphate rock in the rhizosphere of lupins grown in alkaline soil. Aust. J. Soil Res. 33, 477–489. Hinsinger, P., and Gilkes, R. J. (1996). Mobilization of phosphate from phosphate rock and aluminasorbed phosphate by the roots of ryegrass and clover as related to rhizosphere pH. Eur. J. Soil Sci. 47, 533–544. Hisanobori, S., and Takana, A. (Japanese) (1983). Effect of deficiency of P on rice root growth. Jap. J. Soil Fertil. 54(6), 485–489. Hoffland, E. (1992). Quantitative evaluation of the role of organic acid exudation in the mobilization of rock phosphate by rape. Plant Soil 140, 279–289. Hoffland, E., Findenegg, G. R., and Nelemans, J. A. (1989). Solubilization of rock phosphate by rape. II. Local root exudation of organic acids as a response to P starvation. Plant Soil 113, 161–165. Holfold, I. C. R., and Ratrick, W. H. (1979). Effects of reduction and pH changes on phosphate sorption and mobility in an acid soil. Soil Sci. Soc. Am. J. 43, 292–297. Holford, I. C. R. (1997). Soil phosphorus: Its measurement, and its uptake by plants. Aust. J. Soil Res. 35, 227–239. Hong, C. Q., and Nie, Y. L. (2003). Effects of root exudates on plant nutrition. Ecol. Environ. 12, 508–511. Horst, W. J., Abdou, M., and Wiesler, F. (1993). Genotypic differences in phosphorus efficiency of wheat. Plant Soil 155(156), 293–296. Horst, W. J., Kamh, M., Jibrin, J. M., and Chude, V. O. (2001). Agronomic measures for increasing P availability to crops. Plant Soil 237, 211–223. Huang, Y. J., and Liu, Y. B. (1989). Study on grain protein content and amino acid composition in different rice varieties. Acta Agric. Jiangxi(1), 25–31. Huang, B. Y., Xiao, D. Y., and Yang, Z. Y. (1964). The investigation of the conditions and methods for application of P fertilizer. Sci. Agric. Sin.(3), 25–33. Hue, N. V., Craddock, G. R., and Adams, F. (1986). Effect of organic acids on aluminum toxicity in subsoils. Soil Sci. Soc. Am. J. 50, 28–34. IFA (International Fertilizer Industry Association). (2005). IFA-Production and International Trade. www.fertilizer. org/ifa/statistics.asp. IFDC (International Fertiliser Development Center). (1998). An analysis of the potential demand for phosphate fertilisers: Sources of change and projections to 2025. IFDCwww.ifdc.org. Illmer, P., Barbato, A., and Schinner, F. (1995). Solubilization of hardly soluble AlPO4 with P-solubilizing microorganisms. Soil Biol. Biochem. 27, 265–270. Institute of Ecology. (1971). Man in the Living Environment. The Institute of Ecology, Chicago, Report of the 1971 Workshop on Global Ecological Problems. Institute of Plant Protection and Soil Fertilizer, Hebei Province. (1975). Rational application of N and P fertilizers. Soil Fertil.(1), 44–46. Itoh, S., and Barber, S. A. (1983). Phosphorus uptake by six plant species as related to root hairs. Agron. J. 75, 457–461. Iyamuremye, F., and Dick, R. P. (1996). Organic amendments and phosphorus sorption by soils. Adv. Agron. 56, 139–185. Izaurralde, R. C., McGill, W. B., and Juma, N. G. (1992). Nitrogen fixation efficiency, interspecies N transfer and root growth in barley-field pea intercrop on a black chernozemic soil. Biol. Fertil. Soils 13, 11–16. Jacob, K. D. (1953). Phosphate resources and processing facilities. In “Fertilizer Technology and Resources in the United States” (K. D. Jacob, Ed.), 3, pp. 117–165. Academic Press, New York. Jasinski, S. M. (2006). Phosphate Rock, Statistics and Information. U.S. Geological Survey. Jasinski, S. M. (2007). Phosphate Rock, Mineral Commodity Summaries U.S. Geological Survey, minerals.usgs.gov/minerals/pubs/commodity/phosphate_rock/.

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

235

Jasinski, S. M. (2008). Phosphate Rock, Mineral Commodity Summaries U.S. Geological Survey, minerals.usgs.gov/minerals/pubs/commodity/phosphate_rock/. Jensen, E. S. (1996). Grain yield, symbiotic N2 and interspecific competition for inorganic N in pea-barley Intercrops. Plant Soil 198(1), 25–38. Jiang, B. F. (1988). Studies on the utilization of Chinese rock phosphates in agriculture. Sci. Agric. Sin. 21(4), 62–67. Jiang, B. F., Duan, P. M., and Wang, Z. Y. (1988). A correlation study between the crystal characteristics and the fertilizer effect of powered phosphate rock. Acta Podol. Sin. 25(4), 387–396. Jin, S. L. (1989). Balanced fertilization of spring wheat in the Hexi Corridor. In “Proceedings of the International Symposium on Balanced Fertilization” (Institute of Soil and Fertilizers, Chinese Academy of Agricultural Sciences, Ed.), pp. 179–184. China Agriculture Press, Beijing. China. Jiang, T. H. (1997). Plant phosphorus nutrition and phosphate fertilizer. In “Principle of Plant Nutrition” (X. Sun, Ed.), pp. 97–133. China Agriculture Press. Johnson, S. E., and Loepertt, R. H. (2006). Role of organic acids in phosphate mobilization from iron oxide. Soil Sci. Soc. Am. J. 70, 222–234. Johnson, J. F., Vance, C. P., and Allan, D. L. (1996). Phosphorus deficiency in Lupinus albus: Altered lateral root development and enhanced expression of phosphoenolpyruvate carboxylase. Plant Physiol. 112, 31–41. Johnston, A. E. (1994). The Rothamsted classical experiments. In “Long-Term Experiments in Agricultural and Ecological Sciences” (R. A. Leigh and A. E. Johnston, Eds.), pp. 9–37. CAB International, Wallingford, UK. Johnston, A. E., and Poulton, P. R. (1992). The role of phosphorus in crop production and soil fertility: 150 years of field experiments at Rothamsted, United Kingdom. In “Phosphate Fertilizers and the Environment, Proceedings of an International Workshop” ( J. J. Schultz, Ed.), pp. 45–64. International Fertilizer Development Centre, Tampa, FL. Jones, D. L. (1998). Organic acids in the rhizosphere—A critical review. Plant Soil 205, 25–44. Jones, D. L., and Darrah, P. R. (1994). Role of root derived organic acids in the mobilization of nutrients from the rhizosphere. Plant Soil 166, 247–257. Jones, D. L., and Kochian, L. V. (1996). Aluminum-organic acid interactions in acid soils. I. Effect of root derived organic acids on the kinetics of Al dissolution. Plant Soil 182, 221–228. Ju, X. T., Kou, C. L., Christie, P., Dou, Z. X., and Zhang, F. S. (2007). Changes in the soil environment from excessive application of fertilizers and manures to two contrasting intensive cropping systems on the North China Plain. Environ. Pollut. 145, 497–506. Kababo, I., and Gilkes, R. J. (1988). A comparison of mixed and band applications on the dissolution of North Carolina phosphate rock and on bicarbonate-soluble soil phosphorus. Nutr. Cycl. Agroecosyst. 15(1), 3–12. Keyser, H. H., and Li, F. (1992). Potential for increasing biological nitrogen fixation in soybean. Plant Soil 141, 119–135. Khasawneh, F. E., and Doll, E. C. (1978). The use of phosphate rock for direct application to soils. Adv. Agron. 30, 159–206. Khasawneh, F. E., and Doll, E. C. (1984). Direct application of rock phosphate. Prog. Soil Sci.(2), 29–36. Kirk, G. J. D., Santo, E. E., and Santo, M. B. (1999). Phosphate solubilization by organic anion excretion from rice growing in aerobic soils: Rates of excretion and decomposition, effects of rhizosphere pH and effects on phosphate solubility and uptake. New Phytol. 142, 185– 200.

236

Sheng-Xiu Li et al.

Kirkby, E. A. (1968). Influence of ammonium and nitrate nutrition on the cation-anion balance and bitrogen and carbohydrate metabolism of white mustard plants grown in dilute nutrient solutions. Soil Sci. 105, 133–141. Klee, R. J., and Graedel, T. E. (2004). Elemental cycles: A status report on human or natural dominance. Ann. Rev. Environ. Resour. 29, 69–107. Klindworth, A. (2000). Fertilizer market trends. Presentation at the Kentucky Fertilizer and Chemical Association. The Fertilizer Institute. www.tfi.org. Koutroubas, S. D., Papakosta, D. K., and Gagianas, A. A. (1998). The importance of early dry matter and nitrogen accumulation in soybean yield. Eur. J. Agron. 9, 1–10. Kpomblekou, A. K., and Tabatabai, M. A. (2003). Effect of low-molecular weight organic acids on phosphorus release and phytoavailability of phosphorus in phosphate rocks added to soils. Agric. Ecosyst. Environ. 100, 275–284. Kraffczyk, I., Tolldenier, G., and Beringer, H. (1984). Soluble root exudates of maize: Influence of potassium supply and rhizosphere micro-organisms. Soil Biol. Biochem. 16, 315–322. Kucey, R. M. N., Janzen, H. H., and Leggett, M. E. (1989). Microbially mediated increases in plant-available phosphorus. Adv. Agron. 42, 199–228. Lambers, H., Juniper, D., Cawthray, G. R., Veneklaas, E. J., and Martinez-Ferri, E. (2002). The pattern of carboxylate exudation in Banksia grandis (Proteaceae) is affected by the form of phosphate added to the soil. Plant Soil 238, 111–122. Lei, H. F., and Wang, G. H. (2005). Comparison of the solubility of three rock phosphates with different compositions. J. Zhejiang Univ. 31(1), 17–21. Lewis, D. B., and Quirk, J. P. (1965). Diffusion of phosphate to plant roots. Nature 205, 765–766. Li, Q. K. (1959). The research progress of soil chemistry and agrochemistry in recent years. Acta Pedol. Sin. 7(1–2), 1–8. Li, Y. H. (1962). Study on the factors affecting P fertilizer effect in Guanzhong, Shaanxi Province. 1. Effect of some chemical properties on P fertilizer. Acta Pedol. Sin. 13(1), 39–45. Li, Y. H. (1963). Comparison of some methods for determination of soil available P. Acta Pedol. Sin. 11(2), 215–219. Li, Q. K. (1964). The effect of soil phosphorus components and phosphate sorts on crop growth. Sci. Agric. Sin. 10, 23–26. Li, Y. H. (1965). Investigation on the factors affecting P fertilizer efficiency in calcareous soil in Guanzhong. I. the influence of some soil agrochemical properties on the efficiency of P fertilizer. Acta Pedol. Sin. 13(1), 39–44. Li, Q. K. (1966a). The theory basis for the effect of rock phosphate. Chin. Sci. Bull. 2, 49–57. Li, Q. K. (1966b). The interrelation of phosphate sorts and soil characteristics with plant P nutrition. Acta Pedol. Sin. 14(1), 117–122. Li, Y. H. (1981). Investigation on the factors affecting P fertilizer efficiency in calcareous soil in Guanzhong. II. The influence of application of N and organic fertilizer on the efficiency of calcium superphosphate. Shaanxi Agric. Sci.(2), 7–13. Li, Y. H. (1983). Evaluation of rational application of P fertilizer from soil available P and N-supplying capacities. Chin. J. Soil Sci. 14(4), 17–20. Li, Q. K. (1986). The progress of phosphate fertilizer research in modern times. Prog. Soil Sci. (2), 1–7. Li, S. X. (1999). The current research situation and prospect of plant nutrition and fertilizer science in China. Plant Nutr. Fertil. Sci. 5(3), 193–205. Li, H. T. (2004a). Prescription fertilization in dryland areas of China. In “Dryland Agriculture in China” (S. X. Li, Ed.), pp. 483–503. China Agriculture Press, Beijing, China (in Chinese).

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

237

Li, F. M. (2004b). Mechanism of Interspecific Facilitation on Phosphorus Uptake by Crops in Intercropping Systems. PhD Thesis, China Agricultural University, Beijing, China, pp. 25–80. Li, Z. Y. (2004c). Management of phosphorus in drylands of China. In “Dryland Agriculture in China” (S. X. Li, Ed.), pp. 390–413. China Agriculture Press, Beijing, China. Li, X. L. (2005). Plant phosphorus nutrition and application of phosphate fertilizer. In “Soil and Fertilizer Science” (X. Lu, Ed.), pp. 208–212. China Agricultural University Press. Li, S. X. (2007). Dryland Agriculture in China. Science Press, Beijing, China, pp. 452–455. Li, S. X. (2008). The nitrate and ammonium nutrition of plants. In “Soil and Plant Nitrogen in Dryland Areas of China” (S. X. Li, Ed.), pp. 449–596. Science Press, Beijing, China. Li, Q. K., and Hu, Z. G. (1956). The third report on the effect of Haizhou rock phosphate experiments on agriculture in Ganjia-Shan Experimental Station. Acta Pedol. Sin. 4(1), 43–49. Li, S. X., and Li, S. Q. (1992). Responses of wheat, hairy vetch and pea to phosphate fertilizer. Acta Univ. Agric. Boreali-occidentalis 20(Suppl.), 74–78. Li, Y. K., and Wang, S. Z. (1965). The efficiency and rational utilization of P fertilizer. Chin. J. Soil Sci.(1), 18–20. Li, S. X., and Wang, Z. H. (2010). Comprehensive management of nutrients in dryland soils of China for sustainable agriculture. In “Food Security and Soil Quality” (R. Lal and B. A. Stewart, Eds.), pp. 310–408. CRC Press. New York. Li, S. X., and Xiao, L. (1992). Distribution and management of drylands in the People’s Republic of China. Adv. Soil Sci. 18, 147–302. Li, S. X., and Zhao, B. S. (1990). The effect of soil nitrogen supplying capacity on phosphate fertilizer efficiency for some legume crops and non-legume crops. Soil Fertil. 4, 19–23. Li, Z. P., and Zou, Q. X. (1962). The efficiency of P fertilizer in Shanxi drab soil. Chin. J. Soil Sci.(1), 49–56. Li, Q. K., He, J. H., and Hu, Z. G. (1952). The preliminary report on the effect of Haizhou rock phosphate on agriculture. Acta Pedol. Sin. 2(1), 37–42. Li, Q. K., He, J. H., Lu, R. K., and Hu, Z. G. (1953). The second report on the effect of Haizhou rock phosphate experiments on agriculture. Acta Pedol. Sin. 2(3), 107–117. Li, S. X., Zhao, B. S., and Li, C. W. (1978). Rational application of phosphate fertilizer to wheat, part I. Acta Collegii Septentrionali Occident. Agric. 6(1), 77–85. Li, S. X., Zhao, B. S., and Li, C. W. (1979a). Rational application of phosphate fertilizer to wheat, part II. Acta Collegii Septentrionali Occident. Agric. 7(1), 55–99. Li, S. X., Li, C. W., and Zhao, B. S. (1979b). Effects of application of phosphate fertilizer on yield increases of wheat and pea. Shaanxi Sci. Tech. Inform.(3), 30–31. Li, S. X., Zhao, B. S., Li, C. W., and Li, S. Q. (1988). Some factors affecting phosphate fertilizer to pea. Agric. Res. Arid Areas(1), 56–63. Li, S. X., Zhao, B. S., and Li, C. W. (1990). The effect of application of N, P and organic fertilizer on the N fixation of leguminous crops. Chin. J. Soil Sci. 21(1), 13–15, 29. Li, Q. K., Jiang, B. F., and Lu, R. K. (1992a). Phosphate Rock for Agriculture Use in China Jiangsu Science and Technology Press, Nanjing, China, 12–19, 28–40, 50–53 and 140–142. Li, S. X., Yuan, H. L., He, H. X., and Li, S. Q. (1992b). Evaluation of the effect of N fixation by two leguminous crops through crop uptake amount of N from soil in successive three years. Chin. J. Soil Sci. 23(3), 130–132. Li, S. X., Zhao, B. S., and Li, C. W. (1992c). The relationship between some properties of calcareous soil and crop responses to phosphate fertilizer. Acta Univ. Agric. Borealioccidentalis 20(suppl), 68–73. Li, S. X., Chang, Q., and He, L. (1992d). Nitrogen loss from plants by volatilization II. The relation of nitrogen reduction from aboveground plants at maturity to soils, crops and fertilization. Acta Univ. Agric. Boreali-occidntalis 20(suppl), 7–11.

238

Sheng-Xiu Li et al.

Li, S. X., Zhao, B. S., and Li, S. Q. (1995). The causes of the differences of legume and nonlegume crop responses to P fertilizers. Agric. Res. Arid Areas 13(2), 45–56. Li, L., Yang, S. C., Li, X. L., Zhang, F. S., and Christie, P. (1999). Interspecific complementary and competitive interactions between intercropped maize and faba bean. Plant Soil 212, 105–114. Li, L., Zhang, F. S., and Li, X. L. (2003a). Interspecific facilitation of nutrient uptake by intercropped maize and faba bean. Nutr. Cycl. Agroecosyst. 65, 61–71. Li, L., Sun, J. H., Zhang, F. S., Guo, T. W., and Li, W. X. (2003b). Effect of inter specific interactions between maize and different legumes on phosphorus uptake by maize in legumes/maize intercropping under two levels of nitrogen supply. In “Proceedings of 2nd International Symposium on Phosphorus Dynamics in the Soil–Plant Continuum,” 21–26 September 2003, Perth, Australia, pp. 25–26. Li, S. C., Gong, J., and Wang, J. (2003c). Screening maize inbred lines seedling for tolerance to low phosphate stress. J. Maize Sci. 11, 85–89. Li, F., Pan, X. H., Liu, S. Y., Li, M. Y., and Yang, F. S. (2004). Effect of phosphorus deficiency stress on root morphology and nutrient absorption of rice cultivars. Acta Agron. Sin. 30, 438–442. Li, L., Li, S. M., Sun, J. H., Zhou, L. L., Bao, X. G., and Zhang, H. G. (2007). Diversity enhances agricultural productivity via rhizosphere phosphorus facilitation on phosphorus-deficient soils. PNAS 104, 11192–11196. Li, S. X., Wang, Z. H., Malhi, S. S., Li, S. Q., Gao, Y. J., and Tian, X. H. (2009). Nutrient and water management effects on crop production, and nutrient and water use efficiency in dryland areas of China. Adv. Agron. 102, 223–265. Lin, C. G., Li, Z. P., Zhang, Y. H., and Zou, Q. X. (1964). Study on improvement of P fertilizer effect in the calcareous brown soil areas of Shanxi Province. Chin. J. Soil Sci.(1), 4–12. Lin, Z., Xie, C. T., Zhang, Z. S., and Zhang, X. Y. (1997). Morphology, transformation and availability of inorganic phosphorus in calcareous soil. Chin. J. Soil Sci. 28(6), 274–276. Ling, Q. H., and Ling, L. (1984). The function of different layer roots and their effect on rice yield. Sci. Agric. Sin.(4), 3–11. Liu, Z. X. (1964). The preliminary analysis of increasing N by P fertilization. Chin. J. Soil Sci. (5), 33–35. Liu, Z. X. (1980). Studies on application of P and organic fertilizers to leguminous crops. Chin. J. Soil Sci.(1), 36–40. Liu, Z. Y., and Cao, Y. C. (1980). Study on the characteristics of roots absorbing phosphoric acid anion. Sci. China 7, 707–711. Liu, S. Q., and Song, F. B. (2007). The content of the main free amino acids in root system of maize during reproduction stage. Acta Agric. Boreal-Sin. 22, 35–39. Liu, X. L., and Zhao, R. X. (1988). Response of wheat to N, P and K fertilizer in Loessial soil and their rational application. Chin. J. Soil Sci. 19(4), 168–170. Liu, Z. H., and Zhou, X. F. (1992). Soil P state in Hebei Province and application of P fertilizer. Soils 24(2), 77–79. Liu, G. S., Zhou, W. J., Wu, D. G., and Li, Q. K. (1963). Response of crops to superphosphate as affected by the rate of supply of soil nitrogen and by the application of ammonium sulfate in five main soils of Jiangsu Province. Acta Pedol. Sin. 11(1), 25–34. Liu, Y. Z., Wu, J. M., and Chen, Z. Q. (1981). Crop nutrients and effect of P fertilizer. Chin. J. Soil Sci. 12(1), 23–24. Liu, Z. H., Luo, Y. Y., and Zhou, X. F. (1989). The results of long-term application of N, P and K fertilizers to soils with 2 crops per year and its effect on soil fertility. In “Proceedings of the International Symposium on Balanced Fertilization” (Soil and Fertilizer Institute, China’s Academy of Agriculture Sciences, Ed.), pp. 172–178. China Agriculture Press, Beijing. China.

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

239

Liu, P. L., Lin, Q., Sui, F. G., and Sun, Z. Q. (1994). A study on the characteristics of root system in high-yield upright-leaf maize. Maize Sci. 2(1), 59–63. Liu, H., Liu, J. F., and Liu, W. D. (1999). Studies on difference of root morphological and physiological character of different oilseed variety of phosphorus nutrients efficiency. Plant Nutr. Fertil. Sci. 5(1), 40–45. Liu, Q., Loganathan, P., Hedley, M. J., and Skinner, M. F. (2004). The mobilization and fate of soil and rock phosphate in the rhizosphere of ectomycorrhizal pinus radiate seedlings in an allophonic soil. Plant Soil 264, 219–229. Liu, C. H., Zhang, K. W., Zhang, J. R., and Dong, S. T. (2006). Morphological and physiological characteristics of P-efficient maize hybrid to stress of phosphorus deficiency. Plant Nutr. Fertil. Sci. 12(3), 327–333. Liu, L., Liang, C. H., Wang, Q., Du, L. Y., Wuc, Y. M., and Han, W. (2009). Effects of low-molecular-weight organic acids on soil phosphorus release. Plant Nutr. Fertil. Sci. 15 (3), 593–600. Lo¨bermann, B. E., Ko¨hne, S., and Ko¨ppen, D. (2007). Effect of organic, inorganic and combined organic and inorganic P fertilization on plant P uptake and soil P pools. J. Plant Nutr. Soil Sci. 170, 623–628. Lo´pez-Bucio, J., Nieto-Jacobo, M. F., Ramirezrodrigunz, V., and Herrera- Estrella, L. (2000). Organic acid metabolism in plants: From adaptive physiology to transgenic varieties for cultivation in extreme soils. Plant Sci. 160, 1–13. Lopez-Hernander, D., Siegert, G., and Rodriguez, J. V. (1986). Competitive adsorption of phosphate with malate and oxalate by tropical soil. Soil Sci. Soc. Am. J. 50, 1460–1462. Lu, R. K., and Jinag, B. F. (1962). Some matters for application of P fertilizer in some paddy soils in south China. Acta Pedol. Sin. 10(2), 175–182. Lu, W. L., Wang, J. G., and Cao, Y. P. (1998). Kinetics of phosphorus release from soils as affected by organic acids with low molecular weight. Acta Pedol. Sin. 35(4), 493–499. Lu, R. K. (1998). Soil phosphorus and phosphate fertilizer application. In “Principles of Soil and Plant Nutrition and Fertilization” (R. K. Lu, Ed.), pp. 201–203. Chemical Industry Press, Beijing, China. Lynch, J. M., and Whipps, J. M. (1990). Substrate flow in the rhizosphere. Plant Soil 129, 1–10. Maloth, S., and Prased, R. (1979). Relative efficiency of rock phosphate and superphophate for cowpea (Vigna sinensia Savi.). Plant Soil 45, 295–300. Ma˚rald, E. (1998). I mo¨tet mellan jordbruk och kemi: agrikulturkemins framva¨xt pa˚Lantbruksakademiens Experimentalfa¨lt 1850–1907, Skogsoch lantbrukshistoriska meddelanden nr 21, Umea˚ Universitet, Ide´historiska institutionen, Ide´historiska Skrifter nr. 26. (Stockholm), 91–101. Marschner, H., and Ro¨mheld, V. (1983). In vivo measurement of root-induced pH changes at soil-root interface. Effect of plant species and nitrogen source. Z. Pflanzenphysiol. 111, 241–251. Marschner, H., and Ro¨mheld, V. (1994). Strategies of plants for acquisition of iron. Plant Soil 165, 261–274. Matsui, S. (1997). Night soil collection and treatment in Japan. In “Ecological Alternatives in Sanitation” ( J. O. Drangert, J. Bew, and U. Winblad, Eds.), Publications on Water Resources, No. 9. Sida, Stockholm. Mattson, S. (1948). Laws of ionic exchange III. Ann. Agr. Coll. Sweden 15, 308–316. McBeath, T. M., Smernik, R. J., Lombi, E. Y., and McLaughlin, M. J. (2006). Hydrolysis of pyrophosphate in a highly calcareous soil: A solid-state phosphorus-31 NMR study. Soil Sci. Soc. Am. J. 70, 856–862. McKenzie, R. H., Middleton, A. B., Solberg, E. D., DeMulder, J., Flore, N., Clayton, G. W., and Bremer, E. (2001). Response of pea to rate and placement of triple superphosphate fertilizer in Alberta. Can. J. Plant Sci. 81, 645–649.

240

Sheng-Xiu Li et al.

McKenzie, R. H., Middleton, A. B., Hall, L., and Bremer, E. (2004). Fertilizer response on barley grain in South and central Alberta. Can. J. Soil Sci. 84, 513–523. McKenzie, R. H., Middleton, A. B., and Bremer, E. (2005). Fertilization, seeding date, and seeding rate for malting barley yield and quality in southern Alberta. Can. J. Plant Sci. 85, 603–614. McLean, E. O., Adama, D., and Franklin, R. E. Jr. (1956). Cation exchange capacity of plant roots as related to their nitrogen contents. Soil Sci. Soc. Am. Proc. 20(3), 345–347. Mehlich, A., and Drake, M. (1955). Soil chemistry and plant nutrition. In “Chemistry of the Soil” (F. E. Bear, Ed.). Reinhold Publishing Corporation, USA(Translated in Chinese by K.N. Yuan, Z.X. Zhu, and Z.Y. Yu), Science Press (1959), pp. 212–243. Meng, C. F. (1989). The relationship between the composition of apatite and solubility in acid soils. Chin. J. Soil Sci. 20(2), 72–73, 79. Meng, C. F., Cao, Z. H., Jiang, P. K., Xu, Q. F., and Xu, Q. F. (2007). Nutrient absorption and yield of rape under coupled application of phosphate rock and sulfur. Chin. J. EcoAgric. 15(6), 81–83. Mengel, K. (1984). Erna¨hrung und Stoffwechsel der Pflanze. Gustar Fischer Verlag, Stuttgart, pp. 319–321. Mengel, K., and Kirkby, E. A. (2001). Principles of Plant Nutrition fifth ed. Kluwer Acadeic Publishers, Dordrecht/Boston/London, pp. 47, 452 and 453–464. Mi, G. H., Xing, J. P., Chen, F. J., Liu, X. S., and Liu, Y. (2004). Maize root growth to tolerance to low phosphorus. Plant Nutr. Fertil. Sci. 10(5), 468–472. Ming, F., Mi, G. H., Zhang, F. S., Zheng, X. W., and Zhu, L. H. (2000). Studies on varietal difference of rice in response to low P stress and its physiological adaptive mechanism. Chin. J. Appl. Environ. Biol. 6, 138–141. Ming, F., Lou, Y. X., Liang, B., Shen, D. L., Ye, M. M., and Zhu, L. H. (2002). Genetic and physiological mechanism of P-uptake ability difference between different rice genotypes under low P stress. Chin. J. Appl. Environ. Biol. 8, 473–477. Mitchell, D. B., Vogel, K., Weimann, B. J., Pasamontes, L., and van Loon, A. P. G. M. (1997). The phytase subfamily of histidine acid phosphatases: Isolation of genes for two novel phytases from the fungi Aspergillus terreus and Myceliophthora thermophila. Microbiology 143, 245–252. Mitsch, W. J., and Gosselink, J. G. (2000). Wetlands. third ed. Wiley, NYC, 920pp. Mo, H. D. (1993). Quality improvement of rice grain in China. Sci. Agric Sin. 26, 8–14. Mou, J. M., Li, W. H., Zhang, F. X., and Jiang, Y. (1996). Root exudates and their function. J. Jilin Agric. Univ. 18, 114–118. Mudge, S. R., Smith, F. W., and Richardson, A. E. (2003). Root-specific and phosphateregulated expression of phytase under the control of a phosphate transporter promoter enables Arabidopsis to grow on phytate as a sole P source. Plant Sci. 165, 871–878. Murdock, J. T., Corey, R. B., Bhure, N. D., and Waugh, D. L. (1960). Special techniques for the use of radioisotopes in soil and fertilizer research. Trans. 7th Intern. Congr. Soil Sci. 3, 526–534. Nacamulli, C., Bevivino, A., Dalmastri, C., Tabacchioni, S., and Chiarini, L. (1997). Perturbation of maize rhizosphere microflora following seed bacterization with Burkholderia cepacia MCI 7. FEMS Microbiol. Ecol. 23, 183–193. Nagarajar, S., Posner, A., and Quirk, J. (1968). Desorption of phosphate from kaolinite by citrate and bicarbonate. Soil Sci. Soc. Am. J. 32, 507–510. Nair, V. D., Portier, K. M., Graetz, D. A., and Walker, M. L. (2004). An environmental threshold for degree of phosphorus saturation in sandy soils. J. Environ. Qual. 33, 107–113. Narsian, V., Thakkar, J., and Patel, H. H. (1994). Isolation and screening of phosphate solubilizing fungi. Ind. J. Microbiol. 34, 113–118.

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

241

National Statistics Bureau, P.R. China. (2006a). China Statistics Yearbook. Chinese Statistics Press, Beijing, China, (National Statistics Bureau, P.R. China). National Statistics Bureau, P.R. China. (2006b). China Statistics Yearbook. National Statistics Bureau, PR China, http://www.stats.gov.cn/tjsj/ndsj/2006/indexch.htm3-2. Neumann, G., and Ro¨mheld, V. (1999). Root excretion of carboxylic acids and protons in phosphorus deficient plants. Plant Soil 211, 121–130. Nozu, Y., Tsugita, A., and Kamijo, K. (2006). Proteomic analysis of rice leaf, stem and root tissues during growth course. Proteomics 6, 3665–3670. Nuruzzaman, M., Lambers, H., Bolland, M. D. A., and Veneklaas, E. J. (2005). Phosphorus benefits of different legume crops to subsequent wheat grown in different soils of Western Australia. Plant Soil 271, 175–187. Nye, P. H. (1981). Changes of pH across the rhizosphere induced by roots. Plant Soil 61, 7–26. Oliveira, C. A., Alves, V. M. C., Marriel, I. E., Gomes, E. A., Scotti, M. R., Carneiro, N. P., Guimara˜es, C. T., Schaffert, R. E., and Sa´, N. M. H. (2009). Phosphate solubilizing microorganisms isolated from rhizosphere of maize cultivated in an oxisol of the Brazilian Cerrado Biome. Soil Biol. Biochem. 41(9), 1282–1287. Olsen, S. R., Kemper, W. D., and Jackson, J. D. (1962). Phosphate diffusion to plant roots. Soil Sci. Soc. Am. Proc. 26, 222–227. Onthong, J., Osaki, M., and Nilnond, C. (1999). Phosphorus status of some highly weathered soils in peninsular Thailand and availability in relation to citrate and oxalate application. Soil Sci. Plant Nutr. 45, 627–637. Osborne, L. D., and Rengel, Z. (2002a). Screening cereals for genotypic variation in efficiency of phosphorus uptake and utilization. Aust. J. Agric. Res. 53, 295–303. Osborne, L. D., and Rengel, Z. (2002b). Genotypic differences in wheat for uptake and utilization of P from iron phosphate. Aust. J. Agric. Res. 53, 837–844. Osborne, S. L., and Riedell, W. E. (2006). Starter nitrogen fertilizer impact on soybean yield and quality in the Northern Great Plains. Agron. J. 98, 1569–1574. Otani, T., and Ae, N. (1999). Extraction of organic phosphorus in andosols by various methods. Soil Sci. Plant Nutr. 45, 151–161. Otani, T., Ae, N., and Tanaka, H. (1996). Phosphorus (P) uptake mechanisms of crops grown in soils with low P status. II. Significance of organic acids in root exudates of pigeon pea. Soil Sci. Plant Nutr. 42, 553–560. Ozanne, P. G. (1980). Phosphate nutrition of plants—A general treatise. In “The Role of Phosphorus in Agriculture” (P. E. Khasawneh, E. C. Sample, and E. J. Kamprath, Eds.), pp. 559–589. American Society of Agronomy, Crop Science Society of America, Soil Science Society of America, Madison, WI, USA. Pan, X. W., Li, W. B., Zhang, Q. Y., Li, Y. H., and Liu, M. S. (2008). Assessment on phosphorus efficiency characteristics of soybean genotypes in phosphorus-deficient soils. Agric. Sci. China 7(8), 958–969. Parfitt, R. L. (1978). Anion adsorption by soil and soil materials. Adv. Agron. 30, 1–50. Pate, J., and Watt, M. (2001). Roots of Banksia spp. (Proteaceae) with special reference to functioning of their specialized root clusters. In “Plant Roots: The Hidden Half” (Y. Waisel, A. Eshel, and U. Kafkafi, Eds.), third ed., pp. 989–1006. Marcel Dekker Inc., New York, NY, USA. Pellerin, S., Mollier, A., and Ple´net, D. (2000). Phosphorus deficiency affects the rate of emergent and number of maize adventitious nodal roots. Agron. J. 92, 690–697. Peng, K. M. (1990). In “Agrochemistry” (K. M. Peng, Ed.), second ed., pp. 140–141. China Agriculture Press, Beijing, China. Peng, K. M., and Pei, B. Y. (1980). In “Agrochemistry” (K. M. Peng and B. Y. Pei, Eds.), first ed., pp. 133–134. China Agriculture Press, Beijing, China.

242

Sheng-Xiu Li et al.

Polomski, J., and Kuhn, N. (2002). Root Research Methods. Marcel Dekker, New York, pp. 313–314. Prakash, A., Tandon, V., and Sharma, N. C. (2004). Effect of rock phosphate and VAM inoculation on growth and nutrient uptake in Sesamum indicum L. Physiol. Mol. Biol. Plants 10, 137–141. Qin, S. J., Liu, J. S., and Wang, G. P. (2007). Phosphorus fractions under different land uses in Sanjiang plain. Environ. Sci. 28(12), 2777–2782. Raghothama, K. G. (1999). Phosphate acquisition. Ann. Rev. Plant Physiol. Plant Mol. Biol. 50, 665–693. Raghothama, K. G., and Karthikeyan, A. S. (2005). Phosphate acquisition. Plant Soil 274, 37–49. Rahmatullah, M. Ashraf, Maqsood, M. A., Kanwal, S., Tahir, M. A., and All, M. (2009). Growth responses of wheat cultivars to rock phosphate in hydroponics. Pedosphere 19(3), 398–402. Ramirez, R., and Lizaso, J. I. (2006). Maize inbreds response to Riecito phosphate rock and triple superphosphate on an acidic soil. Commun. Soil Sci. Plant Anal. 37, 1917–1931. Ramirez, R., Fernandez, S. M., and Lizaso, J. I. (2001). Changes of pH and available phosphorus and calcium in rhizosphere of aluminum-tolerant maize germplasm fertilized with phosphate rock. Commun. Soil Sci. Plant Anal. 32, 1551–1565. Raven, J. A., Franco, A. A., De Jesus, E. L., and Jacob-Neto, J. (1990). Hþ extrusion and organic acid synthesis in N2-fixing symbioses involving vascular plant. New Phytol. 114, 369–389. Rengel, Z. (1999). Physiological mechanisms underlying differential nutrient efficiency of crop genotypes. In “Mineral Nutrition of Crops: Fundamental Mechanisms and Implications” (Z. Rengel, Ed.), pp. 227–265. Haworth Press, New York. Richardson, A. E. (1994). Soil microorganisms and phosphorus availability. In “Soil Biota Management in Sustainable Farming System” (C. E. Pankhurst, B. M. Doulse, V. V. S. R. Gupta, and P. R. Grace, Eds.), pp. 50–62. CSIRO, Australia. Richardson, A. E. (2001). Prospects for using soil microorganisms to improve the acquisition of phosphorus by plants. Aust. J. Plant Physiol. 28, 897–906. Richardson, A. E., George, T. S., Hens, M., and Simpson, R. J. (2005). Utilization of soil organic phosphorus by higher plants. In “Organic Phosphorus in the Environment” (B. L. Turner, E. Frossard, and D. S. Baldwin, Eds.), pp. 165–184. CABI Publishing, Wallingford. Riley, H. (2007). Long-term fertilizer trials on loam soil at Mystad, south-eastern Norway: Crop yields, nutrient balances and soil chemical analysis from 1983 to 2003. Acta Agric. Scand. B Plant Soil Sci. 57, 140–154. Riley, D., and Barber, S. A. (1971). Effect of ammonium and nitrate fertilization on phosphorus uptake as related to root-induced pH changes at the root–soil interface. Soil Sci. Soc. Am. Proc. 35, 301–306. Robert, H., and Hans, W. P. (2008). Coastal marine eutrophication: Control of both nitrogen and phosphorus is necessary. PNAS 105(49), 1073. Rock Phosphate Research Group, Institute of Soil Science, Chinese Academy of Sciences. (1966). What are the crops that are suitable for application of rock phosphate? Acta Pedol. Sin. 14(1), 83–84. Rodrigues, H., and Fraga, R. (1999). Phosphate solubilizing bacteria and their role in plant growth promotion. Biotechnol. Adv. 17, 319–339. Ro¨mheld, V., Mu¨ller, C., and Marschner, H. (1984). Localization and capacity of proton pumps in roots of intact sunflower plants. Plant Physiol. 76, 603–606. Rosmarin, A. (2004). The Precarious Geopolitics of Phosphorus. Down to Earth: Science and Environment Fortnightly, pp. 27–31. New Delhi, India.

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

243

Rubæk, G. H., and Sibbesen, E. (2000). Long-term phosphorus fertilization—Effects on crop yield and soil phosphorus status. In “DIAS Report Plant Production”, Danish Institute of Agricultural Sciences, Foulum, p. 31. Runge-Metzger, A. (1995). Closing the cycle: Obstacles to efficient P management for improved global food security. In “Phosphorus in the Global Environment” (H. Tiessen, Ed.). John Wiley & Sons Ltd, NewYork. Saarela, I., Ja¨rvi, A., Hakkola, H., and Rinne, K. (2004). Phosphorus status of diverse soils in Finland as influenced by long-term P fertilization. 2. Changes of soil test values in relation to P balance with references to incorporation depth of residual and freshly applied P. Agric. Food Sci. Finland 13, 276–294. Saleque, M. A., and Kirk, G. J. P. (1995). Root-induced solubilization of phosphate in the rhizosphere of lowland rice. New Phytol. 129, 325–336. Sa´nchez, C. A. (2007). Phosphorus. In “Handbook of Plant Nutrition” (A. V. Broker, D. J. Pilbeam, and F. Taylor, Eds.), pp. 51–82. Group Boca Raton, London, New York. Santalla, M., Amurrio, J. M., Rodin˜o, A. P., and De Ron, A. M. (2001). Variation in traits affecting nodulation of common bean under intercropping with maize and sole cropping. Euphytica 122, 243–255. Sas, L., Rengel, Z., and Tang, C. (2001). Excess cation uptake, and extrusion of protons and organic acid anions by Lupinus albus under phosphorus deficiency. Plant Sci. 160, 1191–1198. Sas, L., Rengel, Z., and Tang, C. (2002). The effect of nitrogen nutrition on cluster root formation and proton extrusion by Lupinus albus. Ann. Bot. 89, 435–442. Schenk, M. K., and Barber, S. A. (1979). Root characteristics of corn genotypes as related to P uptake. Agron. J. 71, 921–924. Schenk, M. K., and Barber, S. A. (1980). Potassium and phosphorus uptake by corn genotype grown in the field as influenced by root characteristics. Plant Soil 54(1), 65–76. Schlesinger, W. H. (1997). Biogeochemistry. An Analysis of Global Change second ed., Academic Press, USA, 588pp. Schmid Neset, T. S., Bader, H. P., Scheidegger, R., and Lohm, U. (2008). The flow of phosphorus in food production and consumption—Linko¨ping, Sweden, 1870–2000. Sci. Total Environ. 396, 111–120. Setanac, H., Mousin, D., and Salsac, L. (1980). Measurement of phosphatase activity in cellulose cell wall material obtained with the help of a non ionic detergent. C. R. Acad. Sci. Paris Ser. D 290, 21. Seyhan, D. (2006). Development of a method for the regional management of long-term use of non-renewable resource: The case of the essential resource phosphorus. Dissertation, TU Wien, Fakulta¨t fu¨r Bauingenieurwesen. Shahandeh, H., Hossner, L. R., and Turner, F. T. (2005). A comparison of extraction methods for evaluating Fe and P in flooded rice soils. Plant Soil 165(2), 219–225. Shanxi Institute of Soil and Fertilizer, the Chinese Academy of Agricultural Sciences. (1963). The effect of phosphate fertilizer. Shanxi Agric. Sci.(2), 11–15. Sharpley, A. N. (1985). Phosphorus cycling in unfertilized and fertilized agricultural soils. Soil Sci. Soc. Am. J. 49, 905–911. Sharpley, A. (1999). Agricultural phosphorus, water quality, and poultry production: are they compatible? Poultry Sci. 78, 660–673. Sharpley, A. (2000). Phosphorus availability. In “Handbook of Soil Science” (M. E. Sumner, Ed.), pp. D18–D37. CRC Press, New York. Sharpley, A. N., and Smith, S. J. (1985). Fractionation of inorganic and organic phosphorus in virgin and cultivated soils. Soil Sci. Soc. Am. J. 49, 127–130. Shen, H., and Yan, X. L. (2001). Root exudation and their induced mechanism. Soil Environ. Sci. 10, 339–342.

244

Sheng-Xiu Li et al.

Shepherd, M. A., and Withers, P. J. (1999). Applications of poultry litter and triple superphosphate fertilizer to a sandy soil: Effects on soil phosphorus status and profile distribution. Nutr. Cycl. Agroecosyst. 54, 233–242. Shi, R. H., Qiu, J. Z., Bao, S. D., Qin, H. Y., and Sun, W. L. (1962). The P supplying status of nine 9 soils in Jiangsu Province and P fertilizer effect. Acta Pedol. Sin. 10(4), 374–379. Shi, Z. Y., Zhu, Y. M., and Gu, Y. C. (1982). The fertilizer effect of 45 powdered rock phosphates for direct application. Soils 14(5), 171–176. Shi, M. T., Cai, Q. H., and Yang, R. C. (1995). Rice protein content and amino acid composition in early indica rice. J. Fujian Agric. Univ. 24, 358–362. Shi, W., Wang, X., and Yan, W. (2004). Distribution patterns of available P and K in rape rhizosphere in relation to genotypic differences. Plant Soil 261, 11–16. Siddiqi, M. Y., and Glass, A. D. M. (1981). Utilization index: A modified approach to the estimation and comparison of nutrient utilization efficiency in plants. J. Plant Nutr. 4, 289–302. Smil, V. (2000a). Phosphorus in the environment: Natural flows and human interferences. Annu. Rev. Energ. Environ. 25, 53–88. Smil, V. (2000b). Feeding the World: A Challenge for the 21st Century. The MIT Press, Cambridge. Soil Science Institute, Liaoning Academy of Agricultural Sciences. (1974). Application of N and P fertilizers for improvement of P fertilizer effect. Soil Fertil.(5), 27–29. Sokolov A. B. (1950) (in Russian). “Phosphorus Agro-Chemistry”. CCCP and 1959 (in Chinese). “Phosphorus Agro-Chemistry” (Translated by the Seven Laboratory of Shanghai Chemistry Industry Institute), pp. 94–95. Chemistry Industry Press, Beijing. Song, H. X., and Li, S. X. (2003). Effect of root-growing space on its absorbing characteristics of maize. Agric. Sci. China 2(8), 867–874. Song, H. X., and Li, S. X. (2006). Root function in nutrient uptake and soil water effect on NO3-N and NH4þ-N migration. Agric. Sci. China 5(5), 377–383. Stamford, N. P., Santos, P. R., Santos, C. E. S., Freitas, A. D. S., Dias, S. H. L., and Lira, M. A. Jr. (2007). Agronomic effectiveness of biofertilizers with phosphate rock, sulfur and Acidithiobacillus for yam bean grown on a Brazilian tableland acidic soil. Bioresour. Technol. 98, 1311–1318. Starling, M. E., Wood, C. W., and Weaver, D. B. (1998). Starter nitrogen and growth habit effects on late-planted soybean. Agron. J. 90, 658–662. Staunton, S., and Leprince, F. (1996). Effect of pH and some organic anions on the solubility of soil phosphate: Implications for P bioavailability. Eur. J. Soil Sci. 47, 231–239. Stecker, J. A., Brown, J. R., and Kitchen, N. R. (2001). Residual phosphorus distribution and sorption in starter fertilizer bands applied in no-till culture. Soil Sci. Soc. Am. J. 65(4), 1173–1183. Steen, I. (1998). Phosphorus availability in the 21st Century: Management of a nonrenewable resource. Phosphorus Potassium 217, 25–31. Steffens, D. (1984). Wurzelstudien und Phosphat—Aufnahme von Weidelgras und Rotklee unter feldbedingungen. Zeitschrift Pflanzenerhna¨rung Bodenkunde 147(1), 85–97. Stephen, R. C. (2008). Phosphorus control is critical to mitigating eutrophication. PNAS 105(32), 11039–11040. Stewart, W., Hammond, L., and Kauwenbergh, S. J. V. (2005). Phosphorus as a Natural Resource. Phosphorus: Agriculture and the Environment, Agronomy Monograph No. 46. American Society of Agronomy, Crop Science Society of America, Soil Science Society of America, Madison. Stro¨m, L., Owen, A. O., Godbold, D. L., and Jones, D. L. (2002). Organic acid mediated P mobilization in the rhizosphere and uptake by maize roots. Soil Biol. Biochem. 34, 703–710.

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

245

Subbarao, G. V., Ae, N., and Otani, T. (1997). Genotypic variation in iron-, and aluminumphosphate solubilizing activity of pigeon pea root exudates under P deficient conditions. Soil Sci. Plant Nutr. 43, 295–305. Sun, G. Y. (2001). Effects of application of ground phosphate rock on rape yield in calcareous soils. Chin. J. Soil Sci. 33(2), 118–119, 125. Sun, X., and Wang, F. W. (1980). “Agricultural Chemistry” (X. Sun, and F. W. Wang, Eds.), pp. 84–89, 93–96 and 107–108. Shanghai Science and Technology Press, Shanghai, China, Sustainable Agriculture “Food Security and Soil Quality” pp. 309–395. (Rattan Lal, Ed.). Taylor & Francis. Tamminga, S. (2003). Pollution due to nutrient losses and its control in European animal production. Livestock Prod. Sci. 84, 101–111. Tan, Y., Liang, Z. S., Wang, W. L., Duan, Q. M., and Cao, R. (2006). Effects of nitrogen, phosphorus and potassium on root vigor and free amino acid content of astragalus membranaceus seedlings. Acta Bot. Boreal-Occident. Sin. 26, 478–483. Tang, C., Barton, L., and Raphael, C. (1998). Pasture legume species differ in their capacity to acidify soil. Aust. J. Agric. Res. 49, 53–58. Tang, C., Hinsinger, P., Jaillard, B., Rengel, Z., and Drevon, J. J. (2001). Effect of phosphorus deficiency on the growth, symbiotic N2 fixation and proton release by two bean (Phaseolus vulgaris) genotypes. Agronomie 21, 683–689. Tang, C., Qiao, Y. F., Han, X. Z., and Zheng, S. J. (2007). Genotypic variation in phosphorus utilization of soybean [Glycine max (L.) Murr.] grown in various sparingly soluble P sources. Aust. J. Agric. Res. 58, 443–451. Tao, Q. L. (1964a). The biological basis for increasing N fixation by P addition and its practical use methods. Sci. Agric. Sin.(1), 33–37. Tao, Q. L. (1964b). A discussion of fertilization by the viewpoint and method of agricultural soil from P fertilization results. Tillage Fertil.(3), 42–47. Tao, Q. L. (1983). The effects of P fertilizers as affected by the balance of N and P in soil. In “Soil Nutrients, Plant Nutrition and Rational Fertilization” (X. Sun, Ed.), pp. 230– 245. China Agriculture Press, Beijing, China. Tarafdar, J. C., and Claassen, N. (1988). Organic phosphorus compounds as a source for P for for higher plants through phosphatases produced by plant roots and microorganisms. Biol. Fertil. Soils 5, 308–312. Tarafdar, J. C., and Jungk, A. (1987). Phosphatase activity in the rhizosphere and its relation to the depletion of soil organic phosphorus. Biol. Fertil. Soils 3, 199–204. Three Institutes (Institute of Shanghai Chemical Industry, Institute of Shandong Soil and Fertilizer, and Nanjing Soil Institute, China Academy. (1977). Processing and Application of Medium and Low Grade Rock Phosphate Petro-Chemistry Industry Press, Beijing, China, pp. 1–14. Tiessen, H. (2008). Phosphorus in the global environment. “The Ecophysiology of PlantPhosphorus teractions (Plant Ecophysiology)”, 7, 1–7. Tisdale, S. L., Nelson, W. L., and Beaton, J. D. (1985). Soil Fertility and Fertilizer. fourth ed., Macmillan Publishing Co., New York. Tong, X. J., and Yan, X. (1999). Study on characteristics of phosphorus efficiency of soybean native germplasm in Guangdong Province: 1. Differences of soybean genotypes in characteristics of phosphorus efficiency and relationship between phosphorus efficiency and content of soil availability phosphorus. Acta Pedol. Sin. 36, 404–412. Tong, X. J., Lu, Y. G., and Yan, X. (2000). Studies on the characteristics of phosphorus efficiency of native soybean (Glycine Max (L.) Merr.) germplasm: Differences in characteristics of phosphorus efficiency of shoot and root among soybean genotypes and correlation analysis. J. Chin. Oil Crops 22, 48–53. Tong, C. L., Charles, A., Hall, S., and Wang, H. Q. (2003). Land use change in rice, wheat and maize production in China (1961-1998). Agric. Ecosyst. Environ. 95, 523–536.

246

Sheng-Xiu Li et al.

Tu, S. X., Sun, J. H., Guo, Z. F., and Gu, F. (2000). On relationship between root exudates and plant nutrition in rhizosphere. Soil Environ. Sci. 9, 64–67. Tunney, H., Csatho´, P., and Ehlert, P. (2003). Approaches to calculating P balance at the field-scale in Europe. J. Plant Nutr. Soil Sci. 166, 438–446. Turner, B. L., Papha´zy, M. J., Haygarth, P. M., and McKelvie, I. D. (2002). Inositol phosphates in the environment. Philos. Trans. R. Soc. Lond. B Biol. Sci. 357, 449–469. Turtola, E., and Yli-Halla, M. (1998). Fate of phosphorus applied in slurry and mineral fertilizer: Accumulation in soil and release into surface runoff water. Nutr. Cycl. Agroecosyst. 55, 165–174. Ule´n, B., Bechmann, M., Fo˝lster, J., Jarvie, H. P., and Tunney, H. (2007). Agriculture as a phosphorus source for eutrophication in the north-west European countries, Norway, Sweden. United Kingdom and Ireland: A review. Soil Use Manag. 23, 5–15. Uusitalo, R., Turtola, E., Gro¨nroos, J., Kivisto¨, J., Ma¨ntylahti, V., Turtola, A., Lemola, R., and Salo, T. (2007). Finnish trends in phosphorus balances and soil test phosphorus. Agric. Food Sci. Finland 16, 301–316. UNEP, 2005. Millennium Ecosystem Assessment Synthesis Report. Pre-publication Final Draft Approved by MA Board on March 23, 2005, The United Nations Environment Programme. Valkama, E., Uusitalo, R., Ylivainio, K., Virkaja¨rvi, P., and Turtola, E. (2009). Phosphorus fertilization: A meta-analysis of 80 years of research in Finland. Agric. Ecosyst. Environ. 130, 75–85. van der Zee, S. E. A. T. M., Fokkink, L. G. J., and van Riemsdijk, W. H. (1987). A new technique for assessment of reversible adsorbed phosphate. Soil Sci. Soc. Am. J. 125, 599–604. Van Ray, B., and Van Diest, A. (1979). Utilization of phosphate from different sources by six plant species. Plant Soil 51(4), 577–589. Vance, C. P., Stone, C. U., and Allan, D. L. (2003). Phosphorus acquisition and use: Critical adaptations by plants for securing a nonrenewable resource. New Phytol. 157, 423–447. Vassilev, N., Medina, A., Azcon, R., and Vassileva, M. (2006a). Microbial solubilization of rock phosphate on media containing agro-industrial wastes and effect of resulting products on plant growth and P uptake. Plant Soil 287, 77–84. Vassilev, N., Vassileva, M., and Nikolaeva, I. (2006b). Simultaneous P-solubilizing and biocontrol activity of microorganisms: Potentials and future trends. Appl. Microbiol. Biotechnol. 71, 137–144. Wagner, M. (2003). Weltweite Phosphatvorra¨te und Kostenstrukturen bei ihrer Nutzung. In “Paper Presented at the Symposium “Ru¨ckgewinnung von Phosphor in der Landwirtschaft und aus Abwasser und Abfall,” Berlin, 6–7 February. Wakelin, S. A., Warren, R. A., Harvey, P. R., and Ryder, M. H. P. (2004). Phosphate solubilization by Penicillium spp. closely associated with wheat roots. Biol. Fertil. Soils 40, 36–43. Wallace, A. (1962). A Decade of Synthetic Chelating Agents in Inorganic Plant Nutrition Edwards Brothers, Inc, Ann Arbor, MI, USA. Wang, Q. R., Li, J. Y., and Li, Z. S. (1999a). Studies on plant nutrition of efficiency utility for soil phosphorus. Acta Ecol. Sin. 19(3), 417–421. Wang, C. Q. (2004). Micronutrient fertilizers. In “Soil and Fertilizer Science” (D. T. Xie, Ed.), pp. 276–277. China Forestry Press. Beijing, China. Wang, Y. Z., Li, W. Y., and Yin, Z. R. (1999b). Progress in relationship between (not both) maize root and its yields. J. Jilin Agric. Sci. 24(4), 6–8. Wang, D. R., Lu, W. F., and Chen, W. (2001). Influence of nitrogen application on protein and amino acid contents in rice. Plant Nutr. Fertil. Sci. 7, 353–356. Wang, Q. R., Li, J., Li, Z., and Christie, P. (2005). Screening Chinese wheat germoplasm for phosphorus efficiency in calcareous soils. J. Plant Nutr. 28, 489–505.

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

247

Wang, D. C., Hu, H. Q., and Li, X. Y. (2006). The release mechanisms of phosphate rock powder in acid soils and the agricultural effect. Chin. Agric. Sci. Bull. 22(9), 242–245. Wang, Y. L., He, Y. Q., and Li, C. L. (2007). Persistent activating effect of citric acid on phosphorus in red soil and its mechanism. Acta Pedol. Sin. 44(1), 130–136. Webb, J. R. (1982). Rock phosphate-superphosphate experiment, pp. 7–9. Northern Research Center Annual Progress Report. ORC81–14, vol. 22. Clarion-Webster Research Center Ames, Iowa State University, Iowa. Webb, J. R. (1984). Rock phosphate-superphosphate experiment, pp. 5–7. Northern Research Center Annual Progress Report. ORC83-29, Iowa State University. Wei, J., Zhou, E. X., Zhang, G. Y., Jiang, C., Yao, Y. R., and Bi, S. Q. (2001). Study on effect of different activators to phosphate rock. J. Agric. Univ. Hebei 24(1), 13–15. Weikard, Hans-Peter, and Seyhan, Demet (2009). Distribution of phosphorus resources between rich and poor countries: The effect of recycling. Ecol. Econ. 68, 1749–1755. White, J. (2000). Introduction to Biogeochemical Cycles. Department of Geological Sciences, University of Colorado, Boulder, (Chapter 4). 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. Wissuwa, M., Wegner, J., Ae, N., and Yano, M. (2002). Substitution mapping of Pup1: A major QTL increasing phosphorus uptake of rice from a phosphorus-deficient soil. Theor. Appl. Genet. 105, 890–897. WSRW (Western Sahara Resource Watch). (2007). The Phosphate Exports, Western Sahara Resource Watch. http://www.wsrw.org/index.php?cat¼117&art¼521. Wu, C. S., Meng, C. F., Lu, X. N., Teng, C. Q., and Zhao, S. R. (2002). The agronomic availability and economic effect of applying ground phosphate rock to acid paddy soil. J. Phosph. Comp. Fertil. China 17(3), 67–69. Xiang, W. S., Huang, M., and Li, X. Y. (2004). Progress on fractioning of soil phosphorus and availabilty of various phosphorus fractions to crops in soil. Plant Nutr. Fertil. Sci. 10 (6), 663–670. Xiao, Y. B., Li, L., and Zhang, F. S. (2004). Effect of root contact on interspecific competition and N transfer between wheat and fababean using direct and indirect 15N techniques. Plant Soil 263, 45–54. Xie, J. C. (1966). The relation of crop characteristics with soil release of rock phosphate. Soil Special Issue 33, 44–48. Xie, R. L., and Tan, H. W. (2001). The present and future demand of phosphate fertilizer in China in the light of agricultural production. J. Phosph. Comp. Fertil. China 16(2), 6–9. Xu, Q. Z. (1989). Wheat leaf acid phosphates and effects of loss phosphorus. Plant Physiol. Commun.(6), 20–23. Xu, M. L., and Liu, Z. Y. (1984). Study on the nutrient uptake in soil-root interface using radioactive autograph technique. Soils 16(2), 65. Xu, D. Y., Jin, J., Du, Y., Hu, S. Y., Wang, Y. J., and Zhu, Q. S. (2003). Effect of NPK fertilizer management on protein and amino acid contents of rice grain. Plant Nutr. Fertil. Sci. 9, 506–508. Xu, M. G., Liang, G. Q., and Zhang, F. D. (2006). Changes of Chinese Soil Fertility. Chinese Agricultural Science and Technology Press, Beijing, China, p. 11. Yadav, R. S., and Tarafdar, J. C. (2003). Phytase and phosphatase producing fungi in arid and semi-arid soils and their efficiency in hydrolyzing different organic P compounds. Soil Biol. Biochem. 35, 1–7. Yan, X. (2005). The roots of phosphorus-efficient soybean: Theories and practices. In “Plant Nutrition for Food Security, Human Health and Environmental Protection” (C. J. Li, et al., Eds.), pp. 36–37. Tsinghua University Press, Beijing.

248

Sheng-Xiu Li et al.

Yan, F., Schubert, S., and Mengel, K. (1992a). Effect of low root medium pH on net proton release, root respiration, and root growth of corn (Zea mays, L.) and broad bean (Vicia faba L.). Plant Physiol. 117, 311–319. Yan, X., Huang, Z. W., Lu, R. J., and He, Y. K. (1992b). Study on genetics of P efficiency in crops. Soils 24, 102–105. Yan, F., Schubert, S., and Mengel, K. (1996). Soil pH increase due to biological decarboxylation of organic anions. Soil Biol. Biochem. 28, 617–624. Yan, X., Liao, H., Beebe, S. E., Blair, M. W., and Lynch, J. P. (2004). QTL mapping of root hair and acid exudation traits and their relationship to phosphorus uptake in common bean. Plant Soil 265, 17–29. Yan, X. L., Liao, H., and Nian, H. (2009). Root breeding for better plant nutrient efficiency on acid soils. In “Plant-Soil Interactions at Low pH: A Nutriomic Approach—Proceedings of the 7th International Symposium on Plant-Soil interactions at Low pH” (H. Liao, X. L. Yan, and L. Kochian, Eds.), pp. 5–6. South China University of Technology Press, Guangzhou, China. Yang, L. P., Jin, J. Y., Bai, Y. L., and Huang, S. W. (2001). Comprehensive evaluation of soil nutrients balanced fertilization technique and its industrialization. J. Phosph. Comp. Fertil. China 16(4), 61–63. Ye, K. H. (2002). Plant phosphorus nutrition and phosphate fertilizer. In “Soil and Fertilizer Science” ( Y. K. Ye and K. H. Ye, Eds.), pp. 103–114. Wuhan University Press. Wuhan, Hubei Province, China. You, X. L. (1983). Evaluation of the ancient Chinese fertilizer science from the book entitled Qi Min Yao Shu (Manual of Important Arts for the People). In “Soil Nutrients, Plant Nutrition and Rational Fertilization” (X. Sun, Ed.), pp. 205–212. China Agriculture Press, Beijing, China. Yuan, K. N., Zhu, Z. X., and Yu, Z. Y. (translators) (1959). Chemistry of the Soil (Firman E. Beer Ed., Reinhold Publishing Compaction, UAS, 1955). Chinese version (-translated), p. 217. Science Press, Beijing, China. Yuan, Z. Y., Huang, C. Y., Wang, Z. W., and Li, S. H. (1962). The relation of P fertilizer effect in phaeozem and albic bleached soil to N supplying levels. Acta Pedol. Sin. 13(4), 418–426. Yuan, S., Peng, Z. P., Sha, X. Q., and Wang, Y. Q. (2010). Physiological mechanism of maize hybrids in response to P deficiency and differences among cultivars. Sci. Agric. Sin. 43(1), 51–58. Zang, H. L., Zhang, Z. Y., Wang, K. T., and Wan, W. L. (1965). The relation of soil properties and crop varieties in hilly areas of Jiangting County, Jiangsu Province to P fertilizer effect. Acta Pedol. Sin. 13(4), 456–458. Zaongo, C. G. L., Hossner, L. R., and Wendt, C. W. (1994). Root distribution water use and nutrient uptake of millet and grain sorghum on West Africa soils. Soil Sci. 157(6), 379–388. Zhang, E. H. (2001). Effects of P application on root acid phosphates of crops in intercropping. Acta Bot. Boreal-Occident. Sin. 21(1), 53–58. Zhang, S. J., Zhang, X. L., Liu, H. X., and Wang, D. L. (1994). Application of phosphate fertilizer to the black soil region of northeast China. Chin. J. Soil Sci. 25(4), 178–180. Zhang, F. S., Ma, J., and Cao, J. P. (1997). Phosphorus deficiency enhances root exudation of low-molecular weight organic acids and utilization of sparingly soluble inorganic phosphates by radish (Raghanus sativus L.) and rape (Brassica napus L.) plants. Plant Soil 196, 261–264. Zhang, E. H., Zhang, X. H., and Wang, H. Z. (2004). Adaptable effects of phosphorus stress on different genotypes of faba-bean. Acta Ecol. Sin. 24(8), 1589–1593. Zhang, J., He, L. Y., Li, J. S., and Xu, S. Z. (2005). Phosphorus nutrient characteristics of different maize inbreds with tolerance to low-P stress. Sci. Agric. Sin. 38(1), 110–115.

Responses of Some Leguminous and Nonleguminous Crops to P Fertilizers

249

Zhang, K. W., Li, K. P., Liu, Z. G., and Zhang, J. R. (2007). Effect of phosphorus level on phosphorus absorption and utilization of different genotype maize seedlings. Plant Nutr. Fertil. Sci. 13(5), 795–801. Zhao, Z. Q. (1962). Analysis of condition beneficial to application of P fertilizer. Hebei Agric. J. 1(1), 2–5. Zhao, R. S., and Tong, J. M. (1964). The effect, role and methods for application of P fertilizer to Chinese vetch milk vetch (Astragalus sinicus) and the effective conditions for its use. Chin. J. Soil Sci.(5), 1–8. Zhao, S. X., Yu, W. T., Zhang, L., Shen, S. M., and Ma, Q. (2004). Progress of soil organic phosphorus studies. Chin. J. Appl. Ecol. 15(11), 2189–2194. Zhou, M. Z. (1980). Investigation on the extraction of soil available P. Chin. J. Soil Sci. 11 (4), 47–48. Zhou, M. Z. (1987). Application of fertilizers based on soil testing in China. Chin. J. Soil Sci. 18(1), 7–13. Zhou, G. Q., Xu, M. L., Tan, Z., and Li, X. Z. (1997). Effects of ecological factors of protein and amino acids of rice. Acta Ecol. Sin. 17, 537–542. Zhou, L. L., Cao, J., Zhang, F. S., and Li, L. (2009). Rhizosphere acidification of faba bean, soybean and maize. Sci. Total Environ. 407, 4356–4362. Zhu, D. W., Dong, Y. Y., and Li, X. Y. (1991). The effect of low-grade phosphate rock powder on the chemical properties of acid soils. Chin. J. Soil Sci. 22(4), 257–259. Zhu, D. F., Lin, X. Q., and Cao, W. X. (2001). Effects of deep roots on growth and yield in two rice varieties. Sci. Agric. Sin. 34(4), 429–432. Zhu, X. J., Yu, W. T., Ma, Q., Zhang, L., and Jiang, Z. H. (2009). Effects of phosphate fertilizer rates on nutrient distribution in crop plants in lower reaches of Liao River Plain. Chin. Soil Fertil.(2), 78–79. Zoysa, A. K. N., Loganathan, P., and Hedley, M. J. (1997). A technique for studying rhizosphere processes in tree crops: Soil phosphorus depletion around camellia (Camellia japonica L.) roots. Plant Soil 190, 253–265. Zoysa, A. K. N., Loganathan, P., and Hedley, M. J. (1998a). Effect of forms of nitrogen supply on mobilization of phosphorus from a phosphate rock and acidification in the rhizosphere of tea. Aust. J. Soil Res. 36, 373–387. Zoysa, A. K. N., Loganathan, P., and Hedley, M. J. (1998b). Phosphate rock dissolution and transformation in the rhizosphere of tea (Camellia sinensis L.) compared with other species. Eur. J. Soil Sci. 49, 477–486. Zuberer, D. A. (1998). Biological dinitrogen fixation: Introduction and nonsymbiotic. In “Principles and Applications of Soil Microbiology” (D. M. Sylvia, J. J. Fuhrmann, P. G. Hartel, and D. A. Zuberer, Eds.), pp. 295–321. Prentice-Hall, Inc., Simon & Schuster/A Viacom Company, Upper Saddle River, NJ.

C H A P T E R

F O U R

The Role of Mineral Nutrition on Root Growth of Crop Plants N. K. Fageria* and A. Moreira† Contents 252 255 256 260 263 265 268 270 271 274 276 278 279 287 299 301 302 303 304 312 312 316 317 318 318

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

Introduction Root-Induced Changes in the Rhizosphere Root Systems of Cereals and Legumes Contribution of Root Systems to Total Plant Weight Rooting Depth and Root Distribution Root Growth as a Function of Plant Age Root–Shoot Ratio Root Growth Versus Crop Yield Genotypic Variation in Root Growth Root Oxidation Activity in Oxygen-Deficient Soils Root Growth in Conservation Tillage Systems Mineral Nutrition Versus Root Growth 12.1. Nitrogen 12.2. Phosphorus 12.3. Potassium 12.4. Calcium 12.5. Magnesium 12.6. Sulfur 12.7. Micronutrients 13. Management Strategies for Maximizing Root Systems 13.1. Soil management 13.2. Plant management 14. Conclusions Acknowledgment References

Abstract Agriculture is going through a profound revolution worldwide due to increasing world demand for food, higher costs of energy and other inputs, environmental pollution problems, and instability of cropping systems. In this context, * Rice and Bean Research Center of Embrapa, Santo Antoˆnio de Goia´s, GO, Brazil { Western Amazon Research Center of Embrapa, Manaus, AM, Brazil Advances in Agronomy, Volume 110 ISSN 0065-2113, DOI: 10.1016/B978-0-12-385531-2.00004-9

#

2011 Elsevier Inc. All rights reserved.

251

252

N. K. Fageria and A. Moreira

knowledge of factors that affect root development is fundamental to improving nutrient cycling and uptake in soil–plant systems. Roots are important organs that supply water, nutrients, hormones, and mechanical support (anchorage) to crop plants and consequently affect economic yields. In addition, roots improve soil organic matter (OM) by contributing to soil pools of organic carbon (C), nitrogen (N), and microbial biomass. Root-derived soil C is retained and forms more stable soil aggregates than shoot-derived soil C. Although roots normally contribute only 10–20% of the total plant weight, a well-developed root system is essential for healthy plant growth and development. Root growth of plants is controlled genetically, but it is also influenced by environmental factors. Mineral nutrition is an important factor influencing the growth of plant roots, but detailed information on nutritional effects is limited, primarily because roots are half-hidden organs that are very difficult to separate from soil. As a result, it is difficult to measure the effect of biotic and abiotic factors on root growth under field conditions. Root growth is mainly measured in terms of root density, length, and weight. Root dry weight is often better related to crop yields than is root length or density. The response of root growth to chemical fertilization is similar to that of shoot growth; however, the magnitude of the response may differ. In nutrient-deficient soils, root weight often increases in a quadratic manner with the addition of chemical fertilizers. Increasing nutrient supplies in the soil may also decrease root length but increase root weight in a quadratic fashion. Roots with adequate nutrient supplies may also have more root hairs than nutrient-deficient roots. This may result in greater uptake of water and nutrients by roots well supplied with essential plant nutrients, compared with roots grown in nutrient-deficient soils. Under favorable conditions, a major part of the root system is usually found in the top 20 cm of soil. Maximum root growth is generally achieved at flowering in cereals and at pod-setting in legumes. Genotypic variations are often found in the response of root growth to nutrient applications, and the possibility of modifying root system response to soil properties offers exciting prospects for future improvements in crop yields. Rooting pattern in crop plants is under multi- or polygenic control, and breeding programs can be used to improve root system properties for environments where drought is a problem. The use of crop species and cultivars tolerant to biotic and abiotic stresses, as well as the use of appropriate cultural practices, can improve plant root system function under favorable and unfavorable environmental conditions.

1. Introduction Roots are important plant organs. They absorb water and nutrients from the soil and translocate them to plant tops (Merrill et al., 1996, 2002; Sainju et al., 2005a; Stone et al., 2001). Roots also give mechanical support to plants and supply hormones that affect many physiological and biochemical processes associated with growth and development. Roots exert control

The Role of Mineral Nutrition on Root Growth of Crop Plants

253

over whole-plant growth and development by controlling the uptake of mineral nutrients (Zobel, 1986). Cytokinins produced in roots are translocated to shoots and participate in the control of leaf protein biosynthesis (Oritani, 1995). In addition, cytokinins may control nitrogen assimilation in the root itself. Oritani (1995) also reported that rice roots synthesize zeatin glucoside, a bound cytokinin, in addition to isopentenyladenine (IP), isopentenyl-adenosine (IPA), trans-zeatin riboside (tZR), and trans-zeatin (tZ), which are free cytokinins in zeatin-related compounds. Zobel (2005a) reported that root system dynamics are instrumental in the maintenance of biological and chemical equilibrium within the soil and modulate changes to soil quality. In addition, genotypes with inherently large root systems have been associated with reduced lodging in cereals and legumes (Stoffella and Kahn, 1986). Soil is knitted together by plant roots, which form complex and structurally diverse reinforcing structures. Near the soil surfaces, fibrous plant roots hold soil aggregates together against the stresses of water and wind. Vigorous root systems are needed for the development of healthy plants and consequently, higher yields. Roots that are left in the soil after crop harvest improve soil organic matter (OM) content and contribute to the nitrogen cycle and microbial activity (Sainju et al., 2005a). All these activities improve soil structure, soil water holding capacity, water infiltration into the soil, as well as reduce soil bulk density and soil erosion, ultimately leading to greater soil productivity. Processes that are largely controlled or directly influenced by roots and often occur in the vicinity of the root surface are often referred to as rhizosphere processes (Cheng and Kuzyakov, 2005). These processes may include root turnover, rhizodeposition, root respiration, and rhizosphere microbial respiration that are a result of microbial utilization of rhizodeposits. Rhizosphere processes play an important role in the global C cycle. Terrestrial ecosystems are intimately connected to atmospheric carbon dioxide levels through photosynthetic fixation of CO2, sequestration of CO2 in plant and soil biomass, and the subsequent release of C through respiration and decomposition of organic matter (Cheng and Kuzyakov, 2005). Carbon cycling belowground is increasingly being recognized as one of the most significant components of the ecosystem C fluxes and pools (Cheng and Kuzyakov, 2005; Jackson et al., 1997; Zak and Pregitzer, 1998). Roots improve soil aggregation, which controls biological and hydrologic properties of the soil. A soil aggregate is a group of primary soil particles that adhere to one another more strongly than to surrounding soil particles (Follett et al., 2009). Root materials remain mixed within the soil as they decompose, providing a gum-like material that cements soil particles into aggregates (Melillo and Gosz, 1983; Tresder et al., 2005). Root exudation occurs when organic acids either are leaked from the root or are released as a means of interacting with microbes in the rhizosphere. These

254

N. K. Fageria and A. Moreira

exudates can influence microbial activity and the subsequent metabolism of root detritus. Mucilages released by the root cap and epidermis link particulate organic residues with mineral fragments. Microbes in the rhizosphere use plant mucilage as a substrate for growth and secrete their own mucilage, thereby producing mucigel, a mucilaginous material of mixed origin which stabilizes finer aggregates. In addition, in many soils, fungal hyphae are important for stabilizing larger structural units (Goss and Kay, 2005). Finally, allocation of carbohydrates and other C-containing molecules directly to mycorrhizal fungi forms another conduit of C into the soil as these microbes secrete their own exudates and as their tissues senesce (Tresder et al., 2005). The amount of C and N supplied by roots can be significant for maintaining or improving soil organic matter (Sainju et al., 2005b). The organic input from plant roots to the surrounding soil is the principal support of the biological activity and abundance of organisms in the rhizosphere (Cheng et al., 1994; Kirchner et al., 1993). As much as 7–43% of the total aboveground and belowground plant biomass can be contributed by roots (Kuo et al., 1997a,b). Roots can supply from 400 to 1460 kg C ha 1 during a growing season (Kuo et al., 1997a; Qian and Doran, 1996). Liang et al. (2002) reported that roots contributed as much as 12% of soil organic C, 31% of water soluble C, and 52% of microbial biomass C within a growing season. Roots may play a dominant role in soil C and N cycles (Gale et al., 2000a; Puget and Drinkwater, 2001; Wedin and Tilman, 1990). Roots may have relatively greater influence on soil organic C and N levels than the aboveground plant biomass (Boone, 1994; Haider et al., 1993; Milchumas et al., 1985; Norby and Cortrufo, 1998; Sanchez et al., 2002). Balesdent and Balabane (1996) reported that corn roots contributed 1.6 times more C to soil organic C than did stover. Root-derived C is retained and forms more stable aggregates than does shoot-derived C (Gale et al., 2000a,b). Rhizodeposition, such as root exudates, mucilages, and sloughed cells, may be a significant source of soil organic C (Balesdent and Balabane, 1996; Buyanovsky et al., 1986; Sainju et al., 2005a). Helal and Sauerbeck (1987) estimated that the amount of C released from roots as rhizodeposit could be more than 580 kg C ha 1. This rhizodeposition increases microbial activity and influences N mineralization in the soil (Bakken, 1990; Texier and Biles, 1990). Carbon contribution from corn root biomass and rhizodeposition to soil organic C can be as much as 1.7–3.5 times greater than from stover (Allmaras et al., 2004; Wilts et al., 2004). The environment is seldom optimum for extensive and effective root growth. Canopy conditions that limit photosynthesis reduce shoot growth and limit assimilate translocation to the roots, thus reducing root growth (Miller, 1986). Root growth is under multi- or polygenic control and is also influenced by environmental factors, including soil temperature, soil moisture content, solar radiation and soil physical, chemical, and biological

The Role of Mineral Nutrition on Root Growth of Crop Plants

255

properties (Fageria, 2002c, 2009; Klepper, 1992; Merrill et al., 1996; Russell, 1977; Sainju et al., 2005b; Zobel, 1991). Most of the root biomass of annual crops is located in the 0–20 cm soil depth. This may be associated with greater organic matter, nutrients, aeration, and water availability in the top soil layer compared to lower soil depths (Merrill et al., 1996, 2002; Qin et al., 2004; Sainju et al., 2005a; Stone et al., 2001). Increased knowledge of root architecture and root development dynamics could help improve crop productivity in agroecosystems. Better understanding of root architecture and growth dynamics of annual crops may lead to a more efficient use of applied nutrients and water. The study of plant roots is one of the most promising, but least explored, areas of research related to plant growth. The aerial portions of plant species have received greater attention and study, probably because of their conspicuousness and easy access, while the subterranean portions have been neglected because of the difficulty of observing and sampling them and the disruption of root systems when they are removed from soil. Many crop root studies have relied on soil cores and more recently on minirhizotron observations (Box and Ramseur, 1993; Zobel, 2005b). The data collected using these methods may not be representative of the crop as a whole (Andren et al., 1991; Hansson et al., 1992; Hoad et al., 2001; Parker et al., 1991). In addition, information about annual field crop root growth dynamics as a function of environmental factors is scattered and often not readily accessible. The primary objectives of this chapter are to review the latest advances in relation to the role of mineral nutrition in the growth and development of roots of annual crops. To make the subject matter as practical as possible, most of the discussion is supported by experimental results. Our approach should enhance understanding on the contribution of roots to total dry matter of crops, to assess the effects of root system size and form on overall crop growth, and to relate the effects of root growth on the environment. This information may be useful for agricultural scientists in the fields of plant nutrition, water use, breeding, and plant physiology who are interested in conducting research to manipulate plant root systems in favor of higher yields.

2. Root-Induced Changes in the Rhizosphere Pinton and Varannini (2001) suggested that the soil layer surrounding roots should be termed the ectorhizosphere and the root inside the layer colonized by microorganisms should be designated as endorhizosphere. The two areas are separated by the root surface known as rhizoplane (Fig. 1). Growing roots release an appreciable amount of organic components into the rhizosphere. Marschner (1995) reported that three major components released by roots are low-molecular weight organic compounds (free

256

N. K. Fageria and A. Moreira

Ectorhizosphere

Root hair Mucigel (plant and bacteria)

Endorhizosphere

Stele (xylem, phloem) Epidermis Cortex

Endodermis

Root cap

Sloughed root cap cell

Figure 1 2006).

Root cross section showing ecto- and endorhizosphere (Fageria and Stone,

exudates), high-molecular weight gelatinous materials (mucilage), and sloughed-off cells and tissues and their lysates (Fig. 1). The rhizosphere is the soil zone adjacent to plant roots which is physically, chemically, and biologically different than bulk or nonrhizosphere soil. Plants influence the physical (temperature, water availability, and structure), chemical (pH, redox potential, nutrient concentration, root exudates, Al detoxification, and allelopathy), and biological properties (microbial association) in the rhizosphere. Their effects include changes in nutrient solubility, transport, and uptake of mineral nutrient, and ultimately plant growth. Major rhizosphere changes are synthesized in Fig. 2 and their influence on nutrient availability is discussed in details by Fageria and Stone (2006).

3. Root Systems of Cereals and Legumes Cereals as well as legume seeds contain relatively large reserves of storage carbohydrates and nutrients which allow the initial root system to grow rapidly to considerable depth (Marschner, 1998). Branching often

The Role of Mineral Nutrition on Root Growth of Crop Plants

Physical changes Temperature Water holding capacity Structure

257

Biological changes Nitrogen fixation PGPR bacterias Mycorrhizal fungi Harmful microorganisms

Chemical changes pH Redox potential Release of organic compounds by roots Nutrient concentration and availability AI detoxification Allelopathy

Figure 2 Major physical, chemical, and biological changes in the rhizosphere (Fageria and Stone, 2006).

begins before the leaves have unfolded, with the result that the plant establishes early contact with moist soil (Hoad et al., 2001). Generally, roots are classified into four groups. These groups are the taproot, basal roots, lateral roots, and shootborn or adventitious roots (Zobel, 2005a). When plants produce secondary shoots (tillers) or shoot branches which develop roots, these roots are commonly called adventitious roots. To indicate the true origin of these adventitious roots, the term shootborn is sometimes used (Zobel, 2005a). The primary function of the taproot, basal roots, and adventitious roots is to establish the most optimum framework from which to initiate small lateral roots to effect water and nutrient uptake (Zobel, 2005b). The taproot penetrates relatively deeply to ensure an

258

N. K. Fageria and A. Moreira

adequate supply of soil water, the basal roots spread out laterally to ensure a structure for lateral roots that take up P and other nutrients that are less abundant in the lower levels of the soil profile (Zobel, 2005b), and to provide a degree of lodging resistance to the plant as it matures and produces seed (Barlow, 1986; Stoffella et al., 1979). For many grasses and other species in which root secondary thickening is not important, the shootborn roots take over the role of the basal roots. The shootborn roots continue to build the framework with larger and larger conducting roots as the plant increases in size (Zobel, 2005b). The basal and shootborn roots probably provide little direct uptake of nutrients and water (St. Aubin et al., 1986). Besides length and weight, surface area is an important parameter of the root system in crop plants. The form of root systems and their development conditions greatly affect the surface area of roots. The surface area of roots has a high positive correlation with the amount of nutrient absorption (Takenaga, 1995). Various studies show that 90–95% or more of the root length of an intact plant is made up of roots <0.6 mm in diameter (Zobel, 2003, 2005b). Monocots and dicots typically have different root system structures. Root systems of monocots are fibrous, whereas dicots often have taproots. The fibrous root systems of monocots consist of seminal, nodal, and lateral roots. Seminal roots develop from primordia within seeds and nodal roots develop adventitiously from lower stem nodes. All adventitious roots of stem origin are called nodal roots to distinguish them from other adventitious roots that emerge from the mesocotyl or elsewhere on the plant. Nodal roots are identified by the node number from which they originate. Nodal roots may be functional or nonfunctional (Thomas and Kaspar, 1997). Functional nodal roots are defined as roots that have emerged from stem nodes, entered the soil, and developed lateral roots and/or root hairs. Nonfunctional nodal roots are defined as roots that have emerged from aboveground stem nodes and have not entered the soil or produced lateral roots (Thomas and Kaspar, 1997). Initial seminal or nodal roots develop laterals that are classed as roots of the first order, roots that develop from first-order roots are classed as second-order roots, and additional roots that develop from these laterals are classed as third-order roots, fourth-order roots, etc. (Yamauchi et al., 1987a,b). Nodal roots are also known as adventitious, coronal, and/or crown roots. Roots of cereals such as rice include mesocotyl, radical (seminal), and nodal or adventitious roots (Yoshida, 1981). Mesocotyl roots emerge from the axis between the coleoptile node and the base of the radical, and they typically develop only when seeds are planted very deep or are treated with chemicals (Yoshida, 1981). Until adventitious roots develop, seedlings must rely on roots which initiate on the subcoleoptile internodes above the seed or seminal roots below the seed. Adventitious roots are important to seedling establishment because they can conduct

The Role of Mineral Nutrition on Root Growth of Crop Plants

259

more water than smaller diameter seminal roots. Adventitious roots may develop as early as 2 weeks after sowing. Seedling survival may increase when seeds are sown at greater soil depths, where greater soil water availability may increase adventitious root development (Fageria et al., 2006). Tiller roots do not form in cereals until tillers have two to three leaves (Kleeper et al., 1984), and until these roots have developed, parent culms must provide nutrients and water. Parent culms may also have to provide hormonal control so essential for tiller survival. Delayed root production by tillers may explain why late tillers often do not survive (Kleeper et al., 1984). Figure 3 shows the radical and adventitious root system of upland rice (Oryza sativa L.) (cereal) and Fig. 4 shows the tap root system of dry bean (Phaseolus vulgaris L.) (legume). In addition to their morphological differences, roots of cereals and legumes have different physiochemical properties. The surface of plant roots has a negative electric charge, mainly due to carboxyl groups in the pectin of the root cell walls. The density of this negative charge is defined as a cation exchange capacity (CEC; Takenaga, 1995). The CEC of cereals

3rd leaf 4th leaf 2nd leaf

1st tiller 1st leaf Adventitious root

Radicle root

Figure 3 Root system of upland rice seedlings (Fageria, 2007).

260

N. K. Fageria and A. Moreira

First trifoliate

Gem

Primary leaf Epicotyl

Hypocotyl Adventitious roots Basal roots Lateral roots

Figure 4

Tap root

Root system of dry bean seedling (Fageria and Santos, 2008).

such as rice, barley, and corn is typically lower than the CEC of legumes like dry bean and broad bean. Roots with high CEC absorb more divalent cations like Ca2þ and Mg2þ than monovalent cations such as Kþ and NH4þ. On the other hand, roots with lower CEC absorb more monovalent than divalent cations. Hence, in grass–legume mixtures, legumes generally suffer with Kþ deficiency due to large uptake of this element by grasses. Essau (1977), Fageria et al. (2006), Klepper (1992), Leskovarant and Stofella (1995), O’Toole and Bland (1987), and Zobel (1991, 2005a,b) have dealt extensively with various types of monocotyledonous and dicotyledonous roots and root hairs, and their growth and morphology.

4. Contribution of Root Systems to Total Plant Weight Crops can accumulate photosynthetic products in their stems, leaves, grains, and roots, and the development of robust root systems is necessary to produce good crop yields. A well-developed root system is needed to absorb adequate amounts of water and nutrients, especially when plants

The Role of Mineral Nutrition on Root Growth of Crop Plants

261

are under abiotic or biotic stress. The traditional view is that a large vigorous root system, through avoidance of plant water deficits, is required for high yields in water-limited environments (Ludlow and Muchow, 1990). In addition, large root systems add more organic matter to the soil, creating better environmental conditions for the growth and development of crops. The distribution of photosynthetic products in the root, shoot, and grain is determined genetically, but it also varies with environmental conditions (Fageria, 1992, 2009; Sainju et al., 2005a). Because roots are so difficult to separate from soils, little is known about the effects of soil and crop management practices on their growth and development (Sainju et al., 2005b). In most annual plant species, only 10–20% of mature plant biomass consists of root tissue (Zobel, 1986). Similarly, Fageria (1989a) also reported that the contribution of roots of food crops to the total plant weight varies from 10–20%, depending on species and cultivars within species. However, Sainju et al. (2005b) reported that C accumulation in cotton and sorghum roots ranged from 1% to 14%. Data in Table 1 show the contribution of the root system of 20 upland rice genotypes grown at two N rates in a Brazilian Oxisol. In this study, the nitrogen genotype interaction was significant because some genotypes were highly responsive to the N application while others were not. Thus, genotype selection is an important strategy for upland rice production in Brazilian Oxisols. In the control treatment, the contribution of the root system to total plant weight varied from 12% to 30%, with an average value of 22%. At the 300 mg N kg 1 soil treatment, the contribution of root weight to total plant weight varied from 3% to 21%, with an average value of 14%. The proportionally lower root dry weight at the higher N rate was associated with a significant increase in the grain and shoot weight of rice genotypes with the addition of N fertilizer (Fageria and Baligar, 2005). The contribution of root systems to total plant weights of tropical legume cover crops under three P levels is shown in Table 2. There was a significant influence of P rate, cover crop species and P species interaction was significant, indicating different responses of cover crops at different P rates. At 0 mg P kg 1 level, the contribution of roots to total plant weight varied from 4.91% to 21.50%, with an average value of 12.98%. With 100 mg P kg 1, the root contribution to the total plant weight varied from 11.30% to 25.16%, with an average value of 18.93%. At the 200 mg P kg 1 P rate, the contribution of roots to the total plant weight varied from 6.36% to 25.91%, with an average value of 14.79%. Overall, increase in root contribution to the total plant weight with the increase in P rate may be associated with the response of legumes to P fertilization. Significant differences have been reported among the crop species and genotypes of the same species in the absorption and utilization of P (Epstein and Bloom, 2005; Fageria, 2009; Marschner, 1995). Tian et al. (1998) reported significant responses of legume cover crops grown on

262

N. K. Fageria and A. Moreira

Table 1 Contribution of upland rice roots in the total plant dry weight (%) as influenced by N rates Genotype

0 mg N kg 1 soil

300 mg N kg 1 soil

BRA01506 BRA01596 BRA01600 BRA02535 BRA02601 BRA032033 BRA032039 BRA032048 BRA032051 BRA042094 BRA042156 BRA042160 BRA052015 BRA052023 BRA052033 BRA052034 BRA052045 BRA052053 BRS Primavera BRS Sertaneja Average F-test N rate (N) Genotype (G) NG CV(%)

18abcd 18abcd 23abcd 25abcd 23abcd 20abcd 14cd 12d 23abcd 25abcd 24abcd 28abc 17abcd 16bcd 22abcd 30a 27abc 29ab 28abc 26abc 22

3g 3g 6fg 14cde 19abcd 15bcde 21ab 19abc 14cde 12e 15cde 15cde 13de 21a 15bcde 12e 17abcde 11ef 12e 14de 14

* ** ** 17

Root dry weight 100: Dry weight of root; shoot; and grain *,** Significant at the 5% and 1% probability levels, respectively. Means followed by the same letter in the same column are not significant at the 5% probability level by Tukey’s test. Contribute of roots in total plant dry weightð%Þ ¼

Alfisols in Africa. Similarly, Ae et al. (1990) reported that pigeon pea as a cover crop was more efficient in utilizing iron-bound P than several other cover crops. These authors also reported that this ability of pigeon pea was attributed to root exudates, in particular, piscidic acid and its p-O-methyl derivative, which release P from Fe–P by chelating Fe3þ. These results also show that P fertilization also improves the root weight of cover crops, which may be beneficial in improving soil organic matter content and soil microbial activities.

263

The Role of Mineral Nutrition on Root Growth of Crop Plants

Table 2 Contribution of tropical legume cover crops root dry weight (%) in the total plant weight as influenced by phosphorus rates Cover crop

0 mg P kg 1 100 mg P kg 1 200 mg P kg 1

Crotalaria breviflora Crotalaria juncea L. Crotalaria mucronata Crotalaria spectabilis Roth Crotalaria ochroleuca G. Don Calopogonium mucunoides Pueraria phaseoloides Roxb. Cajanus cajan L. Millspaugh Cajanus cajan L. Millspaugh Dolichos lablab L. Mucuna deeringiana (Bort) Merr. Mucuna aterrima (Piper & Tracy) Holland Mucuna cinereum L. Canavalia ensiformis L. DC. Average F-test P rate (P) Cover crops (C) PC

20.77a 12.28abc 4.91c 11.66abc 5.18c 11.66abc 7.22bc 21.50a 16.90abc 12.81abc 19.70ab 14.17abc

11.39ef 19.77abcdef 19.29abcdef 24.72ab 22.76abc 12.16def 17.17abcdef 21.12abcde 21.80abcd 22.22abc 15.08bcdef 25.16a

21.22ab 12.49bcd 21.96ab 9.26cd 17.38abc 25.91a 8.15cd 6.36d 13.77bcd 14.71bcd 13.36bcd 16.43abc

11.09abc 11.83abc 12.98

13.44cdef 11.30f 18.93

16.94abc 9.12cd 14.79

** ** **

Root dry weight 100: Dry weight of root and shoot ** Significant at the 1% probability level. Means followed by the same letter in the same column are not significant at the 5% probability level by Tukey’s test. Contribute of roots in total plant dry weightð%Þ ¼

5. Rooting Depth and Root Distribution Rooting depth and distribution are important traits for absorption of water and nutrients from the soil profile. Rooting depth, that is, the maximum depth that roots reach, is difficult to ascertain in the field (Hoad et al., 2001). Hsiao et al. (2009) reported that roots typically reach maximum depth about the time when the canopy begins to senesce under nonstress conditions. Gregory (1994) reported that rooting depth in cereals increases until anthesis. Hoad et al. (2001) reviewed the rooting depth literature for cereals and concluded that individual roots of cereal crops can reach a depth of over 2.0 m under favorable conditions. Soil compaction reduces rooting depth (Lipiec et al., 1991; Unger and Kaspar, 1994).

264

N. K. Fageria and A. Moreira

Martino and Shaykewich (1994) reported that the proportion of roots penetrating the soil is inversely related to the soil penetration resistance. Ueda (1936) observed that wheat cultivars with relatively great cold resistance had roots that penetrated rapidly into deeper soil layer at early stages of growth. Similarly, Sanders and Brown (1978) reported that differences in the yields of indeterminate and determinate soybean varieties mirrored the distribution of their root systems, although the dry weights of their root systems were almost the same. Yamauchi et al. (1987b) compared the root system distributions of 13 species of cereals and reported that there was a significant difference among cereals in the distribution and depth of penetration of root systems. Some of the cereals had “concentrated” root systems and others had “scattered” type root systems. Large amounts of organic matter and immobile nutrients are generally found in the upper soil layers. Hence, a major part of the roots of the most crops is concentrated in the upper 0–20 cm soil depth (Gregory, 1994). Application of nitrogen fertilizer to barley caused an accumulation of 90–97% of the root mass in the top 30 cm soil layer (Hansson and Andren, 1987). Similarly, Haberle et al. (1996) reported only a few unbranched primary roots below a depth of 25 cm in fertilized wheat. Differences in root lengths, dry weights of roots at different soil depths, and the extent of rooting at the seedling stage were related to differences in yield and the ability of wheat cultivars to escape drought (Hurd, 1974). Upland rice cultivars, which are more drought tolerant than lowland cultivars, have deeper and more prolific rooting systems (Steponkus et al., 1980). When soil types did not restrict the rooting potential, deep rooting of bean cultivars was positively associated with seed yield, crop growth, cooler canopy temperature, and soil water extraction (Sponchiado et al., 1989). Because about 90% of the total NH4, P, and K uptake and root length of flooded rice cultivars occur within the surface 20 cm of soil, samples collected for routine soil tests should be taken from the top 20 cm (Teo et al., 1995). Lowland rice plants develop a surface mat of roots in the oxygenated zone near the soil surface soon after application of flood waters (University of Arkansas Cooperative Extension Service Rice Committee, 1990). Durieux et al. (1994) reported that more than half of the root length of maize was located in the surface 0–20 cm depth at all sampling times during a season. Roots of the peanut (Arachis hypogaea L.) cultivar Florunner penetrated to depths up to 280 cm when grown in a sandy soil, and the most extensive root growth occurred in the top 30 cm (Boote et al., 1982). Sharratt and Cochran (1993) reported that 85% and 95% of the root mass of barley was located in interrows of the top 20 and 40 cm of soil, respectively. Welbank and Williams (1968) also found that nearly 80% of barley roots occupied the uppermost 15 cm of soil. A study conducted by Stone and Pereira (1994a,b) of four common bean cultivars and three upland rice

The Role of Mineral Nutrition on Root Growth of Crop Plants

265

cultivars to evaluate rooting depths in an Oxisol showed that 70% of the roots were concentrated in the top 20 cm layer and about 90% were concentrated in the top 40 cm soil depth of both crops. The presence of these roots in surface soil layers may contribute to large amounts of nutrients measured in the upper 20 cm of the soil. Using the Claassen–Barber model to predict nutrient uptake by maize grown in silt loam soil, >90% of K and P uptake occurred in the top 20 cm soil depth (Schenk and Barber, 1980). Silberbush and Barber (1984) reported that about 80% of P and 54% of K uptake by soybean was from 0 to 15 cm depths. Soybean cultivars differ in their rate of downward growth during specific shoot development stages and in their maximum rooting depth on specific. Cultivars selected for rapid taproot elongation rates in a greenhouse trial were found to have greater rooting depths in rhizotron and field trials than cultivars selected for slow taproot elongation (Kaspar et al., 1978, 1984).

6. Root Growth as a Function of Plant Age Root development varies with stages of plant growth and development. The most rapid development of corn (Zea mays L.) roots occurs during the first 8 weeks after planting (Anderson, 1987). As corn plants age, growth of roots generally increases at slower rates than shoots (Baligar, 1986). After silking, corn root length declines (Mengel and Barber, 1974). This decline in root length after silking presumably is due to the high C demand of grain resulting in enhanced translocation of C and N to grain, including some C and N that roots would normally obtain (Wiesler and Horst, 1993). Peanut (A. hypogaea L.) root length density and root weight density increased at each soil depth increment from planting to 80 days after planting (Ketring and Reid, 1993). These authors reported that roots had penetrated to depths of 120 cm 40–45 days after planting and spread laterally to 46 cm in mid-furrow. The 0–15 cm depth increment had the highest mean root length density, which increased to a maximum of 2.1 cm cm 3 at 80 days after planting (Ketring and Reid, 1993). This meant that peanut roots were established both deeply and laterally in the soil profile early in the growing season. This would be advantageous in drought environments and helpful for water management. Sunflower (Helianthus annuus L.) rooting depth reached 1.88 m at the beginning of disk flowering and 2.02 m at the completion of disk flowering ( Jaffar et al., 1993). In a review of depth development of roots with time for 55 crop species (Borg and Grimes, 1986), it was shown that maximum rooting depth for most crop species was generally achieved at physiological

266

N. K. Fageria and A. Moreira

Root dry weight (g plant−1)

maturity. Kaspar et al. (1984) noted that the rate of soybean (Glycine max L. Merr.) root depth penetration reached a maximum during early flowering and declined during seed fill. However, some root growth was observed throughout the reproductive stage until physiological maturity (Klepper and Kaspar, 1994). Slaton et al. (1990) studied root growth dynamics of lowland rice and found that maximum root growth rates were reached between active tillering and panicle initiation, and maximum root length was reached by early booting. Beyrouty et al. (1987) noted that the most rapid rate of root and shoot growth in flooded rice occurred before panicle initiation, which corresponds to the plant transition between vegetative and reproductive growth. Approximately 77% and 81% of total shoot and root biomass, respectively, was achieved before panicle initiation. Following panicle initiation, the length of roots and shoots increased only slightly until harvest (physiological maturity). Beyrouty et al. (1988) also reported that lowland rice root growth was most rapid during vegetative growth, with maximum root length occurring at panicle initiation. Root length either plateaued or declined during reproductive growth. Fageria and Santos (2011) studied the root and shoot growth of lowland rice during its growth cycle (Fig. 5). Root dry weight increased in a quadratic fashion with the advancement of plant age from 19 to 120 days, but shoot dry weight increased linearly during the growth cycle. Development of the root system was slow during the first 40 days after sowing and then it increased almost linearly until physiological maturity. The slow

40

Y = −2.4581 + 0.0952X + 0.00098X 2 R 2 = 0.9063**

30 20 10

Shoot dry weight (g plant−1)

0 60

Y = −12.5474 + 0.5287X R 2 = 0.9702**

50 40 30 20 10 0

20

40

60

80

100

120

140

Plant age (days after sowing)

Figure 5 Root and shoot dry weight of lowland rice as a function of plant age (Fageria and Santos, 2011).

267

The Role of Mineral Nutrition on Root Growth of Crop Plants

Maximum root length (cm)

increase in root dry weight early in the growth cycle may be associated with low translocation of photosynthetic materials due to low leaf area (Fageria, 2007). When there is a low amount of photosynthetic product, a major part goes to the shoot, and very little is translocated to the roots (Fageria, 1992). Root growth generally parallels shoot growth in crop plants. When a large amount of nutrients, especially N, is supplied to leaves from roots, photosynthesis remains high during maturation, which secures the supply of carbohydrates to roots. Hence, the activities of roots and shoots are mutually dependent (Osaki et al., 1997). Figure 6 shows maximum root length and root dry weight of dry bean during the growth cycle of a Brazilian cultivar BRS Valente under greenhouse conditions. Maximum root length was

40 30 20 10

Root dry weight (g plant−1)

0 4

Y = 6.0778 + 0.7862X − 0.0049X 2) R 2 = 0.8353** Y = 0.0267EXP(0.0824X − 0.00032X 2) R 2 = 0.8545**

2

Number of trifoliates (plant−1)

0 20

Y = 0.4461EXP(0.0787X − 0.00041X 2) R 2 = 0.9216**

10

Shoot dry weight (g plant−1)

0 20

Y = 0.0297EXP(0.1161X − 0.00054X 2) R 2 = 0.9430**

10

0

20

40

60

80

100

Plant age (days after sowing)

Figure 6 Relationship between plant age and dry bean growth parameters (Fageria and Santos, 2008).

268

N. K. Fageria and A. Moreira

achieved at 80 days after sowing. However, root dry weight increased quadratically with plant age, much like shoot growth. The youngest growing portions of the roots are most active in water and mineral nutrient uptake; therefore, the root system must continually grow to provide adequate nutrition for the plant (Brown, 1984).

7. Root–Shoot Ratio The partitioning of photoassimilate between roots and shoots has frequently been analyzed as a balance between root and shoot activity (Brouwer, 1966; Davidson, 1969a; Werf, 1996). Different plant species may have different patterns for photosynthate transportation and allocation to shoot and root (Dyer et al., 1991; Freckman et al., 1991). There is an interdependence of shoot and root for growth and development. The shoot relies on the root for water and nutrients, while the roots depend on the shoot for carbohydrates (Hoad et al., 2001). The terms “shoot” and “root” are used here in a botanical sense and refer, respectively, to the entire aerial and subterranean portions of higher seed plants (Aung, 1974). In the early part of the twentieth century, shoot–root ratios were used rather extensively to characterize plant response to imposed nutritional changes. Root growth is closely related to the whole plant growth. This relationship is called “allometry” or relative growth. Root dry weight is related to the total dry weight of a plant using the following equation (Yoshida, 1981): WR ¼ HWTh ; where WR is the root dry weight, WT is the total dry weight (shoot dry weight þ root dry weight), and H and h are constants. The above relationship has been tested for different rice cultivars grown under various environmental conditions, and can be expressed by the following equation (Yoshida, 1981): WR ¼ 0:212WT0:936 : When plants are small (substitute 1 for WT), WR/WT is 0.2; WR/WT values approach 0.1 as plants grow larger (substitute 105 for WT). In other words, ratio of root dry weight to total dry weight ranges from 0.2 at the seedling stage to 0.1 at the reproductive stage (heading) for rice (Yoshida, 1981). The above relationship between root and total dry weights gives an estimate of root mass that remains in soil if shoot weight is known. For example, when plants produce shoot dry weights of 3 Mg ha 1 at heading,

The Role of Mineral Nutrition on Root Growth of Crop Plants

269

root dry weights remaining in soil should be 330 kg ha 1. Partitioning of dry matter in roots relative to shoots is high during the seedling stages of growth and steadily declines throughout development (Evans and Wardlaw, 1976). The shoot:root biomass ratio changes during ontogeny, generally becoming high as the plant approaches flowering, and stabilizing after flowering (Zobel, 1986). Shoot-to-root ratios of common bean, rice, wheat, and cowpea increased as plants advanced in age (Fageria, 1992). Increases in shoot-to-root ratios indicate that shoots have a higher priority for photosynthate accumulation than roots. If shoot–root ratios decrease with time, roots have preferential utilization of photosynthates under the existing plant growth conditions. A relatively high conservation of photoassimilate in shoots may increase the plant’s photosynthetic leaf area while decreasing root biomass and the plant’s capacity for water and nutrient uptake (Werf, 1996). Environmental stresses increase the relative weights of roots compared to shoots (Eghball and Maranville, 1993). Decrease in the availability of N, P, or water increased root–shoot ratios of perennial ryegrass (Lolium perenne L.) (Davidson, 1969b). Although deficiencies of many mineral elements influence plant growth and root–shoot relationships, invariably water and N deficiency limit shoot growth the most. Root–shoot ratios of 28-day-old maize plants were 0.27, 0.15, and 0.18 at volumetric soil moisture contents of 0.22, 0.27, and 0.32 m3 m 3, respectively (Mackay and Barber, 1985). When plants are N-deficient, relatively more photosynthate is used by roots as they develop greater length to aid the plant in obtaining more N. In general, when low nutrient levels do not reduce maize grain yield by more than 20%, addition of N will reduce total root weights even though shoot weights increase (Barber, 1995). Champigny and Talouizte (1981) reported that under N deprivation, translocation of photoassimilates from shoots to roots increased because of increased sink strength of roots compared to shoot sinks. In an experiment with 18-day-old wheat seedlings deprived of N for 7 days, soluble sugar contents in roots were higher than in the corresponding roots of seedlings grown continuously with complete nutrient solutions (Talouizte et al., 1984). Similarly, root–shoot ratios of maize plants were higher when grown with low soil N compared to adequate N (Eghball and Maranville, 1993). Soil salinity is another important soil chemical property that influences shoot–root ratios. The depressing effect of salinity on root growth is generally less severe than its effect on shoot growth. Shalhevet et al. (1995) summarized the results of 10 experiments relating shoot and root growth to salinity. In all the experiments, the root and shoot responses were evaluated by measuring fresh or dry weights at the end of the experimental periods. All the 10 experiments produced either the same or stronger growth responses of shoots than roots because of the imposed osmotic potential. However, Slaton and Beyrouty (1992) observed shoot–root ratios

270

N. K. Fageria and A. Moreira

of rice remained constant as a result of a functional equilibrium in which shoot growth was proportional to root growth. Partitioning of photosynthates and their effects on dry matter distribution is influenced by several environmental factors such as low temperature, drought, and mineral nutrient deficiency (Wardlaw, 1990). The mineral nutrients P and N exerted pronounced influences on photosynthate and dry matter partitioning between shoots and roots (Costa et al., 2002). Phosphorus and N-deficient plants usually produce proportionately more dry matter to roots than shoots, compared with unstressed plants. This probably results from higher export rates of photosynthate to roots in deficient plants. Leaf expansion is highly sensitive to low tissue P concentrations, producing higher concentrations of sucrose and starch in P-deficient leaves because of reduced demand (Fredeen et al., 1989). Thus, roots become more competitive for photosynthates than shoots, which leads to higher export of carbohydrates to roots with correspondingly lower shoot–root ratios (Rufty et al., 1993). Cakmak et al. (1994) reported that dry matter partitioning between shoots and roots of common bean was affected differently by low supplies of P, K, and Mg. Although total dry matter production was somewhat similar in P-, K-, and Mg-deficient plants, K- and especially Mgdeficient plants had greater than normal shoot–root ratios, while P-deficient plants had smaller than normal shoot–root ratios (Cakmak et al., 1994). Shoot–root dry weight ratios were 1.8 in P-deficient, 4.9 in control, 6.9 in K-deficient, and 10.2 in Mg-deficient plants. Upland rice usually has high root/shoot weight ratios than lowland rice, an adaptation improving access to soil water (Dingkuhn and Kropff, 1996).

8. Root Growth Versus Crop Yield Roots are responsible for absorption of water and nutrients which are important resources affecting crop yields. In addition, roots improve soil organic matter content and biological activity in the rhizosphere. Root length and root dry weight are standard root parameters that are measured in many studies, largely because they are more easily determined than other root system properties (Gregory, 1994). Barber and Silberbush (1984) studied the relationship between root length and soybean yield and concluded that yield was significantly related to total root length at the R6 (full seed) stage. These authors concluded that root growth is important in determining the nutrient supply to the shoot which, in turn, affects crop yield. Similarly, Thangaraj et al. (1990) reported that root length density of lowland rice at flowering was directly proportional to grain yield. Leon and Schwang (1992) used the grid intercept method (Newman, 1966) to evaluate differences in total root length between cultivars of oats and barley and

271

Grain yield (g plant−1)

The Role of Mineral Nutrition on Root Growth of Crop Plants

Y = 3.6178EXP(0.9326X − 0.0109X

20

2

)

2

R = 0.5101** 10

0

20

30

40

Maximum root length (cm)

Figure 7 Relationship between maximum root length and grain yield of upland rice. Not the root dry weight.

Grain yield (g plant−1)

20

10 Y = 1.0257EXP(0.9397X − 0.0718X R 2 = 0.8656** 0

1

3

5

7

2

)

8

Root dry weight (g plant−1)

Figure 8

Relationship between root dry weight and grain yield of upland rice.

found that yield stability was correlated with root system length. The first author studied the relationship between maximum root length and root dry weight and grain yield of upland rice (Figs. 7 and 8). Grain yield increased in a quadratic fashion with increasing root length or root dry weight, and root dry weight was a better predictor than root length of yield. Similarly, the author studied the relationships between root length and root dry weight and shoot dry weight of tropical legume cover crops (Figs. 9 and 10). There was a significant increase in shoot dry weight of legume cover crops with increasing root length and dry weight, and as with upland rice, root dry weight was a better predictor than root length of shoot dry weight.

9. Genotypic Variation in Root Growth Variability in root growth among crop species and among genotypes of the same species is widely reported in the literature (Fageria. 2009; Gregory, 1994; Kujira et al., 1994; Marschner, 1998; O’Toole and Bland, 1987). This variability can be used in improving the yield of annual crops by

272

N. K. Fageria and A. Moreira

Y = 2.9808 − 0.2634X + 0.0076X 2

Shoot dry weight (g plant−1)

8

R 2 = 0.8783**

7 6 5 4 3 2 1 0

10

20 30 40 Maximum root length (cm)

50

Shoot dry weight (g plant−1)

Figure 9 Relationship between maximum root length and shoot dry weight of tropical legume cover crops. Values are averages of 14 tropical legume cover crops.

Y = 0.3422EXP(5.5187X − 2.7205X

8 7

2

)

R 2 = 0.9712**

6 5 4 3 2 1 0

0.2

0.4 0.6 0.8 1.0 Root dry weight (g plant−1)

1.2

Figure 10 Relationship between root dry weight and shoot dry weight of tropical legume cover crops. Values are averages of 14 tropical legume cover crops.

incorporating vigorous root growth into desirable cultivars. Vigorous root growth is especially important when nutrient and water stress are significant (Gregory, 1994). Ludlow and Muchow (1990), in their review of traits likely to improve yields in water-limited environments, place a vigorous rooting system high in their list of properties to be sought. O’Toole and Bland (1987) reviewed genotypic variation in root growth of annual crops and reported significant differences in rooting depths, maximum root length, and distribution pattern in the soil profile. Hurd (1974) and Yoshida and Hasegawa (1982) reported rooting depth differences among genotypes of wheat (Triticum aestivum L.) and rice,

The Role of Mineral Nutrition on Root Growth of Crop Plants

273

respectively. Similarly, Brown et al. (1987) and Pan et al. (1985) also reported differences in rooting depths of corn and barley (Hordeum vulgare L.), respectively. Similar variability exists in dicotyledonous species where most work has been undertaken on legumes, including dry bean (Fageria, 2002a; White and Castillo, 1989), soybean (Taylor et al., 1978; Zobel, 2005a), chickpea (Cicer arietinum L.; Vincent and Gregory, 1986), peanut (Mathews et al., 1988), and white clover (Trifolium repens L.; Caradus, 1990). Hamblin and Tennant (1987) compared root growth and water uptake of wheat and lupin (Lupinus albus L.) and Gregory and Brown (1989) did similar work with barley and chick pea. These authors concluded that there were differences between the species in the root length necessary to extract water, and the rate of water extraction was greater in legumes than in cereals. Fageria (1991) studied root dry weight of the Brazilian upland rice cultivars IAC 47 and the International Rice Research Institute (IRRI) cultivar IR 43 and concluded that root dry weight of Brazilian cultivars was almost double that of the IRRI cultivar during the entire growth cycle. Xiaoe et al. (1997) reported that hybrid rice has a more vigorous root system, larger panicle, and more grains per panicle than traditional rice cultivars. The genotypic variability in the root growth of annual crops has been used to identify superior genotypes for drought-prone environments (Gregory, 1994; Hurd et al., 1972). Gregory and Brown (1989) reviewed the role of root characters in moderating the effects of drought and concluded that roots may have a direct effect, by increasing the supply of water available to the crop, or an indirect effect by changing the rate at which the supply becomes available. Where crops are grown on deep soils and water is stored throughout the whole soil profile, the depth of rooting has a major influence on the potential supply of water (Gregory, 1994). Rain may replenish the upper soil during the season, but later growth and grain filling in many crops are accomplished during periods of low rainfall when soil moisture stored deep in the profile must be utilized. Sponchiado et al. (1989) reported that in dry bean, drought avoidance results from root growth and soil water extraction deep in the profile. Atkinson (1990) reported significant variation in the speed of root penetration, specific root length, branching pattern, root density, total root mass, and root hair development of 25 spring barley cultivars. Information reported in the literature on old and new cereal cultivars indicates that more modern cultivars are more responsive than older cultivars to high nutrient availability (Haberle, 1993; Haberle et al., 1995), although they tend to have a lower root fraction (Wahbi and Gregory, 1995). Root physiological characteristics also differ among cultivars and can affect processes like nutrient acquisition (Hoad et al., 2001; Marschner, 1998). The rate of uptake of nutrient per unit root length depends on the nutrient availability but also varies considerably among cultivars (Hoad et al., 2001; Romer, 1985).

274

N. K. Fageria and A. Moreira

10. Root Oxidation Activity in Oxygen-Deficient Soils Roots require O2 for respiration, water and mineral absorption, and other metabolic activities. Brailsford et al. (1993) reported that roots asphyxiate within a few hours or days at normal growing temperatures unless at least a small amount of oxygen (0.1 kPa in a flowing gas phase) is available. The O2 used by roots is replaced largely by molecular diffusion between soil air and the aboveground atmosphere (Miller, 1986). Waterlogging, defined as the soil saturated or nearly saturated with water, significantly reduces oxygen availability to roots, causes the soil to become “reduced,” results in the production of toxic compounds, and generally reduces root growth (Ellis, 1979; Hoad et al., 2001). Cereals are generally more tolerant to waterlogging than legumes. Hoad et al. (2001) reported that cereal roots can survive short periods of waterlogging without adverse effects because they can use the small quantities of oxygen dissolved in the soil water and are capable of anaerobic metabolism when oxygen demand exceeds supply. Oxygen requirements are higher for growing root tips than for the older parts of the root system. The effect of waterlogging is least at low temperatures when more oxygen can be dissolved in the water and biological activity is depressed (Hoad et al., 2001). The effect of waterlogging is reduced with the application of chemical fertilizers, especially K, which can improve the respiration capacity of roots (Fageria, 2009). Most agricultural crops are mesophytes, which, for maximum growth, require an environment that is neither too wet nor too dry (Fageria, 1992). However, some plant species grow well under anaerobic conditions. It has long been known that marsh plants or hydrophytes, such as rice, are genetically adapted to grow in reduced soil environments (Horiguchi, 1995). Adaptation to waterlogging in hydrophytes is the result of their unique ability to translocate O2 from the shoot to the root system. Ando et al. (1983) reported that in rice oxygen absorbed from the atmosphere by the shoots can be translocated to the root system and released into the rhizosphere within 5 min. Generally, flooded rice roots have a reddish brown color due to oxidation of the Fe ions by oxygen release from the roots and their deposition on the root surface (Fageria et al., 2008a,b). Flooding induces many changes in plant roots, of which formation of aerenchyma (large interconnected intercellular spaces) is an important adaptive mechanism (Laan et al., 1989). Some species like corn (Z. mays L.) develop aerenchyma as a response to flooding or anoxia (Armstrong and Drew, 2002), a facultative development, while others develop them routinely regardless of the environmental conditions (constitutive development; Barlow, 2002; Zobel, 2005a). The development of aerenchyma in

275

The Role of Mineral Nutrition on Root Growth of Crop Plants

rice (O. sativa L.), and other species and crops growing in wet soils, allows the roots to grow into saturated soils and still extract nutrients (Zobel, 2005b). In corn, the leaf bases also form aerenchyma when submerged, while the enclosed stem bearing aerenchymatous nodal (adventitious) roots may elongate more rapidly ( Jackson, 1994). Aerenchyma formation in corn roots appears to be triggered by the loss of tonoplast integrity (Campbell and Drew, 1983), although cell wall disintegration is also involved at an early stage, at least in rice ( Jackson, 1994; Webb and Jackson, 1986). Aerenchymatous roots are formed either by some cell wall separation and cell collapse (lysigeny) or by cell separation without collapse (schizogeny). Figure 11 shows cross sections of rice and wheat roots, and lysigenous and schizogenous intercellular space. Both forms result in large longitudinal channels in root cortices, and such structures enhance diffusion of atmospheric or photosynthetic oxygen from shoots to roots so that aerobic respiration and growth can be maintained (Armstrong, 1979). Changes in root morphology occur after flooding for both wetland and nonwetland plant species. Flooding may also increase branching of roots, development of new adventitious roots, and superficial rooting (Laan et al., 1989). Cellular spaces that exist in roots facilitate oxygen diffusion; however, the amount of aeration varies greatly among plant species. In the case of certain marsh plants like rice, root cortical cells are arranged in columns, and when channels form, the spaces become large and continuous, facilitating diffusion. However, in the case of certain terrestrial plants, the cellular arrangement is oblique and the spaces formed are small (Horiguchi, 1995). There are marsh plants, however, that do not display columnar forms, while some terrestrial plants do (Horiguchi, 1995). In the case of rice, large cortical aerenchyma spaces develop schizogenously and lysigenously even in well-aerated soils. Horiguchi (1995) reported that in rice and other graminaceous plants, aerenchymous cells are well developed in both the nodes and the internodes. The oxidizing capacity of roots of crop plants also depends on the soil fertility level. Higher soil fertility improves the oxidation activity of rice roots (Horiguchi, 1995). Application of silicon to rice A Rice

B Lysigenous intercellular space

Wheat

Schizogenous intercellular space

Figure 11 Cross section of a rice (A) and wheat (B) root showing lysigenous and schizogenous intercellular spaces (from Horiguchi, 1995).

276

N. K. Fageria and A. Moreira

crops increases the diffusion of oxygen from the tops to the roots, strengthening oxidation activity and the deposition of ferrous iron on root surfaces, and restricting the transport of excess iron to the top (Horiguchi, 1995). Development of adventitious roots in some crops is related to plant tolerance of soil waterlogging. Such roots developed by corn after one day of flooding were short and porous enough to allow significant O2 movement from the shoots to the root meristems (Miller, 1986; Wenkert et al., 1981). Despite this adventitious root formation and increased root porosity, corn and most other crop plants soon perish under such conditions (Miller, 1986).

11. Root Growth in Conservation Tillage Systems Conservation tillage is defined as any tillage sequence, the object of which is to minimize or reduce loss of soil and water; operationally, a tillage or tillage and planting combination that leaves a 30% or greater cover of crop residue on the surface (Soil Science Society of America, 2008). The benefits of conservation tillage are reducing soil erosion, conserving soil moisture, avoiding fluctuations of soil temperature in the arable soil depth, and reducing the costs of soil preparation. In addition, the use of conservation tillage is being encouraged as part of a strategy to reduce C loss from agricultural soils (Kern and Johnson, 1993). Decomposition rates are generally slower in no-till than conventional tillage in which the decomposition of soil organic matter is promoted by the stirring of the soil and alterations in the soil microclimate (Parton et al., 1996). Holland and Coleman (1987) suggested that C retention is increased in no-till because the surface residue is primarily decomposed by fungi, which have higher assimilation efficiency than the bacteria, which dominate the decomposition processes of residue mixed into the soil. Gale and Cambardella (2000) reported that there was a clear difference in the partitioning of surface residue and root-derived C during decomposition and imply that the beneficial effects on no-till on soil organic C accrual are primarily due to the increased retention of rootderived C in the soil. Knowledge of how plant root systems grow under conservation tillage is important because this practice is widely adopted in many countries around the world, most notably, in countries such as United States, Brazil, Argentina, Canada, and Australia (Bolliger et al., 2006). Fortyfive percent of the total cultivated land in Brazil is now estimated to be managed with conservation tillage, although in southern Brazil, this figure is reported to exceed 80% (Bolliger et al., 2006). Conservation tillage reduces soil erosion, conserves soil moisture, conserves energy, increases soil organic matter content, and consequently, soil

The Role of Mineral Nutrition on Root Growth of Crop Plants

277

quality. However, conservation tillage may compact surface soil horizons and may lead to poor root growth. Adverse effects of soil compaction on crop production have been recognized for many years. Cato the Elder (234–149 B.C.) wrote that the first principle of good crop husbandry is to plow well and the second principle is to plow again, presumably to provide a “mellow” (wellaerated) seedbed (Unger and Kaspar, 1994). Conservation tillage increases soil bulk density (Martino and Shaykewich, 1994) which may inhibit root growth in the upper part of the soil profile (Cannell, 1985; Lampurlanes et al., 2001), reducing nutrient uptake and plant growth (Peterson et al., 1984). Qin et al. (2004) reported that wheat root length density, mean root diameter, and percentage of small-diameter roots were lower in no-till than conventional tillage. Root diameter may be indicative of the effects of soil strength on root growth and affects the utilization of nutrients in the soil. Sidiras et al. (2001) reported thicker barley roots under conventional tillage than under no-till. In general, bulk densities that impede root growth are 1.55 Mg m 3 for clay loams, 1.65 Mg m 3 for silt loams, 1.80 Mg m 3 for sandy loams, and 1.85 Mg m 3 for loamy fine sands (Miller, 1986). Tillage-induced differences in the soil nutrient status may also have a significant impact on root growth (Qin et al., 2004). Conservation tillage often results in the stratification of soil nutrients, especially of immobile nutrients like P (Crozier et al., 1999; Holanda et al., 1998; Logan et al., 1991). This produces greater soil fertility near the soil surface which, in contrast to the effects of compaction described above, causes an increase in root length density near the soil surface under conservation tillage (Cannell and Hawes, 1994; Gregory, 1994). Frequently, root growth is greater from 0 to 5 cm in conservation and no-tillage systems than in conventional tillage systems (Chan and Mead, 1992; Rasmussen, 1991; Wulfsohn et al., 1996). Radial root swelling has been reported for lupins (Lupinus angustifolius L.) grown in compacted soil (Atwell, 1989), for barley (H. vulgare L.) under mechanical impedance (Wilson et al., 1977), and for mustard (Brassica sp.) in drying soil (Vartanian, 1981). Studies on root elongation of cotton as a function of soil strength and soil water content showed that root elongation is more sensitive to soil strength than water content (Taylor and Ratliff, 1969). Root volumes were less affected than length, indicating an increase in root diameter (Ball et al., 1994). Chassot et al. (2001) reported that conservation tillage decreases soil temperature, and this may be the main reason for the poor growth of the roots and shoots of corn seedlings compared with conventional tillage under temperate humid conditions. Considering the many advantages of conservation tillage compared to conventional tillage, the effects of soil compaction produced by conservation tillage can be minimized. Unger and Kaspar (1994) reported that growing deep-rooted crops in rotation will help avoid or alleviate compaction, improving root distribution and increasing rooting depth. These authors also reported that weather conditions and soil moisture can enhance

278

N. K. Fageria and A. Moreira

or diminish the effects of compaction on root growth. Even if compaction limits root growth, subsequent weather events may either enhance or diminish the effect of the root limitation on crop growth. The first author has conducted field experiments on an upland rice-dry bean rotation using conservation tillage on Brazilian Oxisols. The upland rice root system was adversely affected, and rice yields were low (around 2000 kg ha 1), about half that expected in a field experiment with conventional soil preparation and favorable environmental conditions. However, dry bean yields were greater than 3000 kg ha 1 (Fageria, 2008; Fageria and Stone, 2004). Hence, selecting the appropriate crop is important.

12. Mineral Nutrition Versus Root Growth There are 17 nutrients essential for plant growth and development and these are carbon (C), hydrogen (H), oxygen (O), nitrogen (N), phosphorus (P), potassium (K), calcium (Ca), magnesium (Mg), sulfur (S), zinc (Zn), copper (Cu), manganese (Mn), iron (Fe), boron (B), molybdenum (Mo), chlorine (Cl), and nickel (Ni). Among these nutrients, plants take C, H, and O from air and soil water and the remaining from the soil solution. About 95% of the plant weight is C, H, and O, and the remaining 5% is the other 14 nutrients. The C, H, O, N, P, K, Ca, Mg, and S are required by plants in large amounts and for this reason are known as major or macronutrients. The remaining nutrients, Zn, Cu, Mn, Fe, B, Mo, Cl, and Ni, are classified as micronutrients because they are required in small amounts by plants (Fageria et al., 2002). Among the micronutrients, Cl is also absorbed by plants in large amounts, but it is needed in only small amounts, and Cl deficiency is rarely observed in crop plants. All the 17 nutrients are equally important for plant growth. If any of these nutrients is limiting in the growth medium, plant growth will be reduced. An example is given in Fig. 12, where at an adequate fertility level, the relative dry weight of dry bean was 100%. When essential nutrients were omitted from the soil or were not applied, growth was reduced, but the amount of reduction depended on the nutrient. The impacts of deficiencies on growth were in the order of P > Ca > Mg > N ¼ K > S among macronutrients and B > Zn > Cu > Fe > Mn > Mo among micronutrients. Similarly, the influence of N, P, and K on shoot and root growth of dry bean is shown in Fig. 13. Both shoot and root growth were significantly reduced when N, P, or K were omitted from the growth medium. Figures 14 and 15 show root growth of wheat, upland rice, and corn with the addition of N þ P þ K and with the omission of these nutrients from the growth medium. Root growth of these three crop species was decreased in the absence of N, P, and K in the soil. A significant amount of variation

279

The Role of Mineral Nutrition on Root Growth of Crop Plants

Relative shoot dry weight (%)

120 100 80 60 40 20

n

o

−M

−M

−B

u −C

n

−F e

−Z

−S

g −M

a −C

−K

−P

−N

AF L

0

Figure 12 Relative shoot dry weight of dry bean as influenced by adequate fertility level (AFL) and other nutrients were not applied or omitted from the Oxisol. 120 Shoot

Root

Relative dry weight (%)

100 80 60 40 20 0

N+P+K

−N

−P

−K

Figure 13 Relative dry weight of shoot and root of dry bean as influenced by N, P, and K fertilization.

exists, both within and among crop species, in nutrient acquisition and use. This variability reflects differences in root morphology and mechanisms that either aid or prevent ion movement into the root (Gabelman et al., 1986).

12.1. Nitrogen Nitrogen is one of the most yield-limiting nutrients in crop production in most agroecosystems. Nitrogen plays numerous key roles in plant biochemistry, including being an essential constituent of enzymes, chlorophyll,

280

N. K. Fageria and A. Moreira

N+P+K

−N

−P

−K

Wheat

Upland rice

Figure 14 Root growth of wheat at the top and upland rice at the bottom grown at N þ P þ K and N, P, and K levels.

N+P+K

−P

−N

−K

Corn

Figure 15 Corn root growth at N þ P þ K and N, P, and K levels.

nucleic acids, storage proteins, cell walls, and a vast array of other cellular components (Harper, 1994). Consequently, a deficiency of N in crop plants profoundly influences plant growth, development, and yield. The recovery

The Role of Mineral Nutrition on Root Growth of Crop Plants

281

of applied N with chemical fertilizers is lower than 50% for most annual crops. The low recovery of N is associated with loss of this element by leaching, denitrification, volatilization, incorporation into soil microorganisms, and soil erosion (Fageria and Baligar, 2005). To improve the efficiency of N uptake and use by crop plants, root systems play an important role. Root morphology is influenced by the amount of N fertilizer applied (Eghball et al., 1993) and factors such as temperature (Feil et al., 1991) and soil mechanical impedance (Bengough and Mullins, 1990). Eghball et al. (1993) showed that N stress in corn reduced root branching. Similarly, Maizlish et al. (1980) showed greater root branching in corn with increasing levels of applied fertilizer N. Costa et al. (2002) reported that greater root length and root surface area were obtained at an N fertilizer rate of 128 kg N ha 1 compared with either the absence of fertilizer N or the higher rate of 255 kg N ha 1. Nitrogen fertilizer improves root growth in soils having low-OM content (Gregory, 1994; Robinson et al., 1994). Nitrogen fertilization may increase crop root growth by increasing soil N availability (Garton and Widders, 1990; Weston and Zandstra, 1989). Sainju et al. (2001) observed that tomato (Lycopersicon esculentum Mill.) root growth was greater with hairy vetch and crimson clover cover crops and 90 kg N ha 1 than with no cover crops or N fertilization. Nitrogen also improves production of lateral roots and root hairs, as well as increasing rooting depth and root length density deep in the profile (Hansson and Andren, 1987). Hoad et al. (2001) reported that surface application of nitrogen fertilizer increases root densities in the surface layers of the soil. Nitrogen fertilization can increase root length and root surface area and decrease root mass per unit area of corn (Anderson, 1987; Costa et al., 2002). It is well known that roots tend to proliferate in nutrient-enriched soil zones (Drew et al., 1973; Qin et al., 2005). Russell (1977) refers to this as a compensatory response. The results of pot experiments showed that corn roots were longer and thinner in zones that were rich in N (Durieux et al., 1994; Zhang and Barber, 1992, 1993). Root mass was less affected by N than root length, but the effect may depend on the stage of maturity of the crop (Baligar et al., 1998). Higher rates of application of N reduced root growth and depth of rooting in wheat (Comfort et al., 1988) and reduced root:shoot ratio in rye (Brouwer, 1966). In corn, the primary root system was 16% thicker when NH4þ–N was applied rather than NO3–N. The NH4þ–N treatment also increased the diameters of lateral and first- and second-order nodal roots (Anderson et al., 1991). Baligar et al. (1998) reported that relative dry weights of roots of rice, dry bean, corn, and soybean were reduced by 38%, 56%, 35%, and 11%, respectively, when N was omitted from a complete fertilizer. Nitrogen deficiency also reduces branching and root hairs in cereals and legumes (Baligar et al., 1998).

282

N. K. Fageria and A. Moreira

Root dry weight (g plant−1)

Figure 16 shows that increasing the N fertilizer rate from 0 to 400 mg kg 1 increased root dry weight of upland rice in a linear fashion in Brazilian Oxisol, with the rate of N fertilizer explaining 59% of the variation in root weight. Nitrogen fertilization increases production of thinner roots with fine root hairs (personal visual observation). Data in Table 3 show root length and root dry weight of 20 upland rice genotypes grown on a Brazilian Oxisol. These two traits were significantly influenced by N rate and genotype treatments. A significant interaction between genotypes and N rates was found for root growth because some genotypes were highly response to the N application while others were not. Figures 17–20 show root growth of upland rice cultivars/genotypes at two N rates. Root growth of all the cultivars/genotypes improved with the addition of N; however, there were differences among genotypes for root development. Thus, selecting genotypes for N use efficiency may be an important aspect of improving root growth and consequently, the yield of upland rice in Brazilian Oxisols. Nitrogen sources also affect root growth in upland rice (Fig. 21). Root dry weight increased in a quadratic exponential fashion with the application of N in the range of 0–400 mg kg 1 of soil. In the case of urea, maximum root dry weight was obtained with 281 mg N kg 1 of soil. Figures 22 and 23 show how the root growth of upland rice is affected by application of urea and ammonium sulfate in the Brazilian Oxisol. Ammonium sulfate produced more vigorous root systems, especially at higher N rates, than urea, perhaps because upland rice is highly tolerant to soil acidity and ammonium sulfate reduces soil pH more than urea. Fageria (2009) reported that upland rice can tolerate up to 70% Al saturation in the soil. Fageria

4

3

2 Y =1.6407 + 0.0053X R 2 = 0.5944**

1

0

50

100

150

200

250

300

350

400

Rate of N application (mg kg−1)

Figure 16 Influence of nitrogen on root dry weight of upland rice. Values are averages of 20 upland rice genotypes.

283

The Role of Mineral Nutrition on Root Growth of Crop Plants

Table 3 Root length and root dry weight of 20 upland rice genotypes as influenced by nitrogen fertilization Root length (cm)

Root dry weight (g plant 1)

Genotype

0 mg N kg 1 300 mg N kg 1 0 mg N kg 1 300 mg N kg 1

BRA01506 BRA01596 BRA01600 BRA02535 BRA02601 BRA032033 BRA032039 BRA032048 BRA032051 BRA042094 BRA042156 BRA042160 BRA052015 BRA052023 BRA052033 BRA052034 BRA052045 BRA052053 BRS Primavera BRS Sertaneja Average F-test N rate (N) Genotype (G) NG CV(%)

34.67ab 30.00b 27.00ab 31.67ab 28.00b 30.00a 43.00a 28.50b 30.67ab 30.33b 29.33b 29.67b 31.00ab 29.67b 30.33b 29.00b 28.67b 29.00b 29.33b 30.00b 30.49 NS ** ** 14.25

21.00cd 15.67d 25.33bcd 30.00abcd 32.67abc 28.00abcd 34.67abc 32.00abc 35.67ab 38.00ab 33.00abc 32.50abc 35.00abc 26.50abcd 31.33abc 37.67ab 37.67ab 36.50ab 36.33ab 40.33a 31.99

0.92a 0.87a 1.14a 1.33a 1.11a 1.12a 1.24a 1.05a 1.31a 1.78a 1.19a 1.51a 1.67a 1.48a 1.56a 1.78a 1.75a 1.65a 1.68a 1.45a 1.38

0.40f 0.45f 1.03ef 3.25abcd 3.73ab 2.41cd 3.77ab 3.62abc 2.82bcd 2.31de 2.84bcd 3.33abcd 2.83bcd 4.14a 3.38abcd 2.58bcd 3.66abc 2.72bcd 2.49bcd 2.68bcd 2.72

* ** ** 15.86

Source: Fageria (2011). *,**Significant at the 5% and 1% probability levels, respectively. Means followed by the same letter in the same column are not significant at the 5% probability level by Tukey’s test.

(2009) also reported that rice growth was better in 10 mg L 1 Al than with 0 mg Al L 1 in nutrient solution. Brazilian rice cultivars (lowland as well as upland) are highly tolerant to soil acidity (Fageria et al., 2004). Another possible explanation is that ammonium sulfate has about 24% S, which may improve root growth if the extractable soil S level is lower than 10 mg kg 1. The timing of nitrogen application can also influence the root growth of crop plants (Table 4). The treatment T3, which produced maximum grain yield, also produced minimum root length; treatment T2 which produced minimum grain yield produced maximum root length. There was a

BRS Sertaneja N0

N300

Figure 17 Root growth of cultivar BRS Sertaneja at 0 and 300 mg N kg 1 soil (Fageria, 2011).

BRS Primavera N0 N300

Figure 18 Root growth of cultivar BRS Primavera at 0 and 300 mg N kg 1 soil (Fageria, 2011).

The Role of Mineral Nutrition on Root Growth of Crop Plants

285

BRA052053 N0

N300

Figure 19 Root growth of genotype BRA052053 at 0 and 300 mg N kg 1 soil (Fageria, 2011).

negative association between root length and grain yield (Y ¼ 31.2041 0.2718X, R2 ¼ 0.7396**). Root dry weight was significantly related to grain yield (Y ¼ 7.6345 þ 5.8030X 0.4081X2, R2 ¼ 0.8747**; Fageria, 2011). Figure 24 shows root growth of upland rice under different N timing treatments. Root dry weight was least (treatment T2) when all N fertilizer was applied at planting. Root and shoot yields were better in treatments T3 and T4, when N was applied later in the growth cycle (Fageria, 2011). Better root growth may be responsible for higher absorption of nutrients and water in the T3 and T4 treatments which resulted in higher grain and straw yields. The depth of N placement can influence NO3–N distribution in the soil and root growth of wheat (Sharma and Chaudhary, 1984). Root length density decreased abruptly below the 15 cm depth when N was surface applied, whereas root length density decreased more gradually below 15 cm when N was placed at the 10 cm depth. Drew (1975) reported that barley root weight increased in the zone of nutrient localization and decreased in the deficient zone. Murphy and Zaurov (1994) reported that N fertilization at the 5, 10, and 15 cm soil depths produced greater root mass than N fertilization at 0 cm soil depth or surface fertilization.

286

N. K. Fageria and A. Moreira

BRA052045 N0

N300

Figure 20 Root growth of genotype BRA0522045 at 0 and 300 mg N kg 1 soil (Fageria, 2011).

Root dry weight (g plant−1)

5 (NH4)2SO4 (Y ) = 1.9132EXP(0.00093X + 0.000031X

4

2)

R 2 = 0.5296** 3

2 CO(NH2)2 (Y ) = 1.1194+0.0135X − 0.000024X 2 R 2 = 0.4122**

1

0

50

100

150

200

250

300

350

400

Nitrogen application rate (mg kg−1)

Figure 21 Relationship between nitrogen application rate by ammonium sulfate and urea and root dry weight of upland rice (Fageria et al. 2011).

287

The Role of Mineral Nutrition on Root Growth of Crop Plants

CO(NH2)2 0N

50 N

300 N

Figure 22 Root growth of upland rice at 0 (left), 50 (center), and 300 (right) mg N kg 1 soil supplied with urea.

The effects of two N rates on root length and root dry weight of six tropical legume cover crops are showed in Table 5. Root length as well as root dry weight were significantly increased with the addition of N fertilizer. Similarly, root length as well as root dry weight were significantly different among cover crop species. Root growth of four legume cover crops is shown in Figs. 25–28. In all cases, root growth was better at the higher N rate compared to low N rate.

12.2. Phosphorus Phosphorus is one of the most yield-limiting nutrients in tropical highly weathered soils (Fageria, 2009; Fageria and Baligar, 2003, 2008). The deficiency of P in these soils may be related to the low natural level of this element in these soils as well as to the immobilization of P in these soils (Fageria and Baligar, 2008). Phosphate plays many roles in the physiology and biochemistry of pants. It is a component of important compounds like

288

N. K. Fageria and A. Moreira

(NH4)2SO4 0N

300 N

400 N

Figure 23 Root growth of upland rice at 0 (left), 300 (center), and 400 (right) mg N kg 1 soil with ammonium sulfate. Table 4 Root length and root dry weight of upland rice as influenced by nitrogen timing treatments Nitrogen timing treatmenta

Root length (cm)

Root dry weight (g plant 1)

T1 T2 T3 T4 T5 Average F-test CV(%)

28.75b 35.00a 25.75b 27.00b 28.25b 28.95 ** 8.7

4.21bc 2.56c 5.74ab 7.70a 4.98b 5.04 ** 20.9

Source: Fageria (2011). ** Significant at the 1% probability level. Means followed by the same letter in the same column are not significant at the 5% probability level by the Tukey’s test. a T1 (1/2 N applied at sowing þ 1/2 applied at panicle initiation), T2 (total N applied at sowing), T3 (1/3 N applied at sowing þ 1/3 N applied at active tillering þ 1/3 N applied at the panicle initiation), T4 (1/2 N applied at initiation of tillering þ 1/2 N applied panicle initiation), and T5 (2/3 N applied at sowing þ 1/3 N applied at panicle initiation).

289

The Role of Mineral Nutrition on Root Growth of Crop Plants

T1

T3

T2

T5

T4

Figure 24 Root growth of upland rice at harvest under N timing treatments. T1 (1/2 N applied at sowing þ 1/2 applied at panicle initiation), T2 (total N applied at sowing), T3 (1/3 N applied at sowing þ 1/3 N applied at active tillering þ 1/3 N applied at the panicle initiation), T4 (1/2 N applied at initiation of tillering þ 1/2 N applied panicle initiation), and T5 (2/3 N applied at sowing þ 1/3 N applied at panicle initiation). Source: Fageria (2011). Table 5 Influence of nitrogen rate on root length and root dry weight of six tropical legume cover crops Root length (cm)

Root dry weight (g plant 1)

Cover crops

0 mg N kg 1 100 mg N kg 1 0 mg N kg 1 100 mg N kg 1

Showy crotalaria Calopogonio Pueraria Pigeon pea Lablab Gray mucuna bean Average F-test N rate (N) Cover crops (C) NC CV(%)

22

29

0.48

0.77

28 23 30 27 28

34 24 34 32 31

0.30 0.16 0.73 0.74 2.01

0.46 0.15 0.48 1.72 1.84

26b

31a

0.74b

0.90a

** **

** **

NS 6.47

** 16.27

**, NS Significant at the 1% probability level and nonsignificant, respectively. Means in the same line for each growth parameter, followed by the same letter are not significantly different by Tukey’s test at the 5% probability level.

Black mucuna bean 100 N

0N

Figure 25 Root growth of black mucuna bean tropical legume cover crop at 0 (left) and 100 (right) mg N kg 1 soil.

0N

Gray mucuna bean 100 N

Figure 26 Root growth of gray mucuna bean tropical legume cover crop at 0 (left) and 100 (right) mg N kg 1 soil.

The Role of Mineral Nutrition on Root Growth of Crop Plants

291

Lablab 0N

100 N

Figure 27 Root growth of Lablab tropical cover crops at two N levels: 0 (left) and 100 (right) mg N kg 1 soil.

phospholipids, phosphorylated sugars and proteins, DNA (deoxyribonucleic acid), and RNA (ribonucleic acid). It is also a component of ATP (adenosine 5-triphosphate), PEP (phosphoenolpyruvate), NADPH (nicotinamide adenine dinucleotide phosphate, reduced), and other biochemicals that use the phosphate bond in energy utilization and storage (Blevins, 1994). Phosphorus is a key nutrient essential for root development in highly weathered tropical soils. Baligar et al. (1998) reported that P increased the root weight of wheat, dry bean, and cowpea in a quadratic fashion with increasing P rate from 0 to 200 mg kg 1 of soil. The regression equations related to P rates versus root dry weight were Y ¼ 0.4019 þ 0.094X 0.00031X2, R2 ¼ 0.74* for wheat, Y ¼ 0.4813Exp. (0.019X 0.000071 X2), R2 ¼ 0.63* for dry bean, and Y ¼ 0.7351 þ 0.0232X 0.000073 X2, R2 ¼ 0.80** for cowpea. Based on these regression equations, maximum root dry weight for wheat was achieved at 152 mg P kg 1, whereas maximum root dry weight for common bean and cowpea was achieved at 134 and 159 mg P kg 1 soil, respectively. These results indicate that increasing P levels increased root growth, but root growth was reduced at

292

N. K. Fageria and A. Moreira

Crotalaria spectabilis 0N

100 N

Figure 28 Root growth of Crotalaria spectabilis tropical legume cover crops at two N levels: 0 (left) and 100 (right) mg N kg 1 of soil.

higher P levels, and the crops had different P requirements to achieve maximum growth. Overall, the root growth of cereals and legume crops was reduced if P was deficient. Most studies indicate that, within certain limits, both root and shoot growth vary similarly as P level increases. Above certain levels, further increases in P supply do not affect root or shoot growth (Troughton, 1962). Fageria et al. (2006) reported that root dry weight was reduced 62% in rice, 74% in common bean, 50% in corn, and 21% in soybean without added soil P, compared to adequate P in a Brazilian Oxisol. Fageria et al. (2011) studied the influence of phosphorus on root dry weight and root length of 20 upland rice genotypes grown on a Brazilian Oxisol. Phosphorus level and genotype interactions for root dry weight and root length were significant, indicating different responses of genotypes to varying P levels (Table 6). Root dry weight of 20 upland rice genotypes at low P level varied from 2.00 to 5.68 g plant 1, with an average value of 3.41 g plant 1. At the high P level, root dry weight varied from 2.43 to 8.55 g plant 1, with an average value of 4.01 g plant 1. However, the effect

293

The Role of Mineral Nutrition on Root Growth of Crop Plants

Table 6 Root dry weight and root length of 20 upland rice genotypes as influenced by P levels Root dry weight (g plant 1) Genotype

BRA01506 BRA01596 BRA01600 BRA02535 BRA02601 BRA032033 BRA032039 BRA032048 BRA032051 BRA042094 BRA042156 BRA042160 BRA052015 BRA052023 BRA052033 BRA052034 BRA052045 BRA052053 BRS Primavera BRS Sertaneja Average F-test P level (P) Genotype (G) PG

Root length (cm)

Low P (25 mg kg 1)

High P (200 mg kg 1)

Low P (25 mg kg 1)

High P (200 mg kg 1)

3.92ab 2.78ab 2.81ab 3.12ab 4.42ab 3.70ab 2.91ab 3.96ab 2.00b 2.82ab 2.50b 5.68a 3.91ab 4.69ab 2.23b 3.18ab 3.07ab 2.57ab 3.56ab

3.22c 2.73c 3.03c 4.30c 3.20c 3.62c 4.36c 3.91c 2.58c 3.92c 2.91c 8.32ab 2.98c 8.55a 2.43c 3.99c 3.08c 3.87c 5.21bc

26.00ab 35.67ab 36.00ab 28.67ab 31.33ab 23.00b 27.67ab 37.00a 36.00ab 30.00ab 29.00ab 32.67ab 27.00ab 29.67ab 27.00ab 31.00ab 38.33a 28.33ab 29.67ab

26.67a 28.00a 29.00a 33.33a 27.00a 29.67a 27.00a 33.33a 30.67a 27.33a 27.00a 33.00a 27.00a 34.33a 29.00a 24.67a 24.67a 23.67a 25.00a

4.36ab

3.92c

34.00ab

23.67a

3.41

4.01

30.9

28.20

NS **

* **

*

**

Source: Fageria et al. (2011). *,**, NS Significant at the 5% and 1% probability level and nonsignificant, respectively. Means in the same column followed by the same letter are not significantly different at the 5% probability level by the Tukey’s test.

of P level on root system dry weight was not significant. Root length varied from 23.00 to 38.33 cm with an average value of 30.9 cm at low P level. At high P level, root length varied from 23.67 to 34.33 cm, with an average value of 28.20 cm. There was a significant 10% decrease in root length at the high P level compared to the low P level. However, higher P level roots had more fine hairs compared to lower P level (personal observations).

294

N. K. Fageria and A. Moreira

Figures 29–31 show root growth of upland rice under different P rates. Unlike the data cited above, root growth of all the cultivars/genotypes increased with increasing P levels. The first author studied the influence of P levels on the root growth of 14 tropical legume cover crops grown on a Brazilian Oxisol. Root dry weight was significantly influenced by P; crop species and P crop species interactions were significant (Fig. 32). The significant P crop species interaction indicates significant variation in shoot dry weight with the variation in P levels. At the lowest P level (0 mg kg 1), maximum root dry weight of 0.77 g plant 1 was produced by white jack bean (Canavalia ensiformis) and minimum root dry weight of 0.01 g plant 1 was produced by crotalaria (Crotalaria mucronata) and pueraria (Pueraria phaseoloides). At the medium P level (100 mg P kg 1), maximum root dry weight of 1.91 g plant 1 was produced by black mucuna bean (Mucuna cinereum) and minimum root dry weight of 0.07 was produced by crotalaria (Crotalaria

0P

50 P

175 P

Figure 29 Root growth of upland rice under 0 (left), 50 (center), and 175 (right) mg P kg 1 soil (Fageria and Stone, 1999).

0P

25 P

50 P

75 P 100 P 200 P

IAC 164

Figure 30 Root growth of upland rice cultivar IAC 164 at different P levels.

The Role of Mineral Nutrition on Root Growth of Crop Plants

295

BRA 01596

25 P 200 P

Figure 31

Root growth of upland rice genotype BRA01596 at two P levels.

breviflora), with an average value of 0.63 g plant 1. At the highest P level (200 mg P), maximum root dry weight of 1.42 g plant 1 was produced by gray mucuna bean (Mucuna cinereum) and minimum root dry weight of 0.09 g plant 1 was produced by calopogonio (Calopogonium mucunoides) and pueraria (Pueraria phaseoloides). The variation in root dry weight is genetically controlled and also influenced by environmental variables, like the supply of mineral nutrition (Baligar et al., 2001; Caradus, 1990; Fageria et al., 2006). Maximum root length of tropical legume cover crops varied from 15.5 to 36 cm at the low P level, from 20.5 to 50.33 cm at the medium P level and 18.33 to 52.33 cm at the high P level (Fig. 33). Overall, root length also increased with increasing P level. Figures 34–36 show root growth of tropical legume cover crops as influenced by P levels. The improvement in root length by improved P nutrition has been reported by Fageria (2009) in various crop species. Barber (1995), Fageria et al. (2006), Marschner (1995), and Mengel et al. (2001) reported that mineral nutrition has tremendous effects on root growth, development, and function and, subsequently, the ability of roots to absorb and translocate nutrients. These authors further reported that mineral deficiencies induce considerable variations in the growth and morphology of roots and such variations are strongly influenced by plant species and genotypes.

296

N. K. Fageria and A. Moreira

1.6 a

200 mg P kg−1

1.4

a

1.2 b

1.0 0.8

c

0.6

c

0.4 0.2

b b

de

de

d de

e

e

de

−1

Root dry weight (g plant )

0 2.0

a 100 mg P kg−1

1.8 1.6 1.4

b

1.2 1.0

bc

bc

bcd

0.8 cde

0.6 ef

0.4 0.2

b

ef

ef

f

f

df

f

0 0.8

a 0 mg P

0.6

kg−1

b

b

0.4 c cd

Figure 32

de

Canavalia ensiformis

Mucuna cinereum

Mucuna aterrima

e Mucuna deeringiana

Crotalaria ochroleuca Calopogônium mucunoides

de

cde

Dolichos lablab

e

Cajanus cajan (black) Cajanus cajan (mixed color)

de

Pueraria phaseoloides

e

Crotalaria spectablis

Crotalaria juncea

0

Crotalaria breviflora

de

cd

Crotalaria mucronata

0.2

Root dry weight of 14 tropical legume cover crops as influenced by P rates.

The influence of P fertilization on dry bean root dry weight was studied by Fageria (1989b). Root dry weight of three bean genotypes increased significantly in a quadratic fashion but differed from genotype to genotype (Table 7). The variability in root dry weight due to P fertilization was about

297

The Role of Mineral Nutrition on Root Growth of Crop Plants

70 200 mg P kg−1

60

a ab

ab

b

ab

50 40 30

c cd

20

cd d

c cd

cd

c

cd

10

Maximum root length (cm)

0 60 100 mg P kg−1

50

bc

40 cde cde

30 20

a

a

def def f

def

cde

ef

b

bcd def

10 0 40

a

0 mg P kg−1

ab

ab

de

de cde

cd

cd

Cajanus cajan (mixed color)

cde 20

cd

Cajanus cajan (black)

30

cd

bc

de

e

Canavalia ensiformis

Mucuna cinereum

Mucuna aterrima

Mucuna deeringiana

Dolichos lablab

Pueraria phaseoloides

Calopogonium mucunoides

Crotalaria ochroleuca

Crotalaria spectablis

Crotalaria mucronata

Crotalaria juncea

0

Crotalaria breviflora

10

Figure 33 Maximum root length of 14 tropical legume cover crops at three P levels.

52% in genotype Carioca, 35% in genotype CNF10, and 70% in genotype CNF4856. Such information may contribute to the selection of cultivars specific to soil type and management systems, resulting in increasing yields on the soils of variable P fertility (Fageria, 1989b).

Crotalaria mucronata 0P

200 P

Figure 34 Root growth of tropical legume cover crop Crotalaria mucronata at 0 and 200 mg P kg 1 of soil.

Gray mucuna bean 0P

200 P

Figure 35 Root growth of tropical legume cover crop gray mucuna bean at 0 (left) and 200 (right) mg P kg 1 of soil.

The Role of Mineral Nutrition on Root Growth of Crop Plants

299

Crotalaria breviflora 0P

200 P

Figure 36 Root growth of tropical legume cover crop Crotalaria breviflora at 0 and 200 mg P kg 1 of soil.

12.3. Potassium Potassium plays an important role in the growth and development of plants, including the root system. Many enzymes are activated in plants by potassium, and it is also required for photosynthesis, transport of photosynthate, and protein synthesis (Blevins, 1994). Potassium plays a role in cell growth following cell division by serving as a major component in cell turgor. It also maintains ionic balance and electrical neutrality in plants. Crops that produce large quantities of protein per unit area of land require more K than those that produce less protein (Blevins, 1994). Potassium also plays an important role in opening and closing the stomata. Tennant (1976) reported that potassium deficiency stops root growth completely within 10–12 days of planting in wheat. Inadequate K reduces root growth and consequently, crop yields (Baligar et al., 1998). A deficiency of K in an Inceptisol reduced root growth by 23% in lowland rice, by 30% in dry bean, by 12% in corn, and by 11% in soybean (Baligar et al.,

300

N. K. Fageria and A. Moreira

Table 7 Influence of phosphorus fertilization on the root growth of three dry bean genotypes at harvest Root dry weight (g per 2 plants) 1

P rate (mg kg

)

Carioca

CNF10

CNF4856

0 0.29 0.45 0.23 25 1.13 1.79 1.38 50 1.66 1.28 1.83 75 1.31 1.29 1.52 100 1.34 1.60 1.59 125 1.62 1.33 1.78 150 1.31 1.75 1.82 175 1.18 1.29 1.48 200 1.79 1.50 1.67 Regression analysis P rate versus Carioca (Y) ¼ 0.6459 þ 0.0128X 0.000044X2, R2 ¼ 0.5179** P rate versus CNF10 (Y) ¼ 0.8704 þ 0.0109X 0.000042X2, R2 ¼ 0.3509* P rate versus CNF4856 (Y) ¼ 0.6145 þ 0.0193X 0.000075X2, R2 ¼ 0.6953** Source: Adapted from Fageria (1989b). *,** Significant at the 5% and 1% probability levels, respectively.

1998). On an Oxisol, a 35% lower root dry mass was observed in 13 corn genotypes when K levels were 0 versus 200 mg kg 1 of soil (Baligar et al., 1998). Data in Table 8 show that at 0 mg K kg 1 of soil root dry weight of common bean (P. vulgaris L.) genotypes varied from 1.54 to 3.14 g per 3 plants, a variation of twofold. At the 200 mg K kg 1 level, root dry weight varied from 1.50 to 2.30 g per 3 plants, a variation of 1.5-fold. Similarly, maximum root length varied from 42 to 46 cm at low K level and 32–44 cm at higher K level. At the higher K level, there was a slight decrease in the root length of all the genotypes, and the root weight of three genotypes also decreased at the higher K level. However, at the higher K level, there were more root hairs than at the low K level (visual observations). There is widespread evidence for genotype diversity in the root characteristics of many crops in response to the environment and increasing interest in using this diversity to improve agricultural production and consequently, nutrient-use efficiency (Barber, 1994; Gregory, 1994). Mullins et al. (1994) studied K placement effects on the root growth of cotton grown on a fine sandy loam soil. Root density measurements taken in-row showed that root growth at depths >20 cm was improved with inrow subsoil additions of K. Tupper (1992) also observed increased cotton taproot length when K fertilizer was band-applied in the subsoils of

301

The Role of Mineral Nutrition on Root Growth of Crop Plants

Table 8 Root dry weight and maximum root length of six common bean genotypes as influenced by potassium levels applied to a Brazilian Oxisol 0 mg K kg 1

200 mg K kg 1

Maximum Root dry weight root length Genotypes (g per 3 plants) (cm)

Maximum Root dry weight root length (g per 3 plants) (cm)

Apore Perola Ruda IAC Carioca Jalo Precoce Safira Average

1.54 1.97 1.94 3.14

45 42 44 45

1.67 2.04 2.30 1.70

32 39 35 38

2.24

42

1.67

36

1.77 2.10

46 44

1.50 1.81

44 37

Source: Fageria et al. (2008a). Values were determined at physiological maturity.

Mississippi soils with low soil-test K. On the other hand, Hallmark and Barber (1984) and Yibrin et al. (1993) reported that localized applications of K did not promote root growth. However, K has been shown to promote the root growth of some vegetable crops (Zhao et al., 1991). Fageria (1992) determined the root growth of rice grown in nutrient solution as well as in an Oxisols at the stress and nonstress levels of K. At the stress levels of K, rice root growth was reduced compared with nonstress levels. Alfalfa herbage yield, root weight, and root total nonstructural carbohydrates increased with increasing K fertilizer (Kitchen et al., 1990).

12.4. Calcium Absolute Ca deficiency is difficult to identify on plants grown in acidic soils (Kamprath and Foy, 1985). Most acidic soils contain adequate total Ca for most plants, and Ca-deficiency symptoms are rarely observed in the field. Only in highly leached, acidic, low-cation exchange soils (Oxisols and Ultisols) would absolute deficiencies be likely to occur (Garrity et al., 1983). Levels of Ca required for essential growth functions are so low as to approach those of micronutrients. Hence, the major role of Ca in soils and in plants is to exclude or detoxify other elements such as Al, Mn, and heavy metals that might otherwise become toxic (Garrity et al., 1983). Gonzalez-Erico et al. (1979) evaluated the maize response to deep incorporation of limestone on an Oxisol. They reported that incorporation

302

N. K. Fageria and A. Moreira

of limestone to depths of 30 cm improved root growth and increased water utilization and grain yield of maize. Similar results were obtained for maize and cotton when limestone was incorporated to depths up to 45 cm (Doss et al., 1979). Suitable diagnostic indices for prediction of Ca limitations on root growth are either Ca saturation of the effective CEC or Ca activity ratio of the soil solution, which has been defined as the ratio of Ca activity to the sum of the activities of Ca, Mg, K, and Na (Bruce et al., 1988). Values corresponding to 90% relative root length (RRL) of soybean were 0.05 for the Ca activity ratio and 11% for Ca saturation. Calcium activity and Ca concentrations in soil solutions and exchangeable Ca were less useful for diagnostic indices (Bruce et al., 1988). The root growth of soybean was improved with the addition of 12 Mg lime per hectare compared to control treatment (Fig. 37).

12.5. Magnesium Magnesium is an essential macronutrient for plant growth. The most wellknown and important role of Mg is its occurrence in chlorophyll molecules. In addition to this, Mg is required for many essential physiological reactions, especially phosphorylation reactions (Mengel and Kirkby, 1978). Fageria and Souza (1991) determined the effects of Mg levels on root weights of rice, common bean, and cowpea grown in an Oxisols of Central Brazil (Fig. 38). Dry weights of rice roots were higher at the lowest Mg concentration compared with the highest soil Mg concentration. Initial exchangeable Mg levels of surface soils were 0.1 cmol kg 1. They increased to 0.3 cmol kg 1 within 3 days after liming and to 0.75 cmol kg 1 at harvest time (33 days after sowing). The lack of growth responses to applications of Mg indicated that this level of exchangeable Mg was adequate to meet Mg

0 Mg lime ha−1

Figure 37

12 Mg lime ha−1

Root growth of soybean at two lime rates.

303

The Role of Mineral Nutrition on Root Growth of Crop Plants

3 Upland rice 2

Y = 1.0752 − 0.08852X R 2 = 0.85*

1

Root dry weight (g/3 plants)

0 Common bean

4 3 2

2 Y = 2.4497exp(0.3009X − 0.05047X ) 2 R = 0.88*

1 0

Cowpea

4 3 2

Y = 2.6658exp(0.3007X − 0.05983X 2) R 2 = 0.80**

1 0 0

1

2

3

4

5

6

7

Extractable Mg in soil (cmolc kg−1)

Figure 38 Root dry weights of upland rice, common bean, and cowpea grown with different Mg levels in an Oxisol (adapted from Fageria and Souza, 1991).

requirements of upland rice grown in this limed soil. Dry weight of roots of common bean increased with Mg application up to 3 cmol kg 1 of soil. Similarly, significant responses of cowpea root growth to soil Mg levels were observed, and maximum root weight was achieved at 2.5 cmol Mg kg 1 of soil.

12.6. Sulfur Sulfur has long been recognized as an essential element for plant growth and development and classified as a macronutrient. Crop responses to applied sulfur have been reported in a wide range of soils in many parts of the world (Fageria, 2009). Sulfur plays many important roles in the growth and development of plants. Fageria and Gheyi (1999) summarized important functions of the sulfur in the plant. It is an important component of two amino acids, cysteine and methionine, which are essential for protein formation. Since animals cannot reduce sulfate, plants play a vital role in

304

N. K. Fageria and A. Moreira

supplying essential S-containing amino acids to them. Sulfur plays an important role in enzyme activation. It promotes nodule formation in legumes. Sulfur is necessary in chlorophyll formation, although it is not a constituent of chlorophyll. Maturity of seeds and fruits is delayed in the absence of adequate sulfur. Sulfur is required by the plants in the formations of nitrogenase. It increases the crude protein content of forages and improves the quality of cereals for milling and baking. Sulfur increases the oil content of oilseed crops and increases winter hardiness in plants. It increases drought tolerance in plants, controls certain soil-borne diseases, and helps in the formation of glucosides that give characteristic odors and flavors to onion, garlic, and mustard. Sulfur is necessary for the formation of vitamins and synthesis of some hormones and glutathione, and it is involved in oxidation–reduction reactions. Sulfur improves tolerance to heavy metal toxicity in plants, and it is a component of sulfur contain sulfolipids. Organic sulfates may serve to enhance the water solubility of organic compounds, which may be important in dealing with salinity stress, and fertilization with sulfate decreases fungal diseases in many crops. Few studies have assessed the impacts of sulfur on root growth and function; however, the effects of sulfur on root growth may be similar to those of N. Zhao et al. (2008) reported that S application increased the root number and root dry weight of soybean compared to control treatment.

12.7. Micronutrients Micronutrients have also been called minor or trace elements, indicating that their required concentrations in plant tissues are small compared to the macronutrients (Fageria et al., 2002; Mortvedt, 2000). Based on physicochemical properties, except B and Cl, the essential micronutrients are metals. Even though micronutrients are required in small quantities by field crops, their influence can be as great as that of macronutrients in crop production. Micronutrients are normally constituents of prosthetic groups that catalyze redox processes by electron transfer (such as with the transition elements Cu, Fe, Mn, and Mo) and form enzyme–substrate complexes by coupling enzymes with substrates (Fe and Zn) or enhance enzyme reactions by influencing molecular configurations between enzyme and substrate (Zn) (Fageria et al., 2002). Micronutrient deficiencies in crop plants are widespread because of (i) increased micronutrient demands from intensive cropping practices and adaptation of high-yielding cultivars which may have higher micronutrient demand, (ii) enhanced production of crops on marginal soils that contain low levels of essential micronutrients, (iii) increased use of high-analysis fertilizers with low amounts of micronutrients, (iv) decreased use of animal manures, composts, and crop residues, (v) use of many soils that are inherently low in micronutrient reserves, (vi) use of liming in acid soils, and (vi)

The Role of Mineral Nutrition on Root Growth of Crop Plants

305

involvement of natural and anthropogenic factors that limit adequate supplies and create elemental imbalance (Fageria et al., 2002). 12.7.1. Zinc Deficiency of Zn in crop production is spread worldwide (Alloway, 2008). Graham (2008) reported that half of the world’s soils are intrinsically deficient in Zn. Zinc deficiency in annual crops is reported in Brazil (Fageria and Stone, 2008), Australia (Graham, 2008), India (Singh, 2008), China (Zou et al., 2008), Turkey (Cakmak, 2008), Europe (Sinclair and Edwards, 2008), USA (Brown, 2008), and Africa (Waals and Laker, 2008). Micronutrient deficiencies are also a worldwide problem in human health (Welch, 2008). Zinc deficiency is the highest priority among micronutrients for agriculture to address (Graham, 2008). In the Brazilian Cerrado region, Zn deficiency is very common in annual crops, especially upland rice and corn (Fageria, 2009). Figures 39 and 40 show Zn-deficiency symptoms in upland rice and corn grown on Brazilian Oxisols. Data in Table 9 show that Zn application of up to 120 mg kg 1 improved the root growth of upland rice and wheat significantly. Similarly, Figs. 41 and 42 show improvement in the root growth of soybean and dry dean, respectively, with the addition of Zn in Brazilian Oxisols. 12.7.2. Boron Boron deficiency is common for plants grown in arid, semiarid, and heavy rainfall areas in calcareous, sandy, light textured, acid, and low-OM soils (Gupta, 1993). Differences between B sufficiency and toxicity are narrow

Figure 39

Zinc deficiency in upland rice grown on Brazilian Oxisol.

306

N. K. Fageria and A. Moreira

Figure 40

Zinc deficiency in corn grown on Brazilian Oxisol.

Table 9 Root dry weight of upland rice and root length of corn as influenced by zinc fertilization Zn rate (mg kg 1)

Upland rice root dry weight (g per 4 plants)a

Wheat maximum root length (cm)a

0 5 10 20 40 80 120 R2

1.08 1.33 1.48 1.55 1.73 1.45 1.18 0.75**

55 54 52 55 57 58 52 0.24*

Source: Adapted from Fageria (2002a). *,** Significant at the 5% and 1% probability levels, respectively. a Upland rice plants were harvested 6 weeks after sowing and corn plants were harvested 4 weeks after sowing.

(Marschner, 1995), and soils supplied with high amounts of municipal compost, sludge, and biosolids tend to accumulate high amounts of B, which may result in B toxicity. Boron is essential for pollen germination and pollen tube growth in crop plants (Blevins, 1994). Boron requirements

The Role of Mineral Nutrition on Root Growth of Crop Plants

0

5

10

40

307

80 mg Zn kg−1

Soja

Figure 41 Root growth of soybean at different Zn levels.

0

120

5 mg Zn kg−1

Dry bean

Figure 42

Root growth of dry bean at three Zn levels.

of dicots are generally higher than monocots (Fageria, 2000). Fageria (2000) reported that maximum root growth of upland rice can be achieved with the application of 0.4 mg B kg 1 soil, whereas maximum root growth of dry bean required 1.9 mg B kg 1 of soil. Figure 43 shows that B requirements for root growth varied among the crop species. Application of 24 mg B kg 1 soil decreased the root dry weight of upland rice and corn in greenhouse studies. However, application of B at lower rates to the same soil increased the root dry weight in dry bean, soybean, and wheat (Fig. 43).

308

N. K. Fageria and A. Moreira

0.3

Upland rice

0.2 0.1

Y = 0.2681−0.00806X R 2 = 0.83*

0.0

Corn 2

Root dry weight (g/4 plants)

1 Y = 2.0432exp (−0.05957X+0.00105X 2) R 2 = 0.93**

0

Common bean Y = 1.3058exp (0.03466X−0.00961X 2) R 2 = 0.98**

2 1 0

Soybean

0.6

0.3

Y = 0.7473exp (0.0991X−0.00652X 2) R 2 = 0.91*

0.0 Wheat

0.3 0.2

Y = 0.2793exp (0.01993X−0.00328X 2) R 2 = 0.90*

0.1 0.0

0

5

10 15 20 Boron applied (mg kg−1)

25

Figure 43 Root dry weights of upland rice, maize, common bean, soybean, and wheat grown with different B levels on an Oxisols of central Brazil (adapted from Fageria, 2000).

309

The Role of Mineral Nutrition on Root Growth of Crop Plants

12.7.3. Copper Copper deficiency is often observed on plants grown in soils inherently low in Cu (coarse-textured and calcareous soils) and in soils high in OM, where Cu complexes with organic substances (Fageria et al., 2002). Higher-thannormal Cu supplies usually inhibit root growth more than shoot growth (Lexmond and Vorm, 1981). Use of Cu-containing fungicides and anthelmintic compounds in agriculture has resulted in Cu toxicity in some plants, but naturally occurring Cu toxicity is relatively uncommon (Welch et al., 1991). Root dry weight of wheat and root length of dry bean were significantly increased by the application of copper fertilization (Table 10). 12.7.4. Iron Iron deficiency is a worldwide problem and occurs in numerous crops (Fageria et al., 2002; Marschner, 1995). Iron deficiency occurs not because of Fe scarcity in soil or plants, but because various soil and plant factors affect Fe availability, inhibit its absorption, or impair its metabolic use (Marschner, 1995; Welch et al., 1991). Plant species that commonly become Fe-deficient are peanut (A. hypogaea L.), soybean (G. max L.), sorghum (Sorghum bicolor L. Moench), and upland rice (O. sativa L.). Iron deficiency reduces root growth (Table 11). Iron toxicity (indicated by leaf bronzing) can be serious for production of crops in waterlogged soils. For wetland rice, Fe toxicity is the second most severe yield-limiting nutrient disorder, and it has been reported in South America, Asia, and Africa (Fageria et al., 2008b). Iron toxicity decreases the root growth of lowland rice (Fageria et al., 2008b); however, genotypic differences exist (Fig. 44).

Table 10 Root dry weight of wheat and root length of dry bean as influenced by copper fertilization Cu rate (mg kg 1) Wheat (g per 4 plants)a Dry bean (cm)a

0 2 4 8 16 32 64 96 R2

0.53 0.60 0.50 0.48 0.47 0.47 0.43 0.17 0.88**

25 30 27 28 28 24 30 14 0.42**

Source: Adapted from Fageria (2002a). ** Significant at the 1% probability levels. a Wheat and dry bean plants were harvested 5 weeks after sowing.

310

N. K. Fageria and A. Moreira

Table 11 Root dry weight of lowland rice as influenced by iron concentration in nutrient solution Root dry weight (g plant 1) Fe concentration (mg L 1)

20 days age

60 days age

0.0 2.5 5.0 10.0 20.0 40.0 80.0 160.0 Average

0.03 0.09 0.13 0.10 0.08 0.11 0.07 0.02 0.08

0.64 1.19 1.05 1.07 1.18 1.24 0.84 0.12 0.92

Source: Fageria et al. (1981).

CNA810208

CNA808951

Fe100 mg L−1

Figure 44 Root growth development of two lowland rice genotypes at 100 mg Fe L 1 in nutrient solution. Source: Fageria et al. (2008a,b).

The Role of Mineral Nutrition on Root Growth of Crop Plants

311

12.7.5. Other micronutrients In addition to zinc, boron, copper, and iron, the other micronutrients that are essential for the growth of higher plants are manganese, molybdenum, chlorine, and nickel. Manganese deficiency has been reported for plants grown in the coarse-textured and poorly drained coastal plain soils of the United States (Reuter et al., 1988) and in the soils of Central America, Bolivia, and Brazil (Leon et al., 1985). In Europe, Mn deficiency has been reported on plants grown in peaty (England and Denmark), coarse-textured (Sweden and Denmark), coarse/fine textured (Netherlands), podzolic, and brown forest (Scotland) soils (Welch et al., 1991). Manganese deficiency has also been reported on plants grown in the semiarid regions of China, India, southeast and western Australia, Congo, Ivory Coast, Nigeria, and other western African countries (Fageria et al., 2002). Manganese toxicity on crop plants grown in many parts of the world has been reported to be more important than Mn deficiency (Foy, 1984; Welch et al., 1991). Molybdenum is the least abundant of the micronutrients in the lithosphere (Mortvedt, 2000), and soil concentrations range from 0.2 to 5 mg kg 1 (mean of 2 mg kg 1). Mo deficiency usually occurs on plants grown in the broad areas of well-drained acid soils and in soils formed from parent materials low in Mo. In Australia, Mo deficiency occurs on crops grown in soils derived from sedimentary rocks, basalts, and granites (Anderson, 1970). Peaty, alkaline, and poorly drained soils commonly have high Mo. Iron oxides adsorb more Mo than Al oxides (Fageria et al., 2002), and clay mineralogy can affect Mo adsorption, in the order montmorillonite > illite > kaolinite (Goldberg, 1993). Hydrous ferric oxides or ferric oxide molybdate complexes and insoluble ferric molybdates may form in well-aerated soils so that Mo solubility and availability to plants is low (Welch et al., 1991). In poorly drained soils, the formation of soluble ferrous molybdates or molybdites may lead to high Mo availability to plants. Plants grown in high Mo soils of the intermountain valleys of western United States have been reported to accumulate high Mo which has induced “molybdenosis” (Cu deficiency) in cattle (Welch et al., 1991). Chloride is essential to higher plants and is required for the watersplitting reactions in photosystem II (Kelley and Izawa, 1978). Nickel has been shown to be essential for soybean (Eskew et al., 1983), and it is known to be a constituent of urease (Blevins, 1994; Klucas et al., 1983). Since urea is a widely used fertilizer in crop plants worldwide, Ni nutrition could be important (Blevins, 1994). Information on the influence of these micronutrients on the root growth of crop plants is not available, and this aspect is not discussed here.

312

N. K. Fageria and A. Moreira

13. Management Strategies for Maximizing Root Systems Root growth of crop plants can be improved by adopting management practices that modify soils to fit crops, as well as to modify the plants to fit the soil. Both of these strategies can be combined to get maximum economic results.

13.1. Soil management Soil management practices that can improve root growth of crop plants include liming acid soils, use of gypsum, maintenance of organic matter, use of adequate rates, sources and methods of fertilizer application, and deep plowing. 13.1.1. Liming acid soils Liming is the most common and effective practice to reduce soil acidity. Lime requirements of crops grown on acid soils are determined by the quality of liming material, status of soil fertility, crop species and cultivar within species, crop management practices, and economic considerations. Soil pH, base saturation, and aluminum saturation are important acidity indices that are used to determine liming. Liming improves soil pH, Ca, and Mg contents and reduces Al concentrations in the soil solution. In addition, liming improves beneficial microbe populations in the soil. Furthermore, liming improves P concentration in the soil solution by reducing P immobilization by Fe and Al in acid soils (Fageria and Baligar, 2008). All these beneficial effects of liming improve the root growth of crop plants. Nurlaeny et al. (1996) found that liming increased shoot dry weight, total root length, and mycorrhizal colonization of roots in soybean and corn grown on tropical acid soils. Gonzalez (1976) reported that incorporation of lime to a 30 cm soil depth allowed the corn roots to penetrate and use stored water throughout the lime layer. Data in Table 12 show that liming increases soil pH and consequently, the root dry weight of dry bean grown on a Brazilian Oxisol. Similarly, Figs. 45 and 46 show how the root growth of dry bean and soybean is influenced by soil pH. The root growth of both the legumes was significantly influenced by increasing soil pH. Dry bean root growth was maximum at pH 5.9 and soybean produced vigorous root systems at pH 6.4. 13.1.2. Use of gypsum Gypsum (CaSO4 2H2O) or phosphogypsum (e.g., byproducts of phosphoric acid manufacturing processes) applications are used to leach Ca deeper into soil profiles where Ca can replace Al on cation exchange complexes.

The Role of Mineral Nutrition on Root Growth of Crop Plants

313

Table 12 Influence of soil pH on root growth of dry bean grown on a Brazilian Oxisol Soil pH in H2O

Root dry weight (g per 4 plants)

4.1 4.7 5.3 5.9 6.6 7.0 R2

1.90 3.58 4.67 5.40 4.73 3.80 0.99**

Source: Adapted from Fageria (2002b). ** Significant at the 1% probability level.

pH 4.9 pH 5.9

pH 6.4 pH 6.7 pH 7.0

Dry bean

Lowland soil

Figure 45 Root growth of dry bean grown on a Brazilian lowland soil (Inceptisol; Fageria and Stone, 1999).

Much of the Al displaced by Ca can be leached from the root zones. This practice works well in sandy soils or Oxisols with clay loam aggregates which behave hydrologically like sands (Foy, 1992). Poor root growth of crop plants has been frequently observed in highly weathered acid soils in various countries (Alcordo and Rechcigl, 1993). The chemical factor identified as most responsible for poor root growth is excess soluble Al (Alcordo and Rechcigl, 1993; Foy, 1992). Excess Al3þ has been reported to inhibit root growth by binding to the PO4 portion of DNA in the root cell nuclei, reducing template activity and thus cell division (Matsumoto and Morimura, 1980). In legumes, it has been shown to impair the growth of root hairs and rhizobia, reducing root nodule initiation and function (Munns and Franco, 1982). Excess Al may also adversely affect the root and overall plant growth in nonphytotoxic ways by competing with Ca and

314

N. K. Fageria and A. Moreira

Lowland soil pH 4.9

pH 5.9 pH 6.4 Soybean

pH 6.7 pH 7.0

Figure 46 Root growth of soybean in a Brazilian lowland soil (Inceptisol) at different pH.

Mg for uptake by plants (Rengel and Robinson, 1989). The use of gypsum can neutralize subsoil acidity by leaching CaSO4 and forming AlSO4þ which is not toxic for root growth (Alva and Sumner, 1989). Ritchey et al. (1980) reported that application of gypsum increased subsoil Ca and Mg while decreasing Al and improving the root growth of corn in Brazilian Oxisol. 13.1.3. Maintenance of adequate amounts of organic matter The benefits of organic matter addition to soils include improving nutrient cycling and availability to plants through direct additions as well as through modification in soil physical and biological properties. The complementary use of organic manures and chemical fertilizers has proved to be the best soil fertility management strategy in the tropics (Fageria and Baligar, 2005). Enhanced soil organic matter increases soil aggregation and water-holding capacity, provides an additional source of nutrients, and reduces P fixation, toxicities of Al and Mn, and leaching of nutrients (Baligar and Fageria, 1999). Build-up of organic matter through additions of crop and animal residues increases the population and species diversity of microorganisms and their associated enzyme activities and respiration rates (Fageria, 2002c). The use of organic compost may result in a soil that has greater capacity to resist the spread of plant pathogenic organisms. The improvement in the overall soil quality may produce more vigorous root systems and higher crop yields (Fageria, 2002c).

The Role of Mineral Nutrition on Root Growth of Crop Plants

315

13.1.4. Use of adequate nutrient rates, sources, and methods of application Use of adequate rates and effective sources of nutrients are important management practices to improve crop yield and plant root systems. In addition, appropriate methods of fertilizer application are also important for the development of vigorous root systems. Immobile nutrients like P and K should be applied in bands to improve their uptake by plant roots. 13.1.5. Deep plowing Deep plowing improves soil conditions for root growth by breaking compacted layers that roots cannot readily penetrate. If water tables are near the soil surfaces, drainage can also be useful. When depths to root-restricting hardpans are relatively shallow (<0.25 m), chisel plowing can be effective for disrupting compacted layers. 13.1.6. Mulching, greater sowing depth, and sowing larger seeds Root growth may be manipulated through cultural practices like mulching, which can affect soil temperatures. Warmer soil temperatures generally produce larger root systems. In addition, crop residues on soil surfaces decrease soil evaporation and improve water-use efficiency. Greater sowing depths may decrease seedling emergence rate, but it can also increase the survival of emergent seedlings by increasing water availability. Larger seeds generally produce seedlings with more extensive root systems. Sowing good quality seeds could also improve root systems. 13.1.7. Leaching salts from soil profiles Leaching may be required to prevent harmful accumulation of salts in the root zone.. Steady-state leaching (Lr) requirements may be estimated (Reichman and Trooien, 1993) as Lr ¼

Dd ECa ¼ ; Da ECd

where Dd and Da are depths of drainage water and applied water, respectively, and ECa and ECd are electrical conductivity of applied water and drainage water, respectively. 13.1.8. Integrated cropping systems and pest management Reduced tillage, N side-dressing, and early planting can be included in integrated pest management programs with no risk of increasing potential for root damage from western maize rootworm (Roth et al., 1995). Miltner et al. (1991) reported that cyst nematodes suppressed soybean root growth on susceptible cultivars, whereas root growth of tolerant cultivars was

316

N. K. Fageria and A. Moreira

stimulated by the presence of soybean cyst nematodes. The use of tolerant or nutrient-use-efficient cultivars may be an important practice to improve root growth in stress environments. Stress-tolerant genotypes are being identified and bred worldwide to solve some of the most difficult problems of soil fertility such as subsoil acidity, salinity, low plant availability of Fe in calcareous soils, and low P availability in acidic soils (Foy, 1992). Rhizobial species/strains differ markedly in tolerance to low pH, and toxicities of Al and Mn in tropical soils have been identified. The use of improved strains can improve N2 fixation and the root growth of legumes grown in acidic soils.

13.2. Plant management Plant management is another important strategy in improving root systems of crop plants. Adequate plant density and spacing can improve the root growth of crop plants. 13.2.1. Genetic variability An important plant management strategy is to exploit root system genetic variability of crop species or cultivars within species. Plant genetic variability can be defined as the heritable characters of a particular crop species or cultivar that produce differences in the growth or production among species, or cultivars of the same species, under favorable or unfavorable growth conditions (Fageria, 1989a). Cultivar differences in root size are quite common and have been related to differences in nutrient uptake (Baligar et al., 1998; Caradus, 1990; Fageria et al., 2006). Differences between white clover (T. repens L) populations and cultivars in P uptake per plant at low levels of P have been related to differences in root size and absolute growth rate (Caradus and Snaydon, 1986). There is widespread evidence for genotype diversity in the root characteristics of many crops in response to environment and increasing interest in using this diversity to improve agricultural production and consequently, nutrient use efficiency (Barber, 1994; Gregory, 1994). Mineral deficiency and toxicity, mechanical impedance, moisture stress, oxygen stress, and temperature have tremendous effects on root growth, development, and function and, subsequently, the ability of roots to absorb and translocate nutrients (Arkin and Taylor, 1981; Baligar et al., 1998; Barber, 1995; Marschner, 1995; Mengel et al., 2001). Mineral deficiency induces considerable variations in growth and morphology of roots, and such variations are strongly influenced by plant species and genotypes. Overall, the growth of the main axis is little affected by nutrient deficiency, but growth of lateral branches and their elongation rates appear to be substantially reduced. Baligar et al. (1998) summarized the effects of various essential elements as follows: nitrogen deficiency increases root hair length, increases or has no effect on root hair density, and reduces

The Role of Mineral Nutrition on Root Growth of Crop Plants

317

branching. Phosphorus deficiency increases the overall growth of roots and root hair length, increases the number of second-order laterals, and either increases or does not affect root hair density. K and Ca deficiencies reduce root growth; however, high Mg levels reduce the dry mass of roots. The effects of these nutrient stress factors on nutrient-use efficiency have not been well explored. Baligar et al. (1998) states that low pH reduces root mass, length, and root hair formation. In alkaline soils, ammonium toxicity causes severe root inhibition and in general, salinity leads to reduction in the mass and length of roots and dieback of laterals.

14. Conclusions The study of plant roots is one of the most promising, but least explored, areas of research related to mineral nutrition. Roots are poorly studied primarily because of their physical location and growth habit. Understanding plant root growth is important for improving productivity of annual crops in agroecosystems. Root systems are important plant organs because they absorb water and nutrients and also provide mechanical support to the plant. In addition, roots synthesize growth substances and hormones such as cytokinins that are important in leaf function and possibly, grain development. In plant growth analysis, the role of roots is generally ignored due to the difficulty in getting accurate root growth data under field condition. The principal structure of the root system includes four developmentally distinct classes of roots. These are taproots, basal roots, lateral roots, and shootborn roots. Current evidence suggests that the four root classes that make up the primary and secondary root systems are physiologically distinct from each other. Rooting depth and spreading capacity are important traits for uptake of water and nutrients. Root growth varies among plant species and cultivars within species and can be modified by environmental factors. Genotypic differences in root growth among crop species and genotypes of the same species under similar and variable environmental conditions are now well demonstrated, and the possibility of developing genotypes of desirable root systems to soil properties offers exciting prospects for the future. Root number, maximum length, and root dry weight increase with increasing levels of macro- and micronutrients in the soil to a point beyond which root growth is suppressed. Plant roots together with their associated bacteria and fungi play an important role in the formation, maintenance, and turnover of soil aggregates. Currently, the techniques available to measure root systems are laborious and time consuming, and this limits their use in plant physiological research. Hence, it is necessary to develop root measurement techniques that are simple, cheap, and less time consuming.

318

N. K. Fageria and A. Moreira

ACKNOWLEDGMENT The authors thank Dr. Charles Allan Jones, Professor, Texas-A&M University, Texas AgriLife Research and Extension Service, Dallas, Texas for their peer review of the chapter and for giving valuable suggestions for its improvement.

REFERENCES Ae, N., Arihara, J., Okada, K., Yoshihara, T., and Johnson, C. (1990). Phosphorus uptake by pigeon pea and its role in cropping systems of the Indian subcontinent. Science 248, 477–480. Alcordo, I. S., and Rechcigl, J. E. (1993). Phosphogypsum in agriculture: A review. Adv. Agron. 49, 55–118. Allmaras, R. R., Linden, D. R., and Clapp, C. E. (2004). Corn-residue transformations into root and shoot carbon as related to nitrogen, tillage, and stover management. Soil Sci. Soc. Am. J. 68, 1366–1375. Alloway, B. J. (2008). Micronutrients and crop production: An introduction. In “Micronutrient Deficiencies in Global Crop Production” (B. J. Alloway, Ed.), pp. 1–39. Springer, New York. Alva, A. K., and Sumner, M. E. (1989). Alleviation of aluminum toxicity to soybeans by phosphogypsum or calcium sulfate in dilute nutrient solutions. Soil Sci. 147, 278–285. Anderson, B. J. (1970). Tracer elements for sheep, pastures and fodder crops in Australia. J. Aust. Inst. Agric. Sci. 36, 15–29. Anderson, E. L. (1987). Corn root growth and distribution as influenced by tillage and nitrogen fertilization. Agron. J. 79, 544–549. Anderson, D. S., Teyker, R. H., and Rayburn, A. L. (1991). Nitrogen form effects on early corn root morphological and anatomical development. J. Plant Nutr. 14, 1255–1266. Ando, T., Yoshida, S., and Niiyama, I. (1983). Nature of oxidizing power of rice roots. Plant Soil 72, 57–71. Andren, O., Rajkai, K., and Katterer, T. (1991). A non-destructive technique for studies of root distribution in relation to soil-moisture. Agric. Ecosyst. Environ. 34, 269–278. Arkin, G. F., and Taylor, H. M. (1981). Modifying the Root Environment to Reduce Crop StressModifying the Root Environment to Reduce Crop Stress. ASAE, St. Joseph, MI. Armstrong, J. (1979). Aeration in higher plants. In “Advances in Botanical Research” (H. W. Woolhouse, Ed.), Vol. 7, pp. 226–332. Academic Press, London. Armstrong, W., and Drew, M. C. (2002). Root growth and metabolism under oxygen deficiency. In “Plant Roots: The Hidden Half” (Y. Waisel, et al., Eds.), pp. 729–761. Marcel Dekker, New York. Atkinson, D. (1990). Influence of root system morphology and development on the need for fertilizers and the efficiency of use. In “Crops as Enhancers of Nutrient Use” (V. C. Baligar and R. R. Duncan, Eds.), pp. 411–450. Academic Press, San Diego, CA. Atwell, B. J. (1989). Physiological responses of lupin roots to soil compaction. In “Structural and Functional Aspects of Transport in Roots” (B. C. Loughman, Ed.), pp. 251–255. Kluwer Academic Publishers, Dordrecht, The Netherlands. Aung, L. H. (1974). Root–shoot relation-ships. In “The Plant Root and Its Environment” (E. W. Carson, Ed.), pp. 29–61. University Press of Virginia, Charlottesville, VA. Bakken, L. R. (1990). Microbial growth and immobilization/mineralization of N in the rhizosphere. Symbiosis 9, 37–41. Balesdent, J., and Balabane, M. (1996). Major contribution of roots to soil carbon storage inferred from maize cultivated soils. Soil Biol. Biochem. 28, 1261–1263.

The Role of Mineral Nutrition on Root Growth of Crop Plants

319

Baligar, V. C. (1986). Interrelationships between growth and nutrient uptake in alfalfa and corn. J. Plant Nutr. 9, 1391–1404. Baligar, V. C., and Fageria, N. K. (1999). Plant nutrient efficiency: Towards the second paradigm. In “Soil Fertility, Soil Biology and Plant Nutrition Interrelationships” (J. Q. Siqueira, F. M. S. Moreira, A. S. Lopes, L. R. G. Guilherme, V. Faquin, A. E. Furtini Neto, and J. G. Carvallo, Eds.), pp. 183–204. Brazilian Society of Soil Science and University of Lavras, Lavras, MG Brazil. Baligar, V. C., Fageria, N. K., and Elrashidi, M. (1998). Toxicity and nutrient constraints on root growth. HortScience 33, 960–965. Baligar, V. C., Fageria, N. K., and He, Z. L. (2001). Nutrient use efficiency in plants. Commun. Soil Sci. Plant Anal. 32, 921–950. Ball, R. A., Oosterhuis, D. M., and Mauromoustakos, A. (1994). Growth dynamics of the cotton plant during water-deficit stress. Agron. J. 86, 788–795. Barber, S. A. (1994). Root growth and nutrient uptake. In “Physiology and Determination of Crop Yield” (K. J. Boote, J. M. Bennett, T. R. Sinclair, and G. M. Paulsen, Eds.), pp. 95–99. ASA, CSSA, and SSSA, Madison, WI. Barber, S. A. (1995). Soil Nutrient Bioavailability: A Mechanistic Approach.2nd edn.. John Wiley & Sons, New York. Barber, S. A., and Silberbush, M. (1984). Plant root morphology and nutrient uptake. In “Roots, Nutrient and Water Influx, and Plant Growth” (S. A. Barber and D. R. Bouldin, Eds.), pp. 65–85. ASA Special Publication No. 49, ASA, CSSA, and SSSA, Madison, WI. 65–87. Barlow, P. W. (1986). Adventitious roots of whole plants: Their forms, functions and evolution. In “New Root Formation in Plants and Cuttings” (M. B. Jackson, Ed.), pp. 67–110. Martinus Nijhoff, Dordrecht, The Netherlands. Barlow, P. W. (2002). Cellular pattering in root meristems: Its origins and significance. In “Plant Roots: The Hidden Half” (Y. Waisel, et al., Eds.), pp. 49–82. Marcel Dekker, New York. Bengough, A. G., and Mullins, C. E. (1990). Mechanical impedance to root growth: A review of experimental techniques and root growth responses. J. Soil Sci. 41, 341–358. Beyrouty, C. A., Wells, B. R., Norman, R. J., Marvel, J. N., and Pillow, J. A. (1987). Characterization of rice roots using a minirhizotron technique. In “Minirhizotron Observation Tubes: Methods and Applications for Measuring Rhizosphere Dynamics” (H. M. Taylor, Ed.), , ASA Special Publication No. 50, pp. 99–108. American Society of Agronomy, Madison, WI. Beyrouty, C. A., Wells, B. R., Norman, R. J., Marvel, J. N., and Pillow, J. A. (1988). Root growth dynamics of a rice cultivar grown at two locations. Agron. J. 80, 1001–1004. Blevins, D. G. (1994). Uptake, translocation, and function of essential mineral elements in crop plants. In “Physiology and Determination of Crop Yield” (G. A. Peterson, Ed.), pp. 259–275. ASA, CSSA, and SSSA, Madison, WI. Bolliger, A., Magid, J., Amado, T. J. C., Neto, F. S., Ribeiro, M. F. S., Calegari, A., Ralisch, R., and Neergaard, A. (2006). Taking stock of the Brazilian zero-till revolution: A review of landmark research and farmers practice. Adv. Agron. 91, 47–110. Boone, R. D. (1994). Light-fraction soil organic matter: Origin and contribution to net nitrogen mineralization. Soil Biol. Biochem. 26, 1459–1468. Boote, K. J., Stansell, J. R., Schubert, A. M., and Stone, J. F. (1982). Irrigation, water use, and water relationships. In “Peanut Science and Technology” (H. E. Pattee and C. T. Yong, Eds.), pp. 164–205. American Peanut Research & Education Society, Yoakum, TX. Borg, H., and Grimes, D. W. (1986). Depth development of roots with time: An empirical description. Trans. Am. Soc. Agric. Eng. 29, 194–197.

320

N. K. Fageria and A. Moreira

Box, J. E., and Ramseur, E. L. (1993). Minirhizotron wheat root data: Comparisons to soil core root data. Agron. J. 85, 1058–1060. Brailsford, R. W., Voesenek, L. A. C. J., Blom, C. W. P. M., Smith, A. R., Hall, M. A., and Jackson, M. B. (1993). Enhanced ethylene production by primary roots of Zea mays L. in response to sub-ambient partial pressures of oxygen. Plant Cell Environ. 16, 1071–1080. Brouwer, R. (1966). Root growth of grasses and cereals. In “The Growth of Cereals and Grasses” (F. L. Milthorpe and D. J. Irvins, Eds.), pp. 153–166. Butterworth, London. Brown, R. H. (1984). Growth of the green plant. In “Physiological Basis of Crop Growth and Development” (M. B. Tesar, Ed.), pp. 153–174. American Society of Agronomy and Crop Science Society of America, Madison, WI. Brown, P. H. (2008). Micronutrient use in agriculture in the United States of America: Current practices, trends and constraints. In “Micronutrient Deficiencies in Global Crop Production” (B. J. Alloway, Ed.), pp. 267–286. Springer, New York. Brown, S. C., Keating, J. D. H., Gregory, P. J., and Cooper, P. J. M. (1987). Effects of fertilizer, variety and location on barley production under rainfed conditions in northern Syria. I. Root and shoot growth. Field Crops Res. 16, 53–66. Bruce, R. C., Warrell, L. A., Edwards, D. G., and Bell, L. C. (1988). Effects of aluminum and calcium in the soil solution of acid soils on root elongation of Glycine max cv. Forrest. Aust. J. Agric. Res. 39, 319–338. Buyanovsky, G. A., Wagner, G. H., and Gantzer, C. J. (1986). Soil respiration in a wheat ecosystem. Soil Sci. Soc. Am. J. 50, 338–344. Cakmak, I. (2008). Zinc deficiency in wheat in Turkey. In “Micronutrient Deficiencies in Global Crop Production” (B. J. Alloway, Ed.), pp. 181–200. Springer, New York. Cakmak, I., Hengeler, C., and Marschner, H. (1994). Changes in phloem export of sucrose in leaves in response to phosphorus, potassium and magnesium deficiency in bean plants. J. Exp. Bot. 45, 1251–1257. Campbell, R., and Drew, M. C. (1983). Electron microscopy of gas space (aerenchyma) formation in adventitious roots of Zea mays L subjected to oxygen shortage. Planta 157, 350–357. Cannell, R. Q. (1985). Reduced tillage in North-West Europe: A review. Soil Tillage Res. 5, 129–177. Cannell, R. Q., and Hawes, J. D. (1994). Trends in tillage practices in relation to sustainable crop production with special reference to temperate climates. Soil Tillage Res. 30, 245–282. Caradus, J. R. (1990). The structure and function of white clover root systems. Adv. Agron. 43, 1–46. Caradus, J. R., and Snaydon, R. W. (1986). Plant factors influencing phosphorus uptake by white clover from solution culture. I. Populations differences. Plant Soil 93, 153–163. Champigny, M. L., and Talouizte, A. (1981). Photosynthate distribution and metabolic fate in relation to nitrogen metabolism in wheat seedlings. In “Photosynthesis: IV. Regulation of Carbon Metabolism” (G. Akoyunoglou, Ed.), pp. 645–652. Balban International Science Services, Philadelphia, PA. Chan, K. Y., and Mead, J. A. (1992). Tillage-induced differences in the growth and distribution of wheat-roots. Aust. J. Agric. Res. 43, 19–28. Chassot, A., Stamp, P., and Richner, W. (2001). Root distribution and morphology of maize seedlings as affected by tillage and fertilizer placement. Plant Soil 231, 123–135. Cheng, W., and Kuzyakov, Y. (2005). Root effects on soil organic matter decomposition. In “Roots and Soil Management: Interactions Between Roots and the Soil” (R. W. Zobel and S. F. Wright, Eds.), pp. 119–143. ASA, CSSA, and SSSA, Madison, WI. Cheng, W., Coleman, D. C., Carroll, C. R., and Hoffman, C. A. (1994). Investigating short-term carbon flows in the rhizospheres of different plant species, using isotopic trapping. Agron. J. 86, 782–788.

The Role of Mineral Nutrition on Root Growth of Crop Plants

321

Comfort, S. D., Malzer, G. L., and Busch, R. H. (1988). Nitrogen fertilization of spring wheat genotypes: Influence on root growth and soil water depletion. Agron. J. 80, 114–120. Costa, C., Dwyer, L. M., Zhou, X., Dutilleul, P., Hamel, C., Reid, L. M., and Smith, D. L. (2002). Root morphology of contrasting maize genotypes. Agron. J. 94, 96–101. Crozier, C. R., Naderman, G. C., Tucker, M. R., and Sugg, R. E. (1999). Nutrient and pH stratification with conventional and no-till management. Commun. Soil Sci. Plant Anal. 30, 65–74. Davidson, R. L. (1969a). Effect of root/leaf temperature differentials on root/shoot ratios in pasture grasses and clover. Ann. Bot. 33, 561–569. Davidson, R. L. (1969b). Effects of soil nutrients and moisture on root/shoot ratios in Lolium perenne L. and Trifolium repens L. Ann. Bot. (London) 33, 571–577. Dingkuhn, M., and Kropff, M. (1996). Rice. In “Photoassimilate Distribution in Plants and Crops” (E. Zamski and A. A. Schaffer, Eds.), pp. 519–547. Marcel Dekker, New York. Doss, B. D., Dumas, W. T., and Lund, Z. F. (1979). Depth of lime incorporation for correction of subsoil acidity. Agron. J. 71, 541–544. Drew, M. C. (1975). Comparison of the effects of localized supply of phosphate, nitrate, ammonium, and potassium on the growth of the seminal root system and shoot of barley. New Phytol. 75, 470–490. Drew, M. C., Saker, L. R., and Ashley, T. W. (1973). Nutrient supply and the growth of the seminal root system in barley. J. Exp. Bot. 24, 1189–1202. Durieux, R. P., Kamprath, E. J., Jackson, W. A., and Moll, R. H. (1994). Root distribution of corn: The effect of nitrogen fertilization. Agron. J. 86, 958–962. Dyer, M. I., Acra, M. A., Wang, G. M., Coleman, D. C., Freckman, D. W., McNaughton, S. J., and Strain, B. R. (1991). Source-sink carbon relations in two Panicum coloratum in response to herbivory. Ecology 72, 1472–1483. Eghball, B., and Maranville, J. W. (1993). Root development and nitrogen influx of corn genotypes grown under combined drought and nitrogen stresses. Agron. J. 85, 147–152. Eghball, B., Settimi, J. R., Maranville, J. W., and Parkhurst, A. M. (1993). Fractal analysis for morphological description of corn roots under nitrogen stress. Agron. J. 85, 287–289. Ellis, F. B. (1979). Roots and their function in the soil. “The Yield of Cereals”, pp. 3–5. Cereal Unit National Agricultural Centre, Stoneleigh, UK. Epstein, E., and Bloom, A. J. (2005). Mineral Nutrition of Plants: Principles and Perspectives.2nd edn.. Sinauer Associations, Sunderland, MA. Eskew, D. L., Welch, R. M., and Cary, E. E. (1983). Nickel: An essential micronutrients for legumes and possibly all higher plants. Science 222, 621–623. Essau, K. (1977). Anatomy of Seed Plants.2nd edn.. John Wiley & Sons, New York. Evans, L. T., and Wardlaw, L. F. (1976). Aspects of the comparative physiology of grain yield in cereals. Adv. Agron. 28, 301–359. Fageria, N. K. (1989a). Tropical Soils and Physiological Aspects of Crops. EMBRAPA, Brasilia. Fageria, N. K. (1989b). Effect of phosphorus on growth, yield and nutrient accumulation in the common bean. Trop. Agric. 66, 249–255. Fageria, N. K. (1991). Response of rice cultivars to phosphorus fertilization on a dark red latosol of central Brazil. Rev. Bras. Cienc. Solo 15, 63–67. Fageria, N. K. (1992). Maximizing Crop Yields. Marcel Dekker, New York. Fageria, N. K. (2000). Adequate and toxic levels of boron for rice, common bean, corn, soybean and wheat production in cerrado soil. Braz. J. Agric. Eng. Environ. 4, 57–62. Fageria, N. K. (2002a). Micronutrients influence on root growth of upland rice, common bean, corn, wheat, and soybean. J. Plant Nutr. 25, 613–622. Fageria, N. K. (2002b). Nutrient management for sustainable dry bean production in the tropics. Commun. Soil Sci. Plant Anal. 33, 1537–1575.

322

N. K. Fageria and A. Moreira

Fageria, N. K. (2002c). Soil quality vs. environmentally based agriculture. Commun. Soil Sci. Plant Anal. 33, 2301–2329. Fageria, N. K. (2007). Yield physiology of rice. J. Plant Nutr. 30, 843–879. Fageria, N. K. (2008). Optimum soil acidity indices for dry bean production on an Oxisol in no-tillage system. Commun. Soil Sci. Plant Anal. 39, 845–857. Fageria, N. K. (2009). The Use of Nutrients in Crop Plants. CRC Press, Boca Raton, FL. Fageria, N. K. (2011). Root growth of upland rice genotypes as influenced by nitrogen fertilization. J. Plant Nutr. 34, (in press). Fageria, N. K., and Baligar, V. C. (2003). Fertility management of tropical acid soils for sustainable crop production. In “Handbook of Soil Acidity” (Z. Rengel, Ed.), pp. 359–385. Marcel Dekker, New York. Fageria, N. K., and Baligar, V. C. (2005). Enhancing nitrogen use efficiency in crop plants. Adv. Agron. 88, 97–185. Fageria, N. K., and Baligar, V. C. (2008). Ameliorating soil acidity of tropical Oxisols by liming for sustainable crop production. Adv. Agron. 99, 345–399. Fageria, N. K., and Gheyi, H. R. (1999). Efficient Crop Production Federal University of Paraiba, Campina Grande, Brazil. Fageria, N. K., and Moreira, A. (2011). Yield and yield components of upland rice as influenced by nitrogen sources. J. Plant Nutr. 34(in press). Fageria, N. K., and Santos, A. B. (2008). Yield physiology of dry bean. J. Plant Nutr. 31, 983–1004. Fageria, N. K., and Santos, A. B. (2011). Lowland rice growth and development and nutrient uptake during growth cycle. J. Plant Nutr. 34(in press). Fageria, N. K., and Souza, C. M. R. (1991). Upland rice, common bean, and cowpea response to magnesium application on an oxisols. Commun. Soil Sci. Plant Anal. 22, 1805–1816. Fageria, N. K., and Stone, L. F. (1999). Acidity management of cerrado and varzea soils of Brazil. Santo Antoˆnio de Goias, Brazil. Fageria, N. K., and Stone, L. F. (2004). Yield of common bean in no-tillage system with application of lime and zinc. Pesqui. Agropecu. Bras. 39, 73–78. Fageria, N. K., and Stone, L. F. (2006). Physical, chemical, and biological changes in the rhizosphere and nutrient availability. J. Plant Nutr. 29, 1327–1356. Fageria, N. K., and Stone, L. F. (2008). Micronutrient deficiency problems in South America. In “Micronutrient Deficiencies in Global Crop Production” (B. J. Alloway, Ed.), pp. 245–266. Springer, New York. Fageria, N. K., Barbosa Filho, M. P., and Carvalho, J. R. P. (1981). Influence of iron on growth and absorption of P, K, Ca and Mg by rice plants in nutrient solution. Pesqui. Agropecu. Bras. 16, 483–488. Fageria, N. K., Baligar, V. C., and Clark, R. B. (2002). Micronutrients in crop production. Adv. Agron. 77, 185–268. Fageria, N. K., Castro, E. M., and Baligar, V. C. (2004). Response of upland rice genotypes to soil acidity. In “The Red Soils of China: Their Nature, Management, and Utilization” (M. J. Wilson, Z. He, and X. Yang, Eds.), Kluwer Academic Publishers, Dordrecht, The Netherlands. Fageria, N. K., Baligar, V. C., and Clark, R. B. (2006). Physiology of Crop Production. The Howarth Press, New York. Fageria, N. K., Baligar, V. C., and Li, Y. C. (2008a). The role of nutrient efficient plants in improving crop yields in the twenty first century. J. Plant Nutr. 31, 1121–1157. Fageria, N. K., Santos, A. B., Barbosa Filho, M. P., and Guimara˜es, C. M. (2008b). Iron toxicity in lowland rice. J. Plant Nutr. 31, 1676–1697. Fageria, N. K., Moraes, O. P., and Vasconcelos, M. J. (2011). Yield and yield components of upland rice as influenced by nitrogen sources. J. Plant Nutr. 34(in press).

The Role of Mineral Nutrition on Root Growth of Crop Plants

323

Feil, B., Thirapon, R., Geisler, G., and Stamp, P. (1991). The impact of temperature on seedling root traits of European and tropical corn (Zea mays L.) cultivars. J. Agron. 85, 287–289. Follett, R. F., Varvel, G. E., Kimble, J. M., and Vogel, K. P. (2009). No-till corn after bromegrass: Effect on soil carbon and soil aggregates. Agron. J. 101, 261–268. Foy, C. D. (1984). Physiological effects of hydrogen, aluminum and manganese toxicities in acid soils. In “Soil Acidity and Liming” (F. Adams, Ed.), 2nd edn., pp. 57–97. American Society of Agronomy, Madison, WI. Foy, C. D. (1992). Soil chemical factors limiting plant root growth. Adv. Soil Sci. 19, 97–149. Freckman, D. W., Barker, K. R., Coleman, D. C., Acra, M. A., Dyer, M. I., Strain, B. R., and McNaughton, S. J. (1991). The use of the 11C technique to measure plant response to herbivorous soil nematodes. Funct. Ecol. 5, 810–818. Fredeen, A. L., Rao, I. M., and Terry, N. (1989). Influence of phosphorus nutrition on growth and carbon partitioning in Glycine max. Plant Physiol. 89, 225–230. Gabelman, W. H., Gerloff, G. C., Schettini, T., and Coltman, R. (1986). Genetic variability in root systems associated with nutrient acquisition and use. HortScience 21, 971–973. Gale, W. J., and Cambardella, C. A. (2000). Carbon dynamics of surface residue and root derived organic matter under simulated no-till. Soil Sci. Soc. Am. J. 64, 190–195. Gale, W. J., Cambardell, C. A., and Bailey, T. B. (2000a). Root-derived carbon and the formation and stabilization of aggregates. Soil Sci. Soc. Am. J. 64, 201–207. Gale, W. J., Cambardell, C. A., and Bailey, T. B. (2000b). Surface residue and root-derived carbon in stable and unstable aggregates. Soil Sci. Soc. Am. J. 64, 196–201. Garrity, D. P., Sullivan, C. Y., and Watts, D. G. (1983). Moisture deficits and grain sorghum performance: Drought stress conditions. Agron. J. 75, 997–1004. Garton, R. W., and Widders, I. E. (1990). N and P preconditioning of small plug seedlings influences growth and yield of processing tomatoes. HortScience 25, 655–657. Goldberg, S. (1993). Chemistry and mineralogy of boron in soils. In “Boron and Its Role in Crop Production” (U. C. Gupta, Ed.), pp. 3–44. CRC Press, Boca Raton, FL. Gonzalez, E. E. (1976).Effect of depth of lime incorporation on the growth of corn in Oxisols of central Brazil. Ph.D. Thesis, North Carolina State University, Raleigh, NC, University of Microfilms. Ann Arbor, MI (Diss. Abstract 37:5474). Gonzalez-Erico, E., Kamprath, E. J., Nederman, G. C., and Soares, W. V. (1979). Effect of depth of lime incorporation on the growth of corn on an Oxisols of central Brazil. Soil Sci. Soc. Am. J. 43, 1155–1158. Goss, M. J., and Kay, B. D. (2005). Soil aggregation. In “Roots and Soil Management: Interactions Between Roots and the Soil” (R. W. Zobel and S. F. Wright, Eds.), pp. 163–180. ASA, CSSA, and SSSA, Madison, WI. Graham, R. D. (2008). Micronutrient deficiencies in crops and their global significance. In “Micronutrient Deficiencies in Global Crop Production” (B. J. Alloway, Ed.), pp. 41–61. Springer, New York. Gregory, P. J. (1994). Root growth and activity. In “Physiology and Determination of Crop Yield” (G. A. Peterson, Ed.), pp. 65–93. ASA, CSSA, and SSSA, Madison, WI. Gregory, P. J., and Brown, S. C. (1989). Root growth, water use and yield of crops in dry environments: What characteristics are desirable? Asp. Appl. Biol. 22, 235–243. Gupta, U. C. (1993). Factors affecting boron uptake by plants. In “Boron and Its Role in Crop Production” (U. C. Gupta, Ed.), pp. 87–123. CRC Press, Boca Raton, FL. Haberle, J. (1993). The comparison of root-system, nutrient, and water-uptake of two old and present Czech winter wheat cultivars under different nitrogen fertilization. Rostlinna Vyroba 39, 851–863. Haberle, J., Svoboda, P., and Blaha, L. (1995). The comparison of shoot and root production in old and new cultivars of winter cereals. Rostlinna Vyroba 41, 511–516.

324

N. K. Fageria and A. Moreira

Haberle, J., Svoboda, P., and Ruzek, P. (1996). Root length of winter wheat and the content of mineral nitrogen in soil profile. Rostlinna Vyroba 42, 193–197. Haider, K., Moiser, A., and Heinemeyer, O. (1993). Impact of growing crop plants on Nutilization from fertilizer, crop residues, and on organic N and C-mineralization. Soil Sci. 1, 183–191. Hallmark, W. B., and Barber, S. A. (1984). Root growth and morphology, nutrient uptake, and nutrient status of early growth of soybeans as affected by soil P and K. Agron. J. 76, 209–212. Hamblin, A. P., and Tennant, A. P. (1987). Root length density and water uptake in cereals and grain legumes; How well are they correlated? Aust. J. Agric. Res. 38, 513–527. Hansson, A. C., and Andren, O. (1987). Root dynamics in barley, Lucerne, and meadow fescue investigated with a minirhizotron technique. Plant Soil 103, 33–38. Hansson, A. C., Steen, E., and Andren, O. (1992). Root-growth of daily irrigated and fertilized barley investigation with ingrowth cores, soil cores, and minirhizotrons. Swed. J. Agric. Res. 22, 141–152. Harper, J. E. (1994). Nitrogen metabolism. In “Physiology and Determination of Crop Yield” (G. A. Peterson, Ed.), pp. 285–302. ASA, CSSA, and SSSA, Madison, WI. Helal, H. M., and Sauerbeck, D. (1987). Direct and indirect influences of plant root on organic matter and phosphorus turnover in soil. In “Soil Organic Matter Dynamics and Soil Productivity” (J. H. Couley, Ed.), INTECOl Bulletin 15, pp. 49–58. International Association for Ecology, Athens, GA. Hoad, S. P., Russell, G., Lucas, M. E., and Bingham, I. J. (2001). The management of wheat, barley, and oat root systems. Adv. Agron. 74, 193–246. Holanda, F. S. R., Mengel, D. B., Paula, M. B., Carvalho, J. G., and Bertoni, J. C. (1998). Influence of crop rotations and tillage systems on phosphorus and potassium stratification and root distribution in the soil profile. Commun. Soil Sci. Plant Anal. 29, 2383–2394. Holland, E. A., and Coleman, D. C. (1987). Litter placement effects on microbial and organic matter dynamics in an agro ecosystem. Ecology 68, 425–433. Horiguchi, T. (1995). Rhizosphere and root oxidation activity. In “Science of the Rice Plant” (T. Matsuo, K. Kumazawa, R. Ishii, K. Ishihara, and H. Hirata, Eds.), Vol. 2, pp. 221–248. Food and Agriculture Policy Research Center, Tokyo. Hsiao, T., Steduto, L. H. P., Rojas-Lara, B., Raes, D., and Fereres, E. (2009). AquaCrop— The FAO crop model to stimulate yield response to water: III. Parameterization and testing for maize. Agron. J. 101, 448–459. Hurd, E. A. (1974). Phenotype and drought tolerance in wheat. Agric. Meteorol. 14, 39–55. Hurd, E. A., Townley-Smith, T. F., Patterson, L. A., and Owen, C. H. (1972). Techniques used in producing Wascana wheat. Can. J. Plant Sci. 52, 689–691. Jackson, M. B. (1994). Root to shoot communication in flooded plants: Involvement of abscisic acid, ethylene, and 1-aminocyclopropane-1-carboxylic acid. Agron. J. 86, 775–782. Jackson, R. B., Mooney, H. A., and Schulze, E. D. (1997). A global budget for fine root biomass, surface area, and nutrient contents. Proc. Natl. Acad. Sci. USA 94, 7362–7366. Jaffar, M. N., Stone, L. R., and Goodwin, D. E. (1993). Rooting depth and dry matter development of sunflower. Agron. J. 85, 281–286. Kamprath, E. J., and Foy, C. D. (1985). Lime–fertilizer–plant interactions in acid soils. In “Fertilizer Technology and Use” (O. P. Engelstad, Ed.), 3rd edn.. pp. 91–151. Soil Science Society of America, Madison, WI. Kaspar, T. C., Stanley, C. D., and Taylor, H. M. (1978). Soybean root growth during the reproductive stages of development. Agron. J. 70, 1105–1107. Kaspar, T. C., Taylor, H. M., and Shibles, R. M. (1984). Taproot elongation rates of soybean cultivars in the glasshouse and their relation to field rooting depth. Crop Sci. 24, 916–920.

The Role of Mineral Nutrition on Root Growth of Crop Plants

325

Kelley, P. M., and Izawa, S. (1978). The role of chloride ion in photosystem II. I. Effects of chloride ion on photosystem II electron transport and on hydroxylamine inhibition. Biochim. Biophys. Acta 502, 148–210. 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. Ketring, D. L., and Reid, J. L. (1993). Growth and peanut roots under field conditions. Agron. J. 85, 80–85. Kirchner, M. J., Wollum, A. G., and King, L. D. (1993). Soil microbial populations and activities in reduced chemical input agro ecosystems. Soil Sci. Soc. Am. J. 57, 1289–1295. Kitchen, N. R., Buchholz, D. D., and Nelson, C. J. (1990). Potassium fertilizer and potato leafhopper effects on alfalfa growth. Agron. J. 82, 1069–1074. Kleeper, B., Belford, R. K., and Rickman, R. W. (1984). Root and shoot development in winter wheat. Agron. J. 76, 117–122. Klepper, B. (1992). Development and growth of crop root systems. Adv. Soil Sci. 19, 1–25. Klepper, B., and Kaspar, T. C. (1994). Rhizotrons: Their development and use in agricultural research. Agron. J. 86, 745–753. Klucas, R. V., Hanus, F. J., Russell, S. A., and Evans, H. J. (1983). Nickel: A micronutrient element for hydrogen-dependent growth of Rhizobium japonicum and for expression of urease activity in soybean leaves. Proc. Natl. Acad. Sci. USA 80, 2253–2257. Kujira, Y., Grove, J. H., and Ronzelli, P. (1994). Varietal differences of root systems in winter-wheat seedlings. Jpn. J. Crop Sci. 63, 524–530. Kuo, S., Sainju, U. M., and Jellum, E. J. (1997a). Winter cover crop effects on soil organic carbon and carbohydrate. Soil Sci. Soc. Am. J. 61, 145–152. Kuo, S., Sainju, U. M., and Jellum, E. J. (1997b). Winter cover cropping influence on nitrogen in soil. Soil Sci. Soc. Am. J. 61, 1392–1399. Laan, P., Benevoets, M. J., Lythe, S., Armstrong, W., and Blom, C. W. P. M. (1989). Root morphology and aerenchyma formation as indicators of the flood-tolerance of Rumex species. J. Ecol. 77, 693–703. Lampurlanes, J., Angas, P., and Cantero-Martines, C. (2001). Root growth, soil water content and yield of barley under different tillage systems on two soils in semiarid conditions. Field Crops Res. 69, 27–40. Leon, J., and Schwang, K. U. (1992). Description and application of a screening method to determine root morphology traits of cereal cultivars. Z. Acker Pflanzenbau 169, 128–134. Leon, L. A., Lopez, A. S., and Vlek, P. L. G. (1985). Micronutrient problems in tropical Latin America. Fertil. Res. 7, 95–129. Leskovarant, D. I., and Stofella, P. J. (1995). Vegetable seedling root systems: Morphology, development, and importance. HortScience 30, 1153–1159. Lexmond, T. M., and Vorm, P. D. J. (1981). The effect of pH on copper toxicity to hydroponically grown maize. Neth. J. Agric. Sci. 29, 217–238. Liang, B. C., Wang, X. L., and Ma, B. L. (2002). Maize root-induced change in soil organic carbon pools. Soil Sci. Soc. Am. J. 66, 845–847. Lipiec, J., Hakansson, I., Tarkiewicz, S., and Kossowski, J. (1991). Soil physical properties and growth of spring barley as related to the degree of compactness of two soils. Soil Tillage Res. 19, 307–317. Logan, T. J., Lal, R., and Dick, W. A. (1991). Tillage systems and soil properties in North America. Soil Tillage Res. 20, 241–270. Ludlow, M. M., and Muchow, R. C. (1990). A critical evaluation of traits for improving crop yields in water-limited environments. Adv. Agron 43, 107–153. Mackay, A. D., and Barber, S. A. (1985). Effect of nitrogen on root growth of two corn genotypes in the field. Agron. J. 78, 659–703. Maizlish, N. A., Fritton, D. D., and Kendall, W. A. (1980). Root morphology and early development of maize at varying levels of nitrogen. Agron. J. 72, 25–30.

326

N. K. Fageria and A. Moreira

Marschner, H. (1995). Mineral Nutrition of Higher Plants.2nd edn.. Academic Press, New York. Marschner, H. (1998). Role of root growth, arbuscular mycorrhiza, and root exudates for the efficiency in nutrient acquisition. Field Crops. Res. 56, 203–207. Martino, D. L., and Shaykewich, C. F. (1994). Root penetration profiles of wheat and barley as affected by soil penetration resistance in field conditions. Can. J. Soil Sci. 74, 193–200. Mathews, R. B., Harris, D., Nageswara Rao, R. C., Williamds, J. H., and Wadia, K. D. R. (1988). The physiological basis for yield differences between four genotypes of groundnut (Arachis hypogaea) in response to drought. I. Dry matter production and water use. Exp. Agric. 24, 191–202. Matsumoto, H., and Morimura, S. (1980). Repressed template activity of chromatin of pea roots treated by aluminum. Plant Cell Physiol. 21, 951–959. Melillo, J. M., and Gosz, J. R. (1983). Interactions of biogeochemical cycles in forest ecosystems. In “The Major Biogeochemical Cycles and Their Interactions” (B. Bolin and R. B. Cook, Eds.), pp. 177–220. John Wiley & Sons, Chichester, England. Mengel, D. B., and Barber, S. A. (1974). Development and distribution of the corn root system under field conditions. Agron. J. 68, 341–344. Mengel, K., and Kirkby, E. A. (1978). Principles of Plant Nutrition International Potash Institute, Bern, Switzerland. Mengel, K., Kirkbay, E. A., Kosegarten, H., and Appel, T. (2001). Principles of Plant Nutrition.5th edn.. Kluwer Academic Publishers, Dordrecht, The Netherlands. Merrill, S. D., Black, A. L., and Bauer, A. (1996). Conservation tillage affects root growth of dryland spring wheat under drought. Soil Sci. Soc. Am. J. 60, 575–583. Merrill, S. D., Tanaka, D. L., and Hanson, J. D. (2002). Root length growth of eight crop species in haplustoll soils. Soil Sci. Soc. Am. J. 66, 913–923. Milchumas, D. G., Lauenroth, W. K., Singh, J. S., and Cole, C. V. (1985). Root turnover and production by 14C dilution: Implications of carbon portioning in plants. Plant Soil 88, 353–365. Miller, D. E. (1986). Root systems in relation to stress tolerance. HortScience 21, 963–970. Miltner, E. D., Karnok, K. J., and Hussey, R. S. (1991). Root response of tolerant and intolerant soybean cyst nematodes. Agron. J. 83, 571–576. Mortvedt, J. J. (2000). Bioavailability of micronutrients. In “Hand Book of Soil Science” (M. E. Sumner, Ed.), pp. D71–D88. CRC Press, Boca Raton, FL. Mullins, G. L., Reeves, D. W., Burmester, C. H., and Bryant, H. H. (1994). In-row subsoiling and potassium placement on root growth and potassium content of cotton. Agron. J. 86, 136–139. Munns, D. N., and Franco, A. A. (1982). Soil constraints to legume production. In “Biological Nitrogen Fixation for Tropical Agriculture” (P. H. Graham and S. C. Harris, Eds.), pp. 135– 152. Centro Internacional de Agricultura Tropical, Cali, Colombia. Murphy, J. A., and Zaurov, E. (1994). Shoot and root growth response of perennial ryegrass to fertilizer placement depth. Agron. J. 86, 828–832. Newman, E. I. (1966). A method for estimating the total length of root in a sample. J. Appl. Ecol. 3, 139–145. Norby, R. J., and Cortrufo, M. F. (1998). A question of litter quality. Nature 206, 17–18. Nurlaeny, N., Marschner, H., and George, E. (1996). Effects of liming and mycorrhizal colonization on soil phosphate depletion and phosphate uptake by maize (Zea mays L.) and soybean (Glycine max L.) grown in two tropical acid soils. Plant Soil 181, 275–285. O’Toole, J. C., and Bland, W. L. (1987). Genotypic variation in crop plant root systems. Adv. Agron. 41, 91–145.

The Role of Mineral Nutrition on Root Growth of Crop Plants

327

Oritani, T. (1995). Cytokinins. In “Science of the Rice Plant” (T. Matsuo, K. Kumazawa, R. Ishii, K. Ishihara, and H. Hirata, Eds.), Vol. 2, pp. 189–198. Food and Agriculture Policy Research Center, Tokyo. Osaki, M., Shinano, T., Matsumoto, M., Zheng, T., and Tadano, T. (1997). A root-shoot interaction hypothesis for high productivity of field crops. In “Plant Nutrition for Sustainable Food Production and Environment” (T. Ando, K. Fujita, T. Mae, H. Matsumoto, S. Mori, and J. Sekiya, Eds.), pp. 669–674. Kluwer Academic Publishers, Dordrecht, The Netherlands. Pan, W. L., Jackson, W. A., and Moll, R. H. (1985). Nitrate uptake and portioning by corn (Zea mays L.) root systems and associated morphological differences among genotypes and stages of root development. J. Exp. Bot. 36, 1341–1351. Parker, C. J., Carr, M. K. V., Jarvis, N. J., Puplampu, B. O., and Lee, V. H. (1991). An evaluation of the minirhizotron technique for estimating root distribution in potatoes. J. Agric. Sci. 116, 341–350. Parton, W. J., Ojima, D. S., and Schimel, D. S. (1996). Models to evaluate soil organic matter storage and dynamics. In “Structure and Organic Matter Storage in Soils” (M. R. Carter and B. A. Stewart, Eds.), pp. 421–448. CRC Press, Boca Raton, FL. Peterson, C. M., Klepper, B., Pumphrey, F. V., and Rickman, R. W. (1984). Restricted rooting decreases tillering and growth of winter wheat. Agron. J. 76, 861–863. Pinton, R., and Varannini, Z. (2001). The rhizosphere as a site of biochemical interactions among soil components, plants and microorganisms. In “The Rhizosphere: Biochemical and Organic Substances at the Soil–Plant Interface” (R. Pinto, Z. Varanini, and P. Nannipieri, Eds.), pp. 1–17. Marcel Dekker, New York. Puget, P., and Drinkwater, L. E. (2001). Short-term dynamics of root- and shoot-derived carbon from a leguminous green manure. Soil Sci. Soc. Am. J. 65, 771–779. Qian, J. H., and Doran, J. W. (1996). Available carbon released from crop roots during growth as determined by carbon-13 natural abundance. Soil Sci. Soc. Am. J. 60, 828–831. Qin, R., Stamp, P., and Richner, W. (2004). Impact of tillage on root systems of winter wheat. Agron. J. 96, 1523–1530. Qin, R., Stamp, P., and Richner, W. (2005). Impact of tillage and banded starter fertilizer on maize root growth in the top 25 centimeters of the soil. Agron. J. 97, 674–683. Rasmussen, J. K. (1991). Reduced soil tillage and Italian ryegrass as catch crop. II. Soil bulk density, root development and soil chemistry. Tidsskr. Planteavl. 95, 139–154. Reichman, G. A., and Trooien, T. P. (1993). Corn yield and salinity response to irrigation on slowly permeable subsoil. Soil Sci. Soc. Am. J. 57, 1549–1554. Rengel, Z., and Robinson, D. L. (1989). Aluminum effects on growth and micronutrient uptake by annual ryegrass. Agron. J. 81, 208–215. Reuter, D. J., Alston, A. M., and McFarland, J. D. (1988). Occurrence and correction of manganese deficiency in plants. In “Manganese in Soils and Plants” (R. D. Graham, R. J. Hannam, and N. C. Uren, Eds.), pp. 205–224. Kluwer Academic Publisher, Dordrecht, The Netherlands. Ritchey, K. D., Souza, D. M. G., Lobato, E., and Correa, O. (1980). Calcium leaching to increase rooting depth in Brazilian savannah Oxisol. Agron. J. 72, 40–44. Robinson, D., Linehan, D. J., and Gordon, D. C. (1994). Capture of nitrate from soil by wheat in relation to root length, nitrogen inflow, and availability. New Phytol. 128, 297–305. Romer, W. (1985). The influence of shoot and root-growth on the uptake of phosphorus in cereals. Bodenkultur 36, 203–211. Roth, G. W., Calvin, D. D., and Lweloff, S. M. (1995). Tillage, nitrogen timing, and planting date effects on western corn rootworm injury to corn. Agron. J. 87, 189–193. Rufty, T. W. Jr., Israel, D. W., Volk, R. J., Qui, J., and Sa, T. (1993). Phosphate regulation of nitrate assimilation in soybean. J. Exp. Bot. 44, 879–891.

328

N. K. Fageria and A. Moreira

Russell, R. S. (1977). Plant Root Systems. McGraw-Hill, New York. Sainju, U. M., Singh, B. P., and Whitehead, W. F. (2001). Comparison of the effects of cover crops and nitrogen fertilization on tomato yield, root growth, and soil properties. Sci. Hortic. 91, 201–214. Sainju, U. M., Singh, B. P., and Whitehead, W. F. (2005a). Tillage, cover crops, and nitrogen fertilization effects on cotton and sorghum root biomass, carbon, and nitrogen. Agron. J. 97, 1279–1290. Sainju, U. M., Whitehead, W. F., and Singh, B. P. (2005b). Carbon accumulation in cotton, sorghum, and underlying soil as influenced by tillage, cover crops, and nitrogen fertilization. Plant Soil 273, 219–234. Sanchez, J. E., Paul, E. A., Wilson, T. C., Smeenk, J., and Harwood, R. R. (2002). Corn root effects on the nitrogen supplying capacity of a conditioned soil. Agron. J. 94, 391–396. Sanders, J. L., and Brown, D. A. (1978). A new fiber-optic technique for measuring root growth of soybean under field conditions. Agron. J. 70, 1073–1076. Schenk, M. K., and Barber, S. A. (1980). Potassium and phosphorus uptake by corn genotype grown in the field as influenced by root characteristics. Plant Soil 54, 65–75. Shalhevet, J., Huck, M. G., and Schroeder, B. P. (1995). Root and shoot growth responses to salinity in maize and soybean. Agron. J. 87, 512–516. Sharma, B. R., and Chaudhary, T. N. (1984). Nitrogen uptake by wheat in relation to water regime and depth of N placement in a loamy sand soil: Measurement and model simulation. J. Soil Sci. 35, 141–148. Sharratt, B. S., and Cochran, V. L. (1993). Skip-row and equidistant-row barley with nitrogen placement; Yield, nitrogen uptake and root density. Agron. J. 85, 246–250. Sidiras, N., Bilalis, D., and Vavoulidou, E. (2001). Effects of tillage and fertilization on some selected physical properties of soil (0–30 cm depth) and on the root grown dynamic of winter barley (Hordeum vulgare cv. Niki). J. Agron. Crop Sci. 187, 167–176. Silberbush, M., and Barber, S. A. (1984). Phosphorus and potassium uptake of field grown soybeans predicted by a simulation model. Soil Sci. Soc. Am. J. 48, 592–596. Sinclair, A. H., and Edwards, A. C. (2008). Micronutrient deficiency problems in agricultural crops in Europe. In “Micronutrient Deficiencies in Global Crop Production” (B. J. Alloway, Ed.), pp. 225–266. Springer, New York. Singh, M. V. (2008). Micronutrient deficiencies in crop and soils in India. In “Micronutrient Deficiencies in Global Crop Production” (B. J. Alloway, Ed.), pp. 93–123. Springer, New York. Slaton, N. A., and Beyrouty, C. A. (1992). Rice shoot response to root confinement within a membrane. Agron. J. 84, 50–53. Slaton, N. A., Beyrouty, C. A., Wells, B. R., Norman, R. J., and Gbur, E. E. (1990). Root growth and distribution of two short-season rice genotypes. Plant Soil 126, 269–278. Soil Science Society of America. (2008). Glossary of Soil Science Terms. ASA, Madison, WI. Sponchiado, B. N., White, J. W., Castillo, J. A., and Jones, P. G. (1989). Root growth of four common bean cultivars in relation to drought tolerance in environments with contrasting soil types. Exp. Agric. 25, 249–257. St. Aubin, G., Canny, M. J., and McCully, M. E. (1986). Living vessel elements in the late metaxylem of sheathed maize roots. Ann. Bot. 58, 577–588. Steponkus, P. J., Cutler, J. M., and O’Toole, J. C. (1980). Adaptation to water stress in rice. In “Adaptation of Plants to Water and High Temperature Stress” (N. C. Turner and P. J. Kramer, Eds.), pp. 401–418. John Wiley & Sons, New York. Stoffella, P. J., and Kahn, B. A. (1986). Root system effects on lodging of vegetable crops. HortScience 21, 960–963. Stoffella, P. J., Sandsted, R. F., Zobel, R. W., and Hynes, W. L. (1979). Root characteristics of some black beans. I. Relationship of root size to lodging and seed yield. Crop Sci. 19, 823–826.

The Role of Mineral Nutrition on Root Growth of Crop Plants

329

Stone, L. F., and Pereira, A. L. (1994a). Rice-common bean rotation under sprinkler irrigation: Effects of row spacing, fertilization and cultivar on growth, root development and water consumption of common bean. Pesqui. Agropecu. Bras. 29, 939–954. Stone, L. F., and Pereira, A. L. (1994b). Rice-common bean rotation under sprinkler irrigation: Effects of row spacing, fertilization and cultivar on growth, root development and water consumption of rice. Pesqui. Agropecu. Bras. 29, 1577–1592. Stone, L. R., Goodrum, D. E., Jaafar, M. N., and Khan, A. H. (2001). Rooting front and water depletion depths in grain sorghum and sunflower. Agron. J. 93, 1105–1110. Takenaga, H. (1995). Internal factors in relation to nutrient absorption. In “Science of the Rice Plant” (T. Matsuo, K. Kumazawa, R. Ishii, K. Ishihara, and H. Hirata, Eds.), Vol. 2, pp. 294–309. Food and Agriculture Policy Research Center, Tokyo. Talouizte, A. M., Champigny, M. L., Bismuth, E., and Moyse, A. (1984). Root carbohydrate metabolism associated with nitrate assimilation in wheat previously deprived of nitrogen. Physiol. Veg. 22, 19–27. Taylor, H. M., and Ratliff, L. F. (1969). Root elongation rates of cotton and peanuts as a function of soil strength and soil water content. Soil Sci. 108, 113–119. Taylor, H. M., Burnett, E., and Booth, G. D. (1978). Taproot elongation rates of soybeans. Z. Acker Pflanzenbau 146, 33–39. Tennant, D. (1976). Root growth of wheat. I. Early patterns of multiplication and extension of wheat roots including effects of leaves of nitrogen, phosphorus, and potassium. Aust. J. Agric. Res. 27, 183–196. Teo, Y. H., Beyrouty, C. A., and Gbur, E. E. (1995). Evaluation of a model to predict nutrient uptake by field grown rice. Agron. J. 87, 7–12. Texier, M., and Biles, G. (1990). The role of rhizosphere on C and N cycles in a plant–soil system. Symbiosis 9, 117–123. Thangaraj, M., O’Tolle, J. C., and De Datta, S. K. (1990). Root response to water stress in rainfed lowland rice. Exp. Agric. 26, 287–296. Thomas, A. L., and Kaspar, T. C. (1997). Maize nodal root response to time of soil ridging. Agron. J. 89, 195–200. Tian, G., Carsky, R. J., and Kang, B. T. (1998). Differential phosphorus responses of leguminous cover crops on soils with variable history. J. Plant Nutr. 21, 1641–1653. Tresder, K. K., Morris, S. J., and Allen, M. F. (2005). The contribution of root exudates, symbionts, and detritus to carbon sequestration in the soil. In “Roots and Soil Management: Interactions Between Roots and the Soil” (R. W. Zobel and S. F. Wright, Eds.), pp. 145–162. ASA, CSSA, and SSSA, Madison, WI. Troughton, A. (1962). The roots of temperate cereals (wheat, barley, oats and rye). Mimeographed Publication No. 2 Commonwealth Bureau of Pasture and Field Crops, Hurley, Berkshire. Tupper, G. R. (1992). Technologies to solve K deficiency: Deep placement. In “1992 Proceedings of the Beltwide Cotton Production Research Conference,” pp. 73–76. National Cotton Council of America, Memphis, TN. Ueda, S. (1936). On the development of root system of wheat and barley in sandy soil. Proc. Crop Sci. Soc. Jpn. 8, 140–143. Unger, P. W., and Kaspar, T. C. (1994). Soil compaction and root growth: A review. Agron. J. 86, 759–766. University of Arkansas Cooperative Extension Service Rice Committee. (1990). Rice Production Handbook. University of Arkansas Cooperative Extension Service Miscellaneous Publication No. 192 University of Arkansas, Little Rock, AR. Vartanian, N. (1981). Some aspects of structural and functional modifications induced by drought in root systems. Plant Soil 63, 83–92. Vincent, C. D., and Gregory, P. J. (1986). Differences in the growth and development of chickpea seedling roots (Cicer arietinum). Exp. Agric. 22, 233–242.

330

N. K. Fageria and A. Moreira

Waals, J. H. V., and Laker, M. C. (2008). Micronutrient deficiencies in crops in Africa with emphasis on southern Africa. In “Micronutrient Deficiencies in Global Crop Production” (B. J. Alloway, Ed.), pp. 201–224. Springer, New York. Wahbi, A., and Gregory, P. J. (1995). Growth and development of young roots of barley (Hordeum vulgare L.) genotype. Ann. Bot. 75, 533–539. Wardlaw, I. A. (1990). The control of carbon partitioning in plants. New Phytol. 116, 341–381. Webb, J., and Jackson, M. B. (1986). A transmission and cryo-scanning electron microscope study of the formation of aerenchyma (cortical gas filled space) in adventitious roots of rice (Oryza sativa L.) J. Exp. Bot. 37, 832–841. Wedin, D. A., and Tilman, D. (1990). Species effects on nitrogen cycling: A test with perennial grasses. Oecologia 84, 433–441. Welbank, P. J., and Williams, E. D. (1968). Root growth of a barley crop estimated by sampling with portable powered soil coring equipment. J. Appl. Ecol. 5, 447. Welch, R. M. (2008). Linkages between trace elements in food crops and human health. In “Micronutrient Deficiencies in Global Crop Production” (B. J. Alloway, Ed.), pp. 287–317. Springer, New York. Welch, R. M., Allaway, W. H., House, W. A., and Kubota, J. (1991). Geographic distribution of trace element problems. In “Micronutrients in Agriculture” (J. J. Mortvedt, F. R. Cox, L. M. Shuman, and R. M. Welch, Eds.), 2nd edn., pp. 31–57. Soil Science Society of America, Madison, WI. Wenkert, W., Fausey, N. R., and Watters, H. D. (1981). Flooding Responses in Zea mays L. Plant Soil 62, 351–366. Werf, A. V. D. (1996). Growth analysis and photoassimilate partitioning. In “Photoassimilate Distribution in Plants and Crops” (E. Zamski and A. A. Schaffer, Eds.), pp. 1–20. Marcel Dekker, New York. Weston, L. A., and Zandstra, B. H. (1989). Transplant age and N and P nutrition effects on growth and yield of tomatoes. HortScience 24, 88–90. White, J. W., and Castillo, J. A. (1989). Relative effect of root and shoot genotypes on yield of common bean under drought stress. Crop Sci. 29, 360–362. Wiesler, F., and Horst, W. J. (1993). Differences among maize cultivars in the utilization of soil nitrate and the related losses of nitrate through leaching. Plant Soil 151, 193–203. Wilson, A. J., Robards, A. W., and Goss, M. J. (1977). Effects of mechanical impedance on root growth in barley, Hordeum vulgare L.: II. Effects on cell development in seminal roots. J. Exp. Bot. 28, 1216–1227. Wilts, A. A., Reicosky, D. C., Allmaras, R. R., and Clapp, C. E. (2004). Long-term corn residues effects; Harvest alternatives, soil carbon turnover, and root-derived carbon. Soil Sci. Soc. Am. J. 68, 131342–131351. Wulfsohn, D., Gu, Y., Wulfsohn, A., and Mojlaj, E. G. (1996). Statistical analysis of wheat root growth patterns under conventional and no-till systems. Soil Tillage Res. 38, 1–16. Xiaoe, Y., Romheld, V., Marschner, H., Baligar, V. C., and Martens, D. C. (1997). Shoot photosynthesis and root growth of hybrid and conventional rice cultivars as affected by N and K levels in the root zone. Pedosphere 7, 35–42. Yamauchi, A., Kono, Y., and Tatsumi, J. (1987a). Qualitative analysis of root system structure of upland rice and maize. Jpn. J. Crop Sci. 56, 608–617. Yamauchi, A., Kono, Y., and Tatsumi, J. (1987b). Comparison of root system structures of 13 species of cereals. Jpn. J. Crop Sci. 56, 618–631. Yibrin, H., Johnson, J. W., and Eckert, D. J. (1993). No-till corn production as affected by mulch, potassium placement, and soil exchangeable potassium. Agron. J. 85, 639–644. Yoshida, S. (1981). Fundamentals of Rice Crop Science. IRRI, Los Bano˜s, Philippines.

The Role of Mineral Nutrition on Root Growth of Crop Plants

331

Yoshida, S., and Hasegawa, S. (1982). The rice root system: Its development and function. In “Drought Resistance of Crops with Emphasis on Rice” (IRRI, Ed.), pp. 97–114. IRRI, Los Bano˜s, Philippines. Zak, D. R., and Pregitzer, K. S. (1998). Integration of ecophysiological and biogeochemical approaches to ecosystem dynamics. In “Success, Limitations, and Frontiers in Ecosystem Science” (M. L. Pace and P. M. Groffman, Eds.), pp. 372–403. Springer, New York. Zhang, J., and Barber, S. A. (1992). Maize root distribution between phosphorus fertilized and unfertilized soil. Soil Sci. Soc. Am. J. 56, 819–822. Zhang, J., and Barber, S. A. (1993). Corn root distribution between ammonium fertilized and unfertilized soil. Commun. Soil Sci. Plant Anal. 24, 411–419. Zhao, Z. R., Li, G. R., and Hung, G. Q. (1991). Promotive effect of potassium on adventitious root formation in some plants. Plant Sci. 79, 47–50. Zhao, Y., Xiao, X., Bi, D., and Hu, F. (2008). Effects of sulfur fertilization on soybean root and leaf traits, and soil microbial activity. J. Plant Nutr. 31, 473–483. Zobel, R. W. (1986). Rhizogenetics (root genetics) of vegetable crops. HortScience 21, 956–959. Zobel, R. W. (1991). Root growth and development. In “The Rhizosphere and Plant Growth” (D. L. Keister and P. B. Cregan, Eds.), pp. 61–71. Kluwer Academic Publishers, Dordrecht, The Netherlands. Zobel, R. W. (2003). Sensitivity analysis of computer based diameter measurement from digital images. Crop Sci. 43, 583–591. Zobel, R. W. (2005a). Primary and secondary root systems. In “Roots and Soil Management: Interactions Between Roots and the Soil” (R. W. Zobel and S. F. Wright, Eds.), pp. 3–14. ASA, CSSA, and SSSA, Madison, WI. Zobel, R. W. (2005b). Tertiary root systems. In “Roots and Soil Management: Interactions Between Roots and the Soil” (R. W. Zobel and S. F. Wright, Eds.), pp. 35–56. ASA, CSSA, and SSSA, Madison, WI. Zou, C., Gao, X., Shi, R., Fan, X., and Zhang, F. (2008). Micronutrient deficiencies in crop production in China. In “Micronutrient Deficiencies in Global Crop Production” (B. J. Alloway, Ed.), pp. 127–148. Springer, New York.

C H A P T E R

F I V E

Physiology of Spikelet Development on the Rice Panicle: Is Manipulation of Apical Dominance Crucial for Grain Yield Improvement? Pravat K. Mohapatra,* Rashmi Panigrahi,* and Neil C. Turner† Contents 1. Introduction 2. Panicle Structure and Development 2.1. The morphology of spikelets 2.2. Ontogeny 2.3. Heterogeneous development of spikelets 3. Panicle Architecture and Grain Yield 3.1. Heterogeneous architecture of the panicle and its relationship to grain yield 3.2. Modification of panicle morphology for improved grain filling 4. Physiological Factors Regulating Spikelet Development 4.1. Metabolic control of assimilate partitioning 4.2. Starch biosynthesis 4.3. Expression profile of the genes regulating the activity of starch biosynthesis enzymes 4.4. Hormonal regulation of spikelet development 5. Is Manipulation of Apical Dominance Crucial for Grain Yield Improvement? 6. Suggestions for Modification of Apical Dominance Acknowledgments References

334 335 335 335 336 337 337 338 340 340 341 344 345 348 351 352 352

Abstract Cultivated rice, Oryza sativa L., originated in the flood plains of Asia with their variable environmental conditions. Heterogeneous architecture, leading to intergrain apical dominance in spikelet development, was an important strategy * School of Life Science, Sambalpur University, Sambalpur, India { Centre for Legumes in Mediterranean Agriculture and UWA Institute of Agriculture, The University of Western Australia, Crawley, WA, Australia Advances in Agronomy, Volume 110 ISSN 0065-2113, DOI: 10.1016/B978-0-12-385531-2.00005-0

#

2011 Elsevier Inc. All rights reserved.

333

334

Pravat K. Mohapatra et al.

for plant survival under such uncertain conditions. When inclement weather conditions coincide with the sensitive stage of spikelet development, the plant sacrifices some spikelets while preserving the rest to enable some grains to fill to provide seed for the next generation. The nature of the interaction between a genotype and its environment determines the ultimate spikelet number. This heterogeneous architecture was considered a liability for the increased demand for food. Innovative breeding efforts have changed the plant type in favor of extra-heavy panicles bearing a large number of spikelets, mostly in a homogeneous distribution. However, the reduced intergrain apical dominance decreased the strength of the sink for assimilate partitioning to the spikelets, much to the detriment of increased grain production. The metabolic control of assimilate partitioning and the genes controlling starch synthesis are reviewed and the importance of hormonal control of grain filling is emphasized. A model has been constructed to project the role of metabolites, hormones, and other factors in intergrain apical dominance. The panicle architecture of rice is amenable to modification by both intrinsic and extrinsic factors, and there is extensive variation of panicle structure among genotypes. Therefore, we suggest that an architecture with controlled intergrain apical dominance, amounting to production of a greater number of responsive sinks, which is a compromise between heterogeneous and homogeneous architecture of the inflorescence, should benefit spikelet filling in rice with large panicles.

1. Introduction Rice is the staple food for half of the human race. Most rice consumers live in the tropics and subtropics of Asia, where in many cases, the irrigation infrastructure cannot support the year-round cultivation of any major cereal. While heavy monsoonal rains, usually confined to only 2–3 months of the year, allow rice cultivation in the region, cropping is limited to rice because the soils become too waterlogged for other crops. Man and domestic animals thrive on the success of rice cultivation, unless the monsoon is fickle and the crop suffers from high abiotic stress. According to Virk et al. (2004), global rice yield, which is 645 million tonnes today (FAOSTAT, 2010), must reach 800 million tonnes in 2025 to meet the demand for rice consumption. This extra rice will have to come from improvements in irrigated rice because the environmental stresses of rainfed rice are hard to overcome (Cassman, 1999). Currently, irrigated rice occupies just over 57% of the 150 million ha of land under rice production (International Rice Research Institute, 2010), but contributes 76% of the global production. To meet the challenge and boost production, the United Nations declared rice as the crop of the year in 2004; major attention was given to improving the production of irrigated rice. The average yield of irrigated rice will have to increase from 4 to 8.5 tonnes ha 1 to meet this challenge (Peng et al., 1999).

Apical Dominance and Grain Yield of Rice

335

Based on the total amount of incident solar radiation, Yoshida (1981) estimated that the theoretical yield potential of rice is 15.9 tonnes ha 1, but the vagaries of climate scale the yield potential back to 8.5 tonnes ha 1 (Matthews et al., 1995). Ample scope exists for achieving this target, provided researchers successfully pyramid all the known beneficial traits in a new plant type (NPT).

2. Panicle Structure and Development 2.1. The morphology of spikelets The rice plant bears a panicle with determinate inflorescences on the terminal end of the shoot (De Datta, 1981). The panicle bears a central axis (rachis) which extends from the base (neck node) to the tip of the axis, indicated by a minor swelling at the base of the terminal primary branch (Fig. 1). The axis has 8–10 nodes at intervals of a few centimeters from which primary branches emerge (Fig. 1). Secondary branches develop on the primary branches in similar fashion (Fig. 1). Pedicels develop from the nodes of primary and secondary branches; each pedicel culminates with a spikelet at the terminal end (Yoshida, 1981). Unlike other cereals, spikelets are considered the individual floral units of the complex inflorescence of rice. Each spikelet bears one grain at maturity (Xu and Vergara, 1986). A spikelet consists of two sterile lemmas, the rachila, and the fertile floret (De Datta, 1981). The rachila is the small axis between the fertile floret and the sterile lemmas. The fertile floret bears a pistil with uniovulate ovary and feathery and bifid stigma. The androecium has six two-celled anthers borne on slender filaments. The androecium and gynoecium are enclosed in the space between two hardened bracts named lemma and palea; the former is five-nerved and bears an awn compared to the awnless three-nerved palea.

2.2. Ontogeny The initiation of panicle primordia occurs at the tip of the vegetative stem apex inside the boot of the flag leaf sheath 30 days before flowering. The dome of the vegetative apex elongates to mark the change to reproductive development. Neck node differentiation is the first step of panicle development and the event closes with maturation of pollen grains at the time of flowering. Initially, the young panicle remains shrouded with colorless hairs; it becomes visible after becoming about 1-mm long (Matsushima, 1970; Xu and Vergara, 1986; Yoshida, 1981). As the panicle axis extends upward and reaches a length of 50 mm, spikelet primordia are differentiated on the primary and secondary branches (De Datta, 1981) and the total number of

336

Pravat K. Mohapatra et al.

Group I Group II Group III Group IV Group V Group VI Group VII

Spikelet

Secondary branch Primary branch

Panicle axis

Figure 1 Schematic of the rice panicle showing the panicle, primary branches, secondary branches, and the spikelets that flowered each day for 7 days (Group I on day 1 to Group VII on day 7). Groups I–III are considered superior spikelets and Groups IV–VII are considered inferior spikelets. From Mohapatra et al. (1993), with permission.

spikelets in the panicle is determined. The coincidence of inclement weather, resulting in any form of environmental stress at this crucial stage, limits spikelet number, and ultimately, the grain number and yield at maturity.

2.3. Heterogeneous development of spikelets During ontogeny, the spikelet primordia differentiate on the primary and secondary branches of the rice panicle (Xu and Vergara, 1986). Initiation of primary branches on the panicle axis, secondary branches on primary

Apical Dominance and Grain Yield of Rice

337

branches, and spikelets on both the branches occur in acropetal succession from the base to the top of the panicle (Fig. 1; Murata and Matsushima, 1975; Xu and Vergara, 1986). However, the development of these organs does not follow the same order. The upper branches and spikelets (Groups I–III in Fig. 1) emerge from the flag leaf sheath (boot) first and are the first released from the physical constraints of the flag leaf enclosure. Contact with light encourages faster development of these apical branches and spikelets, while the basal spikelets (Groups IV–VII in Fig. 1) and branches develop more slowly while they are still within the sheath of the flag leaf. Therefore, primary branch development is progressively delayed in the basipetal direction, while initiation is the opposite (Xu and Vergara, 1986). Lack of homogeneity in development has led to a wide variation in individual grain weight of the spikelets across the rice panicle. Frequently, many spikelets on the basal primary branches do not bear a mature grain. The poorly or partially filled basal grains are also of poor quality and low market value (Mohapatra et al., 1993; Venkateswarlu et al., 1986a,b).

3. Panicle Architecture and Grain Yield 3.1. Heterogeneous architecture of the panicle and its relationship to grain yield Variation in grain weight and quality within a panicle is dependent on panicle architecture. The number of poorly filled grains increases significantly when the panicle size is large because most of the grains on the spikelets on the secondary branches of the panicle become source-limited for assimilates (Kato, 2004). It is possible that poor translocation and partitioning of assimilates from the source leaves and stems at crucial stages of grain filling fails to sustain development of a large number of spikelets (Yang et al., 2002a). However, starch synthesis in the endosperm cells of the spikelets on secondary branches is poor (Umemoto et al., 1994) and assimilates partitioned to them remain unused so that the sink may be the impediment to the transport and storage of assimilates. Spikelets located on the upper primary branches that flower early exert dominance, accumulate larger amounts of starch, and produce better quality grains than lateflowering spikelets on secondary branches (Ishimaru et al., 2003; Mohapatra et al., 1993; Yang et al., 2006a). It has been speculated that in rice, improvement of panicle size due to increased sink strength would induce greater source activity and biomass production (Toenniessen, 1991). The nature of the source–sink relationship in dry matter partitioning to the panicle from source leaves has been worked out by partial removal of competing sink organs from the developing panicle in contrasting rice cultivars. Using the stable isotope of carbon, Mohapatra et al. (2004) reported that the growth

338

Pravat K. Mohapatra et al.

and development of the spikelets of the high-yielding rice cultivar “Takanari” was sustained by photosynthates from the upper three leaves, whereas in the traditional rice cultivar “Nakateshinsenbon,” spikelet growth was supported by assimilates from the flag leaf only. Nevertheless, at present, the relationship between photosynthesis and grain filling is poorly understood. The so-called NPT of rice (Dingkuhn et al., 1991; Sheehy et al., 2001; Virk et al., 2004) and hybrid rice (Zhu et al., 1997) has been developed for increased biomass production with higher carbon assimilation capacity under conditions of high radiation and temperature in the tropics (Khush, 1996). In spite of this increased source activity, many spikelets remain unfilled or poorly filled in the extra-heavy panicles of these newly improved cultivars. Murchie et al. (2002) showed that there was no association between grain-filling rate and light-saturated photosynthesis in any of the five NPT lines and one indica rice studied, while Scoefield et al. (2002) reported that antisense suppression of the rice sucrose transporter gene that impaired grain filling did not affect photosynthesis.

3.2. Modification of panicle morphology for improved grain filling In cereals, the number of spikelets is one of the major components of grain yield (Feil, 1992). An increase in the size of the panicle by increasing spikelet number enhances spikelet number per unit area in the field as well as the yield potential of the crop (Peng et al., 1994). Because individual kernel weight and panicle number per plant in rice appear to be genetically fixed (Luo et al., 2001; Zhikang et al., 1997) and stable for decades (Evans et al., 1984), seed number per panicle is believed to be amenable to genetic modification. Plasticity in spikelet/seed number per panicle has been reported under a range of agroclimatic conditions in which rice is grown (Kropff et al., 1994). Using an empirical model of yield, Sheehy et al. (2001) predicted that the grain yield potential of rice would double if all of the juvenile spikelets were transformed into grains. An improved reproductive sink capacity along with an extended grain-filling period were two of the traits used in the rice breeding program to improve the semidwarf plant type and develop a new plant ideotype (Dingkuhn et al., 1991). Compared to indica types, the NPTs used tropical japonica germplasm to increase the size of the panicles (Peng et al., 1994). Similarly, intercrossing between japonica and indica rices resulted in hybrids bearing panicles larger than conventional inbred rice (Zhu et al., 1997). However, an increase in the number of spikelets does not always result in high grain number and a yield benefit at maturity because the degree of grain filling of individual spikelets depends on the position of the spikelet within a panicle (Kato, 2006). Spikelets located on distal parts and the primary branches of the panicle (Groups I–III in Fig. 1, hereafter called superior spikelets) possess

Apical Dominance and Grain Yield of Rice

339

grain-filling attributes that are better than those spikelets located on proximal parts and/or secondary branches (Groups IV–VII in Fig 1, hereafter called inferior spikelets; Ishimaru et al., 2003; Mohapatra et al., 1993). Kato (2006) used a panicle centroid index, which is a mathematical index to represent distribution of spikelets on various primary branches of the panicle axis, as a criterion for assessing grain-filling efficiency of four extra-heavy and two non-extra-heavy panicle types of rice cultivars by comparing proximally versus distally laid spikelets of the panicle. Cultivars possessing a higher panicle centroid index, that is more superior spikelets, also showed a higher proportion of high-density (specific gravity) grains in the panicle. However, the number of spikelets on secondary branches of the panicle exhibited a stronger relationship with grain filling than that of the panicle centroid index. The increase in the number of spikelets, on secondary branches was found to contribute to the increase in the total number of spikelets per panicle in extra-heavy panicle types. Padmaja Rao (1987) instead used the percentage of high-density grains on the lowermost primary branch of the panicle, which is the least favored destination for assimilates, as an index for screening the grain-filling potential of a cultivar. The panicle architecture of traditional rice is simple. Most of the spikelets are borne on the primary branches and the percentage of filled grains on the panicle is high, but the overall grain yield of the panicle is poor because of a low spikelet number. Breeding for increased spikelet number per panicle was successful in many cultivars, but the extra spikelets were mostly located on secondary branches of the panicle (Kudo, 1991; Yamamoto et al., 1991), and as these spikelets do not fill as efficiently as the spikelets on primary branches, the objective of achieving high-yield potential was not realized (Kato, 1993). A comparison of the percentage contribution of filled, half-filled, and empty spikelets to the total spikelets at various spatial locations of the panicle of indica inbred, indica hybrid, and the NPT rice cultivars confirmed this (Khush and Peng, 1996); the spikelet number in the NPT is higher than other types, but the number of filled grains did not increase (Yang and Zhang, 2010). Similarly, Yang et al. (2000) reported that the indica inbred cultivar IR72 exhibited a high rate of grain filling, whereas some of the NPT rices derived with tropical japonica background did not. The grain number per panicle in rice is a function of panicle length and grains per unit length of panicle (Sheehy et al., 2001). However, increasing the panicle size or height could have a detrimental effect on light interception and the rate of photosynthesis of the source leaves which are positioned beneath the panicle and supply assimilates to the grains during development (Setter et al., 1996; Xu et al., 1990). Therefore, breeders of the extra-heavy super rice have selected for more spikelets per unit length of panicle to increase spikelet number per panicle without any appreciable increase in panicle length. This increased compactness of the panicle allows large number spikelets to be accommodated on the panicle. However, the

340

Pravat K. Mohapatra et al.

increased compactness reduces the grain quality of the spikelets on secondary branches as a result of increased variation in the length, width, length–width ratio, chalky grain percentage, and amylose content of the grains on the rice panicle (Wang et al., 2008). On the other hand, the spikelet development of rice with lax panicles is not impeded by space restrictions and the cultivars produce grains with larger biomass (Panda et al., 2009). Lack of freedom from either spatial confinement on the panicle axis (compact panicle) or temporal confinement inside the flag leaf sheath (late emergence of basal spikelets) affects spikelet development and results in the production of poor-quality grains.

4. Physiological Factors Regulating Spikelet Development Determination of seed number is an important component of grain yield in all grain crops. Therefore, it is important to understand the relationship between the number of spikelets when they differentiate, and the number of grains at maturity in the panicle of rice. The physiological significance of the processes of production of a large number of reproductive units (spikelets) on the flowering axis (panicle) during ontogeny and elimination of many of them prior to grain maturity, especially in the rice with heavy panicles, has not received enough attention by rice scientists. The survival of spikelets in a cultivar may be related to the degree of interaction between the biological and environmental factors that control partitioning of assimilates to the developing spikelets. Cultivated rice was domesticated on the undulating flood plains of Asia, resulting in both drought and floods (Oka, 1991). The production of an excess of reproductive structures that could be sacrificed depending on the seasonal conditions in the adaptation of rice to its environment may be responsible for the source–sink relations and assimilate partitioning of the present-day rice plant. Several authors have examined the source–sink relations of rice by assessing the metabolic and hormonal control mechanisms of spikelet development, as will be discussed in the next sections.

4.1. Metabolic control of assimilate partitioning Increased partitioning of assimilates into the reproductive organs has contributed significantly to the improvement of the yield potential of crops (Gifford and Evans, 1981). Similarly in rice, there has been a marked increase in harvest index and grain production by modification of the reproductive structures in the modern cultivars, but the source activities like photosynthetic rate and crop growth rate have remained unchanged for decades (Evans et al., 1984). Efforts undertaken to increase biomass production or

Apical Dominance and Grain Yield of Rice

341

the photosynthetic rate of rice have not contributed to any increase in yield potential (Vergara, 1987). Intensive study is necessary to improve the assimilate partitioning in favor of spikelets on the secondary branches and raise the number of high-density grains in the panicle in order to achieve a high yield potential (Padmaja Rao et al., 1985; Vergara, 1987; Vergara et al., 1990) and break the yield barrier of irrigated rice (Khush and Peng, 1996). Nishiyama (1983) observed wide variation in the cross-section of the phloem in the pedicel of spikelets positioned on different parts of the rice panicle. Spikelets bearing high-density grains possessed larger phloem (Nishiyama, 1983) and better developed vascular bundles (Chaudhry and Nagato, 1970). The situation was similar in the panicles of cultivars with extra-heavy panicles that had better developed vascular bundles and phloem. Such variation in vascular structures could lead to differences in assimilate supply between the developing spikelets, resulting in heterogeneous growth and formation of qualitatively different types of grains in the rice panicle (Padmaja Rao et al., 1985). However, the size of the phloem and vascular bundles is not the only constraint to grain filling, as an increased number of large vascular bundles entering the base of primary branches ensured better grain filling in some cultivars, but not in others (Kato, 2008). Sikder and Das Gupta (1976) suggested that growth is reduced in the spikelets on secondary branches because of unsuccessful competition for the limited pool of assimilates, while Egli and Bruening (2003) proposed that assimilate supply was the dominant factor determining seed set and grain number in crops. However, the possibility that supply of substrates discriminates between the spikelets on the secondary and primary branches of the panicle has been discounted as a cause of variation in spikelet growth and development by Mohapatra and Sahu (1991) and Yang and Zhang (2010). Vergara (1987) also reported that many spikelets in the rice panicle do not develop into high-density grains despite the presence of sufficient carbohydrates. Because unused assimilate accumulates in the inferior basal spikelets of the rice panicle, they exhibit higher callusing ability in the anther culture than the superior apical spikelets (Afza et al., 2000). Thus, it is necessary to improve the sink efficiency of the spikelets on the secondary branches and maximize the use of available assimilates to attain the yield potential of rice (Mohapatra et al., 1993). While an increase of primary branches of the panicle would be most advantageous for increasing the number of high-density grains (Padmaja Rao, 1989), there is scope for enhancing grain filling of the spikelets located on the secondary branches by manipulation of assimilate utilization.

4.2. Starch biosynthesis While starch contributes 80–90% of the final dry weight of an unpolished grain (Yoshida, 1972), sucrose is the major phloem solute which is transported to the developing rice seeds. Subsequent to loading in the source

342

Pravat K. Mohapatra et al.

organ, sucrose moves by a hydrostatic pressure-gradient-induced mass flow to reach the sink. The processes of loading and unloading of the phloem sucrose are energy-dependent in rice. At the sink, unloading of sucrose depends on sink strength converting the material first into monosaccharides and later into starch. Loading of sucrose into the phloem at the source and its unloading at the sink are dependent on the proton motive force generated by HþATPase. HþATPase hydrolyses ATP to drive the sucrose–Hþ symporter (Taiz and Zeiger, 2002). Colocalization of both the transporters in the plasma membrane of sieve elements, the nucellus of pericarp and the aleurone layer of the endosperm (Furbank et al., 2001; Langhans et al., 2001), confirms close coupling of action in the transport of sucrose across the cell membrane from symplast to apoplast and vice versa. While sucrose– Hþ symporter (OsSUT1) controls loading into the phloem at the source, coordinated regulation of the symporter with cell wall invertases and monosaccharide transporters is often essential in the sink tissue for unloading and storage (Lim et al., 2006). Kuanar et al. (2010) measured the concentration of assimilates in the apoplasmic space of the developing rice caryopsis and found that it was positively correlated with endosperm cell number and grain weight. It was suggested that the concentration of apoplasmic assimilates could be an indicator of the sink-filling capacity of the caryopsis. Grain filling is impaired when the action of the sucrose transporter gene, OsSUT1, is blocked by antisense suppression (Scoefield et al., 2002). Sucrose partitioned into the endosperm is stored in the form of starch through a pathway of starch biosynthesis which is unique in cereal endosperm. The enzyme isoforms required for starch biosynthesis in rice are not found in other cereal tissues or noncereal plants ( James et al., 2003). Additionally, optimal activity of enzymes processing the incoming sucrose for starch synthesis is required for grain filling. Grain filling and the grain quality of the inferior spikelets of the rice panicle can be affected by the activity of the enzymes of sucrose metabolism in the endosperm, and the enzyme action can determine the sink strength of the endosperm. There are 33 enzymes involved in the carbohydrate metabolism of developing rice endosperm (Nakamura et al., 1989). However, only four of them, namely, sucrose synthase (EC 2.4.1.13 SUS), adenine diphosphoglucose pyrophosphorylase (EC 2.7.7.27, AGPase), starch synthase (EC 2.4.1.21 StSase), and starch branching enzyme (EC 2.4.1.18, BE), are believed to play key roles in the processes that determine starch quality and grain filling (Umemoto et al., 1994; Yang et al., 2001a). Intensive research on the function of individual enzymes has provided an insight into how the amylose and amylopectins of rice starch are synthesized and distributed, and ultimately how grain quality is determined. SUS acts on the incoming sucrose as the first step of starch synthesis (Kato, 1995; Patel and Mohapatra, 1996; Perez et al., 1975). In rice, endosperm SUS activity is

Apical Dominance and Grain Yield of Rice

343

considered as a potential indicator of high grain yield (Counce and Gravois, 2006). Activity of the enzyme is poor in the inferior spikelets compared to the superior spikelets (Patel and Mohapatra, 1996), and it is positively linked to the strength of the developing grain to accept sucrose (Liang et al., 2001; Naik and Mohapatra, 2000; Yang et al., 2003) and convert it into fructose and uridine diphosphate glucose. Although sucrose can be catalyzed into glucose and fructose by the action of invertase, SUS is the predominant enzyme during rice endosperm development (Patel and Mohapatra, 1996). UDPglucose is further metabolized to glucose-1-phosphate by phosphorylase. Glucose-1-phosphate becomes the substrate for AGPase action (Vandeputte and Delcour, 2004). AGPase catalyzes the conversion of glucose-1-phosphate and ATP into ADPglucose and inorganic pyrophosphate. Formation of the former compound is the first committed step of the starch biosynthesis pathway. Smidansky et al. (2003) reported that increased endosperm AGPase activity stimulates seed setting in the inferior spikelets of rice, increasing the grain yield by 20%, while Kato et al. (2007) suggested that sink-directed phloem transport of sucrose and improved grain filling is promoted by high activity of AGPase in the developing endosperm. The glucose moiety of ADPglucose is added to extend the chain length of amylose or amylopectin, the two major components of the starch, by the establishment of a-1, 4-glucosidic linkage to the nonreducing end of the aglucan primer by the action of soluble- and granule-bound forms of starch synthase (SS; Nakamura and Yuki, 1992; Nakamura et al., 1989). The elongation reactions for the a-1, 4 chain of amylose and amylopectin are controlled by a granule-bound form of SS and a soluble form of SS, respectively. Greater emphasis has been given to the activity of the former compared to the latter in the regulation of grain development of rice, because the amylose content and grain quality of superior spikelets are much higher than their inferior counterparts (Matsue et al., 1995). Baun et al. (1970) found that the granule-bound SS may be responsible for the integrity of amylose in the developing starch granule; the enzyme increases its activity for 21 days after flowering, equivalent to the grain-hardening stage. Similarly, Umemoto and Terashima (2002) reported that the activity of granule-bound SS is an important determinant of amylose content for low amylose cultivars, but not for high amylose cultivars. The activity of the enzyme is poor in the inferior caryopses compared to the superior caryopses of the rice panicle, and the low activity of the enzyme is related to reduced amylose content of the former (Umemoto et al., 1994). Similarly, japonica rice that has a better grain-filling ability exhibits higher specific activity of the enzyme compared to indica rice. Recent Chinese research has also revealed that poor expression of the granule-bound SS enzyme is responsible for low grain amylose content and that it can be improved by chemical mutation ( Jeng et al., 2007). The role of three more enzymes in starch biosynthesis, starch debranching enzyme, starch phosphorylase, and starch

344

Pravat K. Mohapatra et al.

disproportionating enzyme (discussed in Section 4.3), has not been studied in relation to the differential pattern of grain filling between inferior and superior spikelets.

4.3. Expression profile of the genes regulating the activity of starch biosynthesis enzymes The expression pattern of both SUS and AGPase genes is lower in the developing inferior spikelets compared to the superior spikelets of the rice panicle during kernel development (Ishimaru et al., 2005). Current research indicates that six genes comprise the rice SUS gene family (Hirose et al., 2008). While SUS1 and SUS2 are more involved in the regulation of vegetative growth, SUS3 and SUS4 are exclusively responsible for partitioning of assimilates to the caryopsis during grain filling; the functions of the last two genes are not known. There are 27 genes controlling the activity of the other six enzymes in the starch synthesis pathway, namely, AGPase, SS, starch branching enzyme, starch debranching enzyme, starch phosphorylase, and starch disproportionating enzyme; the mode of expression of the genes is tissue- and developmental stage-specific (Ohdan et al., 2005). The expression profile of the genes is divided into four groups. Group 1 genes are expressed very early during grain development and involve synthesis of glucan primers for initiation of starch biosynthesis, group 2 genes are expressed throughout endosperm development, group 3 genes are initiated when starch biosynthesis occurs in the endosperm, and group 4 genes have low activity and control starch biosynthesis of the pericarp. The first enzyme, AGPase, is heterotetrameric comprising two large and two small subunits encoded by six genes in total (Ohdan et al., 2005). Two genes, OsAGPS1 and OsAGPS2, code for the small subunits and four genes, OsAGPL1, OsAGPL2, OsAGPL3, and OsAGPL4, code for the large subunits. There are two types of granule-bound SS, namely, GBSSI and GBSSII (Ohdan et al., 2005). Amylose synthesis in the rice grain is dependent on the activity of GBSSI (Vandeputte and Delcour, 2004). GBSSI gene expression reaches its peak level at 5 days after flowering and remains high until maturity. It therefore dominates starch biosynthesis in the endosperm. In contrast, the expression of the GBSSII gene is very poor throughout starch biosynthesis. Soluble-SS is classified into four types, namely, starch synthase I (SSI), starch synthase II (SSII), starch synthase III (SSIII), and starch synthase IV (SSIV) (Hirose and Terao, 2004; Kosar-Hashemi et al., 2007; Ohdan et al., 2005). In total, eight genes control the expression of these isoforms of SS in rice (Ohdan et al., 2005). SSI is the major form of SS, its activity is very high during the first few days after flowering and remains high up to the grain mid-filling stage; one gene (OsSSI) regulates the expression of this enzyme. SSII is coded by three genes, (OsSSIIa, OsSSIIb, OsSSIIc), while SSIII

Apical Dominance and Grain Yield of Rice

345

(OsSSIIIa, OsSSIIIb) and SSIV (OsSSIVa, OsSSIVb) have two genes each (Hirose and Terao, 2004). SSIIa and SSIIIa activities reach a climax 5–7 days after flowering and remain high later during grain filling (Ohdan et al., 2005). In contrast, the activity of SSIIIb is high at the start of seed formation, but decreases significantly later. SSIVb transcripts an increase up to 5 days after flowering and then decrease gradually. Compared to SSI, SSIIa, or SSIIIa, the transcripts of SSIIIb and SSIVb are much lower. The rest of the SS genes (SSIIb, SSIIc, SSIVa) are poorly expressed; their molecular expression is high during the first 5 days after flowering and low during the residual part of the grain-filling period. Ohdan et al. (2005) also studied the expression pattern of genes encoding the starch branching enzyme, starch debranching enzyme, starch phosphorylase, and starch disproportionating enzymes during seed development, but there are few reports correlating the activity of the enzymes to differential action of starch synthesis in the endosperm of rice seeds that vary in grain quality.

4.4. Hormonal regulation of spikelet development While physiological investigations undertaken into the cause of the strong metabolic dominance of the apical spikelets in grain filling and the inhibition of the inferior basal spikelets in the rice panicle have discounted the possibility of a poor assimilate supply to the inferior spikelets, Mohapatra and Sahu (1991) and Mohapatra et al. (1993) have suggested that manipulation of spikelet development with hormones or external growth regulators may be an option for improved partitioning of assimilates in favor of the inferior spikelets. After emergence, the distal spikelets quickly produce anthers, and growth inhibitors produced in the process suppress the development of other spikelets. Since the spikelet at the tip of a branch reaches anthesis (and the capacity to produce inhibitor) first, the penultimate spikelet is the most suppressed, and the order of dominance recedes in a sequence to the base (Mohapatra et al., 1993). Initially, the research on hormones concentrated on the positive manipulation of seed growth. The developing seeds are a rich source of cytokinins (Emery and Atkins, 2006; Morris, 1997) that boost seed growth through improved cell division. In the developing seeds of rice and wheat, the cytokinin content exhibits a transient increase immediately after flowering, and the high concentration of the hormone could enhance the activities of cell-cycle genes (Morris, 1997; Morris et al., 1993). An increased number of cells provide increased sink capacity for storage of assimilates. Additionally, cytokinins improve phloem unloading in the developing seeds (Clifford et al., 1986). In addition to cytokinins, abscisic acid (ABA), indole acetic acid (IAA), and gibberellic acid (GA) are also involved in the regulation of grain development (Davies, 1987). However, the specific action of these hormones in the seed development of inferior basal spikelets of rice has not been fully investigated.

346

Pravat K. Mohapatra et al.

Patel and Mohapatra (1992) found that application of GA and kinetin improved growth and development more on the basal branches than on the apical branches and tended to reduce heterogeneity in the architecture of the indica rice panicle (Yang et al., 2002b). Cell number and cell division of the rice endosperm are regulated by cytokinin levels in the endosperm. In contrast, IAA accelerated growth of the spikelets on the apical branches only and increased the heterogeneity of development between apical and basal spikelets. The role of hormones in grain development of rice was further tested in six contrasting rice genotypes of indica inbreds and tropical japonica/indica hybrid lines at the International Rice Research Institute, The Philippines, by measuring the intrinsic levels of zeatin, zeatin ribosides, IAA, and GA of grains and roots at different stages of grain filling (Yang et al., 2000). While the percentage of filled grains to total grains of the cultivars did not correlate with GA or IAA, changes in zeatin and zeatin riboside contents exhibited a significant relationship with the grain-filling pattern of both superior and inferior spikelets of all cultivars. Zhao et al. (2007) also studied the role of hormones in six rice cultivars, including three two-line hybrid rice combinations showing differences in seed set and grain filling. They attributed poor plumpness and the slow rate of filling of the inferior grains to the low contents of zeatin, zeatin riboside, IAA, and ABA and the high rate of evolution of ethylene in the inferior grains. Cultural conditions also vary the grain-filling pattern in rice, and the poor filling of the inferior spikelets arising from cultural conditions is associated with low contents of zeatin riboside and IAA (Xu et al., 2007). However, ABA levels in these studies were higher in the inferior spikelet, leading Xu et al. (2007) to suggest that the shortened grain-filling period of inferior grain is related to increased ABA and reduced IAA levels in the grains. More recently, this group of researchers (Zhang et al., 2009a) reported that inferior spikelets of japonica/indica hybrids possess a slower rate of grain filling and endosperm cell division than the indica/indica hybrids, and both of these functions were positively and significantly correlated with zeatin, zeatin ribosides, and ABA contents of the developing grains. The net development of a plant organ is regulated by a balance between hormones that promote and those that inhibit development. In the context of spikelet development in rice, the hormone balance becomes crucial for survival when a spikelet is disadvantaged because of its temporal or spatial location on the panicle axis. It is possible that application of GA, cytokinins, or IAA may increase the ratio between hormones that promote and those that inhibit the grain filling of inferior spikelets. However, the role of inhibitory hormones, such as ABA or ethylene, has not been studied as extensively as that of promoter hormones. In rice, the flag leaf and the two leaves below the flag produce considerable ethylene during grain filling (Debata and Murty, 1983; Khan and Choudhury, 1992), and the ethylene production of the leaves exceeds that of the panicle (Saka et al., 1992). Similar

Apical Dominance and Grain Yield of Rice

347

ethylene emissions from the leaves and ear were also reported in wheat (Beltrano et al., 1994; Labrana et al., 1991). Spikelets in the lower part of the panicle that are confined in the flag leaf sheath for a longer period may be subjected to the inhibitory action of ethylene more than those in the upper part of the rice panicle. This concept was tested by following the effects of chemical treatments that either inhibited or encouraged ethylene action on grain development in rice (Mohapatra et al., 2000). Application of the ethylene inhibitors cobalt, silver, and 1-aminoethoxyvinyl glycine (AVG) improved the growth and development of the inferior spikelets on basal branches of the panicle, while application of the ethylene promoters, 1-aminocyclopropane1-carboxylic acid, or 2-chloroethylephosphonic acid (CEPA), inhibited the growth and development of these spikelets. The ethylene inhibitors improved starch biosynthesis and the activity of sucrose synthase and invertase enzymes only in the kernels of inferior spikelets, whereas CEPA application had the opposite effect because of enhanced ethylene production (Naik and Mohapatra, 2000). The inferior basal spikelets produced more ethylene than did the superior apical spikelets (Kuanar et al., 2010; Mohapatra and Mohapatra, 2005), and the high ethylene accelerated the senescence of the pericarp (Mohapatra and Mohapatra, 2006), retarded the development of male gametophytes (Naik and Mohapatra, 1999), diminished the activity of the starch-synthesizing enzymes, AGPase and sucrose synthase (Mohapatra et al., 2009), increased chalkiness and decreased the quality of grains (Yang et al., 2007a), and reduced the rate of cell division and starch biosynthesis, resulting in the accumulation of unused sugars in the endosperm (Panda et al., 2009) during grain filling. Ethylene accumulation in the boot of the flag leaf sheath significantly decreased grain filling in rice (Mohapatra and Mohapatra, 2006). Additionally, the compact arrangement of spikelets in the panicle axis increased ethylene production to the detriment of grain filling (Panda et al., 2009), and the extended exposure to ethylene inside the boot of the flag leaf sheath worsens the situation further for the late-emerging spikelets on the proximal part of the panicle. Yang et al. (2006a) accepted the adverse effect of ethylene on endosperm cell division and cell number and ultimately grain filling, and proposed that the antagonistic interaction between ethylene and ABA may be involved in mediating the postflowering growth and development of rice spikelets. The low ABA content of the endosperm inhibited the grain filling of inferior spikelets of rice and exogenous application has been shown to alleviate the inhibition (Zhang et al., 2009a). ABA application promoted endosperm growth by improving the assimilate partitioning to grains (Kato and Takeda, 1993; Kato et al., 1993; Yang et al., 2002a, 2006b). This has led Yang and Zhang (2010) to suggest that controlled soil drying in the mid- to late-grain-filling stages to initiate senescence and induce assimilate remobilization of stored carbohydrates can lead to improved grain filling of inferior spikelets and increased harvest index (Yang et al., 2007b).

348

Pravat K. Mohapatra et al.

5. Is Manipulation of Apical Dominance Crucial for Grain Yield Improvement? Domesticated rice, Oryza sativa, originated in the monsoonal flood plains of Asia from the wild progenitor Oryza rufipogon; the presence of a continuous array of morphological intergrades between wild and cultivated forms in the Jeypore tract of Orissa bears testimony to this proposition (Oka, 1991). An undulating land surface with deep swamps and temporary marshes on higher parts of the landscape became the habitat of the diverse range of rice ecotypes, with annual types in the seasonally dry upland areas and perennial types in the swamps. The erratic pattern and distribution of rainfall might have further increased the survival of different rice ecotypes. With such variable conditions, the production of many more reproductive units (spikelets) than are usually taken to maturity was likely an important strategy for survival/adaptation in rice (Stephenson, 1981). A heterogeneous panicle architecture containing spikelets temporally and spatially segregated from one another could be an important evolutionary trait, enabling the plant to salvage some spikelets by delaying their development when unfavorable weather coincided with a critical stage of spikelet development. Apical dominance, regulated by the balance of promotive and inhibitory hormones in the nascent juvenile organs, is the best possible way of segregating the growth of individual reproductive units on the panicle axis in terms of the time scale for development and discarding spikelets in excess of resource capacity in the particular environment. While homogeneous architecture may be a useful trait under stress-free situations, it could result in loss of all the reproductive units in unfavorable conditions. With apical dominance, the shoot tip controls the shoot branching pattern by production of IAA, which in turn, regulates cytokinin levels, transport, and action (Bangerth, 1989; Sachs and Thimann, 1967). The role of cytokinins as a second signal has been corroborated by decapitation of the apical bud that removes the source of endogenous IAA and increases the availability of cytokinins in the stem, xylem sap (Bangerth, 1994; Li et al., 1995), and axillary buds by increasing the expression of adenosine phosphateisopentenyl transferase gene, which encodes a key enzyme in cytokinin biosynthesis (Tanaka et al., 2006). In contrast, the increase of IAA inhibits cytokinin synthesis (Nordstrom et al., 2004). More recently, it has been suggested that a shoot multiplication signal is involved in apical dominance (Beveridge, 2006; Dunn et al., 2006). In pea, RAMOSUS branching genes control the synthesis and perception of this signal, a strigolactone or a product of strigolactone (Ferguson and Beveridge, 2009). This branch inhibitory signal is synthesized in the shoot and roots, and IAA supply

Apical Dominance and Grain Yield of Rice

349

from the apical bud controls its availability in the stem and developing axillary branches (Dunn et al., 2006). While most of the experimental work relating to hormone control of branching patterns and apical dominance has been conducted on dicotyledonous species, monocotyledonous species have received little attention. Harrison and Kaufman (1984) reported that IAA, ABA, and ethylene could inhibit tiller bud release in oats by restricting cytokinin transport and promoting its catabolism. Kinetin-promoted tiller bud swelling and elongation was inhibited by ethylene or IAA application (Harrison and Kaufman, 1982). Estimation of endogenous levels of IAA and cytokinins supported the proposition that apical dominance of oats is regulated by a balance of IAA and cytokinins, with the ratio decreasing during tiller bud release due to an increase in the level of cytokinins (Harrison and Kaufman, 1983). With sequencing of the rice genome in the International Rice Genome Sequencing Project (2005), the genes regulating agronomically important traits like tiller and spikelet numbers, grain size, and plant height have been identified, but the mechanisms that regulate each trait have not been characterized (Sakamoto and Matsuoka, 2008). Knowledge of the genes and mechanisms controlling intergrain apical dominance on assimilates partitioning within the inflorescence is largely unknown. In rice, each culm bears a single terminal flowering inflorescence, with strong apical dominance inhibiting the development of axillary reproductive structures. Within the inflorescence, intergrain apical dominance regulates the growth of individual spikelets at the various nodes. Studies of apical dominance have focused on intact, decapitated, or in vitro systems. Ontogenetic studies and evaluation of metabolites in individual branches of the panicle have revealed that the gradient in growth and development between the apical and the basal parts of the intact panicle results from the heterogeneous architecture of the panicle (Mohapatra and Sahu, 1991; Xu and Vergara, 1986). Studies on excision of a part of the developing panicle showed that the growth and development of the inferior spikelets is amenable to physiological manipulation that reduces the heterogeneity of spikelet development and improves grain filling. Removal of some of the primary branches from the axis encouraged partitioning of carbon to residual spikelets of the panicle and improved the grain-filling percentage compared to uncut panicles (Mohapatra et al., 2004). Similarly, grain filling of inferior spikelets improved significantly when dominant spikelets of the primary branches were removed (Kato, 2004). Studies on exogenous treatments and endogenous action of hormones reviewed in Section 4.4 have clarified the physiological mechanisms involved in apical dominance in rice. Yang et al. (2000) showed that high IAA levels in the sink lead to high cytokinin levels in the grain and stimulate the partitioning of assimilates to the grain. While exogenous IAA treatment enhanced the growth of superior spikelets and depressed the growth of inferior spikelets, resulting in increased heterogeneity in spikelet growth and development of the

350

Pravat K. Mohapatra et al.

panicle, applications of kinetin and GA stimulated spikelet growth and development more on the inferior proximal branches, improved homogeneity among grains, and increased the final grain yield (Patel and Mohapatra, 1992). The role of growth-retarding hormones like ABA and ethylene in the regulation of apical dominance is also emphasized in the literature. ABA in the developing grains stimulated assimilate accumulation from the prestored reserves by simulating the effects of water stress (Kato and Takeda, 1993; Yang et al., 2001b), while low levels of ABA cytokinins and IAA in the inferior spikelets restricted their growth (Zhang et al., 2009a). The role of ethylene in the regulation of intergrain signaling is a recent addition to the role of hormones in grain development. The concentration of ethylene is high in the basal branches, much to the detriment of grain filling. Temporal ethylene production is high around flowering in rice spikelets irrespective of their position on the panicle (Mohapatra and Mohapatra, 2005; Naik and Mohapatra, 2000). As the apical spikelets reach flowering nearly a week ahead of the basal spikelets (Mohapatra et al., 1993), the early production of ethylene could have an inhibitory effect on the growth and development of the basal spikelets, simulating apical dominance (Panda et al., 2009; Zhang et al., 2009b). It has been shown that the action of ethylene on grain filling counters that of ABA (Yang et al., 2006a), but is similar to root elongation of maize which is maintained under stress by ABA that restricts ethylene production (Spollen et al., 2000). However, this concept has been challenged recently by Kim et al. (2009) who reported that increased jasmonate concentration during drought stress elevated the production of ABA in rice, leading to alteration of spikelet development and loss of grain yield. In an effort to raise the yield potential of the NPT of rice, priority has been given to selection for low tillering and large panicle size (Kim and Vergara, 1992). Apical dominance is encouraged during vegetative growth in order to restrict the tiller number, but is reduced in the reproductive phase in order to allow production of a large number of spikelets competent to bear grains. Breeding has resulted in pyramiding of these two desired traits in the NPT (Peng et al., 1999; Virk et al., 2004). However, the objective of a high grain yield was not achieved because assimilate transport was not maintained to the growing sinks and the late-initiated tillers failed to thrive and died prematurely (Mohapatra and Kariali, 2008). We suggest that tiller dominance could be increased by inducing a hormonal signal from the early-formed tillers to induce early senescence of the late-formed tillers. However, the manipulation of apical dominance of the spikelets is more complicated. Yamagishi et al. (1996) showed that reduced apical dominance leading to uniform distribution of assimilates was the primary cause of poor grain filling in the NPT rice with large panicles. Thus, achievement of homogeneity in spikelet development at the cost of sacrificing the heterogeneous architecture of the rice panicle may not achieve the goal of breaking the grain-yield barrier in the newly developed rice lines with

Apical Dominance and Grain Yield of Rice

351

large panicles unless we know how to manipulate flower development with hormones and the corresponding gene expression. Chung et al. (1994) reported that the OsMADS1 gene is responsible for flower induction and reduced apical dominance in rice. This gene is only expressed in the floral organs. Empirical evidence linking expression of this gene to the hormone balance is lacking, but the evidence mentioned in Section 4.4 indicates that the inflorescence architecture of rice is amenable to modification by manipulation of the factors affecting hormone levels.

6. Suggestions for Modification of Apical Dominance Until recently, there has been no comprehensive understanding of either the physiology or genetics of apical dominance in grasses, or the use of such traits for improvement of grain yield (Morgan et al., 2002). While models by Dunn et al. (2006) and Ferguson and Beveridge (2009) have been developed based on experiments mostly with the pea plant, McSteen (2009) proposed a model for the role of hormones in axillary meristem initiation and outgrowth during both the vegetative phase and inflorescence branching of grasses. We propose a model in Figure 2 for interspikelet apical dominance within rice panicles that is in agreement with the evidence and the models referred to above. The molecular biology of hormone action is excluded from this model because of insufficient information. The model suggests that IAA traveling basipetally in the rachis from dominant spikelets inhibits the growth of inferior spikelets, while cytokinin moving acropetally promotes the growth of the latter. Both cytokinin and GA stimulate cell division and elongation. Age, nutrients, and decapitation encourage release from apical dominance and make the inferior spikelets responsive to hormone action. Dominant spikelets reach flowering and pollination early and produce ethylene, which is counterproductive to the action of cytokinins or GA. The length of time that the panicle is retained inside the boot of the flag leaf sheath, the compact architecture of the panicle as well as exposure to environmental stresses—all increase ethylene production. Environmental stresses also stimulate ABA production for the benefit of assimilate remobilization and resultant growth of the spikelet. Inclement weather can also limit current photosynthesis and assimilate production to the detriment of grain filling. At the same time, in order to sustain filling of spikelets in the large panicles of modern rice, we suggest a measure of controlled apical dominance, to produce a responsive sink with more spikelets, instead of a strong apical dominance or the absence of dominance, resulting in either heterogeneous or homogeneous architecture, respectively. The number of such spikelets would be determined by the source capacity for assimilate

352

Pravat K. Mohapatra et al.

Superior spikelet (strong sink)

Roots

Compact panicle architecture or long period of spikelet retention inside the boot

Strigolactone IAA

GA

Environmental stress

Early anthesis Current photosynthesis

ABA Cytokinin Age Nutrients Decapitation

Ethylene

Remobilisation of reserve

Assimilates

Inferior spikelet (weak sink)

Responsive sink

Cell multiplication and elongation

Grain filling

Arrowhead lines indicate promotion and flat ended lines indicate inhibition.

Figure 2 Schematic of the factors influencing the development of superior and inferior spikelets on the rice panicle.

production (photosynthesis and remobilization of reserves) in a particular environment and the sink strength (balance of hormones). Since the growth of spikelets can be managed by decapitation, chemical application, and exposure to environmental conditions, and germplasm is available for variation in panicle architecture (Protection of Plant Varieties and Farmers Right Authority, 2007), we suggest that a new ideotype with many spikelets with high-quality grains can be developed by gene manipulation or breeding.

ACKNOWLEDGMENTS This work is supported by the Emeritus Scientist Scheme awarded to P. K. M. by the Council of Scientific and Industrial Research, Government of India, New Delhi. Professor Mohapatra was the recipient of the Yoshida Award for 2010 given by International Rice Research Institute, Philippines for significant contribution to rice physiology.

REFERENCES Afza, R., Shen, M., Zapata-Arias, F. J., Xie, J., Fundi, H. K., Lee, K.-S., BobadillaMucino, E., and Kodym, A. (2000). Effect of spikelet position on rice anther culture efficiency. Plant Sci. (Shannon, Ireland) 153, 155–159.

Apical Dominance and Grain Yield of Rice

353

Bangerth, F. (1994). Response of cytokinin concentration in the xylem exudates of bean (Phaseolus vulgaris L.) plants to decapitation and auxin treatment and relationship to apical dominance. Planta 194, 439–442. Bangerth, F. (1989). Dominance among fruits/sinks and search for a correlative signal. Physiol. Plant. 76, 608–614. Baun, L. C., Palmiano, E. P., Perez, C. M., and Juliano, B. O. (1970). Enzymes of starch metabolism in the developing rice grain. Plant Physiol. 46, 429–434. Beltrano, J., Carbone, A., Montaldi, E., and Guiamet, J. J. (1994). Ethylene as promoter of wheat grain maturation and ear senescence. Plant Growth Regul. 15, 107–112. Beveridge, C. A. (2006). Axillary bud outgrowth: Sending a message. Curr. Opin. Plant Biol. 9, 35–40. Cassman, K. G. (1999). Ecological intensification of cereal production systems: Yield potential, soil quality, and precision agriculture. Proc. Natl. Acad. Sci. USA 96, 5952– 5959. Chaudhry, F. M., and Nagato, K. (1970). Role of vascular bundles in ripening of rice kernel in relation to the locations on panicle. Proc. Crop Sci. Soc. Jpn. 39, 301–309. Chung, Y. Y., Kim, S. R., Finkel, D., Yanofsky, M. F., and An, G. (1994). Early flowering and reduced apical dominance result from ectopic expression of a rice-MADS box gene. Plant Mol. Biol. 26, 657–665. Clifford, P. E., Offler, C. E., and Patrick, J. W. (1986). Growth regulators have rapid effects on photosynthate unloading from seed coats of Phaseolus vulgaris L. Plant Physiol. 80, 635–637. Counce, P. A., and Gravois, K. A. (2006). Sucrose synthase activity as a potential indicator of high rice grain yield. Crop Sci. 46, 1501–1507. Davies, P. J. (1987). The plant hormones. Their nature, occurrence, and functions. In “Plant Hormones and Their Role in Plant Growth Development” (P. J. Davies, Ed.), pp. 1–11. Martinus Nijhoff Publishers, Netherlands. De Datta, S. K. (1981). Principles and Practices of Rice Production. Wiley, New York. Debata, A., and Murty, K. S. (1983). Endogenous ethylene content in rice leaves during senescence. Indian J. Plant Physiol. 26, 425–427. Dingkuhn, M., Penning de Vries, F. W. T., De Datta, S. K., and van Laar, H. H. (1991). Concepts for a new plant type for direct seeded flooded tropical rice. In “Direct Seeded Flooded Rice in the Tropics”, pp. 17–38. International Rice Research Institute, Los Banos, Philippines. Dunn, E. A., Ferguson, B. J., and Beveridge, C. A. (2006). Apical dominance and shoot branching. Divergent opinions or divergent mechanisms? Plant Physiol. 142, 812–819. Egli, D. B., and Bruening, W. P. (2003). Seed number and yield of grain crops—Photosynthesis and sink characteristics. Indian J. Plant Physiol. 49–54 (special issue). Emery, Neil, and Atkins, C. (2006). Cytokinins and seed development. In “Handbook of Seed Science and Technology” (Amarjit Singh Basra, Ed.), pp. 63–93. Haworth Press, New York. Evans, L. T., Visperas, R. M., and Vergara, B. S. (1984). Morphological and physiological changes among rice varieties used in the Philippines over the last seventy years. Field Crops Res. 8, 105–124. FAOSTAT. (2010). Production statistics., http://faostat.fao.org/ (accessed 1 May 2010). Feil, B. (1992). Breeding progress in small grain cereals—A comparison of old and modern cultivars. Plant Breed. 108, 1–11. Ferguson, B. J., and Beveridge, C. A. (2009). Role of auxin, cytokinin, and strigolactone in regulating shoot branching. Plant Physiol. 149, 1929–1944. Furbank, R. T., Scoefield, G. N., Hirose, T., Wang, X.-D., Patrick, J. W., and Offler, C. E. (2001). Cellular localization and function of a sucrose transporter OsSUT1 in developing rice grains. Aust. J. Plant Physiol. 28, 1187–1196.

354

Pravat K. Mohapatra et al.

Gifford, R. M., and Evans, L. T. (1981). Photosynthesis, carbon partitioning and yield. Annu. Rev. Plant Physiol. 32, 485–509. Harrison, M. A., and Kaufman, P. B. (1982). Does ethylene play a role in the release of lateral buds (tillers) from apical dominance in oats? Plant Physiol. 70, 811–814. Harrison, M. A., and Kaufman, P. B. (1983). Estimates of free and bound indole acetic acid and zeatin levels in relation to regulation of apical dominance and tiller release in oat shoots. Plant Growth Regul. 2, 215–223. Harrison, M. A., and Kaufman, P. B. (1984). The role of hormone transport and metabolism in apical dominance in oats. Bot. Gaz. (Chicago) 145, 293–297. Hirose, T., and Terao, T. (2004). A comprehensive expression analysis of the starch synthase gene family in rice (Oryza sativa L.). Planta 220, 9–16. Hirose, T., Scofield, G. N., and Terao, T. (2008). An expression analysis profile for the entire sucrose synthase gene family in rice. Plant Sci. (Shannon Ireland) 174, 534–543. International Rice Genome Sequencing Project. (2005). The map based sequence of rice genome. Nature 436, 793–800. International Rice Research Institute. (2010). World Rice Statistics (WRS). http://beta.irri. org/solutions/index.php?option¼com_content&task¼view&id¼250 (accessed 1 May 2010). Ishimaru, T., Matsuda, T., Ohsugi, R., and Yamagishi, T. (2003). Morphological development of rice caryopses located at the different positions in a panicle from early to middle stage of grain filling. Funct. Plant Biol. 30, 1139–1149. Ishimaru, T., Hirose, T., Matsuda, T., Goto, A., Takahashi, K., Sasaki, H., Terao, T., Ishii, R., Ohsugi, R., and Yamagishi, T. (2005). Expression patterns of genes encoding carbohydrate metabolizing enzymes and their relationship to grain filling in rice (Oryza sativa L.): Comparison of caryopses located at different positions in a panicle. Plant Cell Physiol. 45, 620–628. James, M. G., Denyer, K., and Myers, A. M. (2003). Starch synthesis in the cereal endosperm. Curr. Opin. Plant Biol. 6, 215–222. Jeng, T. L., Wang, C. S., Tseng, T. H., and Sung, J. M. (2007). Expression of granule bound starch synthase in developing rice grain. J. Sci. Food Agric. 87, 2456–2463. Kato, T. (1993). Variation in grain filling process among grain positions within a panicle of rice (Oryza sativa L.). SABRAO J. 25, 1–10. Kato, T. (1995). Change of sucrose synthase activity in developing endosperm of rice cultivars. Crop Sci. 35, 827–831. Kato, T. (2004). Effect of spikelet removal on grain filling of Akenohoshi, a rice cultivar with numerous spikelets in a panicle. J. Agric. Sci. 142, 177–181. Kato, T. (2006). Panicle centroid index: An index to represent the distribution of spikelets in a panicle. Kinki J. Crop Sci. Breed. 51, 31–36. Kato, T. (2008). Variation in the branching system of vascular bundles within a panicle among extra-heavy panicle types. Memoirs of the School of Biology Oriented Science and Technology No. 21, pp. 1–6. Kato, T., and Takeda, K. (1993). Endogenous abscisic acid in developing grains on primary and secondary branches of rice (Oryza sativa L.). Jpn. J. Crop Sci. 62, 636–637. Kato, T., Sakurai, N., and Kuraishi, N. (1993). The changes of endogenous abscisic acid in developing grains of two rice cultivars with different grain size. Jpn. J. Crop Sci. 62, 456– 461. Kato, T., Shinmura, D., and Taniguchi, A. (2007). Activities of enzymes for sucrose-starch conversion in developing endosperm of rice and their association with grain filling in extra-heavy panicle types. Plant Prod. Sci. 10, 442–450. Khan, R. I., and Choudhury, M. A. (1992). Role of endogenous hormones in the regulation of whole plant senescence in rice. Indian J. Exp. Biol. 30, 131–134.

Apical Dominance and Grain Yield of Rice

355

Khush, G. S. (1996). Prospects and approaches to increasing the genetic yield potential of rice. In “Rice Research in Asia: Progress and Priorities” (R. E. Evenson, R. W. Herdt, and M. Hossain, Eds.), pp. 59–71. CAB International, Wallingford, UK. Khush, G. S., and Peng, S. (1996). Breaking the yield frontier of rice. In “Increasing Yield Potential in Wheat: Breaking the Barriers. Proceedings of a workshop held on 26–28 March, 1996 in Ciudad Obregon, Sonora, Mexico”, pp. 36–49. International Maize and Wheat Improvement Center, Mexico, DF. Kim, J. K., and Vergara, B. S. (1992). Grain yield potential of a low tillering large panicle type in rice. Korean J. Crop Sci. 37, 361–371. Kim, E. H., Kim, Y. S., Park, S.-H., Koo, Y. J., Choi, Y. D., Chung, Y.-Y., Lee, I.-J., and Kim, J. K. (2009). Methyl jasmonate reduces grain yield by mediating stress signal to alter spikelet development in rice. Plant Physiol. 149, 1751–1760. Kosar-Hashemi, B., Li, Z., Larroque, O., Regina, A., Yamamori, M., Morell, M. K., and Rahman, S. (2007). Multiple effects of the starch synthase II mutation in developing wheat endosperm. Funct. Plant Biol. 34, 431–438. Kropff, M. J., Cassman, K. G., Peng, S., Matthews, R. B., and Setter, T. L. (1994). Quantitative understanding of yield potential. In “Breaking the Yield Barrier. Proceedings of a Workshop on Rice Yield Potential in Favourable Environments” (K. G. Cassman, Ed.), pp. 21–38. International Rice Research Institute, Manila, The Philippines. Kuanar, S. R., Panigrahi, R., Kariali, E., and Mohapatra, P. K. (2010). Apoplasmic assimilates and grain growth of contrasting rice cultivars differing in grain dry mass and size. Plant Growth Regul. 61, 135–151. Kudo, K. (1991). Studies on the ripening of rice plants. 2. Influence on the density of panicles on the ripening in rice varieties. Rep. Tohoku Breed. Crop Sci. Soc. Jpn. 34, 109– 112. Labrana, X., Vendrel, M., and Araus, J. L. (1991). Ethylene production in wheat flag leaves and ear during grain filling. Plant Physiol. Biochem. 29, 349–354. Langhans, M., Ratajczak, R., Luetzelschwab, M., Michalke, W., Waechter, R., FischerSchliebs, E., and Ulrich, C. I. (2001). Immunolocalization of plasma-membrane HþATPase and tonoplast type pyrophosphatase in the plasma-membrane of the sieve element-companion cell complex in the stem of Ricinus communis L. Planta 213, 11–19. Li, C. J., Herrera, G. J., and Bangerth, F. (1995). Effect of apex excision and replacement by 1-naphthylacetic acid on cytokinin concentration and apical dominance in pea plants. Physiol. Plant. 94, 465–469. Liang, J., Zhang, J., and Cao, X. (2001). Grain sink strength may be related to poor grain filling of indica–japonica rice (Oryza sativa) hybrids. Physiol. Plant. 112, 470–477. Lim, J. D., Cho, J., Park, Y., Hahn, T., Choi, S., and Seong, J. (2006). Sucrose transport from source to sink seeds in rice. Physiol. Plant 126, 572–584. Luo, L. J., Li, Z. K., Mei, H. W., Shu, Q. Y., Tabien, R., Zhong, D. B., Yang, C. S., Stansel, J. W., Khush, G. S., and Paterson, A. H. (2001). Over-dominant epistatic loci are the primary genetic basis of inbreeding depression and heterosis in rice. Genetics 158, 1755–1771. Matsue, Y., Odahara, K., and Hiramatsu, M. (1995). Differences in amylose content, amylographic characteristics and storage proteins of grains on primary and secondary rachis branches in rice. Jpn. J. Crop Sci. 64, 601–606. Matsushima, S. (1970). Crop Science in Rice. Fuji Publication Co. Ltd., Tokyo, Japan. Matthews, R. B., Horie, T., Kropff, M. J., Bachelet, D., Centeno, H. G., Shin, J. C., Mohandass, S., Singh, S., Zhu, D., and Lee, M. H. (1995). A regional evaluation of the effect of future climate change on rice production in Asia. In “Modeling the Impact of Climate Change on Rice Production in Asia” (R. B. Matthews, M. J. Kropff, D. Bachelet, and H. H. van Laar, Eds.), pp. 95–139. CAB International, Wallingford, UK.

356

Pravat K. Mohapatra et al.

McSteen, P. (2009). Hormonal regulation of branching in grasses. Plant Physiol. 149, 46–55. Mohapatra, P. K., and Kariali, E. (2008). Time of tiller emergence determines the pattern of dominance of rice tillers. Aust. J. Crop Sci. 1, 53–62. Mohapatra, P. K., and Mohapatra, R. (2005). Sink–source relationship and yield potential of rice: Effect of ethylene on grain filling of late flowering spikelets. In “Rice is Life: Scientific Perspective for the 21st century” (K. Toriyama, K. L. Heong, and B. Hardy, Eds.), pp. 120–124. International Rice Research Institute, Manila, The Philippines. Mohapatra, R., and Mohapatra, P. K. (2006). Ethylene control of seed coat development in low and high sterile semi-dwarf indica rice cultivars. Plant Growth Regul. 50, 47–55. Mohapatra, P. K., and Sahu, S. K. (1991). Heterogeneity of primary branch development and spikelet survival in rice panicle in relation to assimilates of primary branches. J. Exp. Bot. 42, 871–879. Mohapatra, P. K., Patel, R., and Sahu, S. K. (1993). Time of flowering affects grain quality and spikelet partitioning within the rice panicle. Aust. J. Plant Physiol. 20, 231–241. Mohapatra, P. K., Naik, P. K., and Patel, R. (2000). Ethylene inhibitors improve dry matter partitioning and development of late flowering spikelets on rice panicles. Aust. J. Plant Physiol. 27, 311–323. Mohapatra, P. K., Masamoto, Y., Morita, S., Takanashi, J., Kato, T., Itani, T., AduGyamfi, J. J., Shunmugasundaram, M., Nguyen, N. T., Saneoka, H., and Fujita, K. (2004). Partitioning of 13C-labelled photosynthate varies with growth stage and panicle size in high yielding rice. Funct. Plant Biol. 31, 131–139. Mohapatra, P. K., Sarkar, R. K., and Kuanar, S. R. (2009). Starch synthesizing enzymes and sink strength of grains of contrasting rice cultivars. Plant Sci. (Shannon, Ireland) 176, 256– 263. Morgan, P. W., Finlayson, S. A., Childs, K. L., Mullet, J. E., and Rooney, W. L. (2002). Opportunities to improve adaptability and grain yield in grasses. Lessons from sorghum. Crop Sci. 42, 1791–1799. Morris, R. O. (1997). Hormonal regulation of seed development. In “Cellular and Molecular Biology of Plant Seed Development” (B. A. Larkins and I. K. Vasil, Eds.), pp. 117– 149. Kluwer Academic, Dordrecht, Netherlands. Morris, R. O., Blevins, D. G., Dietrich, J. T., Durly, R. C., Gelvin, S. B., Gray, J., Hommes, N. G., Kaminek, M., Mathews, L. J., Meilan, R., Reinbott, T. M., and Sagavendra-Soto, L. (1993). Cytokinins in plant pathogenic bacteria and cereal grains. Aust. J. Plant Physiol. 20, 621–637. Murata, Y., and Matsushima, S. (1975). Rice. In “Crop Physiology” (L. T. Evans, Ed.), pp. 73–99. Cambridge University Press, London. Murchie, E. K., Yang, J., Hubbart, S., Horton, P., and Peng, S. (2002). Are there associations between grain filling rate and photosynthesis in the flag leaves of field grown rice? J. Exp. Bot. 53, 2217–2224. Naik, P. K., and Mohapatra, P. K. (1999). Ethylene inhibitors promote male gametophyte survival in rice. Plant Growth Regul. 28, 29–39. Naik, P. K., and Mohapatra, P. K. (2000). Ethylene inhibitors enhanced sucrose synthase activity and promoted grain filling of basal rice kernels. Aust. J. Plant Physiol. 27, 997– 1008. Nakamura, Y., and Yuki, K. (1992). Changes in enzyme activities associated with carbohydrate metabolism during the development of rice endosperm. Plant Sci. (Shannon, Ireland) 82, 15–20. Nakamura, Y., Yuki, K., and Park, S. Y. (1989). Carbohydrate metabolism in the developing endosperm of rice grains. Plant Cell Physiol. 30, 833–839. Nishiyama, I. (1983). Male sterility caused by cooling treatment at the young microspore stage in rice plants. XXIV. Jpn. J. Crop Sci. 52, 43–48.

Apical Dominance and Grain Yield of Rice

357

Nordstrom, A., Tarkowski, P., Tarkowski, D., Norbaek, R., Astot, C., Dolezal, K., and Sandberg, G. (2004). Auxin regulation of cytokinin biosynthesis in Arabidopsis thaliana: A factor of potential importance for auxin-cytokinin-regulated development. Proc. Natl. Acad. Sci. USA 101, 8039–8044. Ohdan, T., Francisco, P. B., Sawada, T., Hirose, T., Terao, T., Satoh, H., and Nakamura, Y. (2005). Expression profiling of genes involved in starch biosysnthesis in sink and source organs of rice. J. Exp. Bot. 56, 3229–3244. Oka, H. I. (1991). Genetic diversity of wild and cultivated rice. In “Rice Biotechnology” (G. S. Khush and G. H. Toenniessen, Eds.), pp. 55–81. International Rice Research Institute, Manila, The Philippines. Padmaja Rao, S. (1987). Relationship of filled grain percentage to yield. Int. Rice Res. Newsl. 12, 10. Padmaja Rao, S. (1989). Ideal panicle type—A new concept for improving rice yields. Agric. Sci. Dig. 9, 139–140. Padmaja Rao, S., Venkateswarlu, B., and Acharyulu, T. L. (1985). Screening for differentiating the degree of spikelet filling in rice. Plant Soil 88, 289–293. Panda, B. B., Kariali, E., Panigrahi, R., and Mohapatra, P. K. (2009). High ethylene production slackens seed filling in compact-panicled rice cultivar. Plant Growth Regul. 58, 141–151. Patel, R., and Mohapatra, P. K. (1992). Regulation of spikelet development in rice by hormones. J. Exp. Bot. 43, 257–262. Patel, R., and Mohapatra, P. K. (1996). Assimilate partitioning within floret components of contrasting rice spikelets producing qualitatively different types of grains. Aust. J. Plant Physiol. 23, 85–92. Peng, S., Khush, G. S., and Cassman, K. G. (1994). Evaluation of a new plant ideotype for increased yield potential. In “Breaking the Yield Barrier, Proceedings of a Workshop on Rice Yield Potential in Favourable Environments” (K. G. Cassman, Ed.), pp. 5–20. International Rice Research Institute, Manila, The Philippines. Peng, S., Cassman, K. G., Virmani, S. S., Sheehy, J., and Khush, G. S. (1999). Yield potential trends of tropical rice since the release of IR8 and the challenge of increasing rice yield potential. Crop Sci. 39, 1552–1559. Perez, C. M., Perdon, A. A., Resurreccion, A. P., Villareal, R. M., and Juliano, B. O. (1975). Enzymes of carbohydrate metabolism in the developing rice grain. Plant Physiol. 56, 579–583. Protection of Plant Varieties and Farmers Right Authority. (2007). Guidelines for the conduct of test for distinctiveness, uniformity and stability on rice (Oryza sativa L.). Plant Variety J. 1, 1–25. Sachs, T., and Thimann, K. V. (1967). The role of auxin and cytokinins in the release of buds from dominance. Am. J. Bot. 54, 136–144. Saka, H., Kogen, H., Okumura, M., and Watanabe, S. (1992). Fluctuation of ethylene production in excised panicle and flag leaf blades during grain ripening in rice (Oryza sativa L.). Jpn. J. Crop Sci. 61, 285–291. Sakamoto, T., and Matsuoka, M. (2008). Identifying and exploiting grain yield genes in rice. Curr. Opin. Plant Biol. 11, 209–214. Scoefield, G. N., Hirose, T., Gaudron, J. A., Upadhyay, N. M., Ohsugi, R., and Furbank, R. T. (2002). Antisense suppression of the rice transporter gene, OsSUT1, leads to impaired grain filling and germination but does not affect photosynthesis. Funct. Plant Biol. 29, 815–826. Setter, T. L., Conocono, E. A., and Egdane, J. A. (1996). Possibility of increasing yield potential of rice by reducing panicle height in the canopy. II. Canopy photosynthesis and yield of isogenic lines. Aust. J. Plant Physiol. 23, 161–169.

358

Pravat K. Mohapatra et al.

Sheehy, J. E., Dionora, M. J. A., and Mitchell, P. L. (2001). Spikelet numbers, sink size and potential yield in rice. Field Crops Res. 71, 77–85. Sikder, H. P., and Das Gupta, D. K. (1976). Physiology of grain filling in rice. IV. Role of panicle morphology. Indian Agriculturalist 20, 143–151. Smidansky, E. D., Martin, J. M., Hannah, L. C., Fischer, A. M., and Giroux, M. J. (2003). Seed yield and plant biomass increases in rice are conferred by deregulation of endosperm ADP-glucose pyrophosphorylase. Planta 216, 656–664. Spollen, W. G., LeNoble, M. E., Samuels, T. D., Bernstein, N., and Sharp, R. E. (2000). Abscisic acid accumulation maintains maize primary root elongation at low water potentials by restricting ethylene production. Plant Physiol. 122, 967–976. Stephenson, A. G. (1981). Flower and fruit abortion: Proximate causes and ultimate functions. Annu. Rev. Ecol. Syst. 12, 253–279. Taiz, L., and Zeiger, E. (2002). Plant Physiology. 3rd edn. Sinauer Associates, Inc., Sunderland, USA. Tanaka, M., Takei, K., Kojima, M., Sakakibara, H., and Mori, H. (2006). Auxin controls local cytokinin biosynthesis in the nodal stem in apical dominance. Plant J. 45, 1028–1036. Toenniessen, G. H. (1991). Potentially useful genes for rice genetic engineering. In “Rice Biotechnology” (G. S. Khush and G. H. Toenniessen, Eds.), pp. 258–259. International Rice Research Institute, Manila, The Philippines. Umemoto, T., and Terashima, K. (2002). Activity of granule bound starch synthase is an important determinant of amylose content in rice endosperm. Funct. Plant Biol. 29, 1121–1124. Umemoto, T., Nakamura, Y., and Ishikura, N. (1994). Effect of grain location on the panicle on activities involved in starch synthesis in rice endosperm. Phytochemistry 36, 843–847. Vandeputte, G. E., and Delcour, J. A. (2004). From sucrose to starch granule to starch physical behaviour: A focus on rice starch. Carbohydr. Polym. 58, 245–266. Venkateswarlu, B., Vergara, B. S., Parao, F. T., and Visperas, R. M. (1986a). Enhancing grain yield potential by increasing number of high density grains. Philipp. J. Crop Sci. 11, 145–152. Venkateswarlu, B., Parao, F. T., Visperas, R. M., and Vergara, B. S. (1986b). Screening quality grains of rice with a seed blower. SABRAO J. 18, 19–24. Vergara, B. S. (1987). Raising the yield potential of rice. “Paper presented during 9th Annual Scientific Meeting of the National Academy of Science and Technology, Philippine International Convention Center, July 15th, 1987,” p. 34 pp. Vergara, B. S., Venkateswarlu, B., Janoria, M., Ahn, J. K., Kim, J. K., and Visperas, R. M. (1990). Rationale for a low tillering rice plant type with high density grains. Philipp. J. Crop Sci. 15, 33–40. Virk, P. S., Khush, G. S., and Peng, S. (2004). Breeding to enhance yield potential of rice at IRRI: The ideotype approach. Int. Rice Res. Newsl. 29, 5–9. Wang, F., Chang, F., and Zhang, G. (2008). Impact of cultivar variation in density of rice panicle on grain weight and quality. J. Sci. Food Agric. 88, 897–903. Xu, X., and Vergara, B. S. (1986). Morphological changes in rice panicle development. A review of literature. IRRI Research Paper Series No. 117. International Rice Research Institute, Manila, The Philippines. Xu, Z. J., Chen, W. F., Zhang, L. B., and Yang, S. R. (1990). Comparative study on light distribution in rice canopies with different panicle types. Sci. Agric. Sin. 23, 6–10. Xu, G., Zhang, J., Lam, H. M., Wang, Z., and Yang, J. (2007). Hormonal changes are related to the poor grain filling in the inferior spikelets of rice cultivated under nonflooded and mulched condition. Field Crops Res. 101, 53–61. Yamagishi, T., Peng, S., Cassman, K. G., and Ishii, R. (1996). Studies on grain filling characteristics in “New Plant Type” rice lines developed in IRRI. Jpn. J. Crop Sci. 65, 169–170.

Apical Dominance and Grain Yield of Rice

359

Yamamoto, Y., Yoshida, T., Enomoto, T., and Yoshikawa, G. (1991). Characteristics for the efficiency of spikelet production and the ripening in high yielding japonica–indica hybrid and semidwarf indica rice varieties. Jpn. J. Crop Sci. 60, 365–372. Yang, J., and Zhang, J. (2010). Grain-filling problems in ‘super’ rice. J. Exp. Bot. 61, 1–5. Yang, J., Peng, S., Visperas, R. M., Sanico, A. L., Zhu, Q., and Gu, S. (2000). Grain filling pattern and cytokinin content in the grains and roots of rice plants. Plant Growth Regul. 30, 261–270. Yang, J., Peng, S., Gu, S., Visperas, R. M., and Zhu, Q. (2001a). Changes in activities of three enzymes associated with starch synthesis in rice grains during grain filling. Acta Agron. Sin. 27, 157–164. Yang, J., Zhang, J., Wang, Z., Zhu, Q., and Wang, W. (2001b). Hormonal changes in the grains of rice subjected to water stress during grain filling. Plant Physiol. 127, 315–323. Yang, J., Peng, S., Zhang, Z., Wang, Z., Visperas, R. M., and Zhu, Q. (2002a). Grain and dry matter yields and partitioning of assimilate in japonica/indica hybrid rice. Crop Sci. 42, 766–772. Yang, J., Zhang, J., Huang, Z., Wang, Z., Zhu, Q., and Liu, L. (2002b). Correlation of cytokinin levels in the endosperms and roots with cell number and cell division activity during endosperm development in rice. Ann. Bot. 90, 369–377. Yang, J., Zhang, J., Wang, Z., Zhu, Q., and Liu, L. (2003). Activities of enzymes involved in sucrose to starch metabolism in rice grains subjected to water stress during filling. Field Crops Res. 81, 69–81. Yang, J., Zhang, J., Wang, Z., Liu, K., and Wang, P. (2006a). Post anthesis development of inferior and superior spikelets in rice in relation to abscisic acid and ethylene. J. Exp. Bot. 57, 149–160. Yang, J., Zhang, J., Wang, Z., Zhu, Q., and Liu, L. (2006b). Abscisic acid and cytokinins in the root exudates and leaves and their relationship to senescence and remobilization of carbon reserves in rice subjected to water stress during grain filling. Planta 215, 645–652. Yang, J., Chang, E., Tang, C., Zhang, H., and Wang, Z. (2007a). Relationships of ethylene evolution rate and 1-aminocyclopropane-1-carboxylic acid concentration in grains during filling period with appearance quality of rice. Rice Sci. 14, 33–41. Yang, J., Liu, K., Wang, Z., Du, Y., and Zhang, J. (2007b). Water-saving and high-yielding irrigation for lowland rice by controlling limiting values of soil water potential. J. Integr. Plant Biol. 49, 1445–1454. Yoshida, S. (1972). Physiological aspects of grain yield. Annu. Rev. Plant Physiol. 23, 437– 464. Yoshida, S. (1981). Fundamentals of Rice Crop Science. International Rice Research Institute, Manila, The Philippines. Zhang, H., Tan, G., Yang, L., Yang, J., Zhang, J., and Zhao, B. (2009a). Hormones in the grains and roots in relation to post anthesis development of inferior and superior spikelets in japonica/indica hybrid rice. Plant Physiol. Biochem. 47, 195–204. Zhang, H., Tan, G., Wang, Z., Yang, J., and Zhang, J. (2009b). Ethylene and ACC levels in developing grains are related to the poor appearance and milling quality of rice. Plant Growth Regul. 58, 85–96. Zhao, B., Liu, K., Zhang, H., Zhu, Q., and Yang, J. (2007). Causes of poor grain plumpness of two line hybrids and their relationships to the contents of hormones in rice grain. Agric. Sci. China 6, 930–940. Zhikang, L., Pinson, S. R. M., Park, W. D., Paterson, A. H., and Stansel, J. W. (1997). Epistasis for three grain yield components in rice (Oryza sativa L.). Genetics 145, 453–465. Zhu, Q., Zhang, Z., Yang, J., Cao, X., Lang, Y., and Wang, Z. (1997). Source sink characteristics related to yield in inter-subspecific hybrid rice. Sci. Agric. Sin. 30, 52–59.

Index

A Adequate fertility level (AFL), 278–279 Allometry, 268 Arbuscular mycorrhiza, 176 Aspergillus terreus, 176 B Barley breeding program, 112–113 Bayesian method BayesB, 90–91 BayesCp, 91 expectation-maximization (EM) algorithm, 92 MCMC algorithm, 92 SSVS, 91 Bayesian shrinkage regression, 90 Biodegradation, DOM, 27–28 Burkholderia cepacia, 176 C Cation exchange capacity (CEC), 259–260 calcium uptake, 190 CaO/P2O5 ratio, 190–193 pectinase, root hairs, 189 wheat and pea roots, 189, 190 Chemical oxygen demand (COD), 6 Claassen–Barber model, 265 Complexation. See also Dissolved organic matter clay, 24 DOM–metal(loid) complexes, 23 ESR and Mossbauer spectroscopy, 48 podzolization, 24 soil affinity, 25 total soil mass, 24 Conservation tillage advantages, 277–278 C retention, 276 definition, 276 Lupinus angustifolius, 277 nutrient status, 277 D Denitrification, 4 Deyeuxia angustifolia, 31 Dissolved organic carbon (DOC) hydrological processes, 42–43 leaching, 30

organic forest floor, 20 shallow organic soil horizons, 42 soil and management factors, 41 Dissolved organic matter (DOM) biogeochemical processes, 3 biological fixation, atmospheric CO2, 4 definition, 2, 3 denitrification, 4 DOC, 3–4 dynamics biodegradation, 27–28 Brazilian oxisols, 35 chemical and spectroscopic methods, 35–36 cultivation, 32–33 leaching, 29–30 lime and organic amendments, 33 mechanisms and factors, 23 MWC–FA, 34 photodegradation, 28–29 pig slurry (PS), 35 poultry manure, 34 sewage sludge composts, 34 soil pH, 36–37 sorption/complexation, 23–27 urban waste, 34 urea fertilizer application, 34 vegetation and land use, 31–32 energy substrate, 60 environmental significance atmospheric pollution, 58–60 biosolid application, 48 cattle manure, 49–50 Cd and Cu kinetics, 43 Cu-binding ability, 48 Cu complexation, 48 endogenous, 50, 55–56 erosion control, 37–38 ESR and Mossbauer spectroscopy, 48–49 exogenous, 56–58 gaseous emission, 58–60 metal (loid) bioavailability, 43 metal(loid) complexation capacity, 48 mobility and co-mobility, 43, 48 nutrients mobilization and export, 38–43 pig slurry (PS), 49 soil aggregation, 37–38 soil pH change, 49 macroscale, 61 microscale, 61

361

362

Index

Dissolved organic matter (DOM) (cont.) molecular scale, 61–62 organic matter degradation, 60–61 properties and chemical composition elemental composition, 20–22 FA (see Fulvic acid) structural components, 13–15 sources, pools, and fluxes biological wastes, 5 C cycling, 11 COD, 6 CO2 dry oxidation, 6 in cultivated and pastoral soils, 12 extraction methods, 5, 8–10 freshly deposited litter, 11 input and output pathways, 5, 6 light absorption, 5–6 rhizosphere, 12 soil microbial biomass, 12–13 SOM solubilization, 5 in temperate forests, 11–12 total C budget, 12 undisturbed soil monoliths, 12 E Elastic net regression method, 87 Epistasis and dominance breeding population, 107 genetic model, 107 genome-wide Bayesian method, 108 maize, 108 F Flooding, 274 Fulvic acid acidic properties, 16 aliphatic compounds, 34 chemical and spectroscopic methods, 35 chemical and thermal degradation, 17 complexation capacity (CC), 48 CS-amended soil, 35 elemental composition, 20 ESR spectroscopy, 19 flourescence monodimensional spectroscopy, 18–19 humification, 16 molecular weight, 16–17 NMR techniques, 18 oxidation effect, 19–20 oxygen-containing functional groups, 16 pyrolosis techniques, 17 size and shape, 17 spectroscopic techniques, 18 G Genomic estimated breeding values (GEBV) Bayesian shrinkage model, 106

dairy cattle breeding program, 107 genetic diversity, 104 genetic relationship, 103 maize, 105 Mendelian segregation, 104 mouse population, 104 selection index theory, 106 uniform shrinkage model, 105 Genomic selection (GS) biparental mapping population, 96 dairy cattle, 96–97 GEBV evaluation, 94–95 marker density, type and training population size Arabidopsis family, 98 elite barley, 98 heritability, 99 NIRS, 100 plant height and grain yield, 100 population-trait combination, 101–102 random sampling, 101 selective phenotyping, 99 self-pollinating species, 103 mutation-recombination-drift equilibrium, 95 perennial species, 95 subpopulation structure cow, 111 crops, 111 dendrograms, 110 genetic divergence, 110 GWAS, 109 marker alleles and phenotype, 109 Grain harvest index, 142 H Helianthus annuus, 265–266 I Indole acetic acid (IAA), 349–351 L LASSO. See Least absolute shrinkage and selection operator Leaching, 29–30, 315 Least absolute shrinkage and selection operator (LASSO), 86–87 LECOTM combustion, 6 Leguminous plants crops responses, 209, 210 dinitrogen fixation, 217, 218 pot experiment, 208 soil buffering capacity, 220 soil P availability index, 220 Lolium perenne, 269 Lupinus angustifolius, 277

363

Index M Machigin method, 206–207 Marker-assisted recurrent selection (MARS), 79 Markov chain Monte Carlo (MCMC) algorithm, 92 MARS. See Marker-assisted recurrent selection Micronutrients boron, 305–308 chloride, 311 copper, 309 iron, 309, 310 molybdenum, 311 zinc, 305 Mineral nutrition vs. root growth. See also Micronutrients; Nitrogen; Phosphorus AFL, 278, 279 calcium, 301–302 magnesium, 302–303 potassium, 299–301 sulfur, 303–304 Mulching, 315 Municipal waste compost (MWC), 34 N Next-generation sequencing (NGS) technology, 82–84 Nitrogen in crop plants, 280 cultivar BRS primavera, 282, 284 fertilization, 281 NO3–N distribution, 285 plant biochemistry, 279 root length and root dry weight, 282, 283 root length density, 285, 287 root mass, 281 root morphology, 281 timing of, 283 upland rice, 282 Nitrogen fertilizer crop responses, 197, 199 leguminous, 209, 211 crops responses, 209, 210 dinitrogen fixation, 217, 218 pot experiment, 208 soil buffering capacity, 220 soil P availability index, 220 maize, 197 nodule formation, 202 Nitrogen fixation CEC, 195 gaseous emission and atmospheric pollution, 58 leguminous crops, 145 nodule formation, 199 P fertilizer application, 195, 224–225 P uptake, pea, 201 root exudates, 186 Nonleguminous plants

dryland locations, 205, 206 Machigin method, 206–207 N-supplying capacity, 206 Olson method, 205 O Olson method, 205 Ontogeny, 335–336 Organic contaminants, transformation and transport DOM-induced mobilization and degradation, 50–54 endogenous DOM, 50, 55–56 exogenous DOM, 56–58 Oxidation activity, root adventitious roots development, 276 aerenchyma formation, 274–275 molecular diffusion, 274 oxygen diffusion, 275 rice cross section, 275 waterlogging, 274 P Partial least squares (PLS) and principal component (PC) regression, 89–90 Phosphate fertilizers. See also Soil phosphorus, leguminous and nonleguminous crops in agriculture, 222 CEC, 224 CaO/P2O5 ratio, 190–193 pectinase, root hairs, 189 wheat and pea roots, 189, 190 cereal crops N fertilization, 197, 199 nitrogen addition, 197 nodule formation, 202–203 sensitivity, 193 soil fertility, 195–196 leguminous crops responses, 209, 210 dinitrogen fixation, 217, 218 pot experiment, 208 soil buffering capacity, 220 soil P availability index, 220 maize-maize-soybean rotation sequence, 222–223 N fertilizer nodule formation, 213 nutrient input, 214, 215 pea, 211 root length inhibition, 213 nonleguminous dryland locations, 205, 206 Machigin method, 206–207 N-supplying capacity, 206 Olson method, 205 root exudates, 224

364 Phosphate fertilizers (cont.) amino acids, 183–184 APase activity, 185, 186 carboxylates effects, 182–183 cation and anion balance, 177 citrate-microbial decarboxylation, 182 citric and malic acids, 181 enzyme production, 184 Hedley method, 181 iron oxide-and aluminum oxide, 182 LMW, 180 N2 fixation, 186–187 pH reduction, 187, 188 proton release, 179–180 protons and organic acids, 176–177 rhizophere soil acidification, 177–178 RP utilization, 178–179 root morphology and absorption capacity, 223–224 absorbing surface, 166 amino acids absorption, 164–165 characteristics, 165 diffusion, 163–164 interception, 163 maize genotypes, 165 mass flow, 164 nutrient uptake, 170 orthophosphate, 163 P-efficient genotypes, 165–166 pot experiments, 166–167 red clover and ryegrass, 170 root biomass, 173, 175 root length, 164 root weight, 167–168 ryegrass, 165 TTC, 171, 173 wheat and pea roots, 168–169 Phosphorus components, 287, 291 dry bean, 296–297 dry weight, 294–295 legume, 295 physiology and biochemistry, 287 regression equations, 291–292 root development, 291 root length, 293 upland rice genotypes, 292, 293 uptake amount definition, 141 environmental differences, 142 grain harvest index, 142 manural soils, 143–145 N supplements, 145–146 plant internal characteristics, 142 and P rate, 147 summer-sown crops, 143 Photodegradation, 28–29 Phytate, 131

Index

Plant breeding, genomics epistasis and dominance, 107–109 genomic selection (GS) prediction accuracy biparental mapping population, 96 dairy cattle, 96–97 GEBV evaluation, 94–95 marker density, type and training population size, 97–103 mutation-recombination-drift equilibrium, 95 perennial species, 95 long-term selection, 111–114 MARS, 79 MAS, 79 population and trait characteristics, 80–82 QTL mapping, 80 SNP marker discovery and genotyping de novo sequencing, 83 genetic mapping, 84 Illumina’s GoldenGate assay, 84 NGS technology, 82–84 Sanger’s method, 83 statistical method Bayesian method, 90–92 GEBV accuracy, 103–107 genetic architecture, 93 high-density parallel genotyping technology, 90 LASSO, 86–87 marker effects, 93 partial least squares (PLS) and principal component (PC) regression, 89–90 RKHS and support vector machine, 87–89 RR-BLUP, 85–86 subpopulation structure, 109–111 Plant growth-promoting microorganisms (PGPM), 173, 224 Podzolization, 24 P-solubilizing microorganisms (PSM), 173, 175 Q Quantitative trait locus (QTL), 79 R Random regression-best linear unbiased prediction (RR-BLUP), 85–86 Reproducing kernel Hilbert spaces (RKHS) classical additive genetic model, 87 Gaussian kernel, 89 kernel function, 87–88 Lagrange optimization, 89 standard mixed-model equation, 88 SVR, 88–89 Rhizosphere process, 253 Rice apical dominance axillary meristem initiation, 351

365

Index

cytokinins role, 348 genes and mechanisms, 349 growth-retarding hormones, 350 hormone action, 351 IAA, 348 pattern and distribution rainfall, 348 tiller bud release, 349 during vegetative growth, 350 panicle architecture and grain yield grain number, 339 heterogenous, 337–338 panicle centroid index, 339 spikelet number, 338–339 panicle structure and development heterogenous development, spikelets, 336–337 ontogeny, 335–336 spikelets morphology, 335 spikelet development gene expression profile, 344–345 hormonal regulation, 345–347 metabolic control, 340–341 starch biosynthesis, 341–344 Ridge regression. See Random regression-best linear unbiased prediction RKHS. See Reproducing kernel Hilbert spaces Rock phosphate (RP), 133–134 leguminous and nonleguminous in acidic paddy soil, 152 albic bleach soil, 157 calcareous soil, 153, 155 calcium carbonate, 157 vs. cereal crops, 156 characteristics, 148 in China, 147–148 citric acid extraction, 154 classification, 149–150 corn–oats–alfalfa rotation, 163 crop comparison, 153–154 crop responses, 150–151 N supply, 157–158 P availability, 146–147 physical and chemical properties, 148–149 pot experiments, 157 P-supplying capacity, 150 in red soil, 151 S fertilizer, 153 soil solution pH, 150 SSP, 151–152, 158–161 P fertilizer, 137 Root growth agroecosystems., 255 cereals and legumes carbohydrates and nutrients storage, 256 CEC, 259–260 classification, 257–258 monocots and dicots, 258 parent culms, 259

physiochemical properties, 259 seminal/nodal, 258–259 surface area, 258 conservation tillage systems advantages, 277–278 C retention, 276 definition, 276 Lupinus angustifolius, 277 nutrient status, 277 soil compaction, 277 vs. crop yield, 270–271 cytokinins, 253 depth and distribution Claassen–Barber model, 265 maize, 264–265 nitrogen fertilizer application, 264 soil compaction, 263–264 water and nutrients absorption, 263 wheat, 264 genotypic variation annual crops, 272–273 drought prone environments, 273 nutrient and water stress, 272 penetration speed, 273 vs. mineral nutrition AFL, 278, 279 calcium, 301–302 magnesium, 302–303 micronutrients (see Micronutrients) nitrogen (see Nitrogen) phosphorus (see Phosphorus) potassium, 299–301 sulfur, 303–304 mucilages, 254 organic input, 254 oxidation activity adventitious roots development, 276 aerenchyma formation, 274–275 molecular diffusion, 274 oxygen diffusion, 275 rice cross section, 275 waterlogging, 274 plant age A. hypogaea, 265 dry weight, 266–267 growth parameters, 267 Helianthus annuus, 265–266 panicle initiation, 266 silking, 265 tillering, 266 plant management, genetic variability, 316–317 rhizodeposition, 254 rhizosphere, 253, 255–257 root–shoot ratio allometry, 268 biomass ratio, 269 dry matter partitioning, 269

366

Index

Root growth (cont.) environmental stresses, 269 photosynthates partitioning, 270 photosynthate transport, 268 soil salinity, 269–270 soil aggregation, 253–254 soil management adequate rates and effective sources, nutrients, 315 cropping systems, 315–316 deep plowing, 315 gypsum use, 312–314 leaching, 315 liming acid soils, 312 mulching, 315 organic matter maintenance, 314 pest management, 315–316 total plant weight abiotic/biotic stress, 261 annual plant species, 261 genotype selection, 261 legume cover crops, 261, 263 pigeon pea, 262 P rate, 261 upland rice genotypes, 261, 262 Root morphology and absorption capacity absorbing surface, 166 amino acids absorption, 164–165 characteristics, 165 diffusion, 163–164 interception, 163 maize genotypes, 165 mass flow, 164 nutrient uptake, 170 orthophosphate, 163 P-efficient genotypes, 165–166 pot experiments, 166–167 red clover and ryegrass, 170 root biomass, 173, 175 root length, 164 root weight, 167–168 ryegrass, 165 TTC, 171, 173 wheat and pea roots, 168–169 RR-BLUP. See Random regression-best linear unbiased prediction S Silking, 265 Single nucleotide polymorphism (SNP), 82 Single superphosphate (SSP), 151–152, 158–161 SNP. See Single nucleotide polymorphism Soil organic matter (SOM), 3 Soil phosphorus, leguminous and nonleguminous crops. See also Phosphate fertilizers animal and human excreta, 132–133 CEC

calcium uptake, 190 CaO/P2O5 ratio, 190–193 pectinase, root hairs, 189 wheat and pea roots, 189, 190 cereal N fertilization, 197, 199 nitrogen addition, 197 nodule formation, 202–203 sensitivity, 193 soil fertility, 195–196 crop responses, 139–141 environmental pollution, 136–137 insoluble and souble inorganic P fraction, 131–132 manures, 135–136 vs. metal and fossil fuels, 138 microorganisms, rhizosphere soil, 224 Arbuscular mycorrhiza, 176 Aspergillus terreus, 176 Burkholderia cepacia, 176 microbial population, 173 PSM, 173, 175 organic P fraction, 131 P fertilizers, 135–136 plant cultivation, 138 plant growth and development, 130 P recycling, 135 P uptake amount definition, 141 environmental differences, 142 grain harvest index, 142 manural soils, 143–145 N supplements, 145–146 plant internal characteristics, 142 and P rate, 147 summer-sown crops, 143 P utilization, 138–139 rock phosphate (see Rock phosphate) root exudates amino acids, 183–184 APase activity, 185, 186 cation and anion balance, 177 citrate-microbial decarboxylation, 182 citric and malic acids, 181 enzyme production, 184 Hedley method, 181 iron oxide-and aluminum oxide, 182 LMW, 180 N2 fixation, 186–187 pH reduction, 187, 188 proton release, 179–180 protons and organic acids, 176–177 rhizophere soil acidification, 177–178 RP utilization, 178–179 root morphology and absorption capacity absorbing surface, 166 amino acids absorption, 164–165 characteristics, 165

367

Index

diffusion, 163–164 interception, 163 maize genotypes, 165 soil characteristics, 132 yield and quality, 132 Sorption, DOM andisols vs. other soils, 27 distribution coefficients, 25 hydrophilic fraction, 26 hydrophobic, 25 linear initial mass isotherm, 25 mineral soil horizons, 26 organic C, 25 retention, 23 soil organo-mineral complexes, 26–27 Spikelet. See also Rice gene expression profile, 344–345 hormonal regulation ethylene emissions, 347 GA and kinetin applications, 346 growth regulators, 345 metabolic dominance, 345 in rice, 346 metabolic control, 340–341 starch biosynthesis, 341–344 SSVS. See Stochastic search variable selection

Starch biosynthesis carbohydrate metabolism, 342 enzyme activity, 342–343 gene expression profile, 344–345 phloem loading and unloading, 342 rice endosperm development, 342 sucrose transporter gene, 342 SUS activity, 342–343 Stochastic search variable selection (SSVS), 91 T Total organic carbon (TOC) analyzer, 6 Total plant weight annual plant species, 261 genotype selection, 261 legume cover crops, 261, 263 pigeon pea, 262 P rate, 261 upland rice genotypes, 261, 262 Triphenyltetrazolium chloride (TTC), 171, 173 W Waterlogging, 274 Water-soluble organic matter (WSOM), 3