DVANCES IN
gronomyy
8-
V O L U M5 7E ...-
Advisory Board Martin Alexander
Eugene J. Kamprath
Cornell University
...
23 downloads
1423 Views
25MB 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
DVANCES IN
gronomyy
8-
V O L U M5 7E ...-
Advisory Board Martin Alexander
Eugene J. Kamprath
Cornell University
North Carolina State University
Kenneth J. Frey
Larry P. Wilding
Iowa State University
Texas A&M University
Prepared in cooperation with the American Society of Agronomy Monographs Committee P. S. Baenziger J. Bartels J. N. Bigham L. P. Bush
M. A. Tabatabai, Chairman R. N. Carrow W. T. Frankenberger, Jr. D. M. Kral S. E. Lingle
G. A. Peterson D. E. Rolston D. E. Stott J. W. Stucki
D V A N C E S I N
onomy VOLUME 57 Edited by
Donald L. Sparks Department of Plant and Soil Sciences University of Delaware Newark, Delaware
ACADEMIC PRESS San Diego New York Boston London Sydney Tokyo Toronto
This book is printed on acid-free paper. @ Copyright 0 1996 by ACADEMIC PRESS, INC. All Rights Reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher.
Academic Press, Inc. A Division of Harcourt Brace & Company 525 B Street, Suite 1900, San Diego, California 92101-4495 Uniied Kingdom Edition published by Academic Press Limited 24-28 Oval Road. London NW I 7DX
International Standard Serial Number: 0065-2 I 13 International Standard Book Number: 0-12-000757-6 PRINTED IN THE UNITED STATES OF AMERICA 96 97 9 8 9 9 00 01 BB 9 8 7 6 5
4
3
2
1
Contents CONTRIBUWRS .............................................. PREFACE ...................................................
ix xi
SOILSCIENCEAND ARCHAEOLOGY S. J. Sadder, J. E. Foss, and M. E. Collins Introduction. ....................................... The Archaeological Context .................................... ............. 111. Soils Data Useful in Archaeological Interpretations ............. Paleosols.. . . . . . ............................... .. oil-Archaeological Investig V. Case Studies of Soil. VI. Summary .............................. ................................................. References
I. 11.
w.
2 4 12 21 24 67 68
PHOSPHATE ROCKSFOR DIRECT APPLICATION TO SOILS S. S. S. Rajan, J. H. Watkinson, and A. G. Sinclair Introduction. . . . . . . . . . . Reactivity of PRS . . . . . . 111. Measurement of Phospha Factors Affecting Phosphate Rock Dissolution in Soil and Availability to Plants ..................................... V. Modeling the Rate of Phosphate Rock Dissolution in Field Soil. ......... VI. Agronomic Effectiveness of Phosphate Rock . . . . . . . . . . . . VII. Economics of Using Phosphate Rock Fertilizers ....................... ............. VIII. Soil Testing Where Phosphate Rocks Are Used Ix. Amendments to Phosphate Rocks. ................................... ............. X. Concluding Remarks ................... .............................. . . . . . References
I. 11.
w.
78 79 86 91 111 118 130 133 142 146 146
BREEDINGAND IMPROVEMENT OF FORAGESORGHUMS FOR THE TROPICS R. R. Duncan I.
Introduction..
...........................
................................ ................... 111. Breeding.. . . . .................................. IV. Germplasm.. . ......................... V. Conclusions ....................... . 11. 11.
Genetic Param Parameters tienetlc
References
...
Y
161 162 171 173 175 178
vi
CONTENTS
NITROGEN MINERALIZATION IN TEMPERATE AGRICULTURAL SOILS:PROCESSES AND MEASUREMENT Stephen C. Jarvis, Elizabeth A. Stockdale, Mark A. Shepherd, and David S. Powlson ........................... I. Introduction.. . . ............... ...... 11. 111. Process Controls.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
w. Measurement and Prediction of Mineralization v
The Impact of Mineralization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
VI. Conclusions and Future Progress
References . . . . . . . . . . . . . . . . . . .
188 189 201 208 219 224 226
THEBUFFERINGPOWEROF PLANT NUTRIENTS AND EFFECTSON AVAILABILITY I. 11.
K. P. Prabhakaran Nair Introduction. . . . . . . . . . . . . . . . . . . . . . . . .
Efficient Plant Nutrient Management in Sustainable Soil Management. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111. The Buffer Power and Effect on Nutrient Availability . . . . . . . . . . . . . . . . . . Quantifymg the Buffer Power of Soils and Testing Its Effect on Nutrient Availability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v. The Role of Electro-Ultrafiltration in Measurin K ty for the Construction of Buffer Power Curves.. ... .. VI. Quantifymg the Buffer Power for Precise Availa Prediction - Heavy Metals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII. Influence of Heavy Metal Contamination on Buffering of Major Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VIII. Possible Buffering Effect on Plant Acqu on of Heavy Metals . . . . . . . . . . Concluding Comments and Future Imperatives. . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
w.
Ix.
238 239 242 247 267 270 277 277 278 281
OVERVIEW OF VERTISOLS: CHARACTERISTICS AND IMPACTS ON SOCIETY
I. 11. 111.
Clement E. Coulombe, Larry P. Wilding, and Joe B. Dixon Introduction. . . . . . . . . Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Formation of Vertisols ...............
w. Morphological Propert v.
VI. VII. VIII.
Pedogenic Processes in Classification: From Marbut to Soil Taxonomy . . . . . . . . . . . . . . . . . . . . . . . . Mineralogical Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chemical Properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
290 292 294 301 307 3 16 322 329
CONTENTS M. X. XI . XI1.
Biological Properties .............................................. Physical Properties ................................................ Management of Vertisols ........................................... Summary and Concluding Remarks .................................. References .......................................................
vii 333 336 352 363 364
HYBRID RICE S. S. Virmani Introduction...................................................... Heterosis in ................... Genetic Too ........................... . Breeding Procedures for Developing R m Hybrids ..................... V. Accomplishments ......... VI. Agronomic Management ........................................... ........................ VII. Disease/Insect Resistance .............. VIII . Grain Quality ................................ .............. M . Adaptability to Stress Environm .............. X . Hybrid Seed Production ...... XI . Economic Analysis ................................................ XI1. Technology Transfer and Policy Issues ............................... XI11. Future Outlook ............................. ........................................ X W. Conclusions . References . . . . . . . . . . . . . . . . ................
378 379 393 404 411 419 421 424 425 426 438 442 444 448 449
INDEX .....................................................
463
I. I1. 111.
rv
This Page Intentionally Left Blank
Contributors Numbers in parentheses indicate the pages on which the authors’ contributions begin.
M. E. COLLINS (1) Department of Soil and Water Science, University of Florida, Gainesville, Florida 3261 I CLEMENT E. COULOMBE (289) Department of Soil and Crop Sciences, Texas A 6M University, College Station, Texas 77843 JOE B. DIXON (289) Department of Soil and Crop Sciences, Texas A 6M University, College Station, Texas 77843 R. R. DUNCAN (16 1) Department of Crop and Soil Science, The University of Georgia, Grifin, Georgia 30223 J. E. FOSS (1) Department of Plant and Soil Science, University of Tennessee, Knoxville, Tennessee 3 7901 STEPHEN C. JARVIS (187) Institute of Grassland and Environmental Research, Okehampton, Devon EX20 2DG, United Kingdom K. P. PRABHAKARAN NAIR (237) University of Fort Hare, Alice 5700, Republic of South Africa DAVID S. POWLSON (187) Institute of Arable Crops Research, Rothamsted, Harpenden, Her forshire, A L 5 2yQ United Kingdom S. S. S. RAJAN (77) Department of Earth Sciences, The University of Waikato, Private Bug 31 05, Hamilton, N m Zealand S. J. SCUDDER (1) Department of Anthropology, Florida Museum of Natural History, University of Florida, Gainesville, Florida 3261 I MARK A. SHEPHERD (187) ADAS, Gleadthorpe Research Centre, Mansfield, Nottinghamshire, NG20 9PF, United Kingdom A. G. SINCLAIR (77) Invermay Agricultural Center, 50034 Mosgiel, New Zealand ELIZABETH A. STOCKDALE (18 7 ) Institute of Arable Crops Research, Rothamsted, Harpenden, Her fordshire AL5 2 I Q United Kingdom S. S. VIRMANI (377) International Rice Research Institute, Manila, Philippines J. H. WATKINSON (7 7 ) AgResearch, Ruakura Ayimltural Research Center, 3123, Hamilton, New Zealand LARRY P. WILDING (289) Department of Soil and Crop Sciences, Texas A 6M University, College Station, Texas 77843
ix
This Page Intentionally Left Blank
Preface Volume 57 contains seven interesting and comprehensive reviews on various agronomic topics. The application and uses of archaeology in soil science, including soil data useful in archaeological interpretations and case studies of soil archaeological investigations, are discussed in the first chapter. Advances in the application of phosphate rocks to soils, including their reactivity, aspects of phosphate rock dissolution, and the economics and agronomic effectiveness of phosphate rock additions is thoroughly covered in the second chapter. The third chapter discusses the breeding and improvement of forage legumes in the tropics, with information on genetic parameters, breeding techniques, and germplasm. The processes and measurement of nitrogen mineralization in temperate agricultural soils are the subjects of the fourth chapter. Nitrogen pools and processes, process controls, and the measurement and prediction of mineralization are comprehensively reviewed. The fifth chapter provides a thorough overview of the buffering power of plant nutrients and effects on nutrient availability. Details on nutrient management, quantification of buffering power, and effects of metal contamination on buffering of major nutrients are discussed. The sixth chapter is a comprehensive overview of Vertisols, including discussions on their genesis, morphology, classification, chemical, physical, and biological properties, and management. The seventh chapter is a detailed review on advances in hybrid rice including discussions on heterosis, genetics and breeding tools, hybrids, cultural and management practices, hybrid seed production, economic analysis, and technology transfer. I appreciate the authors’ first-rate reviews.
DONALD L. SPARKS
xi
This Page Intentionally Left Blank
SOILSCIENCE AND ARCHAEOLOGY S. J. Scudder,' J. E. FOSS,~ and M. E. Collins3 'Department of Anthropology, Florida Museum of Natural History, University of Florida, Gainesville, Florida 3261 I *Department of Plant and Soil Science, University of Tennessee, Knoxville, Tennessee 37901 %Departmentof Soil and Water Science, University of Florida, Gainesville, Florida 3261 1
I. Introduction A. New Applications of Soil Science B. Earth Science C. The Land Resource D. Objectives of Article 11. The Archaeological Context A. The Changing Emphasis in Archaeology B. Modem Subfields of Archaeology C. Applications of Soil Science to Archaeology D. Multiple Lines of Evidence E. Consensus and Controversy 111. Soils Data Useful in Archaeological Interpretations A. Soil Surveys and Maps B. Soil Morphology C. Soil Laboratory Analyses D. Landscape Analyses E. Micromorphology IV.Paleosols A. Buried Paleosols: Keys to Archaeological Interpretations B. A Paleosol Case Study V. Case Studies of Soil-Archaeological Investigations A. Alluvial Sequences in the Southeastern United States B. Soil Studies at the El Mirador Bajo C . Chemical Properties of Soils at Hadrian's Villa, ltaly D. Paleosols near Mt. Vesuvius E. Geomorphology and Site Selection a t the Seminole Rest Site, Volusia County, Florida F. Soils and Landscapes: Archaeopedology at the Pineland Site G. Pedoarchaeological Analysis of D Prehistoric Shell-Bearing Island, Florida VI. Summary References
1 Advonces in Agronomy, Volume 57
Copyright 0 1996 by Academic Press, Inc. MI rights of reproduction in any form reserved.
2
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
I. INTRODUCTION A. NEWAPPLICATIONS OF SOILSCIENCE Soil science has traditionally been closely associated with agricultural enterprises such as cropping systems, forestry, and other types of land use where plant growth was the dominating interest. Today, however, soil scientists are using principles developed in the past century or more for many applications in addition to the more conventional plant adaptation and response. One of the first major movements toward applications of soil science outside the agricultural realm was in the soil survey program, which began moving into urban planning and interpretations in the 1960s. These activities were highlighted by the publication in 1966 of Soil Surveys and Land Use Planning (Bartelli et al, 1966). Since that time increasing numbers of soil scientists have been working in numerous other areas such as soil archaeology, forensic science, global cycling of nutrients, geographical information systems (GIs), water pollution, environmental hazards and use of waste products, solute transport, and genetically engineered organisms. Potential developments of the major subdisciplines of soil science were outlined in the golden anniversary publication of the Soil Science Society of America (Boersma et al., 1987).
B. EARTHSCIENCE Earth science is a broad study area that includes numerous subdisciplines such as soil science, geology, hydrology, geophysics, geochemistry, oceanography, climatology, archaeology, and meteorology. It is a relatively new concept in comparison to the disciplines listed above, but it is particularly relevant today due to the widespread interest in a more holistic view of the Earth and its many processes. Earth science has many opportunities to contribute to current issues such as environmental change; evaluation of physical resources such as water, soil, minerals, and wetlands; sustainable development; land use decisions, and understanding basic Earth processes. Archaeological studies provide an excellent avenue to approach some of the Earth science issues mentioned above. Combining subdisciplines such as soil science and geology with archaeological investigations results in detailed environmental studies of the Holocene. The Holocene is generally defined as that portion of the Quaternary period from 10,000 or 12,000 years ago to the present time. Archaeological studies are particularly important in developing the chronology, evidence of environmental change, and the impact of human populations on landscapes during the Holocene. Studies of historical land use and the impact
SOIL SCIENCE AND ARCHAEOLOGY
3
of human populations on the environment have been especially revealing at archaeological sites associated with the Romans (Foss et al., 1994a; Stiles et al., 1994; Olson, 1981), in Mayan cultural areas in Central America (Olson, 1977; Dunning, 1993; Dahlin et al., 1980), the Middle East (Olson, 1981; Pendall and Amundson, 1990); and Peru and New Mexico (Sandor, 1992).
C. THELANDRESOURCE Significant progress has been made in our appreciation of the value of the land resource in the past half century or more. At one time, land was simply thought to be used, sometimes abused, and then abandoned with minimal regard to environmental impact and future use. Today development projects must pay close attention to the land resource, e.g., many projects require the identification and delineation of prime agricultural lands, wetlands, aquifer recharge area, and archaeological potential. The concept of sustainable development requires that concern for the long-term impact on natural resources be considered in development projects. The land resource includes soil, water, vegetation, animals, man-made features, and archaeological resources (Vink, 1975). Archaeological resources are considered so important that construction projects such as roads, bridges, and buildings that are federally funded must include an archaeological assessment and impact statement. If significant findings are recorded, mitigation procedures are then developed. These activities have resulted in a proliferation of archaeological investigations and have thus provided opportunities for joint study by archaeologists and associated scientists, especially in soil science and geology.
D. OBJECTIVES OF ARTICLE The objectives of this article are threefold. First, we stress the mutual advantages of developing new interdisciplinary efforts in the earth sciences, especially in archaeology. Although soil scientists have aided in the understanding of archaeological sites, we have also been able to effectively study and understand soil weathering processes because of the detailed chronology established at these sites. Table 1, for example, shows the approximate age of diagnostic soil horizons that have been identified in a number of archaeological sites. Second, we summarize some of the unique contributions of pedology to archaeology. Many interdisciplinary soil-archaeological studies have been published in the past two decades. A sampling of those studies, including several that we have been associated with, is presented here in order to illustrate the broad range of applications of soil science to archaeology. We feel that these
4
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS Table I Length of Time for the Formation of Various Soil Horizons Observed at Some Archaeological Sites (Foss el al., 1995)
Horizon Cambic Argillic
Fragipan
Approx. age (ye=) 2000 250-3000 3500-4000 3000-4000 3500 3000-4500 8600 8800 6500 8200
Archaeological site Shawnee-Minisink R. B. Russell Thunderbird Fifty Site Flint Run R. B. Russell 38LX338 40PK27 Fifty Site 40CH 162
Location Delaware River, Pennsylvania Savannah River, Georgia-South Carolina Shenandoah River, Virginia Shenandoah River, Virginia Shenandoah River, Virginia Savannah River, Georgia-South Carolina Saluda River, South Carolina Polk County, Tennessee Shenandoah River, Virginia Harpeth River, Tennessee
projects typify the approach and contributions that pedology can make to developing archaeological site history. Third, we hope to encourage other pedologists to become involved in archaeological studies. Interdisciplinary studies are rapidly becoming the norm in science, and soil science has a unique opportunity to collaborate with other earth sciences in understanding the numerous earth processes. Archaeological sites provide perhaps the best and most complete look at the Holocene and the impact of the human population on the environment.
11. THE ARCHAEOLOGICAL CONTEXT A. THECHANGING EMPHASIS IN ARCHAEOLOGY The classical concept of archaeology invokes collection and analysis of cultural artifacts, construction of ceramic typologies and cultural chronologies, and tracing the rise and fall of civilizations by cataloging the material evidence of human endeavor. This artifact-based approach has a venerable history, producing painstaking reconstructions of both the broad scope of cultural evolution and the mundane details of daily life. Increasingly, however, questions of resource use, human impact on natural environments, and the application of archaeologically derived information to the reconstruction of past environments are redefining the focus of archaeology. Human ecology is becoming a unifying theme, placing past cultures in the context of natural landscapes, of changing climate and re-
SOIL SCIENCE AND ARCHAEOLOGY
5
source bases, of developing (or failing) horticultural or agricultural systems (Butzer, 1982; Marquardt, 1992a). The cultural markers that identify human groups are still important, but they are now more often employed not simply to name or categorize those groups, but to track them through complex relationships with other human groups and with a changing mosaic of landscapes and plant and animal communities on which their lives depended. The emergence of this new brand of archaeology-termed “bioarchaeology” by many (Larsen, 1987)-has resulted in the alignment of many earth science disciplines with social sciences, or anthropology. The products of these new alignments emphasize the multidisciplinary nature of modem archaeology.
B. MODERNSUBFIELDS OF ARCHAEOLOGY 1. Geoarchaeology Geoarchaeology places the site under study in a local and regional geomorphic setting. Geomorphology, sedimentology, structural geology, hydrology, and pedology are used to define the natural landscape elements which surrounded or were incorporated into past human settlements. Those elements, such as alluvial terraces, floodplains, deltas, and others, possess characteristic sequences of sediments, drainage patterns, and soils, which offered unique resources to past human settlers (Ferring, 1992; Gartner, 1992; Segovia, 1985). Disruptions in natural processes of landscape evolution, initiated by human alterations and manifest as erosion or accumulation, can be detected by comparing archaeological stratigraphic sequences with modern regional geomorphology. Such information can be used to infer the scale of human impact on the local environment (Holliday, 1985; Farrand, 1975).
2. Archaeometry Archaeometry employs remote sensing methods, techniques for chronometric dating, and materials identification and sourcing. Geophysical methods of remote sensing such as magnetometry and resistivity detect differences in subsurface concentrations of iron and water or salts, or material density, to locate buried foundations, ditches, metal artifacts, and other physical or chemical interfaces (Carr, 1982). Ground-penetrating radar (GPR) reflects radar pulses off of subsurface discontinuities such as extreme soil textural changes (e.g., sand to clay), rock layers, water tables, and buried architectural features (Collins and Doolittle, 1993). Geochemical “prospecting,” such as soil phosphate testing on a radial grid at arbitrary levels, gives a three-dimensional map of subsurface element distribution (e.g., Hassan, 1985).
6
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
These methods, used in soil mapping and landscape analysis, are readily applicable to archaeological situations, particularly as reconaissance tools to indicate where to locate excavation units, to what depths to dig, and where not to waste valuable time and money. (See Section III.B.2 for a further discussion of GPR.)
3. Archaeopedology The “study of old soils” characterizes and interprets both paleosols (which have been formed by natural processes) and anthrosols (which have been created by human activity) in archaeological contexts. Paleosols may underlie archaeological sites as complete soils with all horizons intact, or may be truncatedusually by erosion. They are excellent indicators of climate change or stability, forming under specific conditions which either contrast to or corroborate modem conditions and the soils which result from them. They have been used as markers of Holocene/Pleistocene boundaries in some areas (see, for example, Goodyear and Foss, 1993) and therefore can be used to predict or prospect for early human sites. Paleosols are discussed more extensively in Section IV. Anthropogenic soils or anthrosols are as diverse as the cultures which produced them. They may take the form of a simple midden-an aboriginal garbage heap mantling a native soil surface-or a thick series of strata reflecting periods of habitation, intermittent abandonment, and resettlement (e.g., Bullard, 1985). The taxonomic categories erected for human-influenced soils reflect augmentation of native soil resources: the plaggen epipedon describes long-term manuring of fields, the anthropic epipedon is one enriched in phosphorus contributed by human activities, the agric subsurface horizon forms under cultivation and contains significant amounts of illuvial silt, clay, and humus (U.S.D.A., 1975). Many anthrosols fit this model of human enrichment of soils, whether purposeful, as in the case of plaggen soils, or fortuitous, as in the case of “black earth” midden soils. Settlements, animal corrals, food preparation areas, aboriginal refuse heaps, and urban dumps all result in increased soil contents of organic carbon, nitrogen, phosphorus, calcium, magnesium, potassium, and other elements (e.g., Eidt, 1985; Griffith, 1980; Lillios, 1992; Smith, 1980). Some soils, such as those formed in spoil from ditching for raised-field agriculture, canalizing, and the formation of monumental earthworks, exhibit “reverse stratigraphy,” with material from excavated subsurface horizons mounded or spread on top of the original land surface. In such cases, particle-size and element distributions, as well as clay-size mineral occurrence, may appear as a reverse weathering sequence with more enriched materials situated above more depleted ones (Birkeland, 1984; Johnson and Collins, 1993). The original soil surface may present an abrupt increase in organic carbon, or a discontinuity in particle size or specific mineral content (see case study on Pineland, Section
SOIL SCIENCE AND ARCHAEOLOGY
7
V.F). With the passage of time and continued weathering, these “inverted soils’’ eventually begin to develop a more normal sequence of horizons and element distributions (Sokoloff and Carter, 1952). One end-member of this “anthrosol continuum” is the depleted or eroded soil resulting from cultural factors such as repeated cycles of unfertilized cropping, poor irrigation or tillage practices, and overstocking. Natural drought or climate changes can also result in such soils if humans are unable to adjust management practices to new conditions (Sanchez, 1976). These soils contrast with enriched anthrosols as well as native, non-human-impacted soils, in decreased contents of major and trace elements removed through cropping or overgrazing. In many cases, the structure of these soils is destroyed or weakened by removal of organic matter, by erosion, compaction, or increased acidity or salinity resulting from drainage or improper irrigation. (Hodges and Carlisle, 1979; Jenny 1961). The value of anthrosols to the interpretation of human lifeways and ecology is multifaceted. From an archaeological perspective they chronicle human landuse, recording patterns of settlement and landscape modification on a large scale. On a smaller scale, they identify intrasite features such as hearths, burials, storage pits, and garden plots. Intensity of use or duration of habitation can be inferred by comparing the chemical and physical characteristics of anthrosols with those of local, nonimpacted native soils (Eidt, 1985; Lillios, 1992). Soil inclusions such as pollen, phytoliths, seeds, and microfauna can be extracted from the mineral matrix, identified, and used to interpret past climate, resource use, and postdepositional soil conditions (Pearsall, 1989; Piperno, 1988; Ruhl, 1995; see also case study: Seminole Rest, Section V.E). From a pedologic perspective, anthrosols offer a unique means of studying soil-forming events by “manipulating” one or more of the classically recognized soil forming factors (Jenny, 194I), especially parent material, biological activity, and/or time. Most archaeological deposits have chronologic indicators such as ceramic or tool types characteristic of specific time periods. Styles of shell tool and lithics manufacture serve to relatively date preceramic deposits. Disturbance of site content and contexts has always presented an interpretive challenge to archaeologists, but a site with no indication of relative age is rare. Also, absolute dating of site materials such as bone, shell, and charcoal using radiometric 14C is now routinely accomplished. So the archaeological site presents the pedologist with a deposit of “parent material”-that is, the midden, household ruin, corral, garden plot, etc.-resultant from the biological activity of humans (and their commensal animals and plants) over a specified period of time, under the influence of a definable climate. The study of the effects of time and parent material in the soil-forming equation is further enhanced by the fact that the local modern soil can be compared not only with the anthrosol, but also with the buried soil or paleosol below the archaeological deposit. Catenas, chronosequences, and clay
8
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
mineral changes, have all been defined and studied (e.g., Holliday, 1985, 1992; Goodyear and Foss, 1993; Foss et al., 1993; Scudder, 1993) using sequences of anthrosols created by past humans interacting with specific environments.
C. APPLICATIONS OF SOILSCIENCE TO ARCHAEOLOGY 1. Human-Modified Landscapes and Settlement Patterns Human effects on natural soils and landscapes can be detected using the most basic tools of pedology: soil morphology, particle-size distribution analysis, clay mineralogy, and patterns of chemical element accumulation. Lippi ( 1988), working on the Nambillo ridgetop site in the western rainforest of the Ecuadorian Andes, constructed a “paleotopographic” map based on soil morphological information from augered cores. The abrupt upper boundaries of a series of four superimposed paleosols, developed in volcanic sediments, were plotted to reconstruct changing local paleogeomorphology. Phosphate accumulations and cultural materials, which spanned approximately 3000 years, were also mapped, creating a composite interpretation of land use patterns and ancient topography. Lippi’s subsurface mapping revealed that the ridgetop landform evolved from a sharply peaked, steeply sloped hilltop to a broader-topped, more habitable one. It also chronicled cycles of human settlement, volcanic activity, abandonment, and resettlement of that evolving hilltop. Research by Smith (1980) in the Neotropics also applied the methods of pedology to the detection and interpretation of past human settlements. Smith, working in Brazil, analyzed physical and chemical properties of terra preta, the black earth soils found scattered in a mosaic throughout large areas of the Amazon basin. He found high contents of phosphorus and calcium, higher than normal pH and base saturation, and cultural artifacts scattered throughout the black soil epipedon but missing from the subadjacent horizons. This led him to confirm that the terra preta were indeed anthropogenic and that earlier theories of their origin as volcanic ash, Tertiary lake sediments, or modern pond sediments were inaccurate. Their occurrence in areas of Oxisols, Spodosols, and Ultisols and their wide range of textures-from silty clay to sand-argued against a common pedogenic origin. Smith estimated accumulation rates for terra preta soils and population densities of past village sites based on modern analogy and areal extent of the black earth sites. He then concluded that the strong evidence for large pre-European human populations in the Amazon basin, presented by the widespread occurrence of terra preta soils, should serve as a warning to biologists discussing “virgin” Amazonian ecosystems. [“Potsherds and black earth may lurk under control plots and pristine natural reserves.” (Smith, 1980, p. 566.)]
SOIL SCIENCE AND ARCHAEOLOGY
9
In a different timeframe and landscape, Farrand (1975) interpreted the paleoclimate of southwestern France based on sediments in the Abri Pataud rock shelter at Les Eyzies (Dordogne) (see also Movius, 1975). He combined detailed data from particle-size distribution analysis, heavy mineral identification and quantification, clay weathering sequences, and sedimentary geology and stratigraphy with 14C dates and analysis of cultural remains to reconstruct local conditions in Upper Paleolithic times (35,000 to 20,000 B.P). The shelter itself, a reentrant undercut at the base of a bedrock limestone cliff, was determined to have formed primarily by differential solution. Frost-shattering of the walls and ceiling contributed the majority of the floor sediments (a total of 9 m in depth), with minor contributions from aeolian and anthropogenic sources. Because particle-size class ratios were similar between the fine material contributed from the weathered limestone and that derived from dry-season wind-blown fluvial sediments from the river below, Farrand used clay and heavy mineral assemblages to differentiate the two sediment types. He was then able to delineate periods of relatively dry climatic conditions based on increased content of aeolian sediments in the floor deposits. Changes in ratios of montmorillonite to kaolinite also signalled broad changes in annual precipitation: increased montmorillonite indicated drier times with less weathering, increased kaolinite denoted the reverse. Increases in pedogenic clay in the stratigraphic column were used as a marker of relative climatic stability. Farrand’s study also addressed regional changes in temperature. Variations through time in the size of limestone fragments spalled from the ceiling and walls of the shelter allowed inferrences to be made about changes in intensity of the freeze-thaw cycles, and hence, changes in general seasonal temperature ranges.
2. Traces of Daily Life The same techniques used to interpret ancient landscapes and paleoclimates can yield information on intrasite features such as individual post-holes, hearths and storage pits within rooms, burials (with or without bodies), and small-scale land use applications (e.g., garden plots and corrals). Particle-size distribution analysis, pH, clay mineralogy, and chemical element distribution patterns, all used in the more traditional realm of soil classification, are now used with increasing frequency in anthrosol analyses. These methods, coupled with innovative micromorphological interpretations gleaned from fixed soil-column thin section techniques (Courty et al., 1989; Goldberg and Courty, 1993), offer an an exhaustive array of analytic techniques. Soil phosphorus quantification, in particular, offers a reliable means of assessing the impact of human occupation on native soils. Early work by European geographers revealed positive correlations between abandoned settlements and elevated soil phosphorus content (Arrhenius, 1929; Broadbent, 1981; Proudfoot,
10
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
1976). Refinements in field and laboratory techniques have resulted in the development of a quick, relative spot test well-suited for site reconnaisance work (Eidt, 1985) and in sophisticated fractionation procedures designed to sequentially remove loosely bound, occluded (tightly iron- or aluminum-bound) and residual calcium phosphates from the soil (Chang and Jackson, 1957; Eidt, 1985). Ratios of the various soil phosphate fractions from archaeological soils have been compared with those from modern soils under known land-use regimes to infer past practices (Eidt, 1985; Lillios, 1992). Such detailed fractionation analyses are not without their critics, in terms of both theoretical application (Courty et al. , 1989; White, 1978) and practical considerationsof execution time and cost (Conway, 1983), but they will undoubtedly undergo more refinement and reapplication. Research by Conway (1983) illustrates the use of total phosphorus distribution patterns in the analysis of small-scale occupation deposits. Working on a walled Romano-British hut group in northern Wales, Conway sampled both withinstructure areas and adjacent fields on a linear grid system at 1-m intervals. Because the site presented culturally layered multicomponent deposits, he interpreted the resultant data (drawn as distribution contour maps) using Trend Surface Analysis (Unwin, 1975). That technique compared observed phosphorus distributions with a series of mathematically derived distributions based on natural soils, producing two components: a generalized distribution-the “trend surface’’-and a series of residual values-the differences between the derived and observed values. The variance of the residuals from the derived trend surface provided a measure of significance. Conway found that by matching attributes of known archaeological features (e.g., size, shape, spatial arrangement) with patterns of phosphorus distribution, he was able to greatly increase understanding of the function of those features. For example, one building showed evidence of having been demolished and partially reincorporated into the courtyard of a subsequent structure. The floor area of the remnant original structure was protected by a layer of small stones and contained high levels of phosphorus. That portion of the floor which was subsequently converted into a courtyard, unprotected by stones, had less total phosphorus, having lost it by exposure and erosion. Such an analysis provides not only a means of relatively dating associated structures, but also an approximation of duration or intensity of occupation through comparison of intrasite phosphorus accumulation with the phosphorus content of nonanthropogenic local soils.
D. MULTIPLELINESOF EVIDENCE Archaeopedological studies based on standard soils analyses, such as those discussed above, are becoming integral to archaeological investigations. Still
SOIL SCIENCE AND ARCHAEOLOGY
11
other studies, focusing on biological soil components, are also contributing to the multidimensional approach evolving in modem archaeology. Botanical residues such as pollen, phytoliths, seeds, charcoal, and macroremains (stems, leaves), embodying the discipline of archaeobotany, are being cross-correlated with soil characteristics to infer horticultural practices. For example, Johnson and Collins (1993) hypothesized that the high soil aluminum content of in-filled aboriginal ditches at the Fort Center site in south Florida would have provided poor growing conditions and thus explained the virtual lack of corn pollen in soils of that area. His findings call into question the speculation that corn horticulture was responsible for supporting the dense populations and high cultural complexity of the Fort Center people. Jacob’s work in the Cobweb Swamp of Belize (1991) combined field archaeology, pedology, palynology, geomorphology, stable isotope analysis, and invertebrate biology to trace the “agroecological evolution” of an agricultural area adjacent to the Mayan cultural center of Colha. Soil cores combined with pollen analysis and identification of soil invertebrate infauna and microfossils revealed a series of changing environments in this region of Mayan civilization: freshwater cattail swamp, mature forest or forest edge, disturbed vegetation and cropland, abandoned wetland. Cultigens in the pollen record and ditched fields provided evidence of Mayan horticulture. Jacobs used soil profiles, characterization data, and stable isotope analysis of marl carbonates to define episodes of massive upland erosion marked by the genesis of a clay horizon at the swamp margin known as the Maya clay. This episode of human-induced erosion was followed by site abandonment-or at least a significant reduction of the human population. Jacobs superimposed his reconstruction of the ecological evolution of Cobweb Swamp onto work conducted at the adjacent cultural center of Colha. At Colha the Maya clay also appeared, at the time of major social stress that preceded the collapse of Mayan civilization. Although his archaeopedological work could not definitively answer the question of what caused the collapse-the social or the ecological turmoil-it provided substantial clues to conditions at the time.
E. CONSENSUS AND CONTROVERSY There is no question that archaeopedology is becoming an integral component of environmental archaeology, or geoarchaeology. The evolution of attitude regarding the treatment of archaeological soils, from an inert matrix to be disposed of to a rich source of environmental and cultural information, is nothing short of revolutionary. The formalization of the subdiscipline and the increasing communication among its practitioners may be signalled by the inception of professional meetings such as the First International Pedo-Archaeology Conference in 1992 in
12
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
Orlando, Florida. Proceedings of this and subsequent meetings gather and present the works of geographers, archaeologists, soil scientists, and geologists, all focusing on the interpretation of anthropogenic soils and landscapes. One of the functions of the pioneers in any field is to define terms, formulate procedures, and set protocols so that investigations done in diverse settings by a variety of people will be comparable and usable. Archaeopedology is fortunate to have the entire body of soil science to draw from (see Section 111) in terms of soil taxonomy and nomenclature, models of soil/landscape relationships, and field and laboratory procedures. However, the application of these principles and practices to anthropogenic soils presents some unique problems. For example, how should samples be taken? By cultural level, or natural horizon? By an arbitrary 10- or 20-cm level? How should cultural inclusions in the soil, such as charcoal, lithic debitage, or ceramic sherds, be treated? Should they be screened out of the 2-mm fraction typically used in soil characterization analyses, or quantified as pebbles, cobbles, or soil separates? (See Stein, 1987 for a discussion of this question.) How should soil carbon be quantified? By the standard Walkley-Black procedure which does not address charcoal; by loss-on-ignition, which is not comparable to standard soil characterization data, or by a weight per volume measure? And what analytical procedures should be used on soils as radically altered as some anthrols are: available elements? total elements? fractionated elements? (See Courty et ul., 1989 for a discussion of the applicability of some soils analyses to anthrosols.) Another question that needs to be addressed to aid in linking archaeology and pedology is what to call these human-influenced soils. The terms anthrosol and anthropogenic soil give the reader a general meaning. But there is a world of difference between the black earth (terra pretu) featured in Smith’s work in Amazonia and the spalled cave ceiling detritus and lithic debitage of Farrand’s Abri Pataud sediments. Conways’s Welsh hut floor samples bear no relationship to Jacob’s Maya Clay, but they are all anthropogenic. Perhaps a serious consideration of anthrosols as a recognized soil order is due. However, though there may be questions of method and theory yet to answer, the range of topics addressed by soil science leaves no doubt that it has become an indispensible tool in archaeology.
111. SOILS DATA USEFUL IN ARCHAEOLOGICAL INTERPRETATIONS A. SOILSURVEYSAND MAPS One of the invaluable resources available for archaeological investigations and interpretations is soil maps. Soil maps are part of a soil survey which is an
SOIL SCIENCE AND ARCHAEOLOGY
13
inventory of the soil resources in a specific area. Soil surveys are completed by field soil scientists on a county-wide basis or a particular part of the state, but some soil surveys may include only part of a county or two or more counties. All soil surveys are made by describing and classifying soils in the field and delineating the areas on maps. Aerial photographs have been used as the base map for most of the soil surveys. Soil series are used to name the map units of the soils delineated in the soil surveys. Soil series are defined and differentiated by one or more of the following significant characteristics: kind, thickness and arrangement of soil horizons along with the color, texture, structure, consistence, pH, content of carbonates and other salts, content of coarse fragments, and mineralogy. Each soil map unit is a collection of areas defined and named according to its soil components (Soil Survey Divison Staff, 1993). Each individual area shown on the soil map is a delineation. The areas that can be delineated on a soil survey depend greatly upon the map scale. The map scale chosen depends on projected use. State general soil maps are published at scales of 1:250,000 to 1:3,000,000 (Soil Survey Division Staff, 1993) and are used for regional to state planning board land-use needs. The map scales used in making local soil surveys generally range from 1:12,000 to 1:24,000. A scale of 1:15,840 was very common, but many soil surveys were converted to the 1:24,000 scale to be used more easily in GIs. Soil map scale influences the size of the soil delineation. Some soil maps are used for national planning with minium-size delineations of 252 to 4000 ha. Others are utilized for local decision making with minium-size delineation of 1 ha or less (Soil Survey Division Staff, 1993). Therefore, the archaeological objective(s) for using soil maps must be determined, so that the level of detail is known and the correct map scale is employed. Some soil maps have been stored in a computer (digitized format) and can be used in a GIS format. The GIS can be especially useful, for example, to quickly search for a particular soil map unit throughout a county that may contain cultural features. Duncan and Hurt (1993) used GIS technology to test a predictive archaeological site model. They used digital soil survey spatial data, soil survey attribute data, and digital known archaeological sites in GRASS (Geographical Resources Analysis Support System) to develop a statistical predictive model. Using this model they were able to adaquately predict archaeological site locations.
1. Use of Soil Maps for Archaeological Purposes Although the locations of cultural features are not identified or delineated in soil surveys, many soil properties or characteristics described may be very useful in archaeological investigations. For example, the landform and physiographic position of the soil map units are described in each soil survey. Many investiga-
14
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
tors have noted a relationship between the landform or landscape position and the presence of cultural features. Foss et af. (1995) indicated that 80 to 90% of the known stratified and well preserved archaeological sites in the eastern United States are in alluvial valleys. They also noted that alluvial soils were not shown with much detail in the older soil surveys, but that more recent soil surveys provide much more detailed data on alluvial soils that can be useful in archaeological investigations. Foss et af. (1995) pointed out that the relationship of the location of Holocene alluvial soils and archaeological sites has resulted in a number of cooperative soil-archeological investigations recently. The natural soil drainage of the soil map units is also determined and described in soil surveys. Drainage and the presence of cultural features also may be related. Distribution of secondary carbonates has been useful in reconstructing paleoenvironments. Other important soil characteristics described and interpreted in soil surveys include the parent material which is useful in determining the age of the cultural features. Ferring (1992) noted the importance of understanding alluvial soils in defining and correlating stratigraphic units in archaeology.
B. SOILMORPHOLOGY Pedology is the science of studying soils as natural bodies on the Earth’s surface. Some describe pedology as “the study of weathering of materials” (Morris et af., 1993). To be able to study soil, one must be able to represent the soil continuum as discrete soils. Therefore, the morphology of a soil is the most important diagnostic criterion in soil science. Without an objective description of soil morphology, no correlation can be made among soils. When a soil is properly described, samples are normally taken by horizon designation. These samples are analyzed for specific soil constituents. The results of the laboratory analyses and the morphology allows one to (i) classify the soil at the soil series level, (ii) understand the soil’s genesis, (iii) determine soil variability by comparing its properties to other known soils, and (iv) interpret suitabilities for land use by noting depths to mottles, bedrock, and restrictive horizons, as well as documenting hydric soil indicators for wetland determinations. The soil’s morphology can also be used to determine if the soil has been disturbed or if the features noted are the result of pedogenic processes. Soil morphological descriptions are made best in an excavated pit large enough to examine any soil variations. The pit should be about 1 m in width so that the vertical face can be studied and described. The description begins by noting the depth or location of changes in the soil. These variations include changes in soil color, texture, structure, and stoniness. Boundaries between soil horizons are marked on the face of the pit, and the horizons are described. Master soil
SOIL SCIENCE AND ARCHAEOLOGY
15
horizons and subordinate designation(s) are assigned to a section of the soil. Peds (natural soil aggregates) are removed so that closer observations can be made. All of the information is written on a soil description form that includes information about the soil’s environment. Archaeological descriptions of soils vary considerably from pedological descriptions. Pedological descriptions assumes no human-influenced disturbance except to the soil surface. Archaeologists, of course, are very interested in features in the soil that are the result of human habitation (Fig. 1). In their study of the San Luis Archaeological Site, Collins and Shapiro (1987) discussed morphological features such as abrupt, smooth boundaries between layers; abrupt, laterally discontinuous layers; dark matrix colors extending to depths greater than expected; weakly developed soils as indicated by the lack of argillic (Bt) or cambic (Bw) horizons; and textures high in sand with mixings of clay in lower layers as evidence of human activities. Some archaeologists (such as Bettis, 1992; Scudder, 1993; Johnson, 1991; and Mandel, 1992) use pedologic terminology to describe soil features. Many archaeologists describe soil color according to the Munsell color notation, describe soil texture according to the U.S.D.A. soil textural triangle, and in some cases use horizon nomenclature.
Figure 1 Example of archaeological soil morphology, Ft. San Luis, Leon C o . , FL.
16
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
1. Soil Morphology as a Stratigraphic Marker Holliday (1989) stated that the earliest use of soils in archaeology was probably as stratigraphic markers. Pedologic features such as soil horizons may be most prominent, especially in stratified deposits. Holliday (1989) noted that in North America, archaeological sites in thick, well-stratified deposits with distinct buried soils are relatively common. He provided numerous examples throughout the United States of the importance of the recognition of buried soils in the recovery and interpretation of archaeological records. In the midwestern United States buried soils have been recognized for many years in the study of Quaternary stratigraphy. Follmer (1978) noted that the Sangamon paleosol was the significant stratigraphic marker in loess/till sequences in Illinois in the 1870s, and was considered one of the most significant stratigraphic units in the world. Recognition of the presence of an argillic horizon (horizon with signficant increase in clay content) can be significant in archaeological studies. It is known that it may require thousands of years to form an argillic horizon. In contrast, cambic horizons (horizons with color development but lacking the increase in clay content) may form quite rapidly. Morris et al. (1993) used basic soil science techniques to study the depositional history and pedogenic properties of two archeological sites in Tennessee. They concluded that soil morphology was the key to the interpretation of the validity of the site in the archaeological context. Also, soil morphology helped determine what processes have affected the site since the time of its deposition. In the study by Moms et al. (1993), a backhoe was used to expose the soil profile, and the physical properties of each soil horizon were described, including color, texture, structure, consistence, and presence of clay films, roots, pores, and concretions. Soil samples were collected from each horizon, and the particle size, total carbon, and other chemical analyses were completed. Results of the study showed that although the two sites in their study had a similar depositional history, the soil morphological development at one of the sites was consistent in the archaeological context while the other site was not. One of the sites had a well-developed buried paleosol. The paleosol had a moderate grade of structure and discontinuous clay films in the argillic horizon. The paleosol also contained charcoal and lithic artifacts and the radiocarbon date was consistent with the soil morphology of the paleosol. The other site exhibited a weaker grade of structure in the paleosol and lacked clay films which would be expected to occur in that period of soil development. Morris et al. concluded that the latter soil morphological features indicated that the paleosol was in an early stage of development when buried and, thus, was not as old as the other. Many archaeological investigations have benefited from pedologic interpretations. Foss et al., (1993) outlined a study of archaeological sites with a soil science approach. They discussed a general survey of the site to better under-
SOIL SCIENCE AND ARCHAEOLOGY
17
stand the general soil characteristics and stratigraphy, examined soil morphology; interpreted the morphology together with the laboratory analyses; and assembled all the information including archaeological data to develop a site history plan. Foss et al. (1993) stressed that accurate field descriptions are the key to success in pedoarcheological investigations. They emphasized that soil morphology is the basis for interpretations and that laboratory data are needed to enhance and clarify the field soil descriptions.
2. Geophysical Tools Applicable to Soil Science and Archaeology Geophysical tools such as resistivity, electromagnetics, and ground-penetrating radar (GPR) have been used by both soil scienists and archaeologists in their field investigations. An advantage (but also a disadvantage) in using these tools is that no physical samples are taken. This is an advantage when a site should not be disturbed. It is a disadvantage when you need a sample for other purposes, e.g., laboratory analyses. GPR is increasingly being used by archaeologists for nonintrusive detection of archaeologically significant buried assemblages. GPR has been very useful in archaeological investigations because it can quickly detect possible cultural features without disturbing the site. In the past 10 years there has been a tremendous increase in interest and GPR use by archaeologists and others studying archaeological sites (Batey, 1987; Doolittle, 1988; Doolittle and Miller, 1990; Goodman and Nishimura, 1992; Imai et al., 1987; Kong et al., 1992; Mellett, 1992; Sakayama et al., 1988; Unterberger, 1992; Bauman et al., 1994; Bernabini et al., 1994; Papamarinopoulos and Papaicannou, 1994). Collins and Doolittle ( 1993) discussed GPR and soil science applications to archaeological investigations using case studies and future developments in GPR techniques. Many of the future developments will have a direct effect on archaeological investigations. Software programs, high-speed multiple-channel radar systems, variable frequency antennae, and shorter range antennae (possibly in the range of 3 GHz) will enhance the signal for improved penetration that allows for better resolution. Even though GRP is being used by archaeologists, more scientists would utilize this technology if it were more “user-friendly” and less costly.
C. SOILLABORATORY ANALYSES Laboratory analyses must accompany the soil morphological description. If only the field morphology is used to interpret the archaeological site, only part of the site’s story is told. Laboratory results supplement the field morphology. Laboratory analyses of selected soil properties are routinely done in pedology. These analyses can be separated into physical, chemical, mineralogical, and
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
micromorphological determinations. Soil scientists use the procedures defined in “Soil Survey Laboratory Manual” (Soil Survey Staff, 1992) and in “Methods in Soil Analyses” (SSSA, 1982, 1986). Microbiological analyses are not routinely done on archaeological soil samples. Recently, archaeologists have become interested in particle-size distribution (Timpson and Foss, 1993; Johnson and Collins, 1993), pH (Coultas et al., 1993), organic matter (Stein, 1992), mineralogy (Scudder et al., 1993), and extractable elements (Morris ef al., 1993). An excellent, detailed discussion of analytical methods in soil science that archaeologists may find useful is presented by Courty et al. (1989).
1. Soil Physical Analyses Particle size analysis is a very common procedure in soil science labs. Determining the percent sand, silt, and clay can be done by several procedures (Soil Survey Staff, 1992). These procedures fractionate the sand into very coarse sand (particles 2.0-1 .O mm), coarse sand (1 .O-0.5 mm), medium sand (0.5-0.25 mm), fine sand (0.25-0.10 mm), and very fine sand (0.10-0.05 mm); silt (0.050.002 mm) into very coarse silt to very fine silt; and clay (0.002 mm) into coarse and fine clay. Timpson and Foss (1993) stated that soil particle-size analyses can be one of the most useful laboratory analyses for characterizing soils and parent materials in alluvial systems. Particle-size analyses can be especially useful in determining depositional history and discontinuities in profiles. Bulk density is the weight per unit volume (e.g., g/cc). Bulk density gives an estimation of the pore space within the sample. Changes in bulk density can be attributed to variations in lithology, depositional history, weathering processes, and restrictive layers. Bulk density changes that cannot be explained by pedogenic processes should interest archaeologists.
2. Soil Chemical Analyses Archaeologists have been interested in the chemical environment of their sites as an indicator of site habitation for many years. Soil chemistry is also useful to evaluate settlement patterns and agricultural history. One chemical that has been studied intensively by archaeologists is phosphorus in various forms. The identification of the different forms in phosphorus in soils has been difficult in the past. Determination of total P necessitated the use of dangerous chemicals and special fume hoods, but these procedures have been made easier by the method developed by Dick and Tabatabai (1977). Differentiating between total (TP) and extractable phosphorus (EP) is important. Extractable P content dependents on the soil’s pH and therefore on the ability of the extracting solution to “extract” the P. If a weak acid extractant is used, much of the total soil P may
SOIL SCIENCE AND ARCHAEOLOGY
19
remain unextracted and undetected; therefore, it is important to know which laboratory procedure is used to determine P content. Proudfoot (1976) stated that sampling designs need to be developed to better evaluate the variations in P that occur in natural and artificial features that occur horizontally as well as vertically on the sites. He emphasized that archaeological sites provide pedologists with the unique opportunity to examine variations in P contents of soils within a chronological framework. Further work on the phosphorus content of soils in the archaeological context, especially since many P interactions in soils are time dependent, according to Proudfoot (1976), could help pedologists increase their understanding of soil P. Dauncey (1952) and Cook and Heizer (1965) published reviews on the use of P analyses in archaeology. Collins and Shapiro (1987) used EP (Mehlich 1) and TP to understand the settlement history at San Luis. They compared archaeological and buried soils contents of EP and TP to the amounts in the naturally occurring Orangeburg (fine-loamy, siliceous, thermic Typic Kandiudult) soil. Total phosphorus distributions were difficult to interpret because the Organgeburg soil at San Luis was naturally high in TP. Therefore, using TP as an adjunct to artifact distribution in identifying activity areas (see Sjoberg, 1976) could not be done at San Luis. Distribution of EP, though, was a good indicator of past human activity because the mean EP in the Orangeburg was notably lower than that in the archaeological and buried soils. Other chemical constituents are also being used to discriminate features at archaeological sites. Organic carbon or organic matter content have been examined by soil scientists and archaeologists because of the potential for I4C dating. The amount of organics in soils can be determined by using either wet (e.g., Walkley-Black) or dry combustion (e.g., ashed) laboratory methods. If the organic material is associated with a buried horizon or soil, the absolute age of the material above that layer could be determined if the carbon could be dated. Stein (1992), though, discussed some of the limitations in dating organic matter in archaeological sites, but mentioned that “with a clearer understanding of its potential, the study of organic matter can make a significant contribution to the discipline” (of archaeology). Foss (1991), in a study of alluvial soils along the Delaware River in Pennsylvania, found that the elements Ba, Mn, and Sr were useful in verifying the presence of a buried A horizon. Griffith (198 1) found that exchangeable magnesium was the most efficient discriminator of features at an archaeological site in Ontario, Canada. Magnesium is one of the major chemical constituents of wood-ash, and exchangeable magnesium in Griffith’s study was high in the middens and pits and low in village areas where no chemical-enriching activity occurred. Inorganic and organic P were more uniformly enriched throughout the archaeological site and differences in habitation features were obscured. Griffith (198 1) emphasized, however, that
20
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
although chemical data are important in studying the occupation of archaeological sites, other soil properties should not be overlooked. He noted that the physical characteristics (particle size, structure, general morphology ) and particularly the micromorphology of soils are often neglected, but these are areas of research that benefit from knowledge of pedology.
D. LANDSCAPE ANALYSES Analysis of the landscape is a bridge for geomorphologists, geographers, and pedologists. Archaeologists are also using landscape studies to reconstruct paleoenvironments, to develop a better understanding of the archaeological record, and to predict archaeological site locations (Ferring, 1992). Reconstructing landscapes is important for archaeologists because it is on the old landscapes or surfaces that prehistoric people lived (Stein, 1992). Understanding the different processes that occur on various parts of the landscape is important in making archaeological interpretations. Movement of water across and through the soil profile is one of the major reasons for the differences in soils and one of the primary causes for the movement of materials on slopes. A number of researchers have stressed the horizontal as well as vertical movement of substances on the soil landscape and have tried to relate processes and soils to landscape position. Landscapes are classified as having erosional surfaces if their surfaces are dominated by erosional processes and if their shape is a result of what is left after material has been transported. Depositional surfaces are affected by material being deposited, and their shape depends on the size and shape of material deposited. Landscapes are considered stable if weathering is the dominant process and erosional or depositional processes have little affect on the landscape surfaces (Gerrard, 198I). Archaeologists interested in landscape reconstruction generally study areas dominated by depositional processes and search for buried surfaces on old landscapes that were once stable and have well-developed soils. Ruhe ( 1960) described five landscape positions: summit, shoulder, backslope, footslope, and toeslope. The summit is considered the most stable landscape position. The shoulder generally has the most surface runoff and thus is a relatively unstable landscape position. The backslope is also an unstable landscape position and slumping of material may occur along with surface creep, flow or wash. Recognizing the footslope position is important in archaeological studies as the footslope is a concave landscape position and deposition of material from upslope occurs. Soils on footslopes can be quite complex and heterogeneous due to this movement of materials and deposition as well as to irregular seepage. Palesosols are common on footslope as well as toeslope landscape positions. Toeslopes are unstable because these positions are subject to periodic flooding
SOIL SCIENCE AND ARCHAEOLOGY
21
and may be receiving materials from multiple sources. This position may have the thickest A horizon. Underwater archaeology and soil science may seem like strange bedfellows but a recent study by Kuehl and Benson (1993) discussed how both disciplines can compliment each other. In a study of a 15.5-km stretch of the Oklawaha River in north-central Florida, divers and archaeologists collaborated with a soil scientist to make archaeological interpretations along and under the river. The multidisciplinary study used soil morphological features to quantify the erosional and depositional processses occurring along the river and the resulting impact of these processes on archaeological sites. The study also illustrated the importance of understanding the relationship of soils to landscapes and landforms. The recognition of the footslope landscape position and the buried soil associated with this position was useful in predicting the location of excellent archaeological sites. The study also produced a good correlation between the total phosphorus level in the soils and the areas in which human activity had occurred.
E. MICROMORPHOLOGY Micromorphology, as the name implies, considers the morphology of the soil at a microscopic scale. It is the study of soils (or sediments) in thin sections (2530pm) (Courty et al., 1989; Buol ef al., 1989; Goldberg, 1992). Courty er al. (1989) in their book “Soils and Micromorphology in Archaeology” go into great detail on (i) basic principles of soils and micromorphology, (ii) processes and features in archaeological contexts, and (iii) case studies in which observations were made from field level to the micromorphological level to reconstruct the history of the area of interest. Micromorphology was initiated by Kubiena in 1938 when he demonstrated how the processes involved in soil genesis could be resolved with a soil sample prepared as a petrographic thin section examined under a microscope.
IV. PALEOSOLS Paleopedology is the historical branch of pedology whose aims include the retracing of the developmental stages of soils, particularly in the Quaternary period, and the study and interpretation of relict pedological characteristics (Ruellan, 1971). Paleopedology includes the study of paleosols, the term “sol” being Latin for solum or soil. Paleosol is a term that is widely used in archaeology and paleopedology, but is often poorly defined (Fenwick, 1985). Definitions include an-
22
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
cient soils, and soils formed in an environment of the past under conditions generally different from today. The term paleosol, as currently used in North America, refers to any soil that formed on a landscape of the past. Three kinds of paleolandscapes and paleosols are generally recognized: buried soils, relict soils, and exhumed soils (Ruhe, 1965; Ruhe et al., 1971; Ruellan, 1971). In all cases, these soils developed on landscapes in the past. Buried soils were formed at the Earth’s surface and subsequently covered by younger sediments. These soils may crop out on slopes of hills, road cuts, or excavations. Relict soils began forming on a preexisting landscape but were never buried by younger materials. These soils can be very significant in paleopedology in that they date from the initiation of the original land surface. The exhumed paleosols formed, were buried at some time, and later were reexposed at the surface following the erosion of the younger surface. Pedologically, a buried soil may be defined as an identifiable soil profile (A horizon and/or underlying B and/or C horizons that underlie a mantle of pedologically differing material(s)). Pedologically differing materials may be C horizons or B horizon material overlying a buried A horizon which developed under different soil forming periods.
A. BURIEDPALEOSOLS: KEYSTO ARCHAEOLOGICAL INTERPRETATIONS Much of the early interest in paleosols in the late 19th century seems to have derived from the use of paleosols as stratigraphic markers in geological sections (Valentine and Dalrymple, 1976). The most extensive stratigraphic application of buried paleosols was by pedologists and geologists in the loess-covered landscapes in the United States as well as in Europe. Archaeology is the other principal discipline in which buried soils have been used as stratigraphic markers. Archaeological research generally involves the study of soils buried at some depth, and thus, it is the buried paleosols that receive the most attention. Properties of the buried paleosol are commonly used to make assumptions about the climate and vegetation of the preexisting landscape. However, numerous researchers who have studied the physical and chemical characteristics of paleosols warn about potential errors that could result from these assumptions (Ruhe, 1965, 1975; Gerasimov, 1971). Scholtes et al. (1951) emphasized that care must be taken to choose soils from similar topographic positions when comparing buried and surface soils. Past and present-day soils are frequently the result of different soil-forming processes. It can be a mistake to equate the color of paleosols with the color of soils developed in present day environments. For example, thick reddish paleosols are often equated with the present-day reddish-
SOIL SCIENCE AND ARCHAEOLOGY
23
colored soils that developed in warm, humid environments. Ruhe (1969) studied the buried and relict grayish Yarmouth-Sangamon paleosols in Iowa and stressed that the paleosols developed in a much wetter environment than the present-day soils, and the gray paleosol does not reflect the water table depth of the soils in today’s climate and environment. Once a soil is buried, it may undergo significant chemical and physical changes. Some of the initial soil properties are lost while new properties are acquired. Workers in paleopedology generally agree that the same methods used in pedology must be used in the study of paleosols (Working Group on the Origin and Nature of Paleosols, 1971). Field evidence of pedogenesis, more than one pedogenic feature or diagnostic horizon, and strict adherence to stratigraphic and geomorphic principles are needed for recognizing and validating paleosols. Detailed field descriptions and laboratory analyses are essential in paleopedological studies along with an understanding that past as well as current processes have led to the development of soils and paleosols. Tracing the upper and lower boundaries of a paleosol is essential in making interpretations in paleopedology. Defining and tracing the boundaries can be difficult, however, as the boundaries are frequently gradational and few paleosols are exposed continuously in any cut or excavation (Follmer, 1982). In addition, the physical, chemical, and morphological properties of any pedostratigraphic unit may vary greatly both vertically and laterally across the landscape. Rolph et al. (1994), using seismic refraction, were able to trace the variations in depth of a paleosol underlying loess in China. The absolute dating of paleosols for stratigraphic studies in archaeology has been attempted by radiocarbon dating both organic and inorganic carbon in the soil. Dating carbon in the soil is difficult as there are continual additions of carbon as well as a constant recycling. The true age of the soil, in terms of when organic matter started to accumulate, may be impossible to determine. Instead, in the case of buried soils, the age represents the time elapsed since burial. An additional complication is encountered in dating archaeological sections in which the date is required in calendar years rather than radiocarbon years. One of the major complicating factors in studying paleopedology is climatic changes. It is well-known that climatic changes have occurred since the Pleistocene time period. One major and widespread contributor to climatic change that resulted in landscape changes was the Pleistocene glaciation. The glaciation resulted in the deposition of loess as well as till deposits and the formation of glacial lakes, terraces, and outwash valleys (Flint, 1976). An understanding of the climatic changes since the Pleistocene is essential to the study of the processes involved in paleopedology. There is no clear distinction between the processes involved in pedology and paleopedology, and thus, interdisciplinary studies are important.
24
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
B. A PALEOSOL CASESTUDY Goodyear and Foss (1993) illustrated the stratigraphic significance of paleosols along the South Carolina coastal plain in their detailed soil morphological study of an archeological site located on the flood plain of the Savannah River. The soil characteristics were described in each soil horizon and comparisons were made between the pedologic features and the archeological time period. Four major lithologic and pedologic discontinuities were noted. Of particular significance was that the Paleoindian occupation during 11,000to 10,000 B.P. is situated in sands that are essentially pedogenically unmodified. In the three soil profiles studied, the Holocene sands overlie two well-developed paleosols that were described as argillic horizons. Artifacts were observed in the Holocene sands, but no cultural features were associated with the older deposits that were presumed to be late-Pleistocene or older. The presence of these paleosols helps to define the early Holocene alluvial deposits which may contain Paleoindian cultural features in other areas of the Savannah River valley as well as other river valleys in the southeast United States. In this study, it was determined that based on the clay content and structure of the argillic horizon in the lower Pleistocene paleosol, the argillic horizon had weathered for a long period of time (>50,000 years; Foss er al., 1981). Goodyear and Foss (1993) indicated that this argillic horizon is a dominating horizon on many landscapes in the Savannah River valley and it provides an excellent index horizon to help interpret the early Holocene and Pleistocence landscapes in this Southeastern region. When the number of cultural features in alluvial deposits is low, the capability of recognizing the initial Holocene deposits and terminal Pleistocene deposits will help expedite the search for deposits that are old enough to support cultural features during this time period.
V. CASE STUDIES OF SOIL-ARCHAEOLOGICAL INVESTIGATIONS A. ALLUVIAL SEQUENCES INTHE SOUTHEASTERN UNITED STATES 1. Introduction Soils in alluvial valleys have been of great interest in soil-archaeological studies because some 80 to 90% of the known stratified and well preserved archaeological sites in the southeastern part of the United States are found in these landscapes. One of the keys to understanding the erosion, depositional, and
SOIL SCIENCE AND ARCHAEOLOGY
2s
weathering sequences in alluvial environments is through interdisciplinary studies in archaeology-geology-pedology. Although archaeologists have been studying alluvial environments for decades, recent interdisciplinary activities with soils and geology have provided further insight into alluvial sedimentation patterns and chronology. Several examples of these types of studies will be given in the discussion below.
2. Methods Profile descriptions of major soils at the archaeological sites mentioned in this section were prepared using nomenclature and methods outlined in the Soil Survey Manual (Soil Survey Division Staff, 1993). Profile descriptions were made in excavation units and by using a bucket auger. Laboratory analyses included particle-size distribution (Kilmer and Alexander, 1949), organic carbon (Nelson and Sommers, 1982), and elemental analyses (Lewis et al., 1993).
3. Results a. Thunderbird Site, Virginia The Thunderbird site is located along the South Fork of the Shenandoah River in Virginia. It was perhaps one of the first Paleo-Indian sites studied by a multidisciplinary team of archaeologists, geologists, pedologists, palynologists, and other scientists working on developing site chronology and history. That study was described in a publication produced by Gardner (1974). The pedologic portion of the study included a model of landscape development, soil distribution and characteristics, and chronology of the Thunderbird Site and surrounding areas (Foss, 1974). Figure 2 illustrates the landscape and soil development sequence along the South Fork of the Shenadoah River Valley near Front Royal, Virginia (Foss, 1974). This stratified site included artifacts from Paleo-Indian to Woodland (1 1,000 to 2000 B.P.) cultural periods; thus, the stratigraphy and soil development sequences provided an opportunity to interpret site history and observe pedologic processes over well-defined periods of time. The soils varied, showing minimal development along the levee, moderate development on the Holocene terrace, and strong argillic horizons in an old terrace and in the limestone residuum. The Paleo-Indian layer was identified as the Clovis clay (an argillic horizon). This soil had developed on an old terrace of the Shenandoah River (perhaps 13,000-20,000 years) and was subsequently covered with Holocene alluvium. The soil sequence above the Paleo-Indian artifacts included an argillic horizon with clay increases and clay-flow surfaces. Figure 3 shows the clay content with depth in three profiles (a-c) located on the Holocene terrace and underlain by the
Figure 2 Cross section showing soils at the Thunderbird site in Virginia (Foss, 1974).
Clay
-
.-+
(%I
20
-
AP -
40
-
BAIBt 1
60 -
0
5 00 : 100
-
120
-
140
\ \
t
.\
Figure 3 Distribution of clay in three soil profiles at the Thunderbird site in Virginia (Foss, 1974).
26
Table II Chronology of soils and Sediments and General Morpbology of Soils in the Savannah River Valley (Adapted from Foss ef uf., 1981)
Unit N -4
Estimated age of soils (years B. P. x 103)
I IIa
0.25 0.25-4
IIb
4-6 6-8 8-10.3 10.3-30 100-250
IIC
III IVa Ivb
Elevation of soils above Savannah River (m)
Soils characteristics
Mean
Diagnostic B
Colora
B-horizon thickness (m)
0-6 3-6
3 4.6
Variable 10YR-7.5YR
0-0.5 0.5-0.8
Historic Historic/ Woodland
3-6 3-6 3-6 4.6-7.6 6-15.2
4.6 4.6 4.6 6
None to weak Cambic or weak argillic Weak argillic Moderate argillic Moderate argillic Strong argillic Strong argillic
0.8-1.2 0.8-1.2 1.O- 1.5 1.0-2.0 1.5-2.5
Archaic Archaic Early ArchaiclPaleo-Indian
Range
Color is for well-drained soils.
9.1
7.5YR-10YR 7.5YR 7.5YR-5YR
5YR-7.5YR 2.5YR-5YR
Expected archaeological components
Recent to Paleo-Indian
28
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
Clovis clay. Note the uniformity of the clay curves for the three profiles on the Holocene terrace; these soils are much more uniform than generally expected on alluvial terraces. b. Savannah River Studies, Georgia and South Carolina A detailed study of the soils occurring along the Savannah River in the Piedmont of Georgia and South Carolina was made in association with archaeological investigations of the R.B. Russell dam (Foss and Segovia, 1984; Foss et al., 1985). Table I1 shows the range of soils and some of the morphological characteristics associated with the age of the soils. Qpes of diagnostic B horizons, such as argillic or cambic, were closely associated with the age of the terraces. Figure 4 shows clay distribution curves for three soils illustrating the range of soil development in the Savannah River Valley. Soils on the ancient terrace show clay maxima greater than 30%, while soils 4000 years or younger have maximum clay contents of less than 10%. Figure 5 summarizes five of the general characteristics of soils in the Savannah River Valley under different weathering periods. These data were useful in developing strategies for the archaeological investigations and subsequent interpretations of sites.
Clay (70)
nnn
I
-
2.5
Figure 4 Clay distribution curves for a chronosequence in the Savannah River Valley in Georgia and South Carolina (Foss and Segovia, 1984).
29
SOIL SCIENCE AND ARCHAEOLOGY
0
0
4
8 12 1
100200
Time tx 10%
n m e tx lo3)
0
4
8 12 1 n m e tx
100200
lo3)
nme tx l o 3 )
Figure 5 General characteristics of soils in the Savannah River Valley with age or length of weathering (Foss and Segovia, 1984).
c. Hiwasse River, Tennessee A landscape model was developed for the Hiwasse River in Pope County, Tennessee, in conjunction with an archaeological investigation. As noted in Fig. 6 , Pleistocene and Holocene terraces dominate the alluvial landscape, with recent
Figure 6 Cross section of landscape along the Hiwasse River in Pope County, Tennessee (Foss er al., 1995).
30
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
sediments occurring on the bench terrace next to the Hiwasse River and in the flood chute. Soils on the Pleistocene terrace consisted of strong, well-developed, reddish argillic horizons. The Holocene terrace had soils with a weakly developed argillic; a I4C date of 8800 B.P. was obtained at the base of this argillic horizon. The bench terrace had more than a meter of recent sediment overlying late Holocene alluvium that contained numerous Woodland period artifacts. In contrast, bench terraces next to rivers in other areas of the Southeast are normally composed entirely of recent sediment.
B. SOILSTUDIESAT
THEEL MIRADOR BAJO
1. Introduction A great deal of interest has been generated in Mayan civilization during the century or more that scientists have been studying it in Central America. Soil resources were probably a major factor in the environmental difficulties experienced by the Mayans and in the subsequent decline of their civilization. Because of densely populated cities, they were evidently forced to use soils of bajos (low areas) for some of their agricultural sites as indicated by raised fields and other landscape modifications. The soils in a bajo near El Mirador, a major Mayan site in the Pettn region of Guatemala, were investigated in conjunction with the archaeological studies by Dahlin et al. (1980). Earlier soils studies at the large Mayan City of Tikal by Olson (1977) showed 12% of the 9 km* central area was classified as swamp; these soils would be classified as Vertisols. The Vertisols near El Mirador are classified with the Yaloch series, which, according to Simmons et al. (1959), composes 229,563 hectares or 6.3% of northern Pettn. Cowgill and Hutchinson (1963) studied the chemistry and mineralogy of a single profile in a bajo near Tikal. More recent studies by Hammond (1994) and Coultas et al. (1993) in Belize and by Dunning (1993) in Guatemala have further elucidated the agricultural systems and soil characteristics in areas of Mayan activity.
2. Methods Twenty-four soil profiles were described and sampled in a 473-ha bajo north and west of El Mirador. The profiles were described and sampled according to procedures in the Soil Survey Manual (Soil Survey Division Staff, 1993). Extractable ions were determined using a weak acid extractant (0.05 N HCl and 0.025 N H2S04) and analyzing for each ion on a Technicon auto-analyzer (Bandel and Rivard, 1975). The pH of a 1:l soil-water ratio was made on a Beckman Zeriomatic pH Meter, and organic matter was determined by the
SOIL SCIENCE AND ARCHAEOLOGY
31
Walkley-Black method. The electrical conductivity method of Bower and Wilcox (1965) was used for soluble salt measurements and the acid-neutralization method described by Allison and Moodie (1965) for determining calcium carbonate equivalents.
3. Results Figure 7 shows the landscape of the El Mirador archaeological site and the bajo that was studied. Figure 8 is a soil map developed from observations made at the archaeological excavations. The landscape at El Mirador is composed of the upland area on which the large city Mayan city was built and the bajo that was used for agriculture and water storage, and as a conduit for transportation to other areas near El Mirador. The upland area consists of limestone-derived soils (Rendolls); in the bajo area, soils were developed on gypsiferous clays and marl of Eocene age and sediments derived from the uplands. Causeways were built across the bajos in several localities, using limestone fragments carried from the uplands (Dahlin ef al., 1980). The causeways ranged from perhaps 0.3 to 1.75 m in height above the surface of surrounding areas. The soils developed on these structures had coarse limestone fragments, dark-colored surfaces, neutral to slightly acid pH values, and weakly developed B horizons (cambic or weakly developed slickensides). Soils of the bajo were classified as Vertisols, although some variation in characteristics was observed. Soil units P1, P3, P10, P17, and P29 on the landscape model and map were classified as Vertisols but with differences in the
Flgure 7 Geopedologic relationships at the bajo at El Mirador, Peten, Guatemala (Dahlin er al., 1984).
32
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
Figure 8 Soil map of the Bajo at El Mirador, Guatemala (Dahlin et al., 1984).
parent materials and development of slickensides, color, depth of organic matter in subsoils, and chemical properties. Table I11 gives the morphological characteristics of several soils included in the legend. Those soils in the central part of the bajo (e.g., P1, P3, P29) were grayish brown to light brownish gray, while soils along the edge of the bajo near the Rendolls were dark-colored and more typical of the classical Vertisols (e.g., P17). The pH values of the Bss horizons of Vertisols developed on gypsiferous clays were less than 3.5 (soil units 1 and 3).
4. Archaeological Implications The soils investigation at the El Mirador bajo provides information on the general characteristics of the soil resource available to the Mayans. While the Mayans built their city on the more productive (though more fragile) landscapes, some agriculture was undoubtedly practiced in the bajo. With the high clay content, difficult physical conditions, high salt potential in some areas, and very low pH values in some cases, the soil environment of the bajos would be very challenging for agricultural production. Combining these soil characteristics with the 6-month dry period would further limit agricultural production potentials. Modifications of landscape, e.g., raised fields and terraces, although not noted at El Mirador, would provide some improvement of drainage, but they would not modify the overall physical and chemical properties of the soils. Profile 1, shown in Fig. 9, had low pH values which would definitely cause problems in growing corn or other crops. The extent of these highly acid soils (soil units 1 and 3) in the El Mirador area would be of great interest in evaluating the agricultural potential of landscapes for the Mayan civilization and for future development as well.
Table 111 Characteristics of Soils Occurring at the El Mirador Bajo
Horizon
Depth (cm)
A1 BA
0-18 18-40
IOYR 312
Bssl
40-62
I O Y R 5/2
Bss2
62-82
2.5Y 512
Bss3
82- 103
2.5Y 512
2Bss4
103- 137
5Y 613
2Bss5
137-168
2.5Y 612
2Bss6
168- 185
2.5Y 612
A1 A2 Bw
0-20 20-3 1 31-52
c1 c2
52-73 73- 120
lOYR 311 IOYR 411-311 IOYR 5/1, 411 lOYR 811 carbonate coatings lOYR 511, 311 2.5Y 512, 311
Color
lOYR 513
Mottling
Texture
Soil Unit P1 (S78G01)a None C cld C 7.5YR 518 cld C 7.5YR 518 c Id C 7.5YR 518 ClP C 2.5YR 418 m3d C 7.5YR 518 mlP C 2.5YR 418 mlP C 10R 418 Soil Unit P2 (S78GU2)b None C None C None C
Structure 2mabk-2fabk 2msbk
Slickensides
Bound
None None
1-2csbk
Mod.devel.
lcsbk
Mod. devel.
Om
Weak devel.
Om
Weak devel.
Om
Strong devel.
Om
Strong devel.
2mgr 3mgr lmsbk-Om
None None None
None None
1
I
lmpr, lmsbk 3mpr
None Weak
None
1
1mpr
None
(10%)
c3
120-166
2.5Y 512, 311 (10%)
(continues)
Table I11 (condnued) ~~
Depth (cm)
Color
2Ab 2Bssb 2BCb 2Cr
166-188 188-224 224-248 248-255
N 310 l0YR 511 2.5Y 612 lOYR 811, 712
A1 A2 2Bsslb
0-25 25-46 46-70
IOYR 312 lOYR 411
2Bss2b
70-89
2.5Y 512
0-9 9-39 39-72 72- 148 148-160 160-173
2.5Y 210 2.5Y 210 2.5Y 210 2.5Y 210 lOYR 612 2.5Y 210, 510
Horizon
w
P
A1 A2 BA Bss 2c 3Bss
5Y 511
Mottling
Texture
None C None C None C None sil Soil Unit P2 (S78GU5)C C None None gc flf C 2.5Y 516 d P C 5YR 518 Soil UnitP 17 (S78GU17)d C None None C None C None C None gsl None C
Structure
Slickensides
Bound
Strong Strong None None
cw cw cw
3mgr 2msbk lcsbk
None None Strong devel.
cs as gs
lcsbk
Strong devel.
-
None None None
cs cs cw 2s 2s
Om Om
Om Om
2-3f~ 2-3 mgr 2msbk Om 0% Om
Mod.devel. None Strong devel.
-
a Sampled near Bullard Causeway; gypsum 2% of Matrix at 82-103 cm, 5% at 103-185 cm; there appear to be two sets of slickensides, with moderate to weakly developed slickensides 40-103 cm and strong development 137-185 cm. Sampled on Gifford Causeway; soil developed on calcareous fill materials over limestone (marl); the buried 2Ab ranges in thickness from 12 to 74 cm; appears to be a wave cycle of 140 cm; small amounts of gypsum (1-2%) are present in the 2B; calcium carbonate equivalents in upper fill material ranged from 7.6% in A1 to 39.7% in C1; the Cr horizon had 58.6% calcium carbonate equivalent. Sampled on Bullard Causeway; soil developed on calcareous fill materials over limestone (marl); 40% coarse limestone fragments occur in A2; slight effervescence in A1 and strong effervescencefrom 25 to 89 cm; calcium carbonate equivalents in A1 and A2 are 3.1 and 21.6%, respectively. Sampled in Aquada Limon; slight effervescence in upper 39 cm, mod. 39-72 cm, and strong 72-173 cm; lense of limestone pebbles at 148-160 cm.
Depth (cm)
EC (dS/m)
PH
2
1
Figure 9 Guatemala.
EC (dS/m)
PH
-
6
A1
7
8
0
1
2
1
I
3
4
5
\
50
i
Soil morphology, pH, and conductivity of soil unit PI at El Mirador bajo, Peten,
Depth (cm) 0
3
Bw
c1
100
150
c2
c3
....
I
........
2Ab
200
2Bsst PBCb
250
ZUL
Figure 10 Soil morphology, pH, and conductivity of soil unit P2 at El Mirador bajo, Peten, Guatemala. 35
36
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
The preserved buried soils under the causeways provide an indicator of the conditions prior to construction of these features. For example, Fig. 10 shows the pH and electrical conductivity measurements of Profile 2, in which the original buried surface began at 166 cm (Table 111). The high salt content in the buried Ab (greater than 4 ds/m) would be deleterious to plant growth; the extractable Na content in the buried Ab and Bss horizons (ranging from 2112 to 4080 ppm) would also be a problem for many biological systems.
C. CHEMICAL PROPERTIES OF SOILSAT HAD RIA"^ VILLA, ITALY 1. Introduction
The use of chemical properties of soils in archaeological interpretations has been increasing in the last decade or more in the United States. The introduction of the inductively coupled argon plasma-atomic emission spectrometer (ICAP) to laboratories has greatly facilitated the elemental analyses of soils from archaeological sites (Lewis et al., 1993). The Villa, near Rome, Italy, was built by Emperor Hadrian from A.D. 118 to 133 and occupies approximately 121 ha. Some initial work on heavy metal distribution, in conjunction with soil characterization of gardens at the Villa, showed increased Pb concentrations in several gardens and adjacent areas (Foss el al., 1994). This resulted in more detailed sampling of soils from gardens, agricultural fields, and areas outside the Villa to determine the locations of soils with high levels of Pb and other elements. 2. Methods
Some 49 sites were sampled at Hadrian’s Villa; samples were obtained from surface and subsoils to represent the major soils and landscapes of the Villa. The elements analyzed included Al, As, B, Ba, Ca, Cd, Co, Cr, Cu, Fe, K, Mg, Mn, Mo, Na, Ni, P, Pb, S, Sr, Ti, and Zn. The elemental distribution was determined using an extracting solution of HCI-HNO, acid of 0.61 and 0.16 M , respectively, and analyzing the extract using an ICAP (Lewis el al., 1993).
3. Results Figure 11 shows the distribution of Pb at Hadrian’s Villa. The Pb content in the Villa ranged from 47 to 953 mg/kg; off-site the range was 19 to 29 mg/kg. The areas with the highest Pb contents were in the older portion of the Villa and near some of the major buildings. In the agricultural fields some distance from the buildings, the Pb levels dropped to near background values. The source of the
Figure 11 Areal distribution of Pb at Hadrian’s Villa in Italy.
Figure 12 Areal distribution of Zn at Hadrian’s Villa in Italy.
SOIL SCIENCE AND ARCHAEOLOGY
37
increased Pb at the Villa is still unknown, but the data on Pb distribution may aid in developing the answer to this vexing question. In addition to Pb, other elements show individualized distribution patterns. Figure 12 gives the distribution of Zn at the Villa. In general the Zn levels are higher near the buildings than in outlying areas, but the distribution pattern is different from that of Pb. Other elements such as As, Cu, Mn, Ni, and P showed increased levels in the gardens and areas surrounding buildings in comparison to off-site locations or to the agricultural fields in the Villa. These data, in conjunction with analyses of heavy metals at other Roman sites in Italy and Tunisia, have shown that Roman cultural levels can actually be identified by the increased concentrations of elements such as Pb, Cu, Zn, Ni, and P (Foss et al., 1994). Information such as this is vital to the interpretation and development of site history.
D. PALEOSOLS NEAR MT. VESUVIUS 1. Introduction
The large catastrophic volcanic eruption in A.D. 79 covered Pompeii with 5 to 6 m of volcanic material. However, significant eruptions preceded the eruption of A.D. 79, and the paleosols developed during intervals of volcanic activity have been preserved. The objective of this investigation was to study paleosols developed in volcanic deposits of Mt. Vesuvius dating from 1871 B.P. (A.D. 79) to 17,000 B.P. The interpretation of these paleosols can provide information on the number of eruptions and length of weathering time between eruptions.
2. Methods Soil profiles were described and sampled at the following locations around Mt. Vesuvius: Pompeii, Villa Oplontis, Herculaneum, Boscoreale, Ottaviano quarry, and Pozzelle quarry. Description and sampling were in accordance with procedures outlined in the Soil Survey Manual (Soil Survey Division Staff, 1993). The elemental distribution was determined using an extracting solution of HClHN03 of 0.61 and 0.16 M, respectively, and analyzing the extract using an ICAP (Lewis et al. 1993).
3. Results Figure 13 shows the general morphological characteristics of soil profiles sampled below the A.D. 79 level at Pompeii (Polybius and Pompeii garden sites), Boscoreale, and Oplontis. All profiles exhibited numerous pedologic (de-
Polybius
(ssntl)
Pompeii Garden
Boscoreale
Oplontis
(S8.W (S87It2) (ssnt4) Hgure 13 Morphological characteristics of soil profiles described at Pompeii, Boscoreale, and Oplontis near Mt. Vesuvius, Italy.
SOIL SCIENCE AND ARCHAEOLOGY
39
velopmental) and lithologic (textural) discontinuities and buried surfaces (Ab horizons). The Ab horizons indicate previous surfaces on which organic matter from plants and animals decomposed into humus and developed into an A horizon. Some of the profiles had cambic (Bw) horizons that indicated longer periods of weathering. The major paleosols were the Pompeii (1871 B.P.), Avellino (3800 B.P.), and the Mercato (7900 B.P.). Correlation of paleosols was possible using the chemical properties of the soils. Figure 14 shows the Ba/Pb contents of the Avellino and Mercato paleosols at the Ottaviano quarry. Figure 15 presents the Ba/Pb ratios of soils located near Pompeii. Note the high Ba/Pb ratios of the Mercado paleosol at each of the four sites. The chemical properties of the Mercado paleosol were also closely associated with a 14Cdate of 8200 2 230 years B.P. at Boscoreale. Thus, the combination of soil chemical properties, 14C dates, and soil morphology is an effective tool in mapping and interpreting paleosols in the Mt. Vesuvius area. Figure 16 shows the detailed morphology of a 15-m section at the Ottaviano quarry. Both the Avellino and Mercato paleosols were complex profiles in that discontinuities were present within each; thus, the soil-weathering sequence of these paleosols was interrupted by additional deposition of volcanic sediment. For example, the Mercato paleosol had two additional Ab horizons below the upper Ab. Using the numerous I4C dates, cultural artifacts, and soil morphology, a solum thickness based on age (i.e., duration of soil weathering) was proposed (Fig. 17). These data permit an approximation of soil age based on morphologic charac-
Wgure 14 Content of extractable Ba and Pb in paleosols at the Ottaviano quarry near Mt. Vesuvius, Italy (Lewis et al., 1993).
40 P
Boscoreale
Oplontis
Pompeii Garden
Figure 15 Ba/Pb ratios of soils at Pompeii, Boscoreale, and Oplontis near Mt. Vesuvius, Italy.
Polybius
figure 16 Soil morphological characteristics of column at Ottaviano quany near Mt. Vesuvius.
42
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
Age of Sol1 (yrs.x103)
Figure 17 Solum thickness versus age for soils developed on volcanic materials near MI.Ves-
uvius.
teristics. For example, the Pompeii paleosol (soil at the A.D. 79 surface) was less than 30 cm in most locations. Using soil morphology (solum thickness) to determine its approximate age, it can be seen that Pompeii had little volcanic activity for perhaps 600 to 700 years prior to the massive eruption in A.D. 79. The P content of soils has been used as an indicator of human activity in a number of archaeological sites. Phosphorous is of special interest because of the amount generated by human populations, and because of its use in fertilizer; it is also relatively immobile in many soils. Figure 18 shows the acid-extractable P in soil profiles at Boscoreale, Pompeii (Garden and Polybius), and Oplontis. The P contents in the Pompeii paleosol (upper 0.5 m) and the Avellino are generally between 500 and 2600 Fg/g; these values are greater than the background values for P found in the Bw and C horizons and at sites outside populated areas. The Mercato paleosol has P values less than 200 pg/g, except at Polybius where the Mercato paleosol is close to the surface and the Avellino paleosol is absent. The P values provide evidence of human activity throughout the time period of the Pompeii paleosol (1871 B.P.) and the development of the Avellino paleosol (3360 B.P.).
E. GEOMORPHOLOGY AND SITE SELECTION AT THE SEMINOLE REST SITE, VOLUSIA COUNTY, FLORIDA 1. Introduction
The Seminole Rest archaeological site is located on the shore of Mosquito Lagoon in Volusia County on Florida’s east-central coast. The site consists of a
Ext. P(ug/g)
Ext. P(ug/g) 1000 2000
0
1
2
I
8
b
Y
9
8
3
4
Boscoreale
5
3000
I
Pompeii Garden
Ext. P(ug/g) 1000 2000
Ext. P(ug/g) 3000
1000 2000
- 0
P
I
A
7 \ 4 \ \
1
\ Pompeii Polybius
1 '
I
Oplontis
Figure 18 Extractable P in soils sampled at Pompeii, Boscoreale, and Oplontis near Mt. Vesuvius, Italy.
3000
1
44
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
large clam-shell mound (Snyder’s Mound) and several smaller interlayered shell/sand mounds bordered by salt marsh. Several aboriginal cultural periods are represented (Horvath er al., 1995) with inclusive calibrated radiocarbon dates ranging from 210 to 1460 A.D. East of the mound is Mosquito Lagoon which is separated from the Atlantic Ocean by a barrier island and mangrove keys. West of the mound is a small water-control canal, then coastal marsh and low Pleistocene terraces now covered by oak woodland (Schmalzer and Hinkle, 1990; White, 1970). The mound today is 227 m long and 155 m wide, and extends from 2 m below to 4.5 m above sea level (Horvath et a l ., 1995). In 1993 archaeological investigations were conducted at Seminole Rest by the National Park Service. Field archaeologists collaborated with faunal analysts, paleobotanists, and soil scientists to reconstruct the cultural and ecological history of Snyder’s mound, one of the smaller ancillary mounds (Fiddle Crab Mound), and the adjacent lagoon and upland habitats (Horvath er al., 1995). One of the questions posed by archaeologists of the Seminole Rest site was its relationship to sea level: was the mound constructed on dry land, or was moundbuilding begun by throwing clam shells into shallow water adjacent to a living area or shellfish-processing site? The 1993 excavations of the site encountered midden material below present sea level. Cores, placed at intervals on Snyder’s Mound and penetrating to the culturally “sterile” sediments below, also indicated below-sea-level accumulations of shells. To address this question, particle-size distribution analysis of the mound substrate and of adjacent marsh surface samples from approximately contiguous topographic positions was used. Two adjuncts to particle-size analysis that were also employed in understanding the geomorphology of the area and site selection by past human inhabitants were clay mineralogy and an examination of the infauna of the submound sediments.
2. Methods Mechanically extracted core samples were taken by geochemical engineers using a split-spoon, 1.5-in. core which extracted 46-cm core intervals. Placement of the cores at various points on the top of Snyder’s Mound was directed by archaeologists. Cores penetrated the shell layers and terminated in the submound sandy sediments. Thirteen auger sites were also chosen, including the east and west margins of Snyder’s Mound, areas below and adjacent to the smaller Fiddle Crab Mound, and marsh areas south and southwest of the excavation units. Samples were taken at 10-cm intervals to depths ranging from 75 to 140 cm. Sampling terminated at the upper boundary of the water table. Particle-size distribution was determined using the pipette method (Day, 1965). Samples with an estimated 1% or more organic carbon (judged by a dark
SOIL SCIENCE AND ARCHAEOLOGY
45
brown or black color) were pretreated with hydrogen peroxide and heat to remove organic matter. Samples for clay-size mineral identification were plated onto ceramic tiles, saturated with either magnesium chloride and glycerol or potassium chloride, and X-rayed at room temperature and at 1 10°C using a Nicolet diffractometer and Cu K-alpha radiation. Faunal material from each core and auger sample was identified using comparative specimens housed in the zooarchaeology laboratory at the Florida Museum of Natural History. Particular attention was paid to small invertebrate infauna that actually resided in the soils and sediments.
3. Results Grain-size distributions in all areas sampled were dominated by fine sand: from 52 to 77% by weight of the sand-to-clay size fraction. Subrnound (Snyder’s Mound) core samples and samples taken beneath and immediately adjacent to Fiddle Crab Mound contained more fine sand (mean of 70%)than soils from the area between those two features, adjacent to the modem canal. In those areas, mean fine sand content was 58%. Extremely high silt contents were encountered in all levels of the tests taken on the western margin of Snyders’s Mound (mean of 25%) and in the surface levels of the southern margin. In contrast, samples from beneath Fiddle Crab Mound contained an average of 4.8% silt. The submound core samples contained only 5.6 to 12.5%silt, with an average of 7.5%. Clay content was less than 1% in all samples analyzed. Clay-size mineral species from selected Seminole Rest soil samples appeared as two somewhat overlapping assemblages: (i) high smectite with kaolinite and quartz, and (ii) low or minimal smectite with hydroxy-interlayer vermiculite (HIV), kaolinite, gibbsite, and quartz. The latter grouping was found in the lower levels of auger samples in the vicinity of Fiddle Crab Mound and in 75% of the submound core samples tested. Broad smectite peaks, indicating the presence of a poorly crystalline, minimally weathered smectite, were seen in all upper-level marsh samples, in upper Fiddle Crab Mound zones, and throughout the samples from the western margin of Snyder’s Mound. Soil faunal analysis revealed that submound core samples and auger samples on the eastern margin of the mound bordering the lagoon contained high numbers (>25% by volume) of both articulated and disarticulated shells of the genus Parastarte, the Brown Gem Clam. Shell numbers decreased with depth in the auger samples. Samples from the western margin of the mound and the Fiddle Crab Mound zone samples contained no Parastarte. The auger test to the south of the mound had a few Parastarte at 90-100 cmbs.
46
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
4. Discussion Although particle size was dominated by fine sand in all samples, relationships between the coarser (medium and coarse sands) and finer (especially silt) fractions helped outline subsurface landscape features and aided in understanding the original rationale for mound placement. In the basal core sediments beneath Snyder’s Mound, relatively high coarse and medium sand contents and low silt contents suggest a mixing of finer aeolian sands with coarser marine sediments, resulting in a texture characteristic of a beach ridge or sand bar or bank. The finer textural class on the western margin of the mound and in the upper 50 cm of the test to the south defines an accumulation of silt over sand. The Volusia County soil survey (Baldwin et al., 1980) offers a basis for these differences. The soil map unit directly south of Snyder’s Mound is the Canaveral fine sand series, which forms on dune flanks and in interdune areas. Prior to modem dredging and canalization, this unit probably extended northward and provided either a low, subaerially exposed ridge or a shallowly submerged sand bar, which represented a suitable substrate for mound building. North and west of the mound is the Turnbull muck soil series. This is a mucky- or clayey-over-sandy series, the development of which has been augmented by the gradual closing of the inlet and subsequent silting of the lagoon (Mehta and Brooks, 1973). The silty areas west of the mound may have been a low swale behind a bar, or a tidal creek, which gradually infilled with fine sediments. During the period of occupation of Snyder’s Mound, the combination of a low bar and swale or tidal creek would have presented the local residents with a comfortable “landing” or haven for canoes filled with shellfish gathered from the lagoon. Extracting the meat from the heavy-shelled clams would have made good sense energetically if the clams had to be carried inland for any distance. The shells would have simply been tossed out at the shore. Two other lines of evidence support the premise that the mound is underlain by a sand feature such as a low ridge or bar. These are clay mineralogy and invertebrate soil fauna. Clay-size mineral distributions in the submound core samples reflect a relatively low smectite content and the presence of HIV and gibbsite, both highly weathered forms compared to smectite. Similar mineral assemblages were found only in the deepest levels of Fiddle Crab Mound. Mound-peripheral and marsh auger samples contained an abundance of smectite. Highly weathered minerals such as HIV and gibbsite are commonly found in old, leached soils, and in reworked marine sediments such as dune or beach deposits which have been transported by wind and water and repeatedly chemically and physically altered. Those soils and sediments are a product of Pleistocene and early Holocene climatic conditions; they form Florida’s inland sand ridges and underlie or mantle its coastal areas. The clay assemblages in the submound core samples and some deep auger samples suggest that the sediments
SOIL SCIENCE AND ARCHAEOLOGY
47
there are composed of such materials. The cation-enriched smectites mantling the more weathered forms are common in quiet-water lagoonal sediments which are high in soluble salts and have a neutral to high pH. The composition of the invertebrate infauna of the submound and peripheral samples is the second line of evidence supporting a sand feature beneath Snyder’s Mound. A high concentration of the Brown Gem Clam, Purasturte triquetru, was found in 6 of 10 of the submound core samples and in four of the auger samples, particularly on the eastern margin of the mound. Virtually no Purusturte were found in samples to the west of the mound. Parustarte is a small bivalve which is common in sand bars (Abbott, 1974). The combination of high numbers of Parustarre, a low smectite and high HIV/gibbsite mineralogy, and coarser grained sediments in the submound samples suggests a sand bar or submerged ridge. In contrast, the finer grained, high smectite environment lacking Purustarte west of the mound indicates a low swale area landward of such a bar or ridge. It is evident that “modern” constructions and excavations (e.g., mounds and canals) have been superimposed over older landforms (sand bars and tidal creeks). The question of whether mound construction began on dry land or whether debris was initially thrown into shallow water can be answered directly only by the clam evidence. Grain-size distributions and clay mineral types might be similar in both subaerially exposed and shallowly submerged environments. But the presence of articulated clam valves, particularly of immature individuals of such a diminutive species, is strong evidence that they were in situ and alive when covered by midden debris. It is difficult to construct a scenario in which they could colonize sediments already covered by coarse midden debris. Considering their aquatic lifestyle, then, it would seem that at least a minimal layer of water covered the site at the onset of midden accumulation.
5. Summary and Conclusions The archaeopedologic study at Seminole Rest focused on broad questions of geomorphology and site/substrate relationships. In particular, the nature of the soil or sediment on which Snyder’s and Fiddle Crab Mounds rested and the character of the sediments which separated the two mounds were addressed. The accumulated evidence suggests that both of these mounds were built on a sandy substrate much different than the marsh sediments found in the area today. Clay mineralogy, particle-size distribution analysis, and invertebrate infauna distributions contributed to this interpretation. The Seminole Rest site offers the opportunity to gain understanding of human adaptation to a dynamic environment. The links between the natural coastal environments, resources, and processes, and early humans’ response to them, are the mounds and middens themselves. The botanical and faunal remains are clear
48
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
documentation of resource availability and choice. Carbon dates, clam seasonality studies, and the stratigraphy of the mound are records of the temporal relationships between humans and their resources (Horvath et al., 1995). The sediments, soils, and geomorphology of the area are tangible clues to the rationale behind site selection for habitation (or deposition), and to the conditions present at the onset of such activities.
F. SOILSAND LANDSCAPES: ARCHAEOPEDOLOGY AT THE PINELAND SITE
1. Introduction The Pineland archaeological site, once home to Florida’s Calusa Indians (Marquardt, 1992; Walker, 1992) is a study in complexity, incorporating monumental earthworks and canals, peopled by accomplished and highly socially stratified inhabitants, and dependent on a rich, natural mosaic of terrestrial and marine resources. Huge shell and sand mounds flank-or are encircled by-hand-dug canals and lakes (Cushing, 1897; Luer, 1989; Marquardt, 1992b, 1995) covering an area of approximately 25 ha (Luer, 1991). Generations of archaeologists have focused attention on interpretations of the site’s cultural features (Cushing, 1897; Douglas, 1885; Gilliland, 1975; Goggin and Sturtevant, 1964; Luer, 1989). More recently, advances in techniques of bioarchaeology (incorporating studies of plants, animals, soils, paleonutrition, and human osteology), and a redefinition of research questions and goals (Marquardt, 1992a), have highlighted environmental issues: how do humans exploit and manipulate their surroundings? What environmental effects of human habitation are in evidence and, conversely, what indications of environmental change are found in abandoned human settlements? One question about the Pineland site posed by archaeologists is: What was the initial relationship between settlement elements and the natural landscape? Did the original inhabitants begin mound construction on the flat, marshy soils adjacent to the shore, or did they use relict sand dunes, beach ridges, or other remnant geomorphic features as the bases for their architectural creations? This study focuses on the soils and sediments that underlie and surround two of the architectural features of Pineland. It characterizes physical and chemical attributes of natural soils and compares them with anthropogenic landscape features, giving evidence of human creation and manipulation of landforms and clues to the original rationale for placement of those features. One of the archaeological features is Smith Mound, the sand burial mound whose imposing size and shape are thought to be a central element in the display
SOIL SCIENCE AND ARCHAEOLOGY
49
of social and political power evoked by the site as a whole (Luer, 1989). The entire mound, some 70 m in diameter and 7 m high, was completely encircled by an artificial lake. Was this mound completely human-constructed, or is it an augmentation of a relict dune? The second feature is known as the Citrus Grove, a low northwest- to southeasttrending feature which today supports an orange grove. One small midden was uncovered beneath the surface horizon, as well as scattered, eroded shells, and intermittent shell lenses at various depths. Calusa canals and modem pasture ponds flank the long sides of this small rise. Is it simply a spoil bank thrown up during canal construction, or is its core an old beach ridge, much attenuated by human use and/or continued sedimentation through time? This study addresses the question of the original forms of Smith Mound and the Citrus Grove.
2. Methods Smith Mound soil samples were taken from three excavated squares and two points along a GPR transect. Four sets of Citrus Grove samples were augered from ground surface to approximately 2 m depth. Complete morphologic descriptions were made for each trench profile sampled, including horizon presence, thickness, and arrangement; color and mottling; boundary thickness and topography; approximate texture; structure; and presence and size of roots and other inclusions such as shells, plant material, and potsherds. Soil content of extractable phosphorus, calcium, magnesium, zinc, and copper was determined by ICAP spectroscopy using an extracting solution of 0.05 N HCl in 0.025 N H,SO,. Total phosphorus content of soils from two of the three squares excavated in Smith Mound was determined by the alkaline oxidation method of Dick and Tabatabai (1977).Organic carbon content was determined using the Walkley-Black potassium dichromate/ferrous ammonium sulfate digestion method (Soil Survey Laboratory Staff, 1992).A 2:l water-soil solution was used to measure pH. Iron and aluminum contents of selected samples were determined using citrate-dithionate extraction (Soil Survey Laboratory Staff, 1992) and atomic absorption. Particle-size distribution was analyzed using the pipette method of Day (1965).Samples with an estimated 1% or more organic carbon (judged by a dark brown or black color) were pretreated with hydrogen peroxide and heat to remove organic matter.
3. Results Particle-size distribution in all of the Pineland soil samples was dominated by the fine sand fraction, from 79.6to 89.5% in the Citrus Grove and from 82 to
50
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
91% in Smith Mound. The second most common particle-size class, very fine sand, averaged approximately 5.5% in the Citrus Grove and 5.2% in Smith Mound. Clay content was less that 1% in all samples. A ratio of “fine” particles (very fine sand, silt, and clay) to “coarse” particles (very coarse, coarse, and medium sands) was calculated for all samples. This ratio excluded the dominant fine sand fraction in order to highlight the more subtle shifts that may occur among grain size classes. In all four Citrus Grove tests, the ratio of fine to coarse particles decreased with depth until the water table was reached, at which point it increased. The ratio fluctuated in the Smith Mound samples, depending on which soil was under consideration (see discussion below). Soil profile descriptions are simplified here to include only features necessary to discuss the genesis of native soils and the creation and manipulation of anthropogenic ones. The Citrus Grove soil profile is described by one auger transect and is therefore a limited approximation of the subsurface morphology of that landscape feature. Figure 19 depicts the subsurface morphology derived from plotting field horizon descriptions. The Ap horizon, subdivided by slight color changes, was dark to light gray, single grain (structureless), dry, fine sand with individual scattered and worn shells. A few lamellae were recorded where the Ap was thickest and formed the flat ridge or “spine” of the Citrus Grove area. The topography of the boundary between the Ap and subadjacent horizons differed from the surface contours of the Ap horizon. Figure 19 shows that boundary, and all subsequent ones, to be flattened in the central portion of the profile, draping downward on the western edge and rising at an angle of 30 to 45” on the eastern edge. The Smith Mound soil profiles recorded a sequence of buried soils (Fig. 20) capped in some areas by a young soil forming in disturbed sediments thrown up by looters’ activities. The square in the topographically lowest position, B-1, exhibited a profile typical of a Spodic Quartzipsamment: salt and pepper A horizon, white E horizon with albic leaching into the light brown Bw below. Auger samples below the floor of the square retrieved a possible E’ horizon above a reddish Bw that articulated with the water table. The intermediately situated square B-2 had a similar horizon sequence, with the Bw horizon occurring at the same depth as in B- 1, and a thicker E‘ horizon. Square B-3 was uppermost on the flank of the mound and excavated deeper due to the discovery of a human burial at 2.18 m below the surface. This square revealed traces of at least two buried soils below the upper 24 cm, which was comprised of an Inceptisol forming in disturbed soil. Below this, a strong A’ horizon was underlain by a Bw’ that gradually faded to a thick C horizon. Below this sequence, the second buried soil exhibited an intact A horizon (A”) overlying thin E’ and Bw” horizons. A second sequum followed, in this case an E” horizon and subadjacent
SOIL SCIENCE AND ARCHAEOLOGY
51
TEST NUMBER
Citrus Grove Figure 19 Profile of Citrus Grove feature, Pineland Archaeological Site, Lee County, FL (Permission for use of figure granted by W. Marquardt, I.A.P.S. Books).
Bw”, which contained the human burial. The topography of the buried horizon boundaries, except for the intrusive interruption, were smooth, parallel, and angled toward the top of the mound, as if following a former hillslope surface. The final auger sample, at 390 cmbs, recovered a portion of the reddish Bw (in this case the Bw’””!) horizon. Soil chemical data are published in Scudder (1995) and are summarized briefly here. Organic carbon (OC) content of Citrus Grove soils was less than 1% by weight in all but the Ap horizon. There, OC content averaged 1.4%. Organic carbon generally decreased with depth to the Bw (or Bhs) horizon which articulated with the water table, at which point it increased. The A horizon of Smith Mound squares contained from 1.1 to 1.7% OC. Organic carbon content of all subsurface horizons there was less than 1%, with slight increases in some A‘, Bw, and Bw‘ horizons.
52
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
Smith Mound Figure 20 Profile of Smith Mound, Pineland Archaeological Site, Lee County, FL (Permission for use of figure granted by W. Marquardt, I.A.P.S. Books).
Extractable calcium (Ca) content of the Citrus Grove horizons varied considerably, from 249 mg-kg-' in the Ap horizon of Test 1 to 4080 mg-kg- in the AE horizons of Test 2. The surface horizon of Smith Mound contained abundant Ca when compared with native soil values-from 450 to 970 mgmkg-1 in the three A horizon samples versus 52 to 72 mg-kg-I in native soils. Contents of all other major and
SOIL SCIENCE AND ARCHAEOLOGY
53
trace element, with the exception of extractable P, were also high. The contents of most of these elements decreased abruptly in the subadjacent E horizon, then increased again in the lowest Bw horizon of each square. Extractable phosphorus (P) content of Citrus Grove Tests 1 and 3 ranged from 31.7 to 180 mgekg-’ and showed no pattern of accumulation with depth. P content of Tests 2 and 4 showed a “mirror image” pattern of accumulation as did Ca, with higher amounts in the upper horizons of Test 2 and in the lower horizons of Test 4. Extractable P content of the surface horizon of Smith Mound was comparable to that of the native soil samples. In square B-3, which contained the repeating sequences of horizons, extractable P was generally lower in the A or A‘ horizons, and increased with depth to the Bw or Bw’. Citrate-dithionite (CDT) extractable aluminum content of the Citrus Grove samples increased with depth in Tests 1, 3, and 4, diminishing only in the E horizon. Aluminum content decreased with depth from surface through the E horizon of Test 2, then increased in the Bw’ horizon. Aluminum content of Smith Mound samples was determined using the Mehlich-1 extraction method, not the citrate-dithionate method used for the Citrus Grove, so only relative increases and decreases in aluminum content are discussed. In all squares, surface horizon samples contained more A1 than the subadjacent E horizon but less than the Bw horizon. In the multiple soils of square B-3, A1 content followed the same trends.
4. Discussion a. The Citrus Grove The existence of Pleistocene dunes and Holocene beach ridges and barrier islands in Pine Island Sound, and of coast-parallel soil series on the island itself (Gagliano, 1977; Widmer, 1988), offers insight into the possible origin of the Citrus Grove feature at Pineland. Figure 21 is a tracing of the Pineland area soil survey sheet (Henderson, 1984) superimposed over a 1940 aerial photo of the site and vicinity. The two NW- to SE-trending hatched black lines on the tracing follow vegetation changes evident on the photo. Although the lower of the two lines bisects the Matlacha soil series, this series is of human construction and does not reflect original soil configurations. The lower line marks the boundary between the marshy Peckish soil (a Typic Sulfaquent) and the sandier Myakka (an Aeric Alaquod). It also passes through the Citrus Grove, placing that feature on the transition between the two soils. This suggests that the Grove-or at least its core-is a product of natural coastal processes at work on a former shoreline. The profile of the Citrus Grove (Fig. 19) shows two distinct horizon boundary topographies. The first is the ground surface itself, following a gently mounded convex line. The second is seen at the boundaries between all subsequent sub-
54
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
Figure 21 Pineland site soils and vegetation from 1940 aerial photo (Permission for use of figure granted by W. Marquardt, I.A.P.S. Books).
horizons. There, the horizons are level and subparallel in the central portion of the profile, but rise toward the east and slope down toward the west. The profile resembles a beach and berm or terrace feature buried under loose sediments. The sediments composing the Ap horizon are not characteristic of unmodified A horizons of native soils in the Pineland area. They are loose, dry, and uniformly gray to a depth of almost a meter, and contain scattered shells and occasional lamellae, which indicate incipient pedogenesis in recently deposited sediments. They also differ from native soils in some chemical and particle-size characteristics. The Citrus Grove area is composed primarily of fine sand, as is Smith Mound and the native soils examined. This indicates an aeolian origin for the sediments from which the soils formed. The resemblance of the subsurface horizons’ topography to a terrace or beach may mean that those features were cut into a preexisting dune, which may be identified in future work by looking for sedimentary
SOIL SCIENCE AND ARCHAEOLOGY
55
structures characteristic of dune deposition. The difference between the texture of the Citrus Grove and the native soils is apparent in the fine-to-coarse-grain ratios. Native soils generally contain more fine particles deeper in the profile: the ratio increases with depth. This is a manifestation of the mechanical translocation of small particles downward through a matrix of larger ones-the normal pedogenic process of eluviation through time. Citrus Grove soils contain more fine particles in the surface horizon than in the subsurface E horizon, suggesting that relatively fresh sediments with a high proportion of “fines” have been piled onto the original soil surface. The chemistry of the Citrus Grove soils also departed from that of the local native soils in several ways. Aluminum content, as measured by CDT extraction, was approximately 10 times higher in the Ap horizon of the Citrus Grove than in the A horizon of the nonmidden soils. It decreased with depth in Test 2, in a reversal of the normal pattern of accumulation of aluminum with depth due to weathering (Birkeland, 1984). This also suggests the piling of fresh-or at least Al-enriched-sediments on the original surface of the Citrus Grove, especially on the highest point, which was sampled at Test 2. A plausible hypothesis for the unusual physical and chemical character of the Citrus Grove A horizon is that it is derived from the poorly drained soils flanking the preexisting, now buried, sand ridge. Spoil from excavation of the Calusa canals would have contained fine particles and aluminum, particularly the subsurface B horizons transected during excavation. Spoil piled on the low sand ridge would form a dark, aluminum-enriched horizon. Calcium content of the Citrus Grove soils was higher in all cases than the native soils. In particular, the Ap horizon of Test 2 and the AE through EBw horizons of Test 4 contained 3500 to 4000 mg.kg-’, compared with 20 to 70 mgekg-1 in the A horizon of the native soils. Extractable phosphorus contents in the same provenances showed similar proportional relationships. The Ca and P contents of the Citrus Grove soils indicate an augmentation of these elements over normal levels. More interesting is the pattern of accumulation in Tests 2 and 4. High contents of these elements occur in Test 4 below th Ap horizon, in what is indicated by soil morphology to be a former surface horizon. Similar high contents of the same elements are found in the upper (Ap) horizon on the highest part of the Citrus Grove. These proveniences correspond to shell and bone accumulations encountered in an independent auger survey of the Citrus Grove. The findings here corroborate those results, although no actual faunal or cultural remains were recovered from the soil samples. Based on the evidence it appears that the Citrus Grove was a natural feature such as a beach and berm, wave-cut into well-sorted fine sands, with loose sands heaped upon it. The accumulation of fine particles and aluminum in the mounded Ap horizon suggests that the source of that horizon was the poorly drained soils that surround the Grove, and particularly the subsurface horizons, which would
56
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
have been transected during excavation of canals and other waterways in the vicinity of the Citrus Grove. b. Smith Mound This sand burial mound is located to the north of the Citrus Grove in a lobe of Matlacha soil (Henderson, 1984). The human-made Matlacha soil was probably based on either the Myakka or the Immokalee soil (an Arenic Alaquod), according to the projected soil map unit boundaries in Fig. 21. Although there are no round remnant Holocene sand features in the vicinity of Pineland today, the upper boundary of the deepest Bw(') horizon from the three squares forms a convex surface that rises below square B-2 (Fig. 20). All of the horizon boundaries are subparallel and angled upward, as if following the rise traced by the underlying Bw' horizon. Three soils are depicted in the figure: (1) an Inceptisol forming in the disturbed surface spoil, (2) a Spodic Quartzipsamment with weak Bw development capped by a dark gray buried A horizon, and (3) a Spodosol with an intact A horizon underlain by a human burial intruding into what appears to be an E'-Bw" and Err-Bw'" bisequum. The continuous gray former surface horizon of that soil is evident in field photographs, as is a clear break in the E'/Bw" boundary leading to the burial at 2.18 m below the ground surface. Four additional sets of samples taken on the opposite side of the mound corroborate a buried A or darker horizon at approximately the same depth as the former surface of the burial-containing soil. Ceramic analysis (Cordell, 1995) showed an increase in two ceramic styles, and the first appearance of Weeden Island Style ceramics, in the excavation levels corresponding to the burial-containing soil. The burial has been 14C dated to A.D. 1020-1170 (Beta-72995) (approximately 800 years B.P.), so the A" horizon of the Spodosol had a maximum of 800 years to reform over the intrusion. That is of course an overestimation of the time that soil was left subaeirally exposed, since the Spodic Quartzipsamment that developed in sediments mounded over it would have taken some part of the 800 years to form. The very thin and weak appearance of the A" horizon, generated in a region which could have supplied abundant organic plant remains to the soil surface, suggests that minimal time was allowed for the development of that horizon. The chemical characteristics of the two buried soils differed considerably. Organic carbon, calcium, magnesium, potassium, phosphorus, and aluminum contents decrease by about 50% between the Bw' of the upper Spodic Quartzipsamment and the A" horizon of the lower one. The relative enrichment of the overlying soil may indicate that the soils heaped upon the burial-containing soil were derived from the surrounding low-lying areas, as was hypothesized for the Citrus Grove. Spoil from lake or canal construction, particularly as it cut into the deepest Bw horizon at the water, was a likely source of these sediments. Additional augering revealed the presence of yet another set of soil horizons below the burial soil, beginning at about 297 cmbs. The difficulty of interpreting
SOIL SCIENCE AND ARCHAEOLOGY
57
augered soil samples near the zone of saturation makes it unclear whether those lower horizons are another entire soil, beginning with a buried surface horizon (A”‘-E”’-Bw””), or whether they are a third sequum of the burial soil, separated from the upper two sets of horizons by a diffuse E horizon. Whatever the origin and genesis of this fourth set of horizons, the chemical and physical data, along with photographic documentation, clearly indicate that Spodosol was an intact soil when disturbed by an intrusive human burial. It remained subaerially exposed long enough for a thin A horizon to reform and was then covered by sediments that developed into a Spodic Quartzipsamment. The chemical nature of that overlying Spodic Quartzipsamment, its topographic position, and the absence of sediments of like thickness and horizonation anywhere else on the site, lead to the conclusion that the sediments were probably derived from the immediate vicinity and that they were human-deposited.
5. Summary Soil science techniques and analyses were used to distinguish natural soils and landscape features from anthropogenic ones at Pineland. Those techniques, applied to the Citrus Grove area, revealed a body of parallel subsurface horizons whose boundaries differed from the surface topography. Chemical and grain-size analyses indicated a core of reworked dune sands in the form of a wave-cut beach or terrace mantled by a surface (Ap) horizon unlike local natural A horizons. This study suggests that the Ap horizon was derived from the surrounding landscape, based on aluminum content, grain-size ratios, and patterns of calcium accumulation. Smith Mound was underlain by a series of convex, subparallel horizons and was composed of at least three soils of varying maturity. A minimally developed Inceptisol forming within surficial spoil mantled a Spodic Quartzipsamment with weak Bw horizonation. Below this was a third, more strongly developed soil incorporating a human burial. The A horizon of this soil was complete, having reformed subsequent to the intrusion of the burial into the E’ and Bw“ horizons. A fourth set of horizons below the burial-containing soil was found, but was neither correlated with that soil nor identified as the beginning of a fourth complete soil due to the saturated condition of the samples.
G. PEDOARCHAEOLOGICAL ANALYSIS OF A PREHISTORIC SHELL-BEAFUNG ISLAND,FLORIDA 1. Introduction
A.B.’s Midden (8-Lv-65) is located on North Key; one of a series of small islands comprising the Cedar Keys National Wildlife Refuge (Fig. 22). North
58
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
Figure 22 Location of the Cedar Keys region.
Key has remained in near pristine condition throughout historic times and has been virtually insulated from human influence. A.B.'s Midden, a prehistoric shell midden, fringes the southeast shore of the island (Fig. 23) There is no evidence to suggest that the soils in A.B.'s Midden have ever been plowed or chemically altered through fertilization or liming. A multidisciplinary research team (Seahorse Key Maritime Adaptations Program) was initiated to investigate the archaeological resources in the Cedar Keys
LOCATION MAP
SEABREEZE I.
Kilometers
RAULESNAKE I.
NOR7 ATSENA OTlE KEY
Gardiner's Midden
~
~~~
~~~
Figure 23 Cedar Keys region location map.
60
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
region. The investigation included the excavation of two sites in addition to A.B.’s Midden. The objectives of this study were to determine the type and extent of human influence on soil formation by characterizing and comparing the site’s archaeological soils to the island’s natural soils. Using the pedologic information, it was anticipated that site perimeters, intense occupation areas, and a site development would be identified.
2. Methods a. Field Methods Six conventional 1 x 1-m excavation units were placed across the site, following the crest of the midden along the natural coastline (Fig. 24). The unexpected thickness of the cultural deposits required the enlargement of three of the test units (1, 3, and 6) to 1 X 2 m. Zones of stratigraphically uniform material within the units were noted and excavated in 5-cm levels. If a soil horizon could be differentiated within a zone, each horizon was excavated separately. Every unit was excavated to a minimum of 25 cm below the last cultural material recovered, except when intrusion of the water table made this impossible. Soil samples for chemical and particle-size distribution analysis were taken from both upper and lower areas of thicker zones. Two pedons of the naturally occurring soil Zolfo sand (sandy, siliceous, hyperthermic Grossarenic Entic Haplohumod) were sampled as controls to examine the midden’s influence on soil development. b. Laboratory Methods Organic carbon content was determined for those samples which were estimated to have more than 1 % organic carbon. A modified version of the WalkleyBlack method was used in the determinations (Nelson and Sommers, 1982). The pH was measured in a 1:1 suspension of water and soil. Total phosphorus (TP) determinations were made using the alkaline-oxidation method (Dick and Tabatabai, 1977). A variation of Chang and Jackson’s (1957) method was used to fractionate phosphorus. Particle size was determinated using the pipette method (Day, 1965).
3. Results and Discussion a. Archaeological Interpretations Analysis of cultural and zooarchaeological samples from A.B .’s Midden indicated that the site was occupied year-round, but intermittently. There was no evidence to support the presence of a permanent occupation. Stratigraphy indicates the site was used less intensively in its initial formation stages. Discrete
Figure 24 Location of excavation units at A.B.’s Midden.
62
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
depositional episodes were distinguished in the zones beneath the dense shell midden. The dense midden component of the site was separated from the previous component by a zone containing very little shell which was interpreted as being a period of very low activity, possibly of site abandonment. Following this period, the site became heavily used as a refuse area, resulting in the shell overlay. The dense accumulation of shell at A.B.’s Midden can be viewed as a discontinuity in the soil, essentially functioning as new parent material for the formation of soil above it. It is also suggested that the presence of the midden has altered the development of the island’s geomorphology through differential erosion patterns. Accelerating shoreline erosion over the last half century, coupled with the effects of major humcanes (i.e., Elena in 1982) severely impacted North Key. The shell midden sites may have helped stabilize the shoreline by retarding the effects of marine erosion. b. Morphological and Chemical Features of Archaeological Soils Morphological features in the excavation units revealed characteristics typical of human-altered soils. Horizons within site boundaries had deeper and darker matrix colors than those in the naturally occurring soils. These colors ranged from very dark gray (lOYR 3/ 1) in the upper 10-cm horizon of some units to yellowish brown (IOYR 5/8) below 2 m. Fine and medium sands made up the largest percentages of the sand fraction in A.B.’s Midden soils. Samples from the dense shell midden components of the site revealed larger percentages of very coarse and very fine sands than from the below the shell deposit (Table IV). These differences reflect the discontinuity between the shell midden component and the occupation levels. Units closer to the edge of the site exhibited particle-size distributions closer to those of the natural soils (Table V). Clay and silt contents in the archaeological soils were higher than in North Key’s natural soils and were highest in the upper shell zones. The large percentages of silt and clay at such shallow depths would be unusual for the island’s natural sandy soils and, thus, are clearly a result of human activity. Organic carbon content of the dune sand parent material of North Key is naturally very low (< 1%); therefore, organic carbon levels were predicted to be very low across the site. However, the highest amount of organic carbon was 3.77% at a depth of 75-80 cm in Unit 1 (Table VI). This reflects the intensity of occupation during these depositional episodes, and is substantiated by the large amounts of phosphorus (Table VII). Organic carbon levels dropped in zones beneath the shell midden, but remain relatively high. The pH levels across the site were significantly higher than those of the natural soils (6.3 to 7.4) and varied. The average site pH was 8.1. In midden soils the pH
Table IV Particle-Size Distribution (%) in Selected Archaeological Soils at A. B.’s Midden
Sand
VCa
C
M
F
VF
Silt
Clay
20-25 45-50 75-80 100-105 130-135 205-210
87.3 80.2 77.6 94.4 94.8 95.9
5.5 4.6 3.0 1.o 0.4 0.4
12.2 11.5 9.6 7.7 5.9 8.4
44.6 23.5 29.6 44.9 43.2 44.7
32.2 23.7 43.4 43.5 49.0 45.3
3.4 25.8 11.9 3.2 1.5 1.2
6.3 9. I 12.6 2.9 5.3 3.3
6.4 10.7 9.8 2.8 0.0 0.9
50-55 70-75 95-100 125-130
89.9 98.3 98.4 98.2
0.9 0.3 0.3 0.3
9.0 6.7 7.2 7.1
48.1 43.2 44.5 45.9
38.6 48.1 46.4 45.5
2.7 1.6 1.5 1.5
6.9 0.9 0.0 0.7
3.2 0.8 1.6
10-15 25-30 85-90
89.0 82.8 86.2 97.1 99.7 99.8 98.4 97.8
2.3 2.4 3.1 0.4 0.3 0.4 0.4 0.4
6.5 5.2 8.5 5.8 6.0 6.1 6.4 5.4
40.6 21.7 25.9 41.3 40.7 39.4 45.1 42.0
45.8 50.9 52.3 49.6 50.0 50.8 46.3 50.2
3.5 20.0 9.9 2.6 2.1 3.2 I .7 2.0
5.8 6.9 3.0 3.5 0.5 0.3 0.3 0.5
5.2 10.3 10.9 0.0 1.8 0.0 1.5
85-90
94.1 97.2 98.4
0.5 0.2 0.3
7.4 8.0 7.5
48.4 50.6 48.0
41.8 39.8 42.4
1.6 1.4 1.8
3.4 0.0 0.9
2.5 2.8 0.7
40-45 80-85 95-100
98.2 98.2 95.9
0.3 0.2 0.2
1.4 6.8 6.5
49.7 49.0 41.4
40.9 42.0 44.0
1.8 1.7 I .9
0.0 1.8 4.1
1.8 0.0 0.0
5-10 20-25 40-45 75-80 100-105 125-130 140-145 150-155 185-190
90.7 94.7 84.8 95.1 95.8 93.2 91.6 97.9 97.6
0.8 1.2 3.2
5.0 5.3 8.2 10.2 4.6 5.4 5.3 4.7 6.2
43.4 44.2 18.9 46.8 35.6 35.5 38.5 36.1 43.3
46.1
1.3 3.8 20.6 2.4 5.5 6.5 3.5 3.5 1.8
3.9 0.0 7.6 2.4 3.0 5.8 2.4 2.1 2.4
5.4 5.3 1.7 2.5 1.2
Depth (cm) Unit 1
B C D F G I Unit D E F G Unit B B E F F G J J Unit B C C Unit B B B Unit A B C E F H I J K
2
I .o
3
105-110
110-115 135-140 180-185 195-200
1.9
4 15-20 50-55
5
6
1.o
0.4 0.2 0.3 0.3 0.3
a Very coarse (VC), coarse (C), medium
44.2 48.2 39.5 53.8 52.3 52.4 53.7 41.1
(M), fine (F), and very fine (VF) sand fractions.
I .o 0.0 0.0 0.0
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
64
Table V Particle-Size (%) and Total Phosphorus (TP) Distribution in Natural Soils at A.B.’s Midden
Horizon
A1 A2 E2 E2 E2 E3 E3 E4 Bh BC Al
A2 E2 E2 E3 E3 E3 E4 Bh Bh
TP
Depth (cm)
Sand
VCa
C
M
F
VF
Silt
Clay
(pg/g)
0-10 40-50 80-90 100-105 105-115 120- 130 135-145 155-166 180- 190 225-235 25-35 45-50 70-80 90- 100 100- I 10 110-120 135-145 150- 160 200-210 2 15-220
92.8 96.9 99.1 97.2 97.7 97.5 97.4 97.7 99.1 98.5 97.9 98.2 98.4 98.8 98.3 97. I 97.4 98.5 99.1 99.2
0.8 0.4 0.4 0.6 0.4 0.5 0.7 1.o 1.7 0.8 0.4 0.3 0.2 0.4 0.4 0.7 0.6 0.9 0.5 0.3
7.0 6.4 7.3 8.4 9.1 8.2 8.4 9.2 9.0 5.2 8.3 8.6 8.4 8.7 8.5 9.2 9.6 12.5 10.8 11.6
49.0 45.9 48.9 50.5 49.6 46.2 46.6 48.5 44.8 42.2 50.0 51.4 51.7 48.2 48.6 46.4 46.2 47.5 48.0 52.4
41.3 46.0 42.2 39.3 40.0 43.9 43.3 43.4 43.7 50.8 39.2 37.8 38.0 40.9 41.5 40.2 41.9 37.2 39.2 39.6
1.3 1.3 1.2 0.9
3.1 0.4 0.0 0.8 0.0 0.2
4.0 2.8 0.9 2.0 2.3 2.3 1.2 1.2 0.9 1.5 0.0 2.8 0.0 0.9 1.5 2.5 2.3 0.0 0.0 0.8
145 160 100 133 139 166 180 124 180 99 78 108 64 67 112 215 316 254 364 304
a Very coarse
1.0 1.1
1.0 1.0 0.8 0.7 1.8 1.7 1.5
1.8 1.4 1.6 1.8 1.8 1.2 1.1
1.5
1.2 0.0 0.0 2.1 0.4 1.6 0.0 0.3 0.4 0.3 1.6 0.9 0.0
(VC), coarse (C), medium (M), fine (F), and very fine (VF) sand fractions.
tended to increase with depth, which is the result of weathering and leaching of Ca from the thick overlying shell zones. Total phosphorus levels were extremely high across the extent of A.B.’s Midden, but varied with depth, reflecting changes in depositional episodes and pedogenesis (Table VII). As expected, TP levels in the dense shell midden component of the site were the highest due to the presence of large quantities of food refuse, primarily bone. The soft parts of mollusks are particularly high in P; however, their shells contain almost no P, and do not contribute to the high P levels found in the shell midden (Cook and Heizer, 1965; Carr, 1982). Phosphorus levels dropped dramatically as the midden thinned out. It should be noted that P levels in natural soils vary widely. Therefore, concentrations from archaeological sites from differing regions should ideally be correlated with local, natural concentrations before being compared. For example, phosphorus levels in the natural soils ranged from 64 mg/g in the eluvial (E) horizons to 364 mg/g in the spodic horizon (Table V). In the natural soils, the P increased gradually with depth to the spodic horizon, then decreased.
SOIL SCIENCE AND ARCHAEOLOGY
65
Table VI Organic Carbon Content in Selected Archaeological Soils at A.B.’s Midden
Depth Zone Unit 1 B C
D F Unit 2 D Unit 3 B B E F
F
(cm)
% Organic carbon
20-25 45-50 75-80 100-105
3.23 3.06 3.77 1.20
50-55
1.76
10-15 25-30 85-90 105-1 10 110-1 15
3.42 3.37 2.71 2.39 1.43
5-10 20-25 40-45 75-80 125-130
3.17 2.95 2.77 3.62 0.76
Unit 6 A
B C
E H
4. Summary
The pedological analysis of A.B.’s Midden contributed to the overall understanding of site use and formation processes, yielding data on the impact of prehistoric human activities on pedogenesis. Morphologically, site zones were evidenced by deeper, darker colors developed from the accumulation of organic material during occupation. Stratigraphic breaks indicated discrete occupation sequences. Textures from the site suggest the A.B.’s Midden was used irregularly during its early formation, allowing natural accumulation of aeolian or storm-driven sands to separate occupations. Later, more intense occupations resulted in the rapid accumulation of shell zones, deposited sequentially with little accumulation of soil. Although sand dominated both natural and site soil textures, archaeological zones were higher in fine and very fine sand, while the natural soils had more medium sand. Some archaeological zones were high in silt- and clay-sized fractions.
Table VIl Total and FractionatedPhosphorus Contents in Selected ArchaeologicalSoils at A.B.'s Midden Zone Unit 1 B C F G H I Unit 2 D E F G Unit 3 B B E F F G J J Unit 4 B C C Unit 5 B B B Unit 6 A B C E F G
H I
J K
Depth (cm)
HzOu
Al'
Fe"
Caa
20-25 45-50 100-105 130-135 170-175 205-210
54 48 75 6 8 9
906 2000 1190 14 38 23
76 186 65 11 23 22
50-55
93 40 30 31
1443 32 25 36
55 138 I28 71 56 36 19 18
P subtotal
TP
3,264 21,208 1,326 12 40 22
4,300 23,541 2,656 43 100 76
4,778 26,157 2,951 48 110 84
269 6 14 37
7,296 89 18 18
9,101 167 86 122
10,112 185 96 135
372 1155 2162 118 52 57 54 60
85 128 150 68 70 74 51 89
1,977 7,992 9,422 350 91 57 51 43
2,489 9,413 1 1,863 607 269 224 175 210
2,766 10,459 13,118 674 299 249 195 322
85-90
92 65 19
198 52 44
59 29 28
266 59 22
614 204 113
682 227 125
40-45 80-85 95-100
2 4 5
3 15 11
2 3 9
8 0 10
14 22 35
16 24 39
172 94 115 38 57 80 88 15 39 34
948 1259 I340 475 111 67 626 100 118 22
128 75 122 23 37 41 60 61 26 23
1,635 3,420 11,529 647 417 200 1,126 54 1 222
2,885 4,847 13,106 1,183 62 1 388 1,901 716 405 130
3,206 5,386 14,562 1,314 690 43 1 2,112 796 449 144
70-75 95-100 125-130 10-15
25-30 85-90 105-1 10 110-1 15 135-140 180-185 195-200 15-20 50-55
5-10 20-25 40-45 75-80 100-105 110-1 15 125- 130 140-145 150-155 185-190
50 ~~
(I
Renodes water-soluble (HzO), aluminum (Al), iron (Fe), and calcium (Ca) phosphates.
SOIL SCIENCE AND ARCHAEOLOGY
67
The higher organic carbon content of the site produced morphological and chemical changes in the diagnostic spodic horizon compared to the spodic horizon occurring in the natural soils. Within site boundaries, particularly under the most densely occupied areas, the spodic horizon exhibited darker colors, firmer consistence, more developed structure, and a higher P content than samples of the spodic horizon in the natural soils. The natural acidity of North Key’s soils was raised by 1 to 3 units within the site perimeter, caused by large concentrations of calcium carbonate in shelldense horizons of the site. Total P values indicate relative occupation intensity of the site through time, and suggest that dense shell zones were associated with habitation areas. This interpretation is supported by zooarchaeological and sclerochronological assessments of A.B.’s Midden. The presence of shell-bearing sites on North Key, including A.B.’s Midden, has had a dramatic effect on the erosional history of the island (Borremans, 1990), while the addition of human-introduced nutrients to the soil environment has altered specific aspects of its character, including the expression of diagnostic subsurface horizons in the soil.
VI. SUMMARY Fundamental changes in the focus of modem archaeological research have resulted in the integration of the earth sciences with archaeology. This new focus emphasizes the synthesis of cultural and environmental information into a cohesive interpretation of human ecology, including settlement patterns, land use practices, and evidence of human impact on soils and landscapes. This chapter defines the role of the emerging disciplines of geoarchaeology, pedoarchaeology, and archaeometry in the archaeological context. It reviews recent works in pedoarchaeology at a wide range of sites, some with a broad, regional landscape approach and some with a more narrowly focused intrasite view. The foundation of modem earth science disciplines includes traditional soil science techniques: soil morphological descriptions, particle-size distribution analysis, chemical element distributions, clay mineralogy, landscape analysis, and micromorphology. These techniques are outlined, as are uses of soil maps and modem geophysical tools such as resistivity, electromagnetic survey, and ground-penetrating radar. Paleosols are defined and their use in the interpretation of archeological sites is illustrated. The use of soil morphology as a stratigraphic marker in both terrestrial and underwater sites is also discussed. The synthesis of pedology and archaeology is illustrated in a series of case studies selected from the authors’ works. Topics as diverse as the content of toxic
68
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
metals in garden soils at Hadrian’s Villa, the role of soil resources in the collapse of Mayan culture in Guatemala, relationships of alluvial soil sequences to early archaeological sites in the southeastern United States, the origin of monumental sand earthworks in southwest Florida, and the effects of middens on the soils and geomorphology of a coastal Florida island are addressed. Though the analytic techniques used in these studies are basic, the information they yield is as varied as the terrains and cultures they describe. The first two objectives of this chapter-to stress the mutual advantages of new interdisciplinary efforts in earth sciences and to summarize unique contributions of pedology to archaeology-are met in the examples of the work itself. The third objective-to encourage other pedologists to become involved in archaeological studies-can only be evaluated by future chapters such as this.
REFERENCES Abbott, R. T. (1974). “American Seashells,” second edition. Van Nostrand-Reinhold, New York. Allison, L. E., and Moodie, C. D. (1965). Carbonates, In “Methods of Soil Analysis,” Part 2, Agronomy 9 (C. A. Black et al., Eds.), pp. 1387-1388. American Society of Agronomy, Madison, WI. Arrhenius, 0. (1929). “Die Phosphatfrage.” Zeitschrift fur Pflanzenemahrung, Dundung und Bodenkunde 14A, 185-194. Baldwin, R., Bush, C. L., Hinton, R. B., Huckle, H. F., Nichols, P., Watts, F. W., and Wolfe, J. A. (1980). ‘‘Soil Survey of Volusia County, Florida.” USDA-Soil Conservation Service, U.S. Govt. Printing Office, Washington, DC. Bandel, V. A., and Rivard, C. E. (1975). “Soil Testing Methods.” Univ. of Maryland Soil Testing Lab. Agronomy Mimeo 37. Bartelli, L. J., er al. (Eds.) (1966). Soil surveys and land use planning. Soil Science SOC.Am. and Am. SOC.of Agronomy, Madison, WI. Batey, R. A. (1987). Subsurface interface radar at Sepphoris, Israel, 1985. J. FieldArch. 14(1), 1-8. Bauman, P. D., Heitzmann, R. J . , Porter, J. E.,Sallomy, J. T., Brewester, M. L., and Thompson, C. J. (1994). The intergration of radar, magnetic, and terrain conductivity data in an archaeogeological investigation of a Hudson’s Bay Company fur trade post. In “Proceedings of the 5th International Conference on Ground-Penetrating Radar,” pp. 53 1-546. Waterloo Centre for Groundwater Research Kitchener, Ontario, Canada. Bernabini, M., Brizzolari, E., Orlando, L., and Santellani, G . (1994). Application of groundpenetrating radar on collosseum pillars. I n “Proceedings of the 5th International Conference on Ground-Penetrating Radar,” pp. 547-558. Waterloo Centre for Groundwater Research Kitchener, Ontario, Canada. Bettis 111, E. A. (1992). Soil morphologic properties and weathering zone characteristics as age indicators in Holocene alluvium in the upper midwest. I n “Soils in Archaeology” (V. T. Holliday, Ed.), pp. 119-144. Smithsonian Institution Press, Washington, DC. Birkeland, P. W. (1984). “Soils and Geomorphology.” Oxford Univ. Press, New York. Boersma, L. L., et al. (Eds.) (1987). “Future Developments in Soil Science Research.” Soil Science SOC.Am., Madison, WI. Borremans, N. (1990). Shell middens and coastal paleogeography on the Florida Gulf coast. In “6th
SOIL SCIENCE AND ARCHAEOLOGY
69
International Congress of Archaeozoology, Abstracts.” Smithsonian Institution, Washington, DC . Bower, C. A., and Wilcox, L. V. (1965). Soluble salts. In “Methods of soil analysis,” Part 2, Agronomy 9 (C. A. Black et al., Eds.), pp. 936-940. American Society of Agronomy, Madison, WI. Broadbent, N. (1981). Phosphate analysis in archaeology: Anthropological uses of an old method. Soc. Am. Archaeol. Proc. 42, 1-16. Bullard, R. G. (1985). Sedimentary environments and lithologic materials at two archaeological sites. In “Archaeological Geology.” (G. Rapp and J. Gifford, Eds.). Yale Press, New Haven. Buol, S. W., Hole, F. D., and McCracken, R. J. (1989). “Soil Genesis and Classification.” 3rd ed. Iowa State University Press, Ames, IA. Butzer, K. W. (1982). “Archaeology as Human Ecology.” Cambridge University Press, New York. Cam, C. (1982). “Handbook on Soil Resistivity Surveying.” Center for American Archaeology Press, Evanston, IL. Chang, S. C., and Jackson, M. L. (1957). Fractionation of soil phosphorus. Soil Sci. 84,133-144. Collins, M. E., and Doolittle, J. A. (1993). Applications of ground-penetrating radar and soil science to archaeological investigations. In “Proceedings of the First International Conference on PedoArchaeology” ( J. E. Foss, M. E. Timpson, and M. W. Morris, Eds.), pp. 117-124. Special Pub. 93-03, University of Tennessee Agricultural Experiment Station, Knoxville. Collins, M. E., and Shapiro, G. (1987). Comparisons of human-influenced and natural soils at the San Luis Archaeological Site, Florida. Soil Sci. Soc. Am. J . 51, 171-176. Conway, J. S. (1983). An investigation of soil phosphorus distribution within occupation deposits from a Romano-British hut group. J . Archaeol. Sci. 10, 117-128. Cook, S. F., and Heizer. R. F. (1965). “Studies on the Chemical Analysis of Archaeological Sites.” University of California Publications in Anthropology 2, Berkeley. Cordell, A. (1996). Technological investigations of pottery variability at the Pineland Site Complex. In “The Archaeology of Pineland: A Coastal Southwest Florida Village Complex, A.D. 1001600,” Chap. 1 1 (K. Walker and W. Marquardt, Eds.) Inst. of Archaeology and Paleoenvironmental Studies, Monograph 3. Coultas, C. L., Collins M. E., and Chase, A. F. (1993). Effect of ancient Maya agriculture on terraced soils of Caracol, Belize. In “Proceedings of the First International Conference on PedoArchaeology‘‘ (J. E. Foss, M. E. Timpson, and M. W. Moms, Eds.), pp. 191-200. Special Pub. 93-03, University of Tennessee Agricultural Experiment Station, Knoxville. Courty, M. A,, Goldberg, P., and Macphail, R. (1989). “Soils and Micromorphology in Archaeology.” Cambridge Manuals in Archaeology, Cambridge University Press, Cambridge, England. Cowgill, U. M., and Hutchinson, G. E. (1963). El Bajo de Santa Fe. Trans. Am. Phil. SOC.53, 344. Cushing, F. H. (1897). Exploration of ancient key dweller remains on the Gulf coast of Florida. Am. Phil. SOC. Proc. 35, 329-448. Dahlin, B. H. (1984). A colossus in Guatemala: The Preclassic Maya city of El Mirador. Archaeology 37, 18-25. Dahlin, B. H., Foss, J. E., and Chambers, M. E. (1980). Project Acalches: Reconstructing the natural and cultural history of a seasonal swamp at El Mirador, Guatemala. N. World Arch. Found. 45, 37-58. Dauncey, K. D. M. (1952). Phosphate content of soils on archaeological sites. Advmr. Sci. 9(33), 33-37. Day, P. R. (1965). Particle fractionation and particle-size analysis. In “Methods of Soil Analysis,’’ Part I , Agronomy 9 (C. A. Black, Ed.) pp. 548-567. American Society of Agronomy, Madison, WI. DeWitt, T. A. (1984). “Soil Survey of Story County, Iowa.” USDA-Soil Conservation Service, U.S. Gov. Printing Office, Washington, DC.
70
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
Dick, W., and Tabatabai, M. A. (1977). An alkaline oxidation method for determination of total phosphorus in soils. Soil Sci. SOC. Am. J. 41, 51 1-514. Doolittle, J. A. (1988). Ground-penetrating radar (GPR) survey at Tell Halif, Israel. In “Technical Proceedings of the 2nd International Symposium on Geotechnical Applications of GroundPenetrating Radar.” IFAS, Gainesville, FL. Doolittle, J. A,, and Miller, W. F. (1991). Use of ground-penetrating radar in archaeological investigations. In “Conference Proceedings of Applications of Space-Age Technology in Anthropology,” (C. A. Behrens and T. L. Sever, Eds.) pp. 81-94. NASA-Stennis, Mississippi. Douglas, A. E. (1885). Ancient canals on the southwest coast of Florida. Am. Anriquarian Orienral J. 7, 277-285. Duncan, R. S., and Hurt, G. W. (1993). Geographic information system archaeological site model. In “Proceedings of the First International Conference on Pedo-Archaeology” (1. E. Foss, M. E. Timpson, and M. W. Moms, Eds.), pp. 125-132. Special Pub. 93-03, University of Tennessee Agricultural Experiment Station, Knoxville. Dunning, N. P. (1993). Ancient Maya anthrosols: Soil phosphate testing and land use. In “Proceedings of First International Conference Pedo-Archaeology” (J. E. Foss et al., Eds.), pp. 203-209. Special Pub. 93-03, University of Tennessee Agricultural Experiment Station, Knoxville. Eidt, R. C. (1985). Theoretical and practical considerations in the analysis of anthrosols. In “Archaeological Geology’’ (G. Rapp and J. A. Gifford, Eds.), pp. 155-191. Yale Univ. Press, New Haven. Farrand, W. R. (1975). Sediment analysis of a prehistoric rock shelter: The Abri Pataud. Quaternary Res. 5 , 1-26. Fenwick, I. M. (1985). Paleosols: Problems of Recognition and Interpretation. In “Soils and Quaternary Landscape Evolution” (J. Boardman, Ed). John Wiley, Chichester, United Kingdom. Ferring, C. R. (1992). Alluvial pedology and geoarchaeological researh. In “Soils in Archaeology” (Holliday, V. T., Ed.), pp. 1-39. Smithsonian Institution Press, Washington, DC. Flint, R. F. (1976). Physical evidence of quaternary climatic change. Quarernury Res.6, 519-528. Follmer, L. R. (1978). The Sangamon soil in its type area-A review. In “Quaternary Soils’’ (W. C. Mahoney, Ed.). Geoabstracrs, Norwich. Follmer, L. R. (1982). The geomorphology of the Sangamon surface: Its spatial and temporal attributes. In “Space and Time in Geomorphology” (C. E. Thorn, Ed.), The Binghampton Symposia on Geomorphology: International Series No. 12. Allen and Unwin, London. Foss, J. E. (1974). Soils of the Thunderbird site and their relationship for cultural occupation and chronology. In “The Flint Run Paleo-Indian Complex” (W. E. Gardner, Ed.), pp. 66-83. Occasional Pub. No. 1, The Catholic University of America. Foss, J. E. (1991). Genesis of soils on alluvium near the Delaware water gap. In “The People of Minisink, Papers from 1989 Delaware Water Gap Symposium” (D. G. Orr and D. V. Campana, Eds.). Natural Park Services, Department of the Interior, Washington, DC. Foss, J. E., Lewis, R. J., Timpson, M. E., Moms, M. W., and Ammons, J. T. (1993). Pedological approaches to archaeological sites of contrasting environments and ages. In “Proceedings of the First International Conference on Pedo-Archaeology” ( J. E. Foss, M. E. Timpson, and M. W. Moms, Eds.), pp. 19-22. Special Pub. 93-03, University of Tennessee Agricultural Experiment Station, Knoxville. Foss, J. E., Lewis, R. J., Stiles, C. A., Lee, S . Y., and Timpson, M. E. (1994a). Evidence of environmental contamination at several Roman archaeological sites in Italy and Tbnisia. In “Environmental Contamination, 6th International Conference,” pp. 189- 191. CEP Consultants, Edinburgh, UK. Foss, J. E., Timpson, M. E., Lewis, R. J., and Stiles, C. A. (1994b). Soil interpretation of some Roman archaeological sites in Italy and ’hnisia. ISrh Int. Cong. Soil Sci. 66, 172-173.
SOIL SCIENCE AND ARCHAEOLOGY
71
FOSS,J. E., Timpson, M. E., and Lewis, R. J. (1995). Soils in alluvial sequences: Some archaeological implications, pp. 1-14. Soil Science SOC.Am., Special Pub. 44. Madison, WI. Foss, J. E. and Segovia, A. V. (1984). Rates of soil formation. In “Groundwater as a Geornorphic Agent” (R. G. LaFleur, Ed.), pp. 1-17. Rensselaer Polytechnic Institute. Allen & Unwin, London. Foss, J. E., Wagner, D. P., and Miller, F. P. (1981). “Soils of the Savannah River Valley.” Russell Papers 1985, Archeological Service, Atlanta, GA. Gagliano, H. (1977). Cultural resources evaluation of the northern Gulf of Mexico continental shelf. In “Interagency Archaeological Services,” Vol. I . National Parks Service, Washington, DC. Gardner, W. E. (1974). The Flint Run complex: Pattern and process during the Paleo-Indian to early Archaic. In “The Flint Run Complex’’ (W. M. Gardner, Ed.), pp. 5-47. Occasional Pub. No. 1, The Catholic University of America. Gartner, W. G. (1992). Soils and sediments as artifacts in prehistoric Wisconsin. In “Proceedings of the First International Conference on Pedo-Archaeology,” pp. 113-1 17. Special Pub. 93-03, University of Tennessee Agricultural Experiment Station, Knoxville. Gerasirnov, I. P. (1971). Nature and origin of paleosols. In “Paleopedology: Origin, Nature and Dating of Paleosols” (D. H. Yaalon, Ed.). International Society of Soil Science and Isreal Universities Press, Jerusalem. Gerrard, A. J. (1981). ‘‘Soils and Landforms.” Allen & Unwin, London. Gilliland, M. S. (1975). “The Material Culture of Key Marco, Florida.” University Presses of Florida, Gainesville. Goggin, J. M. and Sturtevant, W. T. (1964). The Calusa: A stratified non-agricultural society (with notes on sibling marriage). In “Explorations in Cultural Anthropology: Essays in Honor of George Peter Murdock” (W. Goodenough, Ed.), pp. 179-219. McGraw-Hill, New York. Goldberg, P. (1992). Micromorphology, soils, and archaeological sites. In ‘‘Soils in Archaeology” (V. T. Holliday, Ed.), pp. 145-167. Smithsonian Institution Press, Washington, DC. Goldberg, P., and Courty, M. (1993). Micromorphology and the geo-pedo-anthropogenic interface at archaeological sites. In “Proceedings of the First International Conference on Pedo-Archaeology“ (J. E. Foss, M. E. Timpson, and M. W. Moms, Eds.), pp. 107-112. Special Pub. 93-03, University of Tennessee Agricultural Experiment Station, Knoxville. Goodman, D., and Nishimura, Y. (1992). 2-D synthetic radargrams for archaeological investigations. In “Technical Proceedings of the 4th International Conference on Ground-Penetrating Radar, Rovaniemi, Finland” (P.Hanninen and S. Autio, Eds.), pp. 339-343. Geological Survey of Finland Special Paper 16, Espoo, Finland. Goodyear, A. C., and Foss, J. E. (1993). The stratigraphic significance of paleosols at Smith Lake Creek (38AL135) for the study of the Pleistocene-Holocene Transition in the Savannah River Valley. In “Proceedings of the First International Conference on Pedo-Archaeology,” pp. 27-4 1. Special Pub. 93-03, University of Tennessee Agricultural Experiment Station, Knoxville. Goodyear, A. C., and Foss, J. E. (1994). “Field Trip Guidebook, 2nd Int. Conf. of Pedo-Archaeology.” South Caroline Institute of Archaeology and Anthropology. Univ. of South Carolina, Columbia. Griffith, M. A. (1980). A pedological investigation of an archaeological site in Ontario, Canada. I . An examination of the soils in and adjacent to a former village. Geoderma 24, 327-336. Griffith, M. A. (1981). A pedological investigation of an archaeological site in Ontario, Canada. 11. Use of chemical data to discriminate features of the Benson site. Geoderma 25, 27-34. Hammond, N. (1994). The emergence of Maya civilization. Sci. Am., Special Issue on Ancient Cities. Hassan, F. A. ( 1985). Paleoenvironments and contemporary archaeology: A geoarchaeological approach. In “Archaeological Geology” (G. Rapp and J. A. Gifford, Eds.), pp. 85-101. Yale Univ. Press. New Haven.
72
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
Henderson, W. G. (1984). “Soil Survey of Lee County, Florida.” USDA-Soil Conservation Service, U.S. Government Printing Office, Washington, DC. Hodges, S. C., and Carlisle, V. W. (1979). Genesis of acid sulfate soils. Soil Crop Sci. SOC.froc. 30, 62-68. Holliday, V. T. (1985). Morphology of Late Holocene soils at the Lubbock Lake archaeological site. SSSAJ 49, 938-946. Holliday, V. T. ( 1989). Paleopedology in archaeology. In “Paleopedology: Nature and Application of Paleosols” (A. Bronger and J. A. Catt, Eds.), Catena Supplement 16. Cremlingen-Destedt, Germany. Holliday, V. T. (1992). Soil formation, time, and archaeology. In “Soils in Archaeology” (V. T. Holliday, Ed.), pp. 101-1 18. Smithsonian Institution Press, Washington, DC. Horvath, E. A,, Chapman, A. A., Martin, D. B., and McGowan, R. (1995). “Final report on the Archaeological Investigations at the Seminole Rest Site (CANA-O63/8VO124):’ Southeast Archaeological Center, National Park Service, Tallahassee, FL. Imai, T., Sakayama, T., and Kanemori, T. (1987). Use of ground-probing radar and resistivity surveys for archaeological investigations. Geophysics 52(2), 137- 150. Jacobs, J. S. (1991). The Agroecological Evolution of Cobweb Swamp, Belize” (Preliminary Report to the National Geographic Society for work supported by Grant 4274-90). Texas A & M Univ., College Station, TX. Jenny, H. (1941). “Factors of Soil Formation.” McGraw-Hill, New York. Jenny, H. (1961). Reflections on the soil acidity merry-go-round. Soil Sci. SOC. froc., 428-432. Johnson, W. G. (1991). Remote sensing and soil science applications to understanding Belle Glade cultural adaptions in the Okeechobee basin. Ph.D. dissertation, Univ. of Florida, Gainesville. [Diss. Abstr. DEZ 92020091 Johnson, W. G., and Collins, M. E. (1993). Can soil evidence prove that prehistoric maize was not the basis for complex cultural developments in the Okeechobee basin? In “Proceedings of the First International Conference on Pedo-Archaeology“ (J. E. Foss, M. E. Timpson, and M. W. Moms, Eds.), pp. 151-160. Special Pub. 93-03, University of Tennessee Agricultural Experiment Station, Knoxville. Kilmer, V. J., and Alexander, L. T. (1949). Methods of making mechanical analysis of soils. Soil Sci. 68, 15-24. Kong, F-N., Kristiansen, J., and By, T. L. (1992). A radar investigation of pyramids. In “Technical Proceedings of the 4th International Conference on Ground-Penetrating Radar, Rovaniemi, Finland” (P. Hanninen and S. Autio, Eds.), pp. 345-349. Geological Survey of Finland Special Paper 16, Espoo, Finland. Kubiena, W. L. (1938). “Micropedology.” Collegiate Press, Ames, IA. Kuehl, R. J. and Benson, R. (1993). Application of soil science to underwater archaeology: Oklawaha River project. In “Proceedings of the First International Conference on PedoArchaeology” (J. E. Foss, M. E. Timpson, and M. W. Moms, Eds.). University of Tennessee Agricultural Experiment Station, Knoxville. Larsen, C. S. (1987). Bioarchaeological interpretations of subsistence economy and behavior from human skeletal remains. In “Advances in Archaeological Method and Theory” (M. B. Schiffer, Ed.), pp. 339-345. Academic Press, San Diego. Leatherman, S. P. (1988). “Barrier Island Handbook,” Coastal Publication Series. Laboratory for Coastal Research. University of Maryland, College Park. Lewis, R. J., Foss, J. E., Moms, M. W., Timpson, M. E., and Stiles, C. A. (1993). Trace Element Analysis in Pedo-Archaeology Studies. In “Proceedings of the First International Conference on Pedo-Archaeology” (J. E. Foss, M. E. Timpson, and M. W. Moms, Eds.), pp. 81-88. Special Pub. 93-03, University of Tennessee Agricultural Experiment Station, Knoxville.
SOIL SCIENCE AND ARCHAEOLOGY
73
Lillios, K. T. (1992). Phosphate fractionation of soils at Agroal, Portugal. Am. Antiquity 57(3), 495506. Lippi, R. D. (1988). Paleotopography and phosphate analysis of a buried jungle site in Ecuador. J . Field Archaeol. 5 , 85-97. Luer, G. M. (1989). Calusa canals in southwestern Florida: Routes of tribute and exchange. Florida Anthropol. 42(2), 89-130. h e r , G. M. (1991). Historic resources at the Pineland Site, Lee County, FL. Florida Anthropol. 44(I ) , 59-76. Mandel, R. D. (1992). Soils and Holocene landscape evolution in Central and Southwestern Kansas: Implications for archaeological research. In “Soils in Archaeology” (V. T. Holliday, Ed.). Smithsonian Institution Press, Washington, DC. Marquardt, W. H. (1992a). Recent archaeological and paleoenvironmental investigations in southwest Florida. In “Cultural and Environment in the Domain of the Calusa” (W. H. Marquardt, Ed.), Institute of Archaeology and Paleoenvironmental Studies, Monograph No. 1. University of Florida, Gainesville. Marquardt, W. H. (1992b). The Calusa domain. In “Culture and Environment in the Domain of the Calusa” (W. H. Marquardt, Ed.), Institute of Archaeology and Paleoenvironmental Studies, Monograph No. I . University of Florida, Gainesville. Mehta, A. J., and Brooks, H.K. (1973). Mosquito Lagoon barrier beach study. Shore and Beach 41(2), 27-34. Mellett, J. S. (1992). Location of human remains with ground-penetrating radar. In “Technical Proceedings of the 4th International Conference on Ground-Penetrating Radar, Rovaniemi, Finl a n d (P. Hanninen and S. Autio, Eds.), pp. 359-365. Geological Survey of Finland Special Paper 16, Espoo, Finland. Missimer, T. (1973). Growth rates of beach ridges on Sanibel Island, Florida. Trans. Gulf Coast Assoc. Geol. SOC.23, 383-388. Moms, M. W., Foss, J. E., Ammons, J. T., and Klippel, W. E. (1993). The use of soil morphology in archaeology: A technique for discriminating the anthropogenic influence of buried soils. In “Proceedings of the First International Conference on Pedo-Archaeology” (J. E. Foss, M. E. Timpson, and M. W. Morris, Eds.), pp. 89-106. Special Pub. 93-03, University of Tennessee Agricultural Experiment Station, Knoxville. Movius, H. L., Jr. (1975). Excavation of the Abri Pataud, Les Eyzies (Dordogne). Bulletin No. 30, American School of Prehistoric Research, Peabody Museum, Harvard University, Cambridge, MA. Nelson, N. C. (1918). Chronology in Florida. Anrhropol. Papers Am. Museum Natural History 22(2), 75-103. Nelson, R. E., and Sommers, L. E. (1982). Total carbon, organic carbon and organic matter. In “Methods of Soil Analysis,” Part 2 (A. L. page er al., Eds.), Chap. 29. Am. SOC.of Agronomy, Madison, WI. Olson, G. W. (1977). The soil survey of Tikal. Cornell Agronomy Mimeo 77-13. Dept. of Agronomy, Cornell Univ., Ithaca, NY. Olson, G. W. (1981). “Soils and the Environment.” Chapman and Hall, New York. Papamarinopoulos, S. P., and Papaicannou, M. G. (1994). Geophysical investigations with the georadar in the middle of Athens at Syntagma Square and the discovery of the subterrean river Eridenos. In “Proceedings of the 5th International Conference on Ground Penetrating Radar,” pp. 569-576. Waterloo Centre for Groundwater Research, University of Waterloo, Waterloo, Ontario, Canada. Pearsall, D. M. (1989). “Paleoethnobotany: A Handbook of Procedures.” Academic Press, New York.
74
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
Pendall, E., and Amundson, R. (1990). Soilllandform relationships surrounding Harappa archaeological site, Pakistan. Geoarch. 5 , 301-322. Piperno, D. R. (1988). “Phytolith Analysis: An Archaeological and Geological Perspective.” Academic Press, New York. Proudfoot, B. (1976). The Analysis and Interpretation of Soil Phosphorus in Archaeological Contexts. In “Geoarchaeology: Earth Science and the Past” (D. A. Davidson and M. L. Shackley, Eds.), pp. 93-1 14. Westview Press, Boulder, CO. Pun, H. S . , and Vernon, R. 0. (1964). Summary of the geology of Florida and guidebook to the classic exposures. Florida Geologic Survey Special Publ. No. 5 . Tallahassee, FL. Rolph, T. C., Shaw, J., Derbyshire, E., and Zhisheng, A. (1994). Determining paleosol topography using seismic refraction. Quaternary Res. 42, 360-353. Ruellan, A. (1971). The history of soils: Some problems of definition and interpretation. In “Paleopedology: Origin, Nature and Dating of Paleosols” (D. H. Yaalon, Ed.). Israel University Press, Jersualem. Ruhe, R. V. (1956). Geomorphic surfaces and the nature of soils. Soil Sci. 82, 441-455. Ruhe, R. V. (1960). Elements of the soil landscape. Trans. 7th Int. Congr. Soil Sci. 4, 165170. Ruhe, R. V. (1965). Quaternary paleopedology. In “The Quaternary of the United States” (H. E. Wright and D. G. Frey, Eds.). Princeton University Press, Princeton, NJ/Israel Universities Press, Jerusalem. Ruhe, R. V. (1969). “Quaternary Landscapes in Iowa.” Iowa State University Press, Ames, IA. Ruhe, R. V. (1975). “Geomorphology.” Houghton Mifflin, Boston. Ruhe, R. V., Miller, G. A., and Vreeken, W. J. (1971). Paleosols, loess sedimentation, and soil stratigraphy. In “Paleopedology: Origin, Nature and Dating of Paleosols” (D. H. Yaalon, Ed.). Hebrew Universities Press, Jerusalem. Ruhl, D. (1995). Paleoethnobotanical investigations at Seminole Rest. In “Final Report of the Seminole Rest Site, Volusia County, FL” (E. Horvath. Ed.), Chap. 3, pp. 41-58. Southeast Archaeological Center, National Park Service, Tallahassee, FL. Sakayama, T., Osada, T. M., and Tamura, K. (1988). Some examples of archaeological investigations using ground-probing radar. In “Technical Proceedings of the 2nd International Symposium on Geotechnical Applications of Ground-Penetrating Radar.” IFAS, Gainesville, FL. Sanchez, P. A. (1976). “Properties and Management of Soils in the Tropics.” Wiley-Interscience, New York. Sandor, J. A. (1992). Long-term effect of Prehistoric agriculture on soils: Examples from New Mexico and Peru. In “Soils in Archaeology” (V. T. Holliday, Ed.), pp. 217-245. Smithsonian Institution Press, Washington, DC. Schmalzer, P. A,, and Hinkle, G. R. (1990). Geology, geohydrology and soils of Kennedy Space Center: A review. National Aeronautics and Space Admin. J. F. Kennedy Space Center. Scholtes, W. H., Ruhe, R. V., and Riecken, F. F. (1951). Use of morphology of buried soil profiles in the Pleistocene of Iowa. Proc. Iowa Acad. Sci. 58, 295-306. Scudder, S. J. (1993). Human influence on pedogenesis: Midden soils on a southwest Florida Pleistocene dune island. M.S. Thesis, University of Florida, Gainesville. Scudder, S. J. (1995). Soils investigations at the Seminole Rest archaeological site. In “Final Report of the Seminole Rest Site, Volusia County, FL.” (E. Horvath, Ed.), Chap. 2, pp. 15-40. Southeast Archaeological Center, National Park Service, Tallahassee, FL. Scudder, S. J., Harris, W. G., and Collins, M. E. (1993). Clay-sized mineral composition of antropogenic and native soils on a southwest Florida Pleistocene dune island. In “Proceedings of the First International Conference on Pedo-Archaeology” (J. E. Foss, M. E. Timpson, and M. W. Moms, Eds.), pp. 161-175. Special Pub. 93-03, University of Tennessee Agricultural Experiment Station, Knoxville.
SOIL SCIENCE AND ARCHAEOLOGY
75
Segovia, A. V. (1985). “Archaeological Geology of the Savannah River Valley and Main Tributaries in the Richard Russell Multiple Resource Area.” Interagency Archaeological Services Division, National Park Service, Russell Papers 1985, Atlanta. Simmons, C. S . . Tarano, J. M., and Pinto, H. (1959). Clarificasion de reconocimiento de 10s suelos de la Republica de Guatemala, pp. 555-558. Ministerio de Agricultura, Servicio Cooperativo Interamericano de Agricultura, Guatemala. Sjoberg, A. (1976). Phosphate analysis of anthropic soils. J. Field Archaeol. 3, 447-454. Smith, N. (1980). Anthrosols and human carrying capacity in Amazonia. Ann. Assoc. Am. Geographers 70(4), 553-566. Soil Science Society of America. (1982). “Methods of Soil Analysis,” Part 2: “Chemical and Microbiological Properties” (A. L. Page, R. H. Miller, and D. R. Keeney, Eds.). American Society of Agronomy, Madison, WI. Soil Science Society of America. (1986). “Methods of Soil Analysis,” Part 1: “Physical and Mineralogical Methods” (A. Klute, Ed.). American Society of America, Madison, WI. Soil Survey Division Staff. (1993). “Soil Survey Manual.” USDA Agricultural Handbook No. 18. U.S. Government Printing Office, Washington, DC. Soil Survey Laboratory Staff. (1992). “Methods Manual.” Soil Survey Investigative Report No. 42. Version 2.0. USDA-SCS, Lincoln, NE. Sokoloff, V. P. and Carter, G. F. (1952). Time and trace metals in archaeological sites. Science 116, 1-5.
Stein, J. K. (1987). Deposits for archaeologists. I n “Advances in Archaeological Method and Theory,” Vol 2, pp. 337-395. Academic Press, New York. Stein, J. K. (1992). Organic matter in archaeological contexts. In “Soils in Archaeology” (V. T. Holliday, Ed.), pp. 193-216. Smithsonian Institution Press, Washington, DC. Stiles, C. A , , Foss, J. E., and Lewis, R. J. (1994). Fractions of lead and copper in soils from Roman sites in Italy and Tunisia. In Environmental Contamination, 6th Int. Conference,” pp. 95-97. CEP, Edinburgh, UK. Timpson, M. E.. and Foss, J. E. (1993). The use of particle-size analysis as a tool in pedological investigations of archaeological sites. In “Proceedings of the First International Conference on Pedo-Archaeology” (J. E. Foss, M. E. Timpson, and M. W. Moms, Eds.), pp. 69-80. Special Pub. 93-03, University of Tennessee Agricultural Experiment Station, Knoxville. Unwin, D. (1975). An introduction to trend surface analysis. Geol. Absrr.1,40. Unterberger, R. R. (1992). Ground-penetrating radar finds disturbed earth over burials. In “Technical Proceedings of the 4th International Conference on Ground-Penetrating Radar, Rovaniemi, Finland.” (P. Hanninen and s. Autio, Eds.). pp. 359-365. Geological Survey of Finland Special Paper 16, Espoo, Finland. U.S. Department of Agriculture, Natural Resource Conservation Service. ( 1995). Selected resource data on the current status of the National Cooperative Soil survey. National Soil Survey Center, Lincoln, NE. [Unpublished] U.S. Department of Agriculture, Soil Conservation Service. (1975). “Soil Taxonomy: A Basic System of Soil Classification for Making and Interpreting Soil Surveys,” U.S. Dept. of Agric. Handbook 436. U.S. Govt. Printing Office, Washington, DC. Valentine, K. W. G., and Dalrymple, J. B. (1976). Quaternary buried paleosols: A critical review. Quaternary Res. 6, 209-222. Vink, A. P. A. (1975). Land use. In “Adv. Series in Agricultural Science,” No. 1. Spinger-Verlag, Berlin. Walker, K. J. (1992). The zooarchaeology of Charlotte Harbor’s prehistoric maritime adaptation: spatial and temporal perspectives. In “Culture and Environment in the Domain of the Calusa” (W. H. Marquardt, Ed.), pp. 265-366. Institute of Archaeology and Paleoenvironmental Studies, Monograph No. 1. University of Florida, Gainesville.
76
S. J. SCUDDER, J. E. FOSS, AND M. E. COLLINS
White, E. M. (1978). Cautionary note on soil phosphate data interpretation for archaeology. Am. Antiquity 43, 507-508. White, W. A. (1970). The geomorphology of the Florida peninsula. Geol. Bull. No. 51. Bureau of Geology, Florida Dept. of Natural Resources, Tallahassee. Widmer, R. J. (1988). Evolution of the Calusa-a Non-Agricultural Chiefdom on the Southwest Florida Coast. Univ. of Alabama Press. Working Group on the Origin and Nature of Paleosols. (197 1). Paleopedology. In “Nature and Dating of Paleosols” (D. H. Yaalon, Ed.). Israel University Press, Jerusalem.
PHOSPHATE ROCKSFOR DIRECT APPLICATION TO SOILS S. S. S. Rajan', J. H. Watkinsonl, and A. G. Sinclair2 IAgResearch, Ruakura Agricultural Research Center, 3 123, Hamilton, New Zealand LInvermay Agricultural Center, 50034, Mosgiel, New Zealand
I. Introduction 11. Reactivity of Phosphate Rocks
A. Definition of Reactiviry B. Measurement of Reactivity C. Mineralogy and Reactivity 111. Measurement of Phosphate Rock Dissolution in Soil A. Measurement in Acid Soils B. Measurement in Calcareous Soils IV. Factors Affecting Phosphate Rock Dissolution in Soil and Availability to Plants A. Factors Affecting Rate of P Release from Phosphate Rock Applied to Soil B. Factors Affecting Plant Availability of P from Dissolved Phosphate Rock V. Modeling the Rate of Phosphate Rock Dissolution in Field Soil A. Kirk and Nye Model B. Watkinson Model VI. Agronomic Effectiveness of Phosphate Rock A. Determining Agronomic Effectiveness B. Quantifying Comparative Performance of Phosphate Rocks C. Residual Effectiveness of Phosphate Rocks VII. Economics of Using Phosphate Rock Fertilizers VIII. Soil Testing Where Phosphate Rocks Are Used A. Current Research B. Future Research Needs IX. Amendments to Phosphate Rocks A. Composting with Organic Manures B. Phosphate Rock-Sulfur Assemblages C. Partially Acidulated Phosphate Rocks X. Concluding Remarks References
77 Adilnces in Agrnnmny, Vobine Y7
Copyright 0 1996 by Academic Press, Inc. All rights of reproduction in any form reserved
78
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
I. INTRODUCTION Interest in phosphate rocks (PRs) as direct application fertilizer stems from the facts that (i) per kilogram of P, PR is usually the cheapest fertilizer; (ii) direct application, with or without amendments, enables utilization of PRs which are unsuitable for manufacturing phosphoric acid and other soluble fertilizers such as triple (TSP) or single superphosphate (SSP); (iii) because PRs are natural minerals requiring minimum processing they are environmentally benign (Schultz, 1992); and (iv) PRs could be more efficient than soluble fertilizers in terms of recovery of phosphate by plants, even for short term crops in soils where soluble P is readily leached, as in sandy soils (Yeates, 1993) and possibly for long-term crops also in other soils (Rajan et al., 1994).
In spite of this PRs are not widely used as direct application fertilizers. The reasons are: (i) not all soils and cropping situations are suitable for use of PRs from different sources; (ii) the large number of factors controlling their dissolution in soil and availability to plants coupled with the inability to predict their agronomic effectiveness in a given soil climatic and crop situation; and (iii) their lower P content compared with high-analysis fertilizers which make PRs more expensive at the point of application if long-distance transportation is required. It has been more than 15 years since the last comprehensive review on PR for direct application was published by Khasawneh and Doll (1978). More recently Hammond er al. (1986b) reviewed the use of PRs and amended PRs in tropical soils. Since then considerable progress has been made in several areas of PR research. This includes a better understanding of the factors that affect PR dissolution, critical evaluations of the methods used to measure PR dissolution in soils, and its availability to plants and development of mechanistic models to predict the dissolution and availability to plants of PR-P. We found the literature on PR research rather overwhelming. In this chapter, instead of reviewing the numerous published reports, we will concentrate on the advances made on the fundamentals of PRs dissolution and their agronomic use, with specific examples. We have mostly quoted references published since 1978, although for the sake of continuity and comprehensiveness we have also cited some earlier publications. The philosophy behind this review will be that, paraphrasing Nye (1992), if we really understand the fate of PRs applied to soil and
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
79
develop mechanistic models describing their reactions we should be able to predict their effectiveness for any combination of soil, climate, crop, and management. This should enable decisions regarding PR use to be made without time-consuming and expensive field trials in every location.
11. REACTIVITY OF PHOSPHATE ROCKS
A. DEFINITION OF REACTIVITY There are two occasions when some measure of the agronomic performance of PRs would be desirable. One is the performance of a given PR when added to a given soil/plant system, and the other is the relative performance of a number of PRs when added to a soil suitable for using PRs. It seems preferable to restrict the usage of reactivity to the second occasion because it will then depend on only PR properties. The first would also require a knowledge of whether the soil/plant system would be suitable for even the most reactive PR. Although Khasawneh and Doll (1978) indicated that the reactivity of a PR is related to agronomic effectiveness, they did not define the term explicitly. Rather it was discussed in relation to measured PR properties, such as the relative amount dissolved in a particular organic acid solution. More recently Sinclair et al. (1992) described reactivity as the ability of a PR to release P, or the rate of release of P to soil and plant; Rajan et al. (1 992) defined it as the magnitude and rate of dissolution of a PR. Building on these ideas, it is proposed that Reactivity is the combination of PR properties that determines the rate of dissolution of the PR in a given soil under given field conditions. Reactivity is defined as a property of the PR, and deliberately excludes properties of soils and plants. It is a direct measure of the amount of PR that dissolves in a given time, and hence is related to agronomic effectiveness. Sometimes there is apparently no relation because it is masked by other factors (see Section 1V.B). Also, the relation is discontinuous since the relative agronomic effectiveness (RAE) is not increased by a decrease in size below about 0.15 mm (Khasawneh and Doll, 1978), even though the amount dissolved will still continue to increase with decreasing size (Kanabo and Gilkes, 1988a,b,c,d). Reactivity is also defined in terms of the dissolution in a soil in the field, not in the laboratory or glasshouse, i.e., under the most appropriate conditions for providing information on the use of PR as a fertilizer. Reactivity as defined is a kinetic property, which is also consistent with the use of PR as a fertilizer in the open system of a field soil. Any method of measurement must therefore avoid the establishment of concentrations approaching a
80
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
saturated solution, or conditions approaching a quasi-equilibrium. Such equilibrium values would exclude purely kinetic properties, particularly surface area, which is an important factor in the dissolution rate modeling of RPRs (Kirk and Nye, 1986a,b; Watkinson, 1994a). The definition of reactivity should not be based on methods, as at present (Khasawneh and Doll, 1978), such as the amount extracted by a solution of an organic acid. The emphasis should be on dissolution in field soil, and any method can only give an estimate of this. Relative reactivities could be measured from the amount of PR remaining in a field soil at a set time after application. The relative amounts dissolved would give an indication of their relative effectiveness as P fertilizers. However, such a method would require appreciable time and resources, and so a laboratory method that provides an acceptable estimate would be desirable. As will be shown later, the PR reactivity is the combination of several properties measureable in the laboratory, including particle size. As soils become more acidic (or plants more acidifying), differences in PR reactivity become less important; conversely, as soils become more neutral, amounts dissolved are less and differences in reactivity become more important (Khasawneh and Doll, 1978). Consequently it is important to choose a soil for measurement of dissolution that gives a good range of amounts dissolved from a range of PRs. The above definition of reactivity is useful in that it is precise and the property can be measured specifically, i.e., by measuring residual PR in a soil (Section Ill), or calculated from relevant PR properties using mechanistic dissolution rate models (Section V).
B. MEASUREMENT OF REACTMTY Several types of measurement have been proposed, more particularly: dissolution in soil; dissolution in acid or salt solutions; measurement of crystal unit cell dimensions; and calculation of dissolution in soil using parameters from mechanistic dissolution models. Since rapid results for reactivity measurements are required, only aqueous extractions in the laboratory have been used routinely (Chien, 1978, 1993; Rajan er al., 1992; Sinclair et al., 1992), except for a recent laboratory test based on fundamental PR properties from dissolution rate model parameters (Watkinson, 1994~). 1. Principles
Most existing methods are purely empirical (although dilute organic acids have been used to simulate the action of root exudates), largely because the concept of
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
8I
reactivity has not been defined explicitly. Without an exact definition, methods based on precise principles are not possible. Generally, reactivity is equated with the solubility of the PR (Khasawneh and Doll, 1978; Chien, 1993), but modified by other factors, notably free carbonate content, crystal and particle size and porosity, and intermixture with silica (Chien, 1993). However, the term solubility itself is used in two senses: in the usual thermodynamic sense of the equilibrium concentration (Chien, 1993), but mostly as the rate of solution into one of several extractants (Chien, 1993). Furthermore, different extractants give different results. For example, Watkinson ( 1 9 9 4 ~ pointed ) out that citric acid apparently dissolves more fluorapatite, but less francolite than formic acid (Rajan et al., 1992; Chien, 1993) (Fig. 1). Largely because of these difficulties Sinclair et al. (1992) put in a plea for an approach to reactivity measurement based on fundamental properties of the PR. Watkinson (1994~)attempted such an approach incorporating the above definition of reactivity, a dissolution rate model (Watkinson, 1994a), and fundamental properties of the PR measureable in the laboratory.
2. Methods a. Empirical Since the review by Khasawneh and Doll (1978), no new chemical methods seem to have been proposed, only improvements to existing methods (Chien, 1993). Those most commonly used are heated neutral ammonium citrate, 2% citric acid, and 2% formic acid (Chien, 1993). Less common methods include absolute citrate solubility and acid ammonium citrate (Chien, 1993). To overcome the problem of appreciable impurity in the PR, adding a constant mass of P Formic-P Citric-P
a
-
4:
CAs FAs
~
0
A
0
A
10 20 30 40 50 PR dissolved in one year, DRF (“YO)
0
Figure 1 Relationship between citric-P and formic-P for carbonate- (CAS) and fluor- (FAS) apatites as shown plotted against dissolution rate function (DRF) (Rajan ef a / . . 1992; Watkinson, 1995).
82
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
or apatite to the extracting solution has been proposed (Axelrod and Gredinger, 1979). The preferential reaction of free carbonate with the extractant was overcome by measuring the amount of PR dissolved in the second rather than the first extract (Chien and Hammond, 1978; Mackay et al., 1984b; Mishra ef al., 1985). Rajan et al. (1992) recognized the importance of the geometric surface area in rate of solution methods, and adjusted the shaking time to include an effect of initial area or particle size. In contrast, Khasawneh and Doll (1978) treated geometric area (i.e., particle size) as being of lesser importance than the much larger specific area, which includes internal surfaces, but without supporting evidence. (Later discussion, Section V, will show that on the diffusion-controlled model, internal surface area is of small importance compared with geometric area.) b. Theoretical Olsen (1975) found that the dissolution kinetics of PRs into EDTA solution, the rate of which was increased through the chelation of the dissolved Ca2+, was consistent with a second-order reaction rate. However, he did not further test the model or use fundamental PR properties as variables. Watkinson (1995) proposed a Dissolution Rate Function (DRF), using fundamental properties of RPRs measurable in the laboratory, for estimating the reactivity of PRs. The DRF represents the amount of PR dissolved in a standard soil in a given time. It was derived from a simple rate equation in a dissolution rate model (Watkinson, 1994a,b), and the standard soil represented the properties of an average New Zealand pastoral soil used for direct application of PRs. In this soil 30% of Sechura (Peruvian) PR (also referred to as Bayovar PR) of particle size 0.075-0.15 mm dissolved in 1 year. For a PR with particles of the same diameter, do, the amount dissolved in time, t , the DRF was given by (Watkinson, 1994c) DRF
=
1 - [ I - 8D,t(C,/F)/(pdo2)]”2,
(1)
Where D, is the mean diffusion coefficient for phosphate in soil (the value for the standard soil mentioned above is 0.5 cm2 year-’), C , is the phosphate concentration at the PR surface (strictly C , - C s , where Cs is the phosphate concentration in the bulk soil, but C , Cs), p is the PR particle density, approximately 3.2 g cmP3, and F is the fractional P content of the PR. For convenience, the time, t , was 1 year, while values for do and F were measured using standard methods. The value for C , was measured as the equilibrium phosphate concentration of the PR in a simulated soil solution of constant pH (held at pH 5.5 using an automatic titrator) and set initial values of calcium (0.5 mM) and ionic strength ( 5 mM) (Watkinson, 1994~).This solution took into account the calcium from the soil and that dissolved from the PR and the calcite impurity in the PR at pH 5.5. Congruent dissolution was assumed because of the very low solubility of PRs (Kirk and Nye, 1986a), and the large ratio of solution to solid of 500: 1 used (Watkinson, 1994~).
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
83
The DRF can therefore compare a PR of high solubility and large size with a contrasting one of low solubility and small size. A large pH buffering capacity in the soil was assumed in that all the alkali generated by the dissolving lattice phosphate and carbonate was neutralized by the automatic titrator in holding the pH constant. This is generally true for New Zealand pastoral soils (Edmeades er al. 1985). The DRF for a fertilizer mixture with sieve analysis of particle sizes resulting in n successive sieve fractions, each of size range bi+, to bi and of weight n
wi = I , is given by,
fraction wi, where i= I
n
DRF = 1 -
wJ(2
+ at/b?)(b: -
- (2
+ at/b:+,)(b?-?_,- at)”2
i= I
+ ( 3 ~ Z ) ( s i n - l ( f i / b ~-) s i n - 1 ( f i t / b ~ - ~ ) ) 1 / 2 (-b ~bi-,),
(2)
where a = 8 D, ( C , / F ) / p , and after 1 year ( t = 1) the smallest particle, b,, has not dissolved. If the smallest particles have dissolved, additional sets of similar equations are required (Watkinson, 1994c). The DRF for 1 1 PRs, ground and unground, correlated with published values of relative response of ryegrass as the test plant in three soils (Rajan et al., 1992) (Fig. 2) at least as well as acid-extractable P using citric and formic acids (Watkinson, 1994c, 1995). A comparison of the ground and unground PRs in a plot of RAE against the solubility function, C,/F, i.e., only size excluded, showed the effect of grinding (particle size) on RAE (Fig. 3). The difference between ground and unground PRs increased with increasing solubility (Watkinson, 1994~).Conversely there was a neglible difference at very low solubility, which is consistent with evidence cited earlier that it was not possible to convert an unreactive PR into a reactive PR by grinding it to a very small size. These data 1201
0
I
I
1
I
10 20 30 40 Lab test, DRF (“7)
1
50
Figure 2 Relative response of unground (UG)and ground (G) PRs (ground Sechura PR = 100) (Rajan ct a / . , 1992). in relation to the Dissolution Rate Function (DRF) (Watkinson, 1995).
84
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
? = 0.87 Ground
1201
:
n
0.91 Unground
r I
I
I
I
I
1
10 20 30 40 50 60 RPR solubility, CR/F (mg L-l)
figure 3 The effect of grinding on the relative response values of PRs (0,unground, and 0, ground) (Rajan et a/.. 1992) having different levels of solubility, CRIF, at pH 5.5 (Ca = 0.5 mmol liter- I ) (Watkinson, 1995).
indicate that DRF was applicable to fluorapatites with and without carbonate substitution and of different size distributions. The effect on dissolution rate of grindinig, in three stages, a PR with size distribution typical of North African PRs has been calculated from the model (Watkinson, 1994b), and is shown in Fig. 4. The soil properties are such that 30% of Sechura PR would be dissolved in the first year. The economics of grinding could be estimated from such data (Sinclair et al., 1990a,b).
C. MINERALOGY AND REACTIVITY 1. Fluorapatite Fluorapatite (FA) is much less soluble than hydroxyapatite (HA) or even the carbonate substituted fluorapatites (CAs) (Khasawneh and Doll, 1978). This very 1001
e l m m (unground) \
8.25
---I
0
2 4 6 8 1 Dissolution time (years)
0
Figure 4 Predicted effect of grinding a PR to different sizes on its dissolution rate in a soil dissolving 30% of Sechura PR in the first year (Watkinson, 1994b).
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
85
low solubility precludes its use as a direct application fertilizer, and therefore it is classed as unreactive. It was not possible to convert an unreactive to a reactive PR, even by ultrafine grinding to a size c 0 . 0 2 mm (Khasawneh and Doll, 1978).
2. Carbonate Apatites (Sedimentary) Sedimentary fluorapatites in which carbonate substitutes for phosphate in the apatite lattice form two distinct series of phosphate rocks on the basis of their physical and chemical properties (McClellan and Van Kauwenbergh, 1992). The most common are those with an excess of fluorine over that in FAP. In this case carbonate plus fluoride together substitute for phosphate to preserve the charge balance within the lattice. The second series comprise those PRs in which phosphate is substituted by carbonate plus hydroxide and/or fluoride leading to a deficit of fluorine over that in FAP. a. Excess Fluorine The properties of those PRs with excess fluorine are controlled by the extent of carbonate substitution (McClellan and Van Kauwenbergh, 1992). The planar carbonate substitution for tetrahedral phosphate makes the lattice less stable, resulting in increasing solubility with increasing carbonate content (Khasawneh and Doll, 1978). Increasing carbonate also decreases the unit cell a-dimension and increases the solubility, and therefore the reactivity compared to FA, all other things being equal (Section 1V.A). b. Deficit Fluorine Where there is a deficit of fluorine, the combinations of substitutions are more complex (McClellan and Van Kauwenbergh, 1992), and the correlations with other properties more diffuse. The unit cell a-dimension is controlled more by the fluorine than the carbonate content, and decreases with increasing fluorine. In contrast to the excess fluorine series, the solubility decreases with decreasing unit cell size and increasing fluorine, the latter in line with the decreasing hydroxyl. The fluorine content ranges from zero (equivalent to hydroxyapatite) to that in FAP, so as a class they are generally more soluble and akin to hydroxyapatite, and more reactive than the excess fluorine class. c. Carbonate Impurity Calcite and, less commonly, dolomite impurities are often present as discrete minerals (McClellan and Van Kauwenbergh, 1992). Both dissolve to completion because the product is evolved as carbon dioxide increasing the local soil pH and calcium. Calcite dissolves much more rapidly (Sverdrup and Bjerle, 1982; Watkinson and Kear, unpublished data). These effects lower the amount of PR dissolved, and therefore the apparent reactivity. For PRs with excess fluorine, lattice carbonate is usually inversely related to calcite impurity (Watkinson and
86
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
Kear, unpublished data), so that these two effects also accentuate differences in reactivity.
111. MEASUREMENT OF PHOSPHATE ROCK DISSOLUTION IN SOIL PR for direct application has been advocated almost exclusively for noncalcareous acid soils. Consequently the techniques developed to measure dissolved PR have notably been for use in acid soils, although the inorganic fractionation scheme of Baifan and Yichu (1989) may pave the way for developing methods for calcareous soils. The extent of PR dissolution, and therefore the release of PR-P, can be measured (i) directly by determining the PR remaining in soil and (ii) indirectly by determining the reaction products released, P and Ca. A third category of methods measure apparently a constant fraction of PR dissolved. Under this category fall NaHCO, (pH 8.5) extractable P, anion resin (with or without cation resin) extractable P and isotopically exchangeable P which is measured either at a given interval after PR application to soil or at intervals. The third category is for estimating plant available P in soils and not for measuring PR dissolution per se. For that reason these methods are discussed in Section VIII. The direct measurement of residual PR is applicable in all circumstances, including field, greenhouse, and incubation studies. The indirect methods are suitable only for closed incubation systems, where the reaction products are not removed from the soil. They are not suitable for field or greenhouse studies unless the amounts of P or Ca removed from the soil by plant and microbial uptake and/or leaching are also measured. The prerequisites for any method used for estimating PR dissolution are that PR should not dissolve during preextraction and, in the case of indirect methods, the extractant should remove all of the reaction product(s) (Bolan and Hedley, 1989).
A. MEA~URFMENT IN ACIDSOILS 1. Measurement of Phosphate Rock Remaining by Inorganic P Fractionation Methods used for measuring the amount of PR remaining in soil are based on the inorganic P fractionation procedure of Chang and Jackson (1957) and later modifications (Petersen and Corey, 1966; Williams el al., 1967; Syers et al., 1972). The fractionation procedure was originally developed to characterize the
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
87
distribution of soil phosphate in various chemical forms, but has been subsequently applied to measure the dissolution products of P fertilizers and the residual PR to calculate the extent of PR dissolution (Chu et al., 1962; Shinde et al., 1978; Chaudhary and Mishra, 1980; Grigg, 1980a,b; Rajan, 1983, 1987a,b; Chien et al., 1987b; Bolan and Hedley, 1989; Perrott, 1992; Perrott et al., 1992; Rajan and Watkinson, 1992; Tambunan et al., 1993; Perrott and Kerr, 1994). Briefly, the method generally adopted to measure PR dissolution consists of prewashing soil (30 min) with NaCl or BaCI, solution buffered to more than pH 7.8 to remove soil-exchangeable Ca, followed by extraction (17 h) with NaOH (0.5- 1 M ) to extract nonoccluded Fe-P and AI-P, and then with an acid (HCI or H,SO,, 0.5-1 M )solution (4 or 17 h for HISO,) to extract Ca-P. Although some modified methods included Chang and Jackson’s technique of citrate dithionate extraction prior to acid extraction (Peterson and Corey, 1966; Williams et a l . , 1967; Syers et al., 1972), omission of this step has not been found to affect the amount of acid extractable P (Rajan, 1983). The amount of PR remaining is calculated from the increase in the Ca-P fraction (acid extractable P) of the PRtreated soil over that of the untreated control. In applying the procedure it is considered that (i) the P in PR is present as calcium apatite, (ii) the apatite P is not soluble in NaOH but is dissolved by HCI and HISO, (Williams, 1937), and (iii) the P dissolved in soil is transformed into AI-P and Fe-P. In the fractionation procedure a prewash with NaCl or similar electrolyte is necessary to remove soil exchangeable Ca which, if present, could result in precipitation of calcium phosphate during NaOH extraction (Syers et al., 1972). Dissolution of the calcium phosphate in the subsequent acid extract overestimates the Ca-P fraction and therefore the PR present (Perrott , 1992). Hughes and Gilkes (1984) reported that a prewash with unbuffered solutions of NaCI, KCI, or NH,CI of soil, incubated with PRs for a week resulted in the release of exchangeable acidity from soil and therefore dissolution of apatite. They concluded that the extracting solution pH should remain above 7.3 to prevent PR dissolution. These authors recommended prewashing with Bascomb solution (Bascomb, 1964) which consists of 2 M BaCI, solution buffered with triethanolamine (TEA) at pH 8.1. Perrott and Ken (1994), using both soil/PR mixtures and field soils collected 8 months after surface application of PRs, found that prewashing mineral soils with 1 M NaCl resulted in significantly less recovery of PR (70-95%) than when using NaCl solution buffered with 0.1 M H,BO, and NaOH to pH 7.8. The loss was significantly related to soil pH(H,O) (Fig. 5A) but not to the exchangeable acidity (Fig. 5B). H,BO,/NaOH buffers are preferred over TEA because they do not interfere with the molybdenum blue/ascorbic acid method used for P analysis and also are easier to prepare. To improve Ca extraction, Perrott and Kerr (1994) added EDTA to the buffered solution in amounts equivalent to the concentration of exchangeable Ca. Rajan (personal communication) determined the residual PR in a volcanic ash soil
88
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
0
"V
-
0
10
20
Titratable acidity (Cmol kg-l) figure 5 Effect of (A) soil pH(H20), and ( B ) titratable acidity on recovery of Sechura PR-P from soil-PR mixture using NaCl prewash. Recovery values are expressed as percentages of those using buffered NaCl prewash (Perrott and Kerr, 1994).
(Typic vitrandept) collected 3 months after surface application of North Carolina PR. He found that soil prewash with unbuffered NaCl solution gave only 8% less acid extractable P than a prewash with EDTA buffer (Fig. 6). It is possible that a higher soil pH of 5.8 resulted in a smaller loss in acid extractable P in the unbuffered solution. Most authors have used H2S04 to extract Ca-P although Williams et al., (1967) and Syers er al. (1972) have used HCI. Tambunan et al. (1993) reported that, for reasons they could not explain, the recovery of PR residues using up to 4 M HCI solution was less than that when 0.5 or I M H2S04 was used and thus H,S04 is a preferred option. While the sequential extraction procedure to determine the extent of PR dissolution has generally been satisfactory, it is too lengthy for routine testing. They also require a sample of unfertilized soil to enable the calculation of the remaining PR. Perrott and Wise ( 1 995) proposed a simpler procedure to determine PR residues remaining in soil. A mild acid extraction (acetate buffer) was used to differentiate between native soil fluorapatite-P and PR-P. The procedure consists of extracting P from two subsamples as follows: the first subsample is shaken with an acetate solution of pH 4 for a specified time after which a NaOH/citrate
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS .c0
89
Y=0.08 + 1.08 X, R2=0.963
$ , -ool
$
300 500 700 Acid-P, NaCl prewash (rng kg-l soil)
100
Figure 6 Acid-extractable P of soil samples collected from Sechura PR treated plots prewashed with either NaCl or buffered EDTA solution (soil pH 5.75) (S.S.S. Rajan, unpublished data, 1992).
solution is added. The suspension is shaken for a further specified time and centrifuged, and P is determined in the supernatant solution. A second subsample is similarly treated except that the intial shaking is with a borate solution of pH 8. The PR-P is calculated by subtracting the amount of P extracted in the second subsample from that of the first subsample. Their results averaged over 1 1 different soils from New Zealand showed a very significant (0.1% level) variation of PR-P recovery with rock reactivity (Table I). The recovery of PR-P from Sechura and North Carolina PRs was complete, which illustrates the usefulness of this method with highly reactive PRs. The recovery with medium reactive (and also the unreactive PR) PR was less than that present in the soil. Soil types had no significant influence on the recovery of
Table I Percentage of PR-P Recovered from PR-Soil Mixtures (Values Averaged across 11 Soils) Phosphate rock
Recovery (96 of added PR)
Sechura North Carolina Arad Egyptian Florida SED
93.7 101.0 80.9 77.3 37.5 4.8***
90
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
the two most reactive PRs, although when the recovery values were averaged across the five PRs, soils did have a significant effect (1 % level).
2. Measurement of Phosphate Rock Dissolved from ANaOH-P Measurement of PR dissolution in soil from the increase in the amount of NaOH-extractable P determined after prewashing soil to remove exchangeable Ca was proposed by Mackay et al. (1986). The underlying principle behind this method is that P released from PR largely forms complexes with soil Fe and A1 which are extracted by NaOH solution (usually 0.5 M . Since apatite P does not dissolve to any significant extent in this reagent (Williams, 1937) the method provides a direct estimate of the PR dissolved in soil. It has the advantage of needing only two instead of three extractions as in the fractionation procedure. Bolan and Hedley (1989) concluded that this method is suitable for use in incubation studies where the reaction products are not removed and where there is no significant active net mineralization or immobilization of P. This method has been used extensively in short-term incubation studies. The usefulness of this method has not been investigated for long term incubation studies where the sorbed P may be converted from NaOH extractable to occluded forms of P (Hagin et al., 1990).
3. Measurement of Phosphate Rock Dissolved from ACa PR dissolution has also been estimated from the increase in the exchangeable Ca (ACa) content of the PR treated soil over that of the control. In the ACa method it is assumed that the Ca released on dissolution of PR accumulates in the soil as exchangeable Ca which is extracted with appropriate electrolyte solutions. Khasawneh (quoted in Khasawneh and Doll, 1978) used neutral 1 M NH,OAC to determine the increase in exchangeable Ca in the PR-treated soil over the control soil. Smyth and Sanchez (1982) measured PR dissolution using a 1 M KCl solution. As pointed out in a previous section, use of unbuffered solutions could dissolve PR during the extraction because of the release of exchangeable acidity from the soil (Hughes and Gilkes, 1984). These authors therefore advocated the use of BaCI, solution which has been buffered to an alkaline pH. Bolan and Hedley (1989b) reported poorer recovery of Ca by NH,OAC and the recovery decreased as the pH increased. The ACa method is the simplest of the three techniques. However, it is not suitable for use in greenhouse, field, or open incubation studies where the Ca released is removed by plants or by leaching. In closed incubation studies also, overestimation of PR dissolution will result where the PR contains appreciable amounts of free CaCO, because of its preferential dissolution and hence increase in the measured exchangeable Ca.
PHOSPHATE ROCKS FOR DIRECT APPLICATION T O SOILS
91
B. MEASUREMENT IN CALCAREOUS SOILS Although not commonly found, there are situations where PR application to calcareous soils could be plant effective (Edwards, 1956; Singaram et al., 1995). Measurement of PR dissolution in calcareous soils is made complicated by the dissolved P forming not only Fe-P and Al-P but also Ca-P (Holford et al., 1975; Hooker et al., 1980). Unlike methods for noncalcareous soils where Ca-P is treated as one component, methods aimed at measuring PR remaining in calcareous soils should distinguish between the calcium apatite applied as PR and the Ca-P formed after reaction with soil components. Reports indicate that at relatively low solution P concentrations (< 10 mg liter-'), which is far higher than would exist at the PR/soil interface, the Ca-P is probably present as adsorbed P on CaCO, and as Ca-P complexes formed with the exchangeable soil Ca (see Sample et al., 1980). Baifan and Yichu (1989) proposed an inorganic P fractionation scheme which included separation of dicalcium, octacalcium and apatite P. They used sequential extractions with NaHCO, (pH 8.5), NH,Ac (pH 7.0), NaOH plus Na,CO,, and H,SO,. Fractionation schemes similar to the above may be appropriate for PR measurement in calcareous soils but are yet to be investigated.
rV. FACTORS AFFECTING PHOSPHATE ROCK DISSOLUTION IN SOIL AND AVAILABILITY TO PLANTS Several factors affect the rate of PR dissolution in soil and its availability to plants. The availability of PR-P to plants largely depends on its rate of dissolution. However, this is not always so because of the influence of soil characteristics, the plant and fertilizer management factors. This section reviews various factors influencing the rate of PR-P release and its availability to plants.
A. FACTORS AFFECTING RUE OF P RELEASEFROM PHOSPHATE ROCKAPPLIED TO SOIL 1. PR Properties
Two important properties determining the rate of PR dissolution in a given soil are chemical composition, which includes apatite lattice composition and the type of accessory materials, and particle size (Section 11).
92
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
PR deposits fall into three broad classes based upon their mineral assemblages in order of their increasing economic importance. These are Fe-A1 phosphates, Ca-Fe-Al phosphates, and Ca phosphates (McClellan and Gremillion, 1980). Commercial mineral Ca phosphates belong to the group of apatite minerals which are similar in crystal structure to fluorapatites but vary significantly in chemical composition. The Ca apatites of sedimentary origin have generally been found to be suitable for direct application as phosphate fertilizers. It has been well established that increasing substitution of C032- for PO,3- in the lattice structure increases the solubility of carbonate apatites. This occurs through decreased unit cell a-dimension and crystal instability on increased incorporation of planar CO,2- and F- for PO,,- tetrahedra (Lehr and McClellan, 1972; Chien, 1977). Unit cell a-dimensions in turn have been found to be closely correlated with the chemical extractability of P from PRs (Figs. 7A-7C) (Dash et al., 1988; McClellan and van Kauwenberg, 1992). For details on the chemistry of isomorphic substitution and its effect on crystallite properties the readers are referred to the review by Khasawneh and Doll (1978). There is a scarcity of studies relating directly measured dissolution of PRs in soil with the chemical composition of the PRs, under conditions where the reactant products were removed from the PR-soil interface. However, numerous reports have been published measuring PR dissolution indirectly as plant P uptake. Chien et a/. ( 1987b) determined residual PR remaining in a Columbian Oxisol 5 years after application. Six PRs (carbonate apatites) of varying citrate solubility were applied and the inorganic P fractionation of Chang and Jackson (1957) was employed to determine the PR remaining. The apatite dissolved, calculated as a difference between the PR applied and that remaining, ranged from 79 to 98% of that applied and there was a positive correlation between the PR dissolved and the citrate-soluble P of the PRs (Fig. 8). Rajan (1987b) reported that in 1 year after surface application to permanent pastures 27% of Florida PR (low carbonate substitution) dissolved compared with 42% for a North Carolina PR (highly carbonate substituted). Several publications provide evidence of a close positive correlation between increasing carbonate substitution in the lattice structure, determined by direct physical and chemical measurements or as indicated by chemical extractable P, and PR dissolution as measured by crop P uptake (Mackay et al., 1984a,b; Anderson et al., 1985; Leon et al., 1986; Dash et al., 1988; Rajan et al., 1992). Calcium carbonate is the most abundant accessory mineral in PRs. Because CaCO, is more soluble than the most chemically reactive apatites (Silverman et al., 1951) and since its dissolution increases the Ca concentration and pH at the apatite surface it is not surprising that accessory CaCO, can reduce the rate of PR dissolution in some soils (Anderson et al., 1985; Robinson et al., 1992a). How-
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
93
0 1
141
2{ 0
9.32
B
,
,
9.34
,
,
,o,
9.36
9.38
Unit-ceIP Figure 7 Relationship between unit-cell a dimension of apatite sample and solubility of P in (A) neutral ammonium citrate. (B) citric acid, and (C)formic acid (McClellan and Van Kauwenbergh, 1992).
ever, under field conditions where Ca may be removed by plant uptake and or leaching this effect will be minimized. Because PRs are relatively insoluble materials their geometric surface area
94
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
0
0,5 1.0 1.5 2.0 2.5 3.0 Citrate-soluble P (“Aof PR)
Hgure 8 Percentage of PR-P dissolved 5 years after application as related to their solubility in ammonium citrate (Chien er a / ., 1987).
will have an important bearing on their rate of dissolution in soil (Section 11). Thus the finer the particle size, the greater the degree of contact between PR and soil and therefore the greater the rate of PR dissolution, provided the PR application rate is such that the zones of PR dissolution between the particles do not overlap. The results of Kanabo and Gilkes ( 1 988c) support the above reasoning. These authors conducted a laboratory incubation study in a lateritic podzolic soil to estimate the dissolution of North Carolina PR ground to four size fractions ranging from 0.15-0.25 mm to <0.04 mm. Their results show that PR dissolution increased with decreasing particle size, down to the smallest size range used. Indirect evidence of greater dissolution of PRs after fine grinding has been presented by researchers working on permanent pastures. Sinclair et al. (1993a) compared unground Sechura and North Carolina PRs with TSP on ryegrass/ clover pastures at four sites over a 6-year period. Soil pH values were less than 6 and the fertilizers were surface-applied annually. Pasture dry matter production results showed that the performances of both PRs were inferior initially and became similar to TSP in Year 6 (Sinclair er a1 ., 1993a). On the other hand reactive PRs, including Sechura and North Carolina, when applied after grinding either in powder or pelletized form, were found to be generally as effective as SSP from the year of application (Mackay et al., 1984a; Rajan, 1987; Rajan et al., 1987; Gregg el al., 1988).
2. Soil Properties a. Soil pH and Titratable Acidity Dissolution of PR may be expressed by the simple equation Ca,,(P0J6F,
+ 12H20+ 10CaZ++ 6H2P0,- + 2F- + 120H-.
(3)
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
95
Although the above reaction is for fluorapatite it applies to other members of the apatite minerals including reactive PRs (francolite). As indicated in Eq. (3), hydrolysis of the dissolving PO$ releases OH- into solution. Neutralization by soil acidity of the OH- released enables continuation of the PR dissolution process. In the case of PRs with isomorphically substituted ions, hydrogen ions may also be needed to neutralize OH- from Cog- hydrolysis and structural OHof PR released into solution (Chien, 1977). At equilibrium, [Ca2+][H2P0g]6 [F-I2 - -Ksp [HI l2 Kw'2
'
(4)
Thus with increasing proton supply one would expect greater dissolution of PR provided the reaction products are removed. A comprehensive summary of the relative influence of the various soil and plant factors presented by Kirk and Nye (1986b) shows that PR dissolution is very sensitive to changes in soil pH (Figs. 9A,B). Indeed early studies established the positive effect of soil acidity on PR dissolution (Ellis et al., 1955; Peaslee et al., 1962; Barnes and Kamprath, 1975). In these studies, however, the pH of soil or plant-growing mediums was adjusted by applying calcium compounds and thus no distinction could be made between the effect of proton concentration per se and that of Ca ions. Khasawneh (1977) (quoted in Khasawneh and Doll, 1978) separated the effect of these two parameters by adjusting soil pH either with CaCO, or SrCO,. The agronomic effectiveness of corn grown in pots was used as an index of the solubility of North Carolina PR. They reported an additional decrease (due to Ca) in the effectiveness of the PR with increasing pH in soils where the pH was adjusted with CaCO,. Kanabo and Gilkes (1987~)investigated the influence of soil pH, as measured in 0.01 M CaCI,, on the dissolution of North Carolina PR in a lateritic soil. The pH was adjusted by incubating soil samples with water alone, HCl, or solid SrCO,, to give a pH range of 3.73-6.83. The increase in exchangeable Ca (ACa method) was used as an index of the amount of PR dissolved. They found that the PR dissolved (ACa) correlated with increasing soil pH: log ACa = a - bpH. Bolan and Hedley (1 990) studied the effect of soil pH on the dissolution of three different PRs: highly reactive North Carolina, medium reactive Jordan, and unreactive Nauru PRs. They used a volcanic ash soil, the pH of which was adjusted to give a range from pH(H,O) 3.9 to 6.5 by treating either with dilute HCI or NaOH. The extent of PR dissolution, after 84 days incubation, was determined from the increase in the amount of 0.5 M NaOH extractable P in the PR treated soil over the control soil. The amount of PR dissolved, expressed either as increase in NaOH extractable P or as a proportion of that added, was
96
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
r"
0 '
S.V.
Figure 9 The effects of the most important variables on the fraction of PR (A) dissolved, and (B) that has been taken up by roots after 80 days. Standard values as in Kirk and Nye (1986b). The ranges in values are: pH (soil pH), 4.25-5.25; AH(root acidity), O-lO-y mol dm-, soil s-1; pCa (Caz+ activity), 2-4; b,, (pH buffer capacity), 0.035-0.14 mol dm-3 soil pH-1; L, (root density), 100-1500 dm d w 3 ; b, (phosphate buffer capacity as Freundlich a), 0.07-0.28 mol dm-3 soil; N (application rate), 0.08-0.32 kg P m - j soil; a, (particle size), 0.05-0.2 mm;AHCO, (bicarbonate in soil solution), 0-5 X I O - ' O mol dm-2 soil s-I; p, degree of carbonate substitation in the PR (Kirk and Nye, 1986b).
found to be almost linearly related to soil pH (Fig. 10). In addition to enhancing neutralization of the OH- released, an increase in soil acidity can increase the P adsorptive capacity of soils with pH dependent charges This can also increase PR dissolution by removing P released from the PR (Bolan and Hedley, 1990). Rajan et al.(1991b) studied the effect of soil pH, adjusted by applying either HCI or Ca(OH),, on the dissolution of Sechura PR under field conditions. The amount of PR-P dissolved, expressed as a fraction of that applied, decreased either exponentially or linearly with increasing soil pH. Soil pH is an intensity measurement and gives the instantaneous concentration of H + in the soil solution. A measurement of the ability of soil to supply H + or to remove OH- from the soil solution is the pH buffer capacity of the soil or
--
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
.-
A
5
-
97
B
'0°1
loool
Y
p, 800
E
v
Q
lu
+d
0
e
H
I
9 z
400-
200-
,,
.
+
A
a
20-
r u!
0
O1
3.5
4.5
5.5
t
6.5
1
KH2P04 NCPR A Jordan 0 Nauru A Control
3.5
4.5
5.5
I
6.5
Soil pH (1 : 2.5 soil : water) Figure 10 Effect of soil pH on the dissolution of PRs in a New Zealand soil (Typic Dystrandept) after 12 weeks of incubation (Bolan and Hedley, 1990).
titratable acidity. Soil buffer capacity has been estimated as the differential of the soil acidity neutralization curves, &equivalent of OH added)ld(pH), after equilibration of soil for 48 h with Ca(OH), (Nye and Tinker, 1977). The pH is measured in 0.01 M CaCI, medium or water (Anderson et al., 1985; Anderson and Sale, 1993). Some researchers defined buffer capacity as the amount of OHrequired per unit of soil to raise its pH from the initial pH to an arbitrarily chosen pH of 6 (Kanabo and Gilkes, 1987d). The pH was measured in 0.1 M KCl medium. It is noteworthy that the relationship between pH and acid or alkali added is nearly linear in the pH range of most agricultural soils (pH 4.5-6.5) (Magdoff and Bartlett, 1985). Anderson et al. (1985) studied the influence of some chemical characteristics of 18 soils on the P release from four PR materials in a glasshouse experiment. They concluded that no single soil characteristic (Solution P, Solution Ca, pH, pH buffer capacity) had a consistent and predominant effect on P release. However, they found that the pH buffer capacity (Nye and Tinker, 1977) was nearly twice as important as any other soil parameter in the case of Huila PR containing 7.9% of free carbonate, but not with Sechura PR which contained a negligible amount . In Western Australian soils a linear relationship was reported between initial soil pH and titratable acidity (Kanabo and Gilkes, 1987~).Under such conditions
98
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
initial soil pH alone was a good predictor of PR dissolution. However, when a large number of soils (228 soils) of highly variable properties was used, stepwise regression indicated that titratable acidity accounted for only 35% of the variance (Hughes and Gilkes, 1994). It is noteworthy that the sensitivity studies of Kirk and Nye (1986b) indicate only a small influence of pH buffer capacity on PR dissolution (Fig. 9). b. Calcium in Soil Solution and Soil Exchangeable Calcium Following the law of mass action, the dissolution of PR [Eq. (311 is favored as long as the Ca concentration in soil solution is maintained at a lower level than that in the film surrounding the dissolving PR particle, and provided the ionic products of the reactants do not exceed the solubility product. A sensitivity analysis of the Kirk and Nye model (Kirk and Nye, 1986b) shows that the rate of PR dissolution is highly sensitive to Ca2+ activity in solution (Fig. 9). Wilson and Ellis (1984) studied in detail the influence of solution Ca activity on the dissolution of six PRs of a range of reactivity. They reported a linear relationship between the log of Ca ion activity and log P in soil solution. Similar results have also been reported by Robinson and Syers (1991). The solubility product relation requires that the Ca of the soil plays a role independent of pH, and needs to be considered in evaluating PRs. The concentration of Ca2+ in soil solution is largely influenced, in addition to other factors, by the cation exchange capacity of the soil and the degree of Ca2+ saturation of the exchange sites (Barber, 1984). Percentage Ca2+ saturation of the cation exchange complex has been identified as an important soil parameter affecting PR dissolution (Mackay et al., 1986). To be precise, the important factor is the amount of the cation exchange sites available to adsorb the Ca2+released from PR (CEC minus initial exchangeable Ca (Bolan et al., 1990; Robinson et al., 199 I , 1992). Analogous to pH buffering capacity, we propose that Ca buffering capacity be used to characterize the relationship between soil-exchangeable Ca2+ and Ca2+ in solution. The Ca buffering capacity will be the differential of the curve relating exchangeable Ca2+ to Ca2+ in solution (dCa2+exchldCa2+so,n). The greater the gradient, the greater the ability of soil to continue to dissolve PR. c. Phosphate in Soil Solution and Phosphate Buffering Capacity The reaction given in Eq. (3) shows that if the ionic product exceeds the solubility product of PR, the dissolution of PR will not proceed. Phosphate in soil solution is very low in agricultural soils (on the order of 10-5 M ) and any small fluctuations in absolute concentration, as under field conditions, will have less effect on the ionic product than Ca2+ (on the order of M).It has also been shown that compared to phosphate, Ca2+ has a much stronger retarding effect on the dissolution of PR than is expected from the influence of Ca2+ on the ionic product (Christoffersen and Christoffersen, 1979, 1982). A sensitivity analysis of the Kirk and Nye model (Fig. 9) shows that P buffer-
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
99
ing capacity has small influence on the rate of PR dissolution (Kirk and Nye, 1986b; Anderson and Sale, 1993). Nevertheless increasing P sorption capacities of soils have been positively correlated with increasing PR dissolution (Chien er al., 1980; Smyth and Sanchez, 1982; Mackay et al., 1986; Syers and Mackay, 1986). It is implicit in these results that it is not the absolute P adsorption capacity per se that was affecting PR dissolution, but rather the number of sites available to adsorb the P released from PR (P buffering capacity) and therefore maintain a lower P concentration in solution at the interface. Utilizing a technique similar to that described by Bolan et al. (1983), Kanabo and Gilkes (1987) increased the P buffering capacity of a lateritic podzolic soil by incorporating synthetic goethite into it. They found that the dissolution of ground North Carolina PR increased linearly with increasing P buffering capacity, the latter expressed as P-sorption maximum (Fig. 11). However, it needs to be mentioned that the H+ produced by ongoing Fe3+ hydrolysis may also be responsible for increased dissolution. Consistent with pH and Ca buffering capacity, we support the use of P buffering capacity defined as dP,,,ldP,,,, to estimate the effect of P sorbing properties of soil on PR dissolution. The Freundlich equation constant has been used for this purpose but the fit may not be linear at low solution concentrations unless soil P that equilibrates with solution is accounted for (Kirk and Nye, 1985). This can be estimated from the “a” value determined using isotopically labeled P (McAuliffe et al., 1947). d. Relative Importance of Soil Acidity, Soil Ca, and Soil P on PR Dissolution Considering PR dissolution as a simple chemical process, for the reaction to proceed, availability of the reactants (PR, H,O and H + ) and the removal of
-.
0
200 400 600 800 P-sorption maximum (mg kg-’ soil)
Figure 11 Effect of soil P sorption maximum, adjusted by addition of goethite, cn dissolution (AP)of North Carolina PR. Incubation times 0 (O), I (O), 7 (A), and 35 (A) days (Kanabo and Gilkes, 1987a).
100
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
reaction products (Ca, P, and other ions) from the point source of dissolution are of prime importance. Discussion on the greater importance of one factor over another (Robinson and Syers 1992b; Wright et al., 1992) appears to arise from the differences in soil chemical characteristics favoring the supply of the reactant (mainly H+) or removal of one or more reaction products. The situation can be seen as analogous to the limited availability of a particular major nutrient in a specific soil, and therefore a greater attention to that nutrient, although all major nutrients are of equal importance for plant growth. Generalizing on the relative importance of the various factors, the proton supply in acid soils should be adequate to dissolve PR applied at normally recommended rates. For example, titratable acidity between pH 5 and 6 in the top 10 cm of New Zealand soils is sufficient to dissolve between 2.3 and 7.8 t of North Carolina PR per hectare (300-1000 kg P ha-'), provided moisture is not limiting (Bolan et al., 1990). Even the lateritic soils of Western Australia, where PR application is not recommended (Bolland et al., 1988b), have sufficient titratable acidity to dissolve 0.8 t of PR (100 kg P ha-'). Mobilities of H + and OH- ions are also far greater than those of the other ions under consideration. Regarding removal of P from the surface of PR particles, the reaction of P with soil colloids and the high concentration gradient in P from the PR surface to the bulk soil solution generally results in effective removal of this ion on release from PR (Section 11). For example the P concentration at the surface of a dissolving North Carolina PR particle at a constant pH of 5.5 is about 8 mg liter-' (J. H. Watkinson, personal communication), whereas phosphate concentration in soil solutions is in the order of 0.2 mg liter-1. In comparison the Ca concentration near a dissolving PR particle is 40 mg liter-' and calcium in soil solution is on the order of 8-25 mg liter-' (Gillman and Bell, 1978; Edmeades et al., 1985). Because of the smaller Ca concentration gradient at the PR surface, the Ca buffering capacity of soil may appear to be more important in influencing PR dissolution. e . Effect of Organic Matter The positive influence of organic matter on PR dissolution has long been recognized (Johnston, 1952, 1954a,b; Drake, 1965). This seems to arise from the (i) high cation exchange capacity of organic matter and (ii) organic acids produced as a result of microbial and chemical transformations of organic debris. The cation exchange capacity of mineral soils, depending on their clay content, may range from a few to 50 or 60 cmol kg-' , whereas that of organic matter may exceed 200 cmol kg-I (Helling et al., 1964). Thus organic matter can enhance PR dissolution by enhancing the Ca buffer capacity of soils. Numerous organic acids (e.g., oxalic, citric, tartaric, gluconic) have been reported to be produced in soils as a result of microbial and chemical transformations of organic debris. Johnston (1952, 1954a,b, 1959) conducted comprehen-
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
101
sive studies on the effect of organic acids on the dissolution of Ca-phosphates, and also Fe- and Al-phosphates. He concluded that the organic acids dissolved the phosphates not only by supplying protons but also by complexing the cations. Kpomblekou and Tabatabai (1994)compared the ability of organic and mineral acids to dissolve P from two PRs: North Florida and Kodjari PRs. They concluded that organic acids dissolved more P than was accounted for by their proton supply, and suggested chelation of the metals associated with P in the PRs. Chien (1979)studying the effect of urea on North Carolina PR dissolution in two soils of contrasting organic matter presented direct evidence of the Ca chelating effect of hydrolyzed soil organic matter. He concluded that urea mixed with soil hydrolyzed the organic matter and the products of hydrolysis chelated Ca ions and enhanced the dissolution of the PR. Incorporation of Mussoorie PR (Indian PR) into compost has been reported to improve the release of P, suggesting the occurrence of reactions similar to those in organic soils (Bangar et a l ., 1985). Evidence of soil organic matter enhancing the availability of Sechura PR to plants relative to TSP, which implies a greater dissolution of the PR, was presented by Chien et al. (1990)(Fig. 12). In this pot trial study two soils, both Ultisols, of the same pH (pH 4 . Q exchangeable Ca and P sorption capacity were used. However, one soil contained lower (1.8%) and the other higher (4.2%) organic matter. The greater dissolution of PR in the soil containing the higher
50 h c)
0 P
lA
3 40
t 10
2 n
0
I
0
100
200
300
400 0
1
100
200
300
400
P applied (mg P kg-l soil) Figure 12 Effect of soil organic matter on the agronomic effectiveness of Sechura PR in relation to TSP; (A) low organic matter soil, (B) high organic matter soil (source: Chien er a / . , 1990).
102
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
level of organic matter could also been due to indirect effects of organic matter such as greater cation exchange capacity and high calcium buffering capacity.
3. Site Factors a. Soil Moisture In the dissolution of PR, the rate limiting process is diffusion of the dissolved products (Ca, P, bases) away from the surface of PR particles (Kirk and Nye, 1986a,b). Increased soil moisture levels can increase the apparent diffusion coefficient of the ions by reducing the tortuosity of the diffusion path and increasing the cross-sectional area of diffusion (Nye, 1979). Therefore it is expected that increasing soil moisture, whether effected by rainfall or irrigation, will increase the dissolution of PR. Kanabo and Gilkes (1987a) found from incubation studies that the water retained at field capacity was sufficient to support near potential maximum dissolution of ground North Carolina PR in a lateritic podzolic soil from Western Australia. Weil ef al. (1994) investigated the effect of soil moisture on the dissolution of North Carolina PR in closed incubation and open plant-soil systems. Their incubation study results on two volcanic ash soils (Typic Vitrandept) showed that PR dissolution increased with increasing moisture of up to 80% of field capacity in a medium P retentive soil (P retention 79%; Saunders, 1965) and up to field capacity in another soil of higher P retention (P retention 91%). Field capacity moisture, measured at 700 mm water tension, was 75% on an oven dry basis for the lower and 66% for the higher P retentive soil. Indirect evidence exists of increasing PR dissolution with increasing moisture supply under field conditions. As early as the 1950s, data obtained in Senegal, Africa, showed that the percentage yield increase (over a control) of groundnut and cereal, resulting from application of PR, correlated linearly with the mean annual rainfall which ranged from 500 to 1300 mm (Hammond et al., 1986). In acid soils of Shillong, India, potato responded better to PR in a wet year than in a dry year (La1 et al., quoted in Tandon, 1987). Such results, however, need to be interpreted with caution because low moisture levels, corresponding to water stress to plants, could also reduce the effectiveness of soluble P fertilizers largely because of reduction in maximum yield potentials (Bolland, 1994). Soils with low P sorption, where the rainfall is also high, will promote PR dissolution by facilitating the removal of the products of dissolution in contact with the PR particles(Sanchez, 1976). But from the point of plant availability this may be an advantage if P is not leached below the rooting zone. b. Temperature There is a paucity of experimental data measuring directly the influence of temperature on the dissolution of PR applied to soil. Smith et al. (1977) mea-
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
103
sured the rate of dissolution of hydroxyapatite in dilute HCI as influenced by temperature. They found that the rate constant increased in an exponential manner with temperature for Arrhenius-type temperature dependence. Extrapolating the results to soil conditions one would expect PR dissolution rates to increase with increasing soil temperature, provided other factors are not limiting. Contrastingly, the data of Watkinson (J.H. Watkinson, unpublished data) show a negligible change in the solubility of PRs within the temperature range 5-35°C.
4. Plant Effects There are marked differences in the ability of plant species to extract P from PRs, which was recognized as early as 1898 (Merrill, quoted in Flach et al., 1987). Plants can influence the rate of PR dissolution by the following processes: (i) secretion of acid or alkali, (ii) uptake of large quantities of Ca, (iii) production of chelating organic acids (citric, malic, and 2-ketogluconic acids) which complex Ca, and (iv) depletion of P in soil solution. A wealth of literature is available on the root-induced pH changes in the rhizosphere. The causes of the pH changes are attributed to the imbalance in the proportion of the anionic (usually NO3-, H,PO4-, SO;-, and C1-) and cationic nutrient ( K + , Ca2+, Mg2+, and Na+) intake by the plants (Van Ray and Van Diest, 1979; Aguilar and van Diest, 1981; Bekele, 1983; Haynes, 1983, 1992; Nye, 1981b, 1986; Hedley et al., 1982, 1983; Moorby et al., 1988; Gahoonia, 1992a,b). If the equivalent sum of cation uptake exceeds that for anions, the plants release H+ to maintain electrical neutrality across the root-soil interface and the pH of the rhizosphere soil decreases (Fig. 13). Conversely, if the sum of anions within the plants exceeds those of cations, a net efflux of OH- and/or HC03- occurs, increasing the pH of rhizosphere (Haynes, 1992). Proton secretion is greater for legumes which accumulate nitrogen through symbiotic nitrogen fixation. Because uncharged N, molecules are a major source of plant N, there is an excess uptake of cations over anions. De Swart and Van Diest (1987) reported that about 0.50 mM of acid is excreted by Pueruriu juvmica per nM of N, fixed. In the case of plant species which depend on soil and fertilizer nitrogen, uptake of nitrogen as NH4+ results in H+ secretion whereas uptake of nitrate results in secretion of OH-. Increased soil acidity in the rhizosphere can enhance PR dissolution. This has been observed directly as increased PR dissolution (Gahoonia et al., 1992; Haynes, 1992) but often indirectly as increased P uptake by those plants which acidify the rhizosphere (Van Ray and Van Diest, 1979; Bekele, 1983; Haynes, 1983; Nye, 1981, 1986; Hedley et ul., 1982, 1983; Moorby et al., 1988; Gahoonia et al., 1992). Gahoonia et ul. (1992) measured the HCl soluble P, a measure of calcium apatite P remaining, in the rhizosphere of rye grass. They found a greater amount of apatite P remaining in the soil supplied with NO3-N compared
104
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
0 TSP
0 AI-PR
0 Apatitic-PR 4.81
@
8HI 4 . 4 . r
c ..-
W - Wheat P Paspalurn grass M Maize L Molasses grass
-
B - Buckwheat
4.44.2-
c
5
4.0-
1 2 Total equivalents cations absorbed Total equivalents anions absorbed
0
Figure 13 pH of the soil as influenced by the ratio of the sum of equivalents of cations to the sum of equivalents of anions absorbed by plants (Van Ray and Van Diest, 1979).
with that supplied with NH,-N. (Fig. 14). The limited results of Haynes (1992) suggest a greater dissolution of Jordan PR (a medium reactive PR) by lupin (Lupinus angustifolius L), which decreased rhizosphere soil pH, than by barley (Hordeum vulgare L), which increased the rhizosphere soil pH. Since PRs are essentially calcium phosphate compounds, removal of calcium from PR will release P into solution. Thus effective utilization of PR by some plant species (e.g., buckwheat and rape) has been attributed to their high Ca uptake (Bekele, 1983; Bekele and Hofner, 1993; Van Ray and Van Diest, 1979). Flach et al. (1987) determined the abilities of maize (Zea mays),pearl millet (Pennisetum typhoides), and finger millet (Eleusine coracana) to utilize P from a Mexican (Zimapa PR) and Moroccan (Khouribga) PR in a pot experiment. They concluded that finger millet utilized most P from the PRs, followed by pearl millet and then maize. This was attributed to a greater Ca uptake by finger millet. The dry matter yields of finger millet were about twice those of maize, but the Ca uptake was about three to five times greater. The greater ability of finger millet than maize to utilize a Mussoorie PR from India has also been reported by Singaram et al. (1995). Plant roots may also enhance PR dissolution by secreting organic acids which can be expected to lower rhizosphere soil pH, some of which also complex the Ca of the PR (Moghimi and Tate, 1978; Hoffland et al., 1989). Hoffland et al. (1992) attempted to quantify the possible effect of organic acid exudation on phosphate uptake from Mali PR and measured the exudation of malic and citric acids from P-deficient rape plants (Brassica napus L). They concluded that rape
PHOSPHATE ROCKS FOR DIRECT APPLICATION T O SOILS
”
105
1
0
1 2 3 Distance from the root (mm)
Figure 14 Effect of N source (NO,-N, Nil-N, or NH,-N) on depletion of acid-soluble P (apatite
P)determined in P fractionation procedure in the rhizosphere of ryegrass grown for 10 days on a soil fertilized with ground PR for 10 years (Gahoonia et a / . , 1992).
plants increased P uptake from PR through reduction in the rhizosphere pH and also by complexation of the Ca with organic acids. Thus, Ca2+ concentration in the soil solution was reduced. Simulation calculations indicated that the exudation rates can provide the roots with more phosphate than is usually taken up. From in vitro studies evidence has been presented of organic acid secretion by Rhizobium and Brudyrhizobium strains (Halder et ul., 1990). These authors reported that the 2-ketogluconic acid secreted by the cultures was the primary factor influencing the dissolution of Mussoorie PR, implicitly through complexing of the dissolved Ca component of PR, while pH per se was less important.
5. Method and Rate of Application a. Method of Application Increasing the area of contact of PR with soil will enhance PR dissolution by removal of dissolution products and supplying H+ . Therefore one would expect a greater dissolution of PR when PRs are mixed with soil rather than applied as a band. Banding will also enhance dissolution zones of PR particles overlapping and thus hinder continual dissolution. This has been observed indirectly as increased P uptake by subterranean clover (Trifolium subterraneum) with increasing depth of mixing PR with soil (Alston and Chin, 1974). Similarly Purnomo and Black ( 1994) reported from a greenhouse study that the dry matter yields of
106
S. S. S. RAJAN, J. H. WATKINSON,AND A. G. SINCLAIR
wheat from North Carolina PR were in the order of mixed > broadcast > banded. A greater dissolution of North Carolina PR mixed with soil compared with banded PR has also been found in laboratory incubation studies (Kanabo and Gilkes, 1988b). b. Rate of Application Increasing the rate of PR application will eventually result in PR particles being so close that the zones of dissolved Ca and P ions overlap, resulting in a slower rate of PR dissolution. Such an effect can be expected at lower rates of application when the PR is added in clumps or is surface applied. Also, the effect of application rate on PR dissolution will be accentuated in the presence of plants. This is expected because the influence of roots diminishes with increasing rates of application, brought about as a consequence of the root system affecting a smaller fraction of PR (Kirk and Nye, 1986b). Experimental evidence under laboratory (Kanabo and Gilkes, 1988d) and field conditions of decreased proportion of PR dissolution has been presented (Bolland and Barrow, 1988; Rajan et al., 1991) (Fig. 15). In the latter study Sechura PR was surface applied to a permanent pasture. It needs to be emphasized that when calculated as a fraction of that applied, PR dissolved may decrease at high rates of application. In absolute amounts, however, the PR dissolved will generally increase (Kanabo and Gilkes, 1988; Rajan et al., 1991) with increasing rates of application up to the point at which the ionic product equals the solubility product of PRs, at which time PR dissolution will cease.
Y4.47-0.36 log x, RC0.90' (6 years) Yd.39-0.28 log x, R2=0.81NS (4 years) Y=O.89-0.21 log x, R2=0.94' (3 years) 1
0
100 200 300 PR added (kg P ha-')
Figure 15 Effect of rate of application on the fraction of Sechura PR dissolved 3,4, and 6 years after application (Rajan et a!., 1991b).
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
107
B. FACTORS AFFECTINGPLANT AVAILABILITY OF P FROM DISSOLVED PHOSPHATE ROCK Phosphate availability to plants in a soil is governed by concentration of P in soil solution, the sustainability of this concentration on absorption of P by plants, and the ability of crops to utilize the phosphorus. In general, increased PR dissolution is expected to result in a measurable increase in soil solution P and therefore increased plant production (Rajan et al., 1991a,b), but this is not always the case because of the influence of soil, crop, and fertilizer management factors. An example of this is illustrated in Fig. 16 (Syers and McKay, 1986), where P uptake by ryegrass was not related to PR dissolved. In the following section we discuss the factors which influence the availability of PR-P.
1. Soil Factors a. Phosphate Buffering Capacity and P Status Soil P sorption capacity has been identified as an important parameter influencing the availability of dissolved PR-P to plants (Smyth and Sanchez, 1982; Hammond et al., 1986a; Syers and Mackay, 1986). Stated differently it is the fraction of vacant sites in relation to the absolute P sorption capacity that will influence the solution P concentration and therefore P availability to plants (Rennie and McKercher, 1959; Rajan, 1973). The greater this fraction, the smaller the
$
0
30-
? = 0.003
7
25-
-E"I
20-
p
15-
E
2
B 10a r
0
al 5 x
0
0
4 0 0
0 0
0
0
c
4 on
1
PR-P dissolved (mg kg-l soil)
Figure 16 Relationship between P uptake (PR treated - control) by ryegrass and Sechura PR dissolved when added to nine soils at 500 mg P kg-I of soil in a greenhouse experiment (Syers and Mackay, 1986).
108
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
solution P concentration. Also, the greater the P buffering capacity, the slower the rate of change in solution P concentration. As stated previously, low solution concentrations of P favor PR dissolution, but the dissolved P will be immediately adsorbed by soil colloids. Three scenarios can be visualized in such circumstances. First, if the solution P level is below the threshold concentration required for P uptake by plants, very little increase in P uptake or yield will be observed in spite of PR dissolution. This will be the case in soils of low P status. In such soils a sigmoid yield response curve for fertilizer application would be expected. A second situation is that in which the P concentration is above the threshold level. However, considering a curved relationship between P sorbed and that remaining in solution, one will find less than a proportionate increase in solution P with increasing PR-P released and subsequently adsorbed by soil. Thus the available P will still be much less than that dissolved. Third, in soils of high P status most of the PR-P is dissolved and should be theoretically available to plants. b. Effect of Soil Temperature The influence of soil temperature on the availability of dissolved PR-P may be of importance in tropical soils. An increase in soil temperature has been found experimentally to have the overall effect of increasing P adsorption by soils and decreasing P in soil solution (Fig. 17), which in turn will reduce P availability to plants. Chien et al. (1982) reported that when 100 mg kg-l P was added to an Ultisol and an Oxisol from Columbia, P concentrations in solutions decreased linearly with increasing temperature. The decrease in P concentration per degree of increase in temperature was 0.63 mg liter-' for the low P retentive Ultisol whereas the value was 0.12 mg liter- I for the high P retentive Oxisol. The low solution concentration was brought about by the high temperature accelerating 8004 Moiokai
3204 Coirnbatore
400
*35°C 25°C -0-
-a-
0
8
10°C
7
0
16 0 Final P concentration (rng C1)
8
16
Figure 17 Effect of temperature on P sorption by Indian (coimbatore) and Hawaiian (Molokai) Oxisols (6 days equilibration) (S.S.S.Rajan, unpublished data, 1970).
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
109
the “slow reaction” that followed the intial adsorption of P (Barrow and Shaw, 1975; Barrow, 1974).
2. Crop Characteristics Early studies from the Tennessee Valley Authority (Huffman, 1962) have shown that the P concentrations near soluble P fertilizer granules could be as high as 3.5 M . Such a high concentration is conducive to diffusion of P to a large soil volume, which could be to a distance of up to 70 mm from the point of fertilizer placement, depending on the size of the granule (Benbi and Gilkes, 1987). In contrast, the concentration of P in a saturated solution of a reactive PR at a pH of 5.5 is in the order of M ,which suggests diffusion of P to a limited volume of soil. In addition the concentration of P in soil solution following PR application could be very low, in the order of one tenth of that to which soluble fertilizer is applied as has been found by Rajan er al. (1991b) 10 months after fertilizer application. Therefore availability of P from dissolved PR will be greater to crop species which have extensive root systems such as perennial grasses (Chien et al., 1990) and those crops which can extract P from soil solution at low concentrations. Indeed the sensitivity analysis of Kirk and Nye (1986b) indicate that the density of roots in the soil influences greatly the proportion of PR-P taken up by plants. Inoculation of plant roots with mycorrhizae will facilitate extension of the root system to a greater soil volume and thus enhance P uptake (Waidyanatha et al., 1979; Tinker, 1980).
3. Management Practices The method and time of application of PR in relation to crop planting time can affect the availability of dissolved PR-P to plants. a. Method of Application Plant availability of P from PR depends on the probability of plant roots encountering the localized higher-concentration pockets of soil P around dissolving PR particles. This probability will be increased if the PR is broadcast and is uniformly incorporated into the surface soil to the required depth. In other words, the greater the volume of P-enriched soil, the greater its availability to plants, provided roots explore. Broadcasting without incorporation may not increase availability of PR-P in spite of the greater amount of PR dissolution occurring at high rates of application, because the dissolved P is likely to be restricted to a shallow depth of soil (Rajan et al., 1991). The negative aspect of incorporation of PR into soil is that it facilitates greater contact between soil colloids and solution P, resulting in greater P adsorption and therefore in low solution P concentration. The net result of increased P adsorp-
110
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLATR
tion, versus the P enrichment of a greater volume of soil, on P availability to plants will depend on soil P status (Fox et al., 1986b) and rate of PR application. In soils of medium P status, or low P sorbing capacity, such as sandy soils, mixing PR may be desirable (Alston and Chin, 1974). On the other hand this may be an unsuccessful strategy in highly P sorbing soils of low P status. b. Time of Application Since there is a time lag between PR application and significant dissolution (Barnes and Kamprath, 1975; Sinclair et al., 1993a) some authors have suggested application of fertilizers in advance of planting crops. Again the advantage of such a practice will depend on the net result of two opposing reactions that are operating: dissolution of PR and adsorption of dissolved PR-P by soil colloids and its slow conversion to nonavailable forms. In soils of low P buffer capacity, dissolution of PR will increase P in solution to a greater concentration, and probably will be maintained for a longer duration than in soils of greater P buffering capacity. Thus application of PR in advance of planting may be an advantage in soils of low P buffering capacity but not necessarily in soils of higher P buffering capacity. The results of Chien ef al. (1990) (Fig.18) and Purnomo and Black (1994) are consistent with the above reasoning. When PR is
80 Low P bufferingsoil
- High P bufferingsoil
70
J
1
P
w
60
.-
50 .c
40 z .-a,
* 30 a,
L
20
P
n 10 before planting
0
100
200
300
. - I
400 0
100
200 300
400
P applied (mg P kg-1 soil) Figure 18 Interaction between soil P buffering capacity and time of application on the plant effectiveness of North Carolina PR in relation to TSP (source: Chien et af.. 1990).
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
I1 1
applied for maintenance purposes, as implied by Engelstad and Terman (1980), early application of PR may be beneficial also in soils of high P buffering capacity.
V. MODELING THE RATE OF PHOSPHATE ROCK DISSOLUTION IN FIELD SOIL Although a number of studies examine the kinetics of PR dissolution in soil, or model PR dissolution in field soil, most do not look at the system as a whole. For example, Chien et al. (1980) examined the kinetics of PR dissolution in soil by agitating suspensions of the two and measuring changes in soluble phosphate. The results, however, probably reflect more the kinetics of P adsorption by the soil from an initially saturated solution of the PR, since the PR was added in large excess, and adsorbed P was not measured. More recently, Robinson et al. (1994) described a conceptual model for predicting the dissolution of Gafsa PR in soil. While the model qualitatively relates soil properties important to dissolution, such as soil acidity, solution calcium, and exchange sites, the transport processes within the soil and PR particle size are not considered, so that time is not a quantitative aspect of the model. Further, it is not supported by field data; such as the rate of pasture P uptake or measurement of residual PR in soil with time which are related to the rate of PR dissolution. Only two published models appear to quantitatively examine the kinetics of dissolution in field soil: the comprehensive model of Kirk and Nye ( 1986a,b) which incorporates individual soil properties, and the simpler model of Watkinson (1994a,b) for pastoral soils, which combines all soil and site effects into an overall rate constant. The qualitative models will not be discussed further. Both the Kirk and Nye (1986a,b) and the Watkinson (1994a,b) models postulate that the dissolution rate of PR in soil is controlled by the diffusion of dissolved products from the PR surface into the bulk soil. Kirk and Nye ( 1986a,b) include all ions involved in a coupled counterdiffusion process, whereas Watkinson (1 994a,b) assumes the rate-limiting step is diffusion of calcium phosphate, which is controlled by the trace diffusion of phosphate, because of the relatively small calcium concentration gradient at the PR surface. Further, Kirk and Nye (1986a,b) consider the consequences of some important situations, in particular those in which dissolved PR mutually affects the dissolution rate of neighbouring particles (Fig. 19A). In contrast, the Watkinson (1994a,b) model considers only the simpler system of dissolution for maintenance rates of PR in New Zealand pastoral soils, which involves the independent dissolution of PR particles. Under these conditions both models give similar results (Watkinson, 1994a) (Fig. 19B). A major assumption of both models, which needs verifica-
112
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR kg P m-3 0.16 0.27 0.45
0.67
0.16
0
i
16 24 Time (days)
32
Figure 19 The dissolution rate of DCPD. (A) Predicted from the Kirk and Nye model, for different application rates (numbers on curves in units of kg P m-3 soil), showing the influence of neighboring particles at higher rates. Initial radius = 0.1 mm, other variables as in Kirk and Nye (1986a). (B) For particles dissolving independently (application rate of 0.16 kg P I I - ~ ) :comparison of Kirk and Nye model (1986a) (-) and Watkinson (1994a) SFOR model for Eq. 5(-), and Eq. 6 (----)(g = 0.19 mm and G = 0.208 day-').
tion, is that the mobility of phosphate adsorbed on the soil is not a limiting factor, i.e., is more rapid than the diffusion process (Watkinson, 1994a). Instead of the interparticle diffusion assumed, some phosphate may also diffuse very slowly into the interior of soil particles. Implicit in the diffusion controlled models is that the geometric area, not the total area including internal surfaces, is the variable controlling the total flux of dissolved PR into the bulk soil. Diffusion through the soil is much slower than that through the solution in contact with the internal surfaces, and the area of soil in contact with the PR particles is the geometric area. Hence a measure of the particle size distribution by sieve analysis is sufficient to define the area variable of a PR (Watkinson, 1994~). Soil moisture is an important field variable (Anderson and Sale, 1993; Weil er al., 1994) because of diffusion (Kirk and Nye, 1986a; Watkinson, 1994a), but soil temperature would be expected to be less important, because PR solubility has a negligible temperature dependence (Watkinson and Kear, unpublished work), and the activation energy of diffusion is small. Smith er al. (1999), however, found a temperature dependence for hydroxyapatite.
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
113
Other rate-limiting steps are possible. For example, the dissolution rate of some complex crystals can be controlled by the rate of transfer from the crystal surface into the adjacent solution (e.g., dolomite and olivine) (Sverdrup and Bjerle, 1982). Although the rate of calcite dissolution in aqueous suspensions is controlled by Nernst film diffusion, the rate of dolomite dissolution is much slower (Sverdrup and Bjerle, 1982). This will have implications for the dissolution of free calcite and dolomite that are present in many PRs. However, the rate of dissolution of the 1 1 PRs examined by Watkinson (1994~)was consistent with Nernst film diffusion, with only a small component a little slower. This supports the earlier conclusion of Huffman et al. (1957) who investigated dissolution rates of calcium phosphates in phosphoric acid solutions. Diffusion in pure solution is much more rapid than in soil, so the basic assumption of the two models is reasonable based on present knowledge. Although Olsen (1975) found that the dissolution of PRs into EDTA solution was consistent with a second-order reaction rate model, he did not test it as he was only interested in comparing the relative dissolution rates of PRs.
A. KIRK AND NYEMODEL 1. Description Kirk and Nye (1986a) developed a model to describe the rates of dissolution in soil of the apatites in PRs, based on one developed for dicalcium phosphate dihydrate (DCPD), which was itself a successful simplification of an earlier model. The rate-limiting step for DCPD dissolution is essentially the steady-state counterdiffusion of phosphate and hydrogen ions between the solution at the mineral surface and that in the bulk soil. An equation for the rate of loss of mass from the mineral particle was formulated using the equality of the diffusive flux of ions across the mineral interface, while maintaining the solubility product of the DCPD and ionic charge balance in the interface solution. The influence of neighboring particles on the concentration of dissolved material in the bulk soil was also included. In this situation, the soil solution reaches concentrations up to the solubility of DCPD which, while consistent with high soil moisture, would not be maintained under high rainwater leaching rates. The DCPD model, involving numerical solutions to the differential equations, was then extended to describe the rates of dissolution of apatites. Increasing carbonate substitution increases the solubility of carbonate apatites (Section ll.C), and therefore the dissolution rate. In contrast, the increased alkali from hydrolysis of lattice phosphate from CAs compared with DCPD lowers the rate of dissolution. Kirk and Nye (1986a) considered that the effect of gangue material was “likely to be very
114
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
small.” However, calcite dissolves rapidly generating both alkali and calcium, so the effect of appreciable calcite impurity would need to be taken into account. In a second paper (Kirk and Nye, 1986a) quantitative applications of the model to field situations were considered. The effects of several factors on the rate of dissolution of calcium phosphates in soil were examined. They included application rates to the soil, particle size distribution, and the distribution of the calcium phosphate particles in the soil. As the size range about a mean value was extended, so the total dissolution rate decreased. This was caused by the increasing fraction of larger particles in the mixture. Clumping of particles in the soil and high application rates also decreased the dissolution rate, through the increased effect of dissolved material on neighbouring particles. The influence of
’7
bHS
A
/0.02
“I ’0°1
f
U‘
0.32
0.13
NO.10
1
0
50
loo
150
200
Time (days) Figure 20 The predicted effect on North Carolina PR dissolution in a soil, pH 5.2, in the glasshouse of (A) the pH buffer capacity (bHs = 0.02, 0.01, 0.005) whenf = 0.13 and (B)the diffusion impedance factor cf = 0.32, 0.13, 0.10) when bHs=O.O1 compared with experimental
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
115
plant roots was also investigated. Dissolution rates were increased by uptake of phosphate, depending on the rooting density and rate of fertilizer application, and by plants that secreted acid from their roots. Because of the high amounts originally present in soils, it was not considered that calcium uptake would greatly affect dissolution rates.
2. Testing Anderson and Sale (1993) reported results of an experiment to test the prediction by the Kirk and Nye model of the rate of dissolution of North Carolina PR incubated in an acidic soil in a glasshouse. Three pH values (4.8,5.2, and 6.3) of the soil (through adding sodium carbonate) were examined. Observed results were reasonably consistent with theory at the two lower soil pH values, considering all the experimental errors involved, but not at the higher value. One difficulty in fitting the data was the continual decrease in soil pH during the course of the experiment, with the effect being greater at the higher pH values. The decreases were probably the result of two factors: the increase in acidity through nitrification during incubation, and the apparent incomplete equilibration of the added sodium carbonate with the soil. Another problem was the instability of the model in fitting data at pH values greater than 6.0, values which are common in many temperate pastoral soils. Anderson and Sale (1993) concluded that the model was very sensitive to the pH buffering capacity (Fig. 20A), and more particularly to soil moisture via the diffusion impedance factor (Fig. 20B). The authors concluded that soil moisture would need to be closely monitored under field conditions, a problem that would need to be overcome for routine use of the model. They also concluded that only methods strictly pertinent to the measurement of the model parameters of soil pH and pH buffering capacity should be used.
B. WATKINSON MODEL 1. Description
The model of Watkinson (1994a) was designed for use on pastoral soils receiving maintenance rates of PR. The concentration (m)of PR particles is such that they dissolve independently of each other. Dissolution is controlled by diffusion of dissolved phosphate from the PR surface solution into bulk soil solution through a boundary layer of soil. Because the concentrations, C,, C s , at both surfaces of the boundary layer are essentially constant, the process can be regarded as a simple steady-state phenomenon. In this regard, the boundary layer is conceptual and equivalent to the Nernst film of stagnant solution surrounding
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
116
particles in a stirred aqueous suspension. The thickness of the layer would be a function of the rainwater leaching rate, which is unrestricted (in contrast to the Kirk and Nye model), and the rainwater may leach phosphate to lower depths. Although the diffusion coefficient of phosphate will be a function of the phosphate concentration because of the nonlinear adsorption by the soil, a mean diffusion coefficient, D,, can be used since the system is at steady-state. Differential equations relating mass of material dissolved and diffusing into the soil are simple enough to be solved analytically. Two submodels were considered; in the first (SFOR), the outer radius, g, of the boundary layer was constant, and in the second (SCT), the thickness of the boundary layer, h, was constant as the particle dissolved. For slowly dissolving PR, the situation for soils of pH greater than 6.0, the models were essentially the same. The particles were taken to compose uniform equivalent spheres of initial radius r,, which were added to topsoil at an initial concentration m,.
SFOR Model: (m/m0)2/3- (m/m, - 1)(2r0/3g) = I - 2Gt. If the boundary layer is thick, i.e., r,/g (m/m,)2/3 where G
=
< 1, then,
=
(5)
from Eq. (S),
1 - 2Gt
(6)
D,(C, - C,)/(pro2F) (See Section II.B.2.b).
SCT Model: 1 - (m/m,)1/3
+ (h/r,)log[{(m/m,)l~~+ h/ro}/(l + h/ro)] = (ro/h)Gt.
(7)
If the layer is a thin coating, i.e., hlr, 6 1, then, from Eq. (7), (m/m,)1/3 = 1 - (r,/h)Gt
(8)
On the other hand, if the layer is very thick, then the SCT model is approximated by (m/m,)2’3
= 1 - 2Gt,
(9)
which is the same as for a thick layer in the SFOR model. For fertilizer mixtures, which contain a range of sizes, the kinetic relationship is described by a set of three equations like Eq. (2) (Watkinson, 1994b). They are based on the approximation of Eq. ( S ) , i.e., Eq. (6), which is appropriate for slow dissolution (see Section II.B.2.b). The above equations can be tested readily using field data, both directly from measurements of residual PR at increasing times from application and indirectly from agronomic data from which substitution values are calculated, as shown in the next section.
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
1I 7
2. Testing Equations from the model have been found consistent with field dissolution rate data for PRs on a New Zealand pastoral soil, where the pH was near the upper limit at which PRs can be used (Watkinson, 1994a) (Fig. 21A). PR treatments included mixtures with elemental sulfur (Rajan, 1987) and the unreacted component of PAPRs (Rajan and Watkinson, 1992) (Fig. 21B). Data for the dissolution rate of Sechura PR in nearly 100 New Zealand pastoral soils, with pH (H,O) values from 4.9 to 6.4, were also consistent with the model (Watkinson, 1995; Perrott el a/., 1996). The model is also supported by agronomic data. Substitution values are directly related to soluble P applied and therefore to PR dissolution rates (Edmeades et al., 1992), provided allowance is made for an annual loss in P availability (Watkinson and Perrott, 1993). More than a dozen field trials related substitution values with PR dissolution rates from the model (Watkinson and Perrott, 1993) (Fig. 22). The relative agronomic effectiveness measured on three soils in greenhouse experiments related well to the dissolution rates of 1 1 unground and ground PRs (Rajan e t a l . , 1992; Watkinson 1994~.1995) (Fig. 2). One disadvan-
0
1
2
3
Time (years)
4
Figure 21 Fit of Eq. ( 6 )to the experimental dissolution rate of PRs in a pastoral soil. (A) North Carolina (O), Chatharn Rise ( O ) ,Florida (0) (Rajan, 3987b). and North Carolina (A)(Rajan and Watkinson, 1992).Regression forced through ( 0 , 100).and (B)the unreacted North Carolina constit40% (A).and 50% uent of partially acidulated North Carolina PR. Unacidulated (0).and 30% (0). ( 0 )acidulated (Rajan and Watkinson, 1992).
118
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
Time (years) Figure 22 Mean cumulative substitution values (0)for annual North Carolina PR applications in 12 field trials for each of 6 years (Sinclair ef a / ., 1993) compared with values calculated assuming an annual decrease in applied P availability of 15%. at different (cubic) rate constant values, K (year-') (Watkinson and Perrott, 1993). (K = 0.1 year-' is equivalent to 30% dissolution in the first year. )
tage of the simplicity of the model is that it does not predict rate constants directly from theory using easily measured soil properties. To overcome this, rate constants have been measured in 95 field trials (K. W. Perrott, personal communication) with the objective of developing relationships between rate constants and soil properties, which will allow the constants to be estimated from easily measured soil properties.
VI. AGRONOMIC EFFECTIVENESS OF PHOSPHATE ROCK A. DETERMINING AGRONOMIC EFFECTIVENESS Agronomic effectiveness of PR as a P fertilizer is ultimately expressed in its ability to supply adequate P for sustaining desired levels of crop production. Thus plant growth rather than merely dissolution in the soil is the final judge of agronomic effectiveness. What matters ultimately is plant growth in the situation where the PR is to be used for practical farming, and a basis for predicting this must be established. Plant growth experiments are essential for establishing this predictive basis. Assessment of PRs as P fertilizers involves comparison of agronomic performance usually with water-soluble P fertilizers such as TSP and SSP. In such comparisons care must be taken to ensure that the ability to supply P is the only factor influencing the results. In addition to ensuring adequate levels of other
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
119
major nutrients, trace elements may also require attention; for example, Sinclair et a / . (1990) found that the molybdenum (Mo) content of Sechura PR was sufficient to eliminate Mo deficiency in clover-based pastures, so failure to apply
Mo with the standard fertilizer could invalidate the comparison. Comparisons could also be affected by the small liming effect of PR (Sinclair ef af., 1993b), but this would be significant only at very high application rates or after a long period of regular application. Plant growth experiments may be conducted in pots or in the field. Each system has its advantages and limitations. 1. Techniques
a. Pot Experiments Pot experiments have the advantage of being relatively inexpensive so that many factors influencing agronomic effectiveness can be examined. A further advantage is that factors may be examined individually while other factors are kept constant, whereas this is often impossible in the field. However, differences in growth conditions and plant behavior between pot and field are generally so great that predicting the absolute performance of a PR in the field is not possible from pot experiments. Also, considering the slow release characteristics of PRs, long-term evaluation is essential but may be difficult to sustain while avoiding significant changes in soil conditions in pot experiments. For example, substantial soil acidification may occur in lengthy pot experiments with legumes. Whereas this may be neutralized by frequent application of bases the fluctuation in pH may still be greater than would occur with the much larger soil volume in the field. With PRs, pot experiments have their main value in ranking PRs from different sources (Mackay et a / ., 1984b). The relative agronomic performance in pots has been used for judging the ability of laboratory tests to rank PRs in order of their agronomic effectiveness (Chien and Hammond, 1978; Kucey and Bole, 1984; Mackay et af., 1984b; Rajan et af., 1992). Few experiments have directly compared PR rankings in the field with those measured in pots, using the same soil and crop, but it is the general experience that PRs which perform best in the field are those which perform best in pots. Chien and Hammond ( I 978) compared the same group of PRs in glasshouse and field experiments, although the soils and crops differed. PR rankings were similar in both experiments, but not identical. Engelstad et a / . (1974) compared six PRs in a glasshouse and two field experiments with flooded rice and found almost identical rankings in the three experiments. Rajan et af. (1992) found that different soils, similar in pH (5.5-5.7), gave similar but not identical rankings of PRs in a glasshouse experiment using the same crop. Pot experiments may also be used to compare the performance of a particular
120
S. S. S. RAJAV, J. H. WATKINSON, AND A. G. SINCLAIR
PR in different soils and thus identify the soils on which it is most likely to perform well (Anderson et al., 1985). Differences between plant species in their ability to utilize P from PRs have also been examined in pot experiments (Haynes, 1992). b. Field Experiments Field experiments are essential to provide a realistic assessment of the likely performance of a PR in practical farming situations. However, it is important to realize that the field experiments merely provide a basis for prediction, since the conditions under which the PR will be used in practice are unlikely to correspond exactly to those in any field experiment. Ideally a matrix of field experiments should define PR performance as affected by PR properties, soil and climatic conditions, and crop characteristics. Interpolation and extrapolation would then permit accurate prediction of performance in any situation. However, the cost of obtaining such detailed information would be prohibitive. Consequently, predictions must be made from the relatively few relevant field experiments available, along with insights gained from laboratory and pot experiments on the effects of site, crop, and PR properties.
2. Measurements Agronomic performance in the field of PRs is generally judged in relation to conventional soluble P fertilizers such as TSP and the measure of performance is usually crop yield. However, crop yield measurement can fail to detect real differences between fertilizers in their ability to supply P for plants for the following reasons: (i) inadequate precision in yield measurements; (ii) high fertilizer application rates, so that even the less effective fertilizer supports near maximum growth, so differences between fertilizers cannot be expressed in yield differences; and (iii) inadequate responsiveness of the trial site. Thus failure to detect a difference between P fertilizers can often arise not because there was no difference but because the experiment was incapable of detecting the difference. Johnstone and Sinclair (1 99 1) provided biometrical guidelines for the comparison of P fertilizers, emphasizing the limitations of field evaluation. They show, for example, that 40 replicates would be required to ensure a 90% probability of detecting a difference between two fertilizers which differ by 10%in P availability when they are compared in a trial with a CV of 3% in which a 100% response to P could be achieved. Crop yield measurements may be supplemented with herbage P concentration and plant P uptake measurements for comparison of fertilizer effectiveness. Generally these are more sensitive to differences in P supply (see for example Gregg et al., 1988) and can distinguish between fertilizers when soil P status or P fertilizer application rates are too high for yield differences to be detected
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
12 1
(Sinclair et al., 1994). Thus, provided it can be demonstrated that the relationship between crop production and P uptake or concentration in herbage is the same for the PR and soluble P fertilizers being compared, these alternative measurements may well be superior to crop production for evaluating PRs. Radioisotope techniques using 32P and 33P provide a further approach to the evaluation of PR fertilizers (Morel and Fardeau, 1990). Labeling is a simple process with synthesized PR, but neutron irradiation is required to label P in natural PR (Fried and MacKenzie, 1949), and this may alter PR properties. Alternatively the labile P pool in the soil may be labeled and the contribution of unlabeled P fertilizers to plant P uptake calculated. Isotope studies allow the P-supplying ability of PRs to be assessed even where there is no yield response to P fertilizer. They can also give a direct measure of the amount of plant P which is derived from the test fertilizer. However, the short half-life of 32P (14.7 days) and 33P(25.3 days) severely limits the value of the technique since PR evaluation is primarily concerned with long-term effectiveness. Measuring agronomic performance of PRs in pastoral systems presents particular problems. First, the output of these systems is in animal products rather than crops, so it could be argued that those should be the parameters to measure. However, appropriate experiments with animal measurements would be prohibitively expensive. Moreover it is generally found that efficient grazing animal production is in direct proportion to pasture production (Morton er al., 1995), so pasture dry matter (DM) production is used almost exclusively in field comparisons of P fertilizers. Pasture DM production is most conveniently measured in small, regularly mown plots, and to simulate the nutrient returns in animal excreta a large proportion of harvested herbage is often spread over the plot from which it originated.
B. QUANTIFYING COMPARATIVE PERFORMANCE OF PHOSPHATE ROCKS 1. Substitution Value, Relative Response, and Isotopic
Relative Agronomic Effectiveness It is important to recognize that the comparative performance of PRs is strongly influenced by the length of time over which plant growth is measured, by fertilizer application method and frequency, by soil conditions, and by plant species. A full account of these factors is essential when reporting agronomic performance of PRs, and they should be taken into account when making predictions from experimental data. The lower agronomic value of PR relative to soluble P may manifest itself in different ways. Two extremes are illustrated in Fig. 23. In Fig. 23A, PR and
IrP-
122
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR I d
. 0
-
sp
PR
I
I
1.o
0.5
D
P
1 0 c n c n o 0
sv
0 RR
0
P applied Figure 23 Patterns of response to P fertilizers and the effects of application rates on measures of agronomic performance. ( A ) Yield from PR is always the same as yield from half the quantity of P applied as soluble P fertilizer (SP). (B)Yield response to PR is always half the yield response to SP. (C) Fertilizer substitution values (SV) and relative response (RR) corresponding to Figure (A). (D) SV and RR corresponding to (B).
soluble P give the same maximum yield but more PR than soluble P is required for yields below the maximum. It has been drawn for the specific case in which yields from any rate of soluble P can always be achieved with twice the rate of PR. In Fig. 23B PR never achieves the same maximum yield as soluble P because maximum growth of the crop requires a higher P concentration in soil solution than the solubility product of PR can permit. In this case the fraction of PR which dissolves declines as the application rate increases. Figure 23B has been drawn for the specific case in which the yield response to PR is always half of that to soluble P applied at the same rate of P. In practice, as well as these two extremes, a full range of intermediate response patterns probably occur.
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
123
The agronomic performance of PRs relative to soluble P fertilizers is generally expressed quantitatively in one of two ways, these being substitution value (SV) or similar expressions (Black and Scott, 1956; White et al., 1956; Barrow, 1985; Colwell and Goedert, 1988; Black, 1993) and relative response (RR). In this review SV is expressed as the ratio of total P applied in standard fertilizer to total P required as test fertilizer to give the same plant yield. Expressed this way, the SV of a poorly performing test fertilizer is low and vice versa. RR is the response to the test fertilizer divided by the response to a standard fertilizer when both are applied at the same rate of P. The term relative agronomic effectiveness (RAE) is often used for this parameter, but we prefer the term RR because it is unambiguous. Both SV and RR can vary with PR application rate. In cases represented by Fig. 23A, where PR and soluble P response curves have the same asymptote, SV remains constant but RR increases toward a value of 1 at high rates (Fig. 23C). Where the PR response curve has a lower asymptote than the soluble P response curve, as in Fig. 23B, SV declines toward zero with increasing PR application rate, while RR may remain constant (Fig.23D). Values of SV and RR converge as application rates are reduced toward zero, irrespective of the type of response pattern. There are clearly dangers in expressing the agronomic performance of a PR, as either S V or RR, if these are measured at high PR application rates or at rates approaching zero. For situations like that of Fig. 23A RR will be high even for a poorly effective PR, while for situations as in Fig. 23B SV will be low even for PRs which perform well at moderate rates. PR performance at rates approaching zero has been calculated from the initial slopes of the PR and soluble P response curves (Kumar et al., 1992b). This may not provide a reliable measure of performance at practical application rates. Also PRs are known to give similar yields to SSP at low levels of crop production comesponding to low rates of PR application. From the above discussion it is clear that in many situations decisions on the use of PR would need to be based on consideration of complete, well-defined response curves. Both SV and RR can also be determined from plant P uptake and P concentration. Provided that the relationship between plant yield and plant P concentration is not affected by fertilizer form, SV based on plant P content should be identical to SV based on plant yield since in both cases one is calculating the relative amounts of the different fertilizers required to produce the same plant condition. But RR will be different for different plant parameters measured, especially at high fertilizer application rates or on unresponsive soils. In these situations there may be no yield difference between fertilizers, so RR based on yield will be 1.O; but differences in plant P content are still likely, giving RR values different from 1.0. Morel and Fardeau ( 1990) introduced the parameter Isotopic Relative Agro-
124
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
nomic Effectiveness (IRAE) to express agronomic performance as determined by their isotope procedure. This procedure involved applying labeled soluble P and PR fertilizers in separate pots at a single P application rate and calculating the amount of fertilizer-derived P in the plant. Then IRAE (%)
=
100*P~l/P~s,
(10)
where Prland P, are the amounts, respectively, of test and standard fertilizer P in the plant when both fertilizers are applied at the same rate of P. Morel and Fardeau (1990) found that IRAE for North Carolina PR compared with DAP was independent of P application rate up to 200 mg P kg-I soil. This result may well have been an artifact of the experiment rather than a basic property of the IRAE. For Fig. 23A type situations, IRAE would tend toward unity at high rates since the maximum contribution of either fertilizer to plant P is 100%.But in Fig. 23B type situations, IRAE would decline as application rates exceeded that at which no further dissolution of PR could occur. An intermediate situation could account for Morel and Fardeau’s (1990) result.
2. Mathematical Functions for Phosphate Rock Response Patterns The use of mathematical functions to describe response patterns facilitates comparisons of PRs with soluble P fertilizers. The Mitscherlich equation has been widely used for this purpose. In its simplest form, y
=
a - bet.
( 1 1)
where y is plant yield, a is the maximum yield, i.e., the asymptote of the response curve, ( a - b) is the yield without fertilizer, c is curvature factor and x is the fertilizer application rate. In situations where PR and soluble P fertilizers have a common asymptote and of course a common value for (a’- b) (Fig. 23A), SV can be calculated from the curvature (Barrow, 1985; Johnstone and Sinclair, 1991):
A slightly different procedure, but with the same outcome, for Fig.23A situations is to combine data from both fertilizer response patterns into a single curve which is expressed by
y = a - b*c(xsp+
hpR),
(13)
where xsp and xPRare application rates of soluble P and PR and k is the substitution value (SV); k is then determined by best-fit procedures. Where asymptotes differ (Fig. 23B) Mitscherlich curves could still be fitted to the individual fertilizer responses but these would differ in all coefficients ( a , b,
PHOSPHATE ROCKS FOR DIRECT APPLICATION T O SOILS
125
and c). Two alternative functions with only a single material-dependent coefficient have been proposed for Fig. 23B situations. These are y =y,
+ b.ln x
(14)
y =y,
+
(15)
and b.x”jn,
where yo is yield without P, y is yield with x units of P, m is a constant, and b is the material-dependent coefficient. Leon e t a / . (1986) used Eq. (14), and Bationo et a / . (1991) used Eq. (15) with m = 2, for comparisons of various PRs with soluble P fertilizers. Both equations assume a constant value of RR and in both cases RR
=
bpRlbsp.
(16)
Equations (13), (14), and (15) are strictly applicable only to the extreme situations of a constant SV or a constant RR and not to intermediate situations, which probably predominate. Barrow and Bolland (1990) have overcome these restraints by replacing the constant k (i.e., the SV) in Eq. (13) by a function whose value can decline as the application rate of PR increases. They used the function
k
=
(I
+
mxt)-)l,
(17)
where m and IZ are coefficients and xt is the amount of test fertilizer applied. Computer programs have been written specifically for fitting this type of model (Black, 1993).
C. RESIDUAL EFFECTIVENESS OF PHOSPHATE ROCKS Reviewing residual effects of PRs, Khasawneh and Doll (1978) concluded that the experimental evidence they presented did not confirm the common assumption that PRs have greater residual effects than soluble P fertilizers. On the contrary, it appeared that the residual effects of soluble P fertilizers were greater than those of PR in the first 3 or 4 years after application. In comparing the residual effects it must be remembered that with soluble fertilizers the residual effect derives from the soil-phosphate reaction products and the reaction of prime importance is conversion of P from labile to nonlabile forms. But with PRs, the PR-P needs to be released into solution before any residual effect can manifest itself. Therefore in the short term one is likely to find a poorer residual effect from PR application. The situation is different in coarse textured acid soils
126
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
where large and rapid leaching of soluble P will show PR giving a greater residual effect in the shorter term (Yeates and Clarke, 1993). Unlike a single application, when PR applications have continued over a period of several years, a large pool of undissolved PR can accumulate. The substantial release from this pool can result in a high residual value of PR in subsequent years. Residual effects of PR relative to those from SP therefore depend on, in addition to the rate of loss of P from the available soil P pool, the previous pattern of application, and the rate of dissolution of PR. Barrow and Campbell (1972) and Barrow and Carter ( I 978), using mathematical models, described the residual effect of soluble fertilizers in relation to freshly applied superphosphate. Probert (1985) presented a similar model, but extended it to include the measurement of the residual effect from fertilizers such as PRs from which a fraction of P is released in a given time. Building on these, Sinclair et al. (1993) and Sinclair and Johnstone (1995) proposed a compartment model to predict the residual effects for various combinations of the factors that influence the residual effect of fertilizers. The soil P part of the model is illustrated in Fig. 24. Fertilizer P applied to the soil enters the first compartment (F) which contains undissolved fertilizer P in the soil. On dissolution, P moves from F into A which is the pool containing plantavailable P. Phosphate is lost from this pool by immobilization andlor leaching processes in the soil and by nonreturn of a proportion of P taken up by crops. P , and P A are the amounts of P in compartments F and A at any time. K , and K , are the rate constants describing the transfer of P from compartment F to compartment A and the loss of P from compartment A , respectively. For simplicity, first order kinetics are applied to both processes. When the fertilizer applied is PR with a range of particle sizes its dissolution rate by the diffusion model (Section V) is given by an arcsin function. This could be approximated by a simple exponential for the first 90% or so of dissolution where the ratio of greatest to least sizes was <2 (Watkinson, 1994b). The differential equations describing the system are
dP,ldt
FertilizerP
I
-
= -
I
KIPF
1
(18)
.
1
I Pin soil I PF
Kl 'A
K2
b
Figure 24 Two compartment model of active P in soil (Sinclair et a / . . 1993).
PHOSPHATE ROCKS FOR DIRECT APPLICATION T O SOILS
127
Solution of these equations gives the recurrence relationships fF(t
+ 8t) = fF(t)e-fi'kl
(20)
+ 8f)= fA(f)e-"k2 + KIfF(t) [e--6rkl - e-s'k2] / (K2 - K l ) , (21) where t is the time since the last application of P to F. The average amount of fertilizer P in F in the time interval ( t , t application at time f is given by fF(f)(l - c K I ) / K I .
+ 1) after an (22)
Similarly the average amount of P in A in the time interval ( t , t + 1) after an application at time t is given by
+
eKz)/K, K I P F ( [ ) [ (-] e K I ) / K l - ( I - e K z ) / K 2 ] / ( K2 Kl).
fA(t)(l-
(23)
In applying the model in this review, r is in years. The DM production part of the model is based on the assumptions that DM yield must be nil when PA is nil and that DM yield is related to PA by a Mitscherlich response curve, i.e., DM yield = Y,,, (I - CPA)
(24)
or
RY/100
=
1 - CPA,
(25)
where Y,,, = DM yield when P is nonlimiting, C is a constant which defines the curvature of the response curve, and RY is the yield expressed as a percentage of Y,,,,. Numerical solutions can be derived from the above equations except in the case of identical values for K , and K 2 , in which case a very small adjustment in one of the values will allow an accurate solution to be obtained. Figure 25 shows predicted plant yield patterns following a single application of P fertilizers to a soil initially devoid of plant-available P. Yields are expressed as the percentage of yield when P is nonlimiting. At 15% per year loss from the labile P pool (Fig. 25A) the residual value of PR is predicted to exceed that from soluble P after 1 1 , 8, and 4 years for PR dissolution rates of 5, 10, and 40% per year, respectively. At 50% per year loss from the labile P pool the corresponding times are 5, 4, and 2 years, respectively (Fig. 25C). Chien and Hammond (1987b) reported that Gafsa, Sechura, and North Carolina PRs were inferior to soluble P for the first crop of beans in a field experiment on an Andisol in Colombia, but were as effective as soluble P for the second crop and superior for the third crop. These data are for single applications of P fertilizer in situations in which loss from the labile P pool was rapid (as evidenced by the rapid decline in residual effect of SP in both experiments). There-
128
S. S. S. RAJAN, J. H. WATKINSON,AND A. G. SINCLAIR 10080 -
60' 40 I
2o
1 / J
Ploss=15%lyr
0
100-
; 80.%
.-c
(II 5
a
.
60-
.
40-
' .
.
.
.
.
P loss = 50% I yr C PR dissolution
100-
-
80 -
5% 10%
60-
-6- 20%
40 20 -
o+ 0
2
4
6
8 1 0 1 2
Years Figure 25 Predicted relative yield (RY) of crop following single applications of 225 kg P ha- I P as soluble P fertilizer (SP) and PR to soils containing no plant-available P. Assumptions are that yields are related to plant-available P at midyear; the Mitscherlich response function applies; plantavailable P required at midyear for 90% RY is 130 kg ha-' (A.G. Sinclair, unpublished data, 1995).
fore these should be considered in relation to Fig. 25C. The good performance of the PRs for the first crop indicates rapid dissolution, in which case the model predicts the early development of a superior residual effect from PRs, as was observed. Bolland er al. (1988b) surveyed results on PR performance in field experiments in Australia. They concluded that during 8 years following single applications of PR and soluble P, the residual effect of PR improved relative to the residual effect of soluble P, but this was due to the decline in the latter. But even
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
129
8 years after application, PR was still less effective than soluble P. Other writers (Quin, 1988; Bolan et al., 1990) have pointed out that the data summarized by Bolland et al. (1988b) related largely to relatively unreactive PRs on soils with low PR solubilizing ability. The data surveyed by Bolland et a1 .would be fairly closely matched by a PR dissolution rate a little less than 5% and a 50% labile pool P loss rate as in Fig. 25C. Figure 26 shows predicted residual effects following 6 years of annual application of P fertilizers at rates of P calculated to maintain about 90-95% relative
Ploss=15%lyr 100-
8060'
40
-
PR dissolution Per Yr -5% 10% 20% -40%
*
20
-0-
80%
.
0
,
I
,
,
,
100-
E a ..9
80 -
60-
- 40Q
,
'
cT 20-
1
20 U
0
P loss = 25% I yr
P loss = 50% I yr "
'
2
'
4
'
6
.
'
8 1 0 1 2
Years Figure 26 Predicted relative yield (RY) of crop receiving soluble P fertilizer (SP) and PR annually for 6 years and no P fertilizer for a further 6 years. Assumptions as for Fig. 25. Annual application rates are ( A ) 25, ( B ) 37.5, and (C) 75 kg ha-1 (A.G. Sinclair, unpublished data, 1995).
130
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
yield when applied as SSP. These are considered to have been imposed on a soil with P status appropriate for about 90% relative yield. The data of Perrott et al. (1992) are for experiments in which the patterns of PR and SSP used correspond to those in Fig. 26A. PR dissolution rates averaged 30% per year (Watkinson and Perrott, 1993) and labile pool P loss rates can be estimated to be about 15-20% per year (Sinclair et a l . , 1994). For those rates the prediction is that the residual value of PR would be about the same as that from SSP in the first year following cessation of annual fertilizer application. Again there is reasonable agreement between observation and prediction. The patterns of residual value of Gafsa and Nauru PR reported by Scott and Cullen ( 1 965) are compatible with model predictions illustrated in Fig. 26B. In summary, experimental data reported since the review of Khasawneh and Doll (1978), as well as some earlier data, show that residual effects of PRs can exceed those of soluble P fertilizers, and are greatly influenced by the PR dissolution rate and the rate of loss of P from the plant-available P pool in the soil. A simple mechanistic model can apparently account well for observed residual effects.
VII. ECONOMICS OF USING PHOSPHATE ROCK FERTILIZERS Several approaches have been proposed for assessing the economics of using PR instead of soluble P. These may also be used for selecting between PR fertilizers. Engelstad (1 978) fitted polynomial functions to the experimental response curves for each fertilizer and used these along with the ratio of the price/kg of P to the valuehnit of crop to calculate the economically optimum fertilizer rates and the corresponding net returndha. PRs were then assessed on the maximum returns ha- I which they could achieve in comparison with soluble P. Engelstad also considered the situation where funds for fertilizer purchase were very limited and calculated the return per dollar spent on fertilizers applied at a low common rate. A sounder comparison would have been between fertilizers applied at a low common cost. Returns per dollar spent at a low rate of application gave a more favorable assessment of PRs relative to soluble P than maximum returns per hectare. Chien et al. (1990) suggested that the SV of a PR relative to a standard fertilizer such as TSP could be used in a simple way to decide which fertilizer is more profitable to use. If the SV is greater than the price ratio (i.e., the ratio of the price kg P- I in PR to that in TSP) then the PR is more profitable than TSP, and vice versa. These calculations relate to single applications of PR and ignore the value of
PHOSPHATE ROCKS FOR DIRECT APPLICATION T O SOILS
13 1
the PR remaining undissolved in the soil at the end of the growth period which will subsequently dissolve and benefit later crops. A conceptual framework for economic evaluation of PRs which takes into account the residual value of fertilizers following a single application has been proposed by Sidhu ( 1978). However, in practice one is likely to be less interested in single applications than in regular, e.g., annual, applications. If ultimately all PR applied to the soil dissolves and, once dissolved, P from PR has the same agronomic value as P from soluble P, then economic evaluations must give equal long-term value to P from both sources. PR should therefore be penalized only because of its slower release of P. The use of PR instead of soluble P should therefore be considered as an investment. Sinclair et al. (1993a) considered that these conditions were obtained when a reactive PR was applied annually to permanent pastures in New Zealand. They assessed the economics of replacing annual applications of soluble P with sufficient PR to maintain an identical level of production. This would require establishing and maintaining a pool of PR in the soil large enough to release annually the same amount of P that would have been supplied in the annual application of soluble P. Assuming a first order dissolution rate model, the size of the pool of PR required equals the annual application rate of soluble P multiplied by the reciprocal of the fraction of PR dissolved per year. Thus, if for example the dissolution rate of PR is 33.3% per annum, annual application of X kg P ha-' as soluble P could be replaced by PR applied at 3X kgP/ha P in Year 1 and X kg P-I thereafter. This represents an initial investment equivalent to the difference in cost between 3X kg P as PR and X kg P as soluble P, followed by annual savings equal to X times the difference in cost per kilogram P between soluble P and PR. In general terms annual savings from using PR instead of soluble P expressed as a percentage of the initial investment is 100
X
( 1 - R)/(R/F - 1)
forR > F,
(26)
where F is the fraction of PR in the soil dissolving per year and R is the ratio of the cost/kg of P in PR to the cost/kg of P in soluble P. If F is equal to or greater than R there is no initial investment cost in replacing soluble P with PR, so the replacement is immediately profitable. Figure 27 shows the percentage return on investment for various combinations of F and R. In permanent grasslands on generally moist, slightly acidic soils in New Zealand F is approximately 0.3-0.4 (Sinclair et al. 1993), so return on investment would exceed 10% with R values < 0.85. A lower return on investment generally would not be considered worthwhile, since there are likely to be alternative types of investment which would yield a better return. Furthermore the farmer may perceive that there is a greater risk involved in the use of PR than in conventional P fertilizer, thus requiring a higher predicted return on investment. Because this type of economic analysis recognizes the subsequent value of
132
S. S. S. RAJAN, J. H.WATKINSON,AND A. G.SINCLAIR 8o 1
I
70 -
C
0
0.1 0.2 0.3 0.4 0.5 Fraction of PR dissolved I year
Figure 27 Effect of PR dissolution rate and ratio ( R ) of cost of P in PR to cost of P in soluble P fertilizer (SP) on the return on investment for replacing SP with PR (Sinclair et a / . . 1993).
undissolved PR in the soil the conclusions are much more favorable for PR use than the methods of Engelstad (1978) and of Chien et al. (1990) would indicate. The above approach to economic analysis is valid only when the dissolution of PR is approximately first order and when P dissolved in the soil from PR is equally effective for plant growth as P from soluble P (i.e., response curves to annual applications of PR and soluble P are identical in the long term). Where the maximum yield that can be achieved with PR is less than that with soluble P economic evaluation could be based on mean long-term response curves using the methods described by Engelstad (1978). If lime is beneficial, the liming effect of PR should be included in economic analysis. PR absorbs two protons per atom of P in the process of dissolution. Thus 31 kg P as PR has a liming effect equivalent to 100 kg CaCO,. The liming effect of PR relative to TSP has been clearly demonstrated in long-term field trials on pastures (Sinclair et al., 1993). If PR is to be used in place of single superphosphate, the sulfur (S) content of superphosphate must be recognized wherever S is required. The price of PR must be raised to include the cost of augmenting it with the minimum requirement for S which would have been met by the S content of single superphosphate.
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
1 33
VIII. SOIL TESTING WJ3ERE PHOSPHATE ROCKS ARE USED Use of PRs which are sparingly soluble brings in a new dimension in the estimation of plant available P and also in interpretation of data in terms of fertilizer requirements. This is because soils which have received PR can contain undissolved PR which usually continues to dissolve and be available to plants during the cropping season under consideration. Thus the soil test should be able to give an estimate of the availability of P from undissolved PR in addition to the availability of soil pool P at the time of testing. Also, because of their slowrelease nature, the reaction products of PR with soil and their availability to plants may differ from those obtained with soluble P fertilizers (Chien et a/., 1987b). Numerous techniques have been proposed to obtain an estimate of plant available P in soils including extraction with (i) chemical solutions, (ii) ion exchange resins, (iii) iron-impregnated paper strips, and (iv) ion exchange with isotopically labeled phosphate ions. Of these, chemical extraction procedures have been used widely, but ion exchange methods are gaining popularity.
A. CURRENT RESEARCH 1. Chemical Extraction
Chemical extraction procedures are rapid, inexpensive, and amenable to automation and handling a large number of soil samples. The extractants used in soil P tests can be grouped under the following general categories: water and weak electrolytes (e.g., 0.01M CaCI, solution), alkaline solutions (Olsen bicarbonate solution and its modifications), weak acids with complexing anions (Bray 1, acetic acid, citric acid), and strong acids (Bray 2, Truog). In this review we have not discussed all the chemical extractants in the literature. Instead we have summarized the reasons underlying the attempt to use different types of chemical extractants in soils where PR has been applied and briefly summarize the results (Table 11). Of the soil tests employed to estimate available P, the Olsen bicarbonate test or its modifications (Colwell, 1963) have been used widely. Because of its success as a P test in soils treated with soluble fertilizers this technique has also been evaluated in soils treated with PR. However, apatites are not soluble in NaOH or NaHCO, (Williams, 1937; Kumar et al., 1991; Perrott er al., 1992; Saggar er a/., 1992), and therefore Olsen bicarbonate solution cannot be expected to predict potential P release from the residual PR or PR to be applied. In fact the
134
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR Table Il Some Soil Tests Used To Estimate Plant Available P Soil test
SoiVsolution ratio
Shaking time 30 min
+
1:20 1:lOO 1:7
60 s
Olsen et al. (1954) Colwell (1963) Bray and Krutz
+ 0. I M
1:7
40 s
Bray and Krutz
1:2OO
30 min
Truog (1930)
Extractant
Olsen Colwell Bray- I
0.5 M NaHCO,, pH 8.5 0.5 M NaHCO,, pH 8.5 0.03 M NH,F 0.025 M HCl
Bray-2
0.03 M NH,F
Citric acid Ammonium oxalate
HCI 0.001 M H,SO, buffered with (NH4)*SO,, pH 3 1% citric acid solution 0.1 M NH,-oxalate
Calcium acetate + calcium lactate
0.1 M Ca-lactate 0.1 M Ca-acetate, pH 4.1
'hog
16 h
Reference
( 1945)
(1945)
1:lO 1:25
7 days 2h
Dyer ( 1894) Joret and Hubert
1:20
2h
Schiiller (1969)
(1955)
+
literature shows that in soils which received annual applications of a reactive PR, Sechura PR, P extracted by the Olsen solution was closely correlated with the soil PR reactant product, Fe-P and AI-P (Perrott et al., 1992) (Fig. 28). When there is an appreciable amount of PR present in soil, Olsen P values underestimate plant yield in comparison with soluble P fertilizers (Fig. 29) (Cornforth et al., 1983; Bolland et al., 1988a; Yost et al., 1982; Rajan et al., 1991b) . However, when the PR applied is of the crandallite/millisite type (Fe-P and AI-P) the alkaline Olsen extract may dissolve potentially unavailable PR and thus overestimate available P (Bolland et al., 1994). Separate calibration curves with yield data may be needed for both kinds of PR and soluble P fertilizer treatments. Rajan et al. ( 1 99 la,b) obtained significant correlations between bicarbonateextractable P and plant yield at different rates of PR application under field conditions, provided the soil samples were taken at least a few months after PR application (Rajan et a/., 1991a,b). This would imply that the bicarbonateextracted P was proportional to the PR already dissolved which in turn apparently correlated to the relative amounts of PR dissolved later on and was available to plants. Perrott et a/. (1992), from results at four New Zealand sites under permanent grasdlegume pasture, suggested that multiplication of Olsen P test by 1.69 for soils with a history of PR use (six annual applications) was appropriate to equate the calibration curve with that for soluble P treatment. However, this correction
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
135
40
P
30
l J '
0
100
260
300
400
560
Alkali-P (mg kg-' soil) Figure 28 Relationship between Olsen P and alkali extractable P (Fe-P + ALP) in TSP treated
(0) and Sechura PR treated (0 ) soils in New Zedland (Benio low P retentive and Manapouri high P retentive soils) (Perrott ef ul.. 1992).
factor will vary with time as the amount of PR in soil varies with time depending on fertilizer application practice. Mackay el a / . (1984a) indeed found that 3 years after fertilizer application, soils treated with Sechura PR and SSP could be described using a single regression relationship, without any multiplication factor for the soil test values obtained from PR-treated plots. It has been suggested that whenever slow-release fertilizers containing PRs are applied other extractants capable of dissolving fertilizer apatites, such as Bray y = -3.22 + 0.39~,R2 = 0.93 (PR) y = 2.43 + O.O6x, R2 = 0.92" (MCP)
2 0 20 40 60 80 100 Bicarbonate extractable P (mg L-l soil)
Figure 29 Relationship between pasture yield and Olsen P values in plots treated with Sechura PR and a commercial monocalcium phosphate (Rajan ef a / . . 1991b).
136
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
extractants or weak acetic or citric acid, may be of greater value than the Olsen extractant (Rajan, 1982). Chien et al. (1978) concluded that the P extracted from acid soils treated with reactive PRs was partially derived from the residual PR as well as from the reaction products. The amount extracted varied with the reactivity of PRs. The question arises whether the amount of PR-P extracted will be in proportion to that to be dissolved and available to plants during the crop growing season, or during a year in the case of permanent pastures. Results indicate that Bray 1 underestimates P availability in PR-treated soils, relative to P test values obtained in soils where soluble P fertilizers have been used (Reinhorn and Hagin, 1978). Some researchers have tried to overcome this problem by using more acidic extractants such as Bray 2, Truog and Dyer citric acid solutions (Bationo et a l . , 1991; Bolland e t a / ., 1994). These extractants, however, overestimated available P presumably by dissolving a greater amount of PR than would dissolve in soil in a cropping period and contribute to plant uptake. The above evaluation leads to the conclusion that in soils where substantial amounts of residual PR-P contribute to P uptake by plants, because of PR dissolution during the cropping period, weak electrolytes and alkaline solutions will underestimate available P. Weak acids may extract a greater amount of P but not necessarily correlate closely with plant available P, and strong acids may overestimate available P. Consequently one often comes across conclusions similar to that stated by Bolland er a / . (1988a) that “All soil tests were equally predictive of yield but usually for each soil test separate calibrations between yield and soil test values were required for the different fertilizers for each combination of fertilizers and plant species for each year.”
2. Ion Exchange Resins Use of ion exchange resins to estimate plant available P was first proposed in the 1950s (Amer et a l . , 1955). Ion exchange resins either used as beads (Sibbesen, 1983; Dalal, 1985) or more conveniently as membranes (Saunders, 1964; Saggar e t a / ., 1990; Schoenau, 1991) have been found to be effective in estimating plant available P for a wide range of soil conditions where soluble P fertilizers have been applied (Sibbesen, 1978). In his review of various soil tests, Sibbesen concluded that the anion exchange resin (AER) method is generally better than the chemical extraction procedures (Sibbesen, 1983). Application of AER as a P test in soil containing PR was attempted in a greenhouse study by Van Raij and Van Diest (1980). They observed a poor correlation between AER P-test values and P uptake by soybean, and attributed this to the very short growing period of 35 days allowed for the crop. The authors found that P in solution, as estimated by a CaCI, solution, was more relevant. A comprehensive study on the use of AER (HC0,- form) with or without
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
13 7
CER (Na form) both in the form of membranes was conducted by Saggar et al. (1990, 1992a,b). These authors found that AER extracted less P from PR or soil PR mixtures than a combination of AER and CER membranes The greater extractability when CER was included was attributed to its capacity to act as a sink for Ca of the PR. Importantly, the mixed resin of AER + CER extracted a representative portion of residual PR in accordance with the reactivity of PRs and dissolved P. These authors evaluated the mixed resin technique in a glasshouse study using four soils of contrasting P sorption capacities and containing soluble P and residual PRs of different reactivities. The reactivity of the PRs ranged from very reactive Sechura (Peruvian) to unreactive Florida (U.S.A) PR. They found that the mixed-resin P was a good predictor of plant available P in soils to which monocalcium phosphate or PRs have been applied (Fig. 30) (Saggar et al., 1992a,b). The mixed-resin test has been evaluated recently using field soils collected from several sites to which either a reactive PR (Sechura) or TSP has been applied over a period of 6 years. The results (Saggar, unpublished data, quoted by Perrott et al., 1993) showed that this test has potential as a routine soil P test, irrespective of whether soluble or PR fertilizers have been used (Fig. 3 1).
3. Iron Oxide Impregnated Paper Strips of iron oxide-impregnated filter paper developed by Sissingh ( 1983) were used as a sink to adsorb P from a suspension with soil in a 0.01 M CaCI, solution. After separation, P adsorbed on the iron oxide was determined after
80 s p.-x 60h Y
0
2
3 40-
c
2000
10
20
30
40
50
AER + CER P (rng kg-l soil)
+
Elgure 30 Relationship between plant dry matter yield and a mixed resin (AER CER) extractable P in four soils treated with PRs. The regression line is the best tit for monocalcium phosphate treated soils (Saggar el ul., 1992b).
138
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR l2
1
10
-
h
v-
'% 8 v-
c m s 6-
rr
.-(u
5
4-
2-
0 TSP treated plots + control 0 SPR treated plots
0
1
0
30 40 AER + CER P (rng k9-l soil) 10
20
Figure 31 Relationship between pasture yield and a mixed resin (AER + CER) extractable P in plots treated with either TSP or Sechura PR (Sagpar et a / . . unpublished data; Perrott ef a / . . 1993).
extraction with H2S0, (Hammond er al., 1985, Menon et a / . , 1989b). The principles underlying this technique (usually refered to as the Pi test) are similar to those of tests using anion exchange resins. The mechanisms of P extraction by the iron oxide impregnated paper have been discussed in detail by Perrott et al. (1993). In brief, when a soil suspension is shaken with the iron oxide strip, because of the high affinity between phosphate ions and oxides and hydroxides of iron, P in the solution is adsorbed by the iron oxide. This results in depletion of P in solution which promotes the release of P mainly from the soil labile pool by desorption. In contrast, AER exchange is nonspecific and dependent only on ionic charge. It has a much lower affinity if P exists only as the monovalent species, H 2 P O ~Sharpley . (1991) studied the relationships between P extracted by iron oxide strips and the P fractions in a range of calcareous and noncalcareous soils. He concluded that the iron oxide strip extracted P equivalent to that by AER, but very little of sparingly soluble soil ALP, Fe-P, or Ca-P compounds. In soils with a history of soluble P fertilizer use, the Pi test values have been found to give a higher correlation with plant dry matter yield or P uptake than chemical extraction techniques (Menon et al., 1989; Menon, 1990). When compared with the AER method the Pi test seems to be comparable (Lin er al., 1991). The use of Pi in soils with a history of PR use has not been adequately tested. In one study, which had as one of its objectives an evaluation of the Pi test for soils containing PR, the PR used was a Florida rock (Menon et al., 1988). The unreactive nature of the PR and the very short duration of the crop growth (6 weeks) would have resulted in dissolution of very little PR during the crop growth and therefore made insignificant contribution to the plant-available pool of P. Under
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
139
such conditions the source of P for the plants, in PR treated soils also, would have been mostly from the soil pool P. This amounts to evaluation of P tests under conditions similar to that existing in soils where only soluble fertilizers have been used. In another study (Kumar et al., 1992b) also short-term crops were grown. Without knowing either how much PR dissolved during this period or its contribution to plant uptake, a reliable assessment cannot be made of the Pi test for application in soils where PR dissolution is a significant factor. Perrott and Wise (1993) identified three problems with the technique itself: (i) soil particles which adhered to the filter-paper surface, even after careful rinsing, contributed 0-85% to the Pi test; (ii) suspension pH was lowered in some cases by 0.5-1.2 units, enhancing dissolution of phosphate rock during shaking; and (iii) PR particles adhered to the strip, contributing to the Pi values during extraction.
4. Isotopic Ion Exchange The isotope exchange method measures the amount of phosphate in soil that is isotopically exchangeable with 12P-labeled phosphate in either of two ways: by measuring the change in specific activity of the solution in a soil suspension (McAuliffe et af., 1947) or by measuring the difference in specific activity between that of labeled P equilibrated with soil and that of P in a plant later grown in it (Larsen, 1952). Isotopic exchange is usually a rapid process in the first few hours but the reaction continues for weeks. Thus the amounts of exchangeable P measured by both methods are a function of time and for comparative purposes the amount estimated is that exchanged within a specified time. Because these methods, whether employed as a single measurement or as a function of time (Morel and Fardeau, 1990; Di et al., 1994; Frossard et af., 1994; Morel et al., 1994; Zapata et af., 1994), estimate a portion of adsorbed P that correlates with plant-available P, they are similar to the AER technique. As such, when used in soils containing PR, the limitations that apply to the AER technique will also apply to the isotopic technique. The other restrictions as a routine test are the safety requirements in handling isotopes and the likely higher costs involved.
B. FUTURERESEARCHNEEDS As one reviews the numerous publications on P tests one becomes aware of the deficiencies in the design of experiments intended to evaluate the various tests for use in soils with a history of PR application. .Inadequate experimental approaches appear to have resulted in not obtaining significant correlations, albeit at different levels, for a number of P tests. Ideally the following conditions should be satis-
140
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
fied for rigorous evaluation of P tests: (i) at the time of testing one set of soils should have a history of only soluble fertilizer use and another set should contain both adsorbed P (which is normally the case) and undissolved PR; and (ii) the cropping experiment should be such that the amount of PR dissolving and contributing to P uptake by the plants during the crop growth season should exceed (preferably by two times) the coefficient of variation obtained when the yield or P uptake function is plotted against soil test values. Information on the amount of PR that has dissolved during the cropping season will also render the results more convincing. When crops are grown for a short period of only a few weeks it is probable that very little PR is dissolved. Not surprisingly, in such circumstances even an alkaline solution such as Olsen extract or weak electrolyte solutions (e.g., CaCI,), which extract exclusively or predominantly adsorbed P, may give significant correlation with plant yield parameters. In the same way when unreactive PRs are used, their contribution to P uptake may be in such a small proportio; compared with soil pool P to distinguish between different soil testing methods. In such cases highly acidic solutions (Bray 2 or Truog solution) may overestimate available P. These solutions can dissolve unreactive PRs which may not dissolve in soil during the cropping duration. To estimate available P in soils containing PR two procedures appear to be promising. One is empirical and is based on the mixed AER + CER technique and the second one is based on a mechanistic model of PR dissolution in soil and its availability to plants. The research procedure adapted by Saggar et al. ( 1992a,b) is probably close to meeting the two criteria mentioned earlier for evaluating P tests rigorously in soils containing PR. In this study, since the mixed resin of AER + CER extracted a representative portion of residual PR in accordance with the reactivity of PRs and also dissolved P, this technique needs to be evaluated further. In addition, it would appear that a single calibration is possible between P extracted and yield function irrespective of whether the soils received soluble P fertilizer or PRs. As mentioned in other sections of this review PR availability to plants is a property not only of the reactivity of the PR but also of soil and climatic conditions and the type of plant species. Therefore, greenhouse studies should be conducted initially to determine which soils can be grouped together. This should be followed by evaluation under field conditions on sites covering a range of fertilizer histories. The ultimate aim will be to provide information on P status and translation of that into fertilizer recommendations for specific soil, cropping, and climatic situations. The AER/CER method needs to be tailored so that the P extracted reflects, in addition to plant-available P from the soil pool-P, the PR-P dissolved during the cropping season for short-term crops, or annually for permanent pastures. If a very short extraction time is used a negligible amount of even
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
141
reactive PR may dissolve, and after a very long time all of the reactive PR will dissolve. An additional advantage of the mixed resin method is the possibility of using this technique not only to determine available P but as a comprehensive method to determine multielements (Van Raij et al., 1986 ; McLaughlin et al., 1994). The mechanistic approach to determine soil P status and availability of P to plants from the PR residues includes: (i) estimation of labile P, (ii) estimation of the residual PR in soil and their particle size distribution, and (iii) prediction of the amount of PR that will dissolve during the crop growth season and the amount that becomes available to plants. Perrott and Wise (1995) proposed a relatively simple alternative to the sequential inorganic fractionation technique to measure PR residues in soil (Section 111). According to these authors this method also had the advantage of not requiring a sample of unfertilized soil to correct for background levels of acid P largely due to native fluorapatite. Once the amount of residual PR in soil is estimated, the next step is to predict how much of the PR will dissolve in the cropping season. Dissolution of PR can be predicted using the comprehensive but rather complex PR dissolution model of Kirk and Nye (1986a,b). Alternatively a simpler model (Watkinson, 1994a,b) can be used to predict the rate of dissolution of PR (Rajan and Watkinson, 1988). If it is assumed that the effectiveness of the PR is equal to the proportion of PR dissolved relative to soluble fertilizers (Section, Vl) (Sinclair et al., 1993; Watkinson and Perrott, 1993) one can predict the fertilizer requirement or the agronomic response. This assumption encompasses two aspects: there is no difference between the sources of P once P is dissolved in the soil (i) in P uptake by plants and (ii) in P immobilization or leaching in the soil. Luxury P uptake by plants immediately following addition of Soluble P could give rise to greater P removal in plant material than occurs with the gradual release of P from PR. Under cropping, however, luxury P uptake from Soluble P at an early stage in crop growth could be followed by P translocation at a later stage giving the same overall efficiency of P use by plants as can be provided by slow release P fertilizer. Bolland et al. (1988) have reported common relationships between plant yield and plant P concentration for SSP and PR with barley, oats, triticale, and clover, thus demonstrating equal internal efficiency of P derived from the two sources. Under permanent pastures a greater forage production per unit P uptake in reactive PR applied than in soluble P applied plots was reported (Mackay et al., 1984; Rajan et a/., 1991). On the other hand, data from long-term field trials (A. G . Sinclair, personal communication) with perennial pastures suggest no difference between soluble P and PR in the relationship between pasture yield and plant P uptake. It appears that provided excessive amounts of P fertilizer are
142
S. S. S. RAJAN, J. H. WATKINSON, AND A. G . SINCLAIR
avoided, only small differences will arise between soluble P and PR in the efficiency with which plants use the P which they take up. The other potential difference in P efficiency between the two sources could arise from differences in their reaction with soil: diffusive migration of adsorbed P into soil (Barrow, 1985; Parfitt, 1989), decomposition of hydrous oxides and aluminum-silicate clays (Rajan, 1975; Rajan and Fox, 1975), and precipitation of sparingly soluble compounds (Chien et al., 1987; Pierzynski, 1990). Such differences are to be expected in view of the very great differences in P concentration, cation concentration and pH surrounding dissolving particles of SP and PR in the soil. The results of Rajan (1991a) and Rajan et al. (1993) suggest a greater efficiency of PR, but further studies are needed to establish different nutrient efficiencies. The fact that there is good agreement between observed and predicted values of SV (Fig. 22) indicates the usefulness of the models to predict the agronomic effectiveness (Sinclair et a / ., 1993; Sinclair and Johnstone, 1995). However, this does not necessarily establish the validity of the underlying relation. More expermental evidence for similar agronomic efficiency of P derived from the dissolution of both fertilizer forms is required before predictions of agronomic performance based on PR dissolution rates can be accepted without reservation in all situations. The mechanistic model based approach requires detailed information on the rate of dissolution of PRs as affected by several factors, namely soil properties, climatic conditions, plant species, and management practices. Although this may appear daunting, modem computer technology has opened that possibility, so that predicting fertil'izer requirement based on this approach may be viewed as a practical proposition in the near future.
IX. AMENDMENTS TO PHOSPHATE ROCKS There are only a limited number of climatic and soil situations in which PRs will be sufficiently reactive for use as direct-application fertilizers, especially for fast-growing annual crops. In view of this, numerous studies have been conducted amending PRs to increase their immediate P availability and also to possibly enhance their rate of dissolution after application to soil. In this regard three processes have often been investigated: (i) composting with organic manures, (ii) combining PR with elemental S (So),with or without Thiobaciffusspp culture, and (iii) partially acidulating with mineral acids or compacting with superphosphate. The principle underlying the first two processes is production of organic or
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
143
mineral acids which will create a localized high acidity in the immediate vicinity of PRs and also complex Ca in the case of some organic acids. Partial acidulation results in conversion of a portion of the PR to a soluble form.
A. COMPOSTING WITH ORGANIC MANURES Cornposting of PRs with agricultural wastes is known to increase the solubility of PRs (Bangar, 1985; Mey e t a / ., 1986; Mishra el a/., 1986; Mishra and Bangar, 1986; Kothandaraman, 1987; Singh and Arnberger, 1990) The extent of solubilization of a given PR varies with the kind of waste and the rate of decomposition (Bangar et a / . , 1985; Mahimairaja et a / . , 1995). For example, Bangar et a/. (1985) reported that composting unreactive Mussoorie PR with farm wastes (chopped grasses and tree leaves) increased the citric solubility of the PR. Their results from a small plot field experiment indicate that the product applied on an equivalent total P basis gave grain and straw yields of clusterbean (Cyrnopsis tetragonoloba L) equal to those on application of SSP. Similar results have also been obtained by Mishra et a/. (1984) on red gram (Cajanus cajan L). The increased P availability from the phospho-compost could have resulted both from conversion of PR-P to water-soluble form and a greater efficiency of the dissolved P in terms of its availability to plants (Khanna et al., 1983). Cornposting PR with poultry manure may not be an attractive option because poultry manures contain large amounts of CaC03 and other basic compounds (Mahimairaja et a / ., 1995). Although phospho-composts contain low amounts of P (e.g., 3.4%; Mishra et a/., 1984) they may still be favored in organic farming systems or where farm wastes are to be utilized effectively.
B. PHOSPHATE ROCK-SULFUR ASSEMBLAGES Experiments were conducted from as early as 1916 (Lipman and Mclean, 1916, 1918) on admixing So with PRs to increase the availability of PR-P. The topic was briefly reviewed by Kucey et a / . (1989). The principle behind the process is that soil microbes oxidize So to H,SO,, which in turn dissolves the PR particles which are in close proximity to So. Germida and Janzen ( 1993) recently presented a comprehensive review of the factors affecting the rate of oxidation of S o . Of the soil bacteria which are able to oxidize S o the chemoautotrophic bacteria Thiobacillus thioxidans and Thiobacillus thioparus are considered to be the most important ones (Starkey, 1966). Inoculating soils already containing Thiobacillus spp. has not been found to be of additional benefit (Rajan, 1982).
144
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
On the other hand, soils with inadequate bacterial activity will benefit from inoculation of the PR-So with the Thiobacillus spp. bacteria. The term “biosuper” was introduced to describe PR-So-inoculated product (Swaby, 1975, 1983). Factors that influence the effectiveness of PR-So assemblages are (i) proximity of PR particles to So (ii) reactivity of the PR, (iii) mode of application, (iv) PR:So ratio, and (v) type of crops. The effectiveness of So will be maximum if the So particles are in intimate contact with PRs since this will facilitate the reaction of the H,SO, produced on the PR. For this reason PR-So has been used after cogranulation (Swaby, 1975, 1983; Rajan, 1983). The almost complete ineffectiveness of So when applied without mixing with PR has been demonstrated (Rajan, 1983). Application of either unreactive or reactive PRs in the form of PR-So granules increases their dissolution in soil and availability to plants. However, when the PR used is unreactive the agronomic effectiveness of PR-So may not equal that of soluble fertilizers, such as superphosphate (Rajan, 1982; Swaby, 1983; Loganathan et al., 1994). This is unlike the results where the PR-So product is prepared from reactive PRs (Rajan, 1982). Application of PR as a band can be expected to conserve the H2S0, produced and permit maximum acidulation of the PR. On the other hand, this may restrict the volume of P-enriched soil that is available for root interception. Studies conducted comparing mixing versus layer application of PR-So products showed that at lower rates of application layering is preferred whereas at higher rates of application mixing gave greater yields of ryegrass (Rajan, 1983). In addition to the less capital intensive technology an advantage of PR-So assemblages is the flexibility it offers in altering the PR:S ratio according to the pH and the P and S nutrient requirement of a given soil. PR-Su of lower sulfur content will be suitable for soils of greater native acidity and vice versa. Attoe and his associates (Attoe and Olson, 1966; Kittams and Attoe, 1965; Nimgade, 1968) found that products containing a PR:S ratio of 1 : I were as effective as superphosphate for ryegrass in soils of pH 6.6 or greater. They used an unreactive Florida PR. Swaby (1983) concluded that PR:S ratio of 5: 1 was effective in areas where there was adequate rainfall (>635 mm). This ratio is similar to the proportion in which So is used as H2S0, to make SSP. The greenhouse study results of Rajan ( 1983) showed that PR-So products of 7: 1 PR:S ratio gave yields of ryegrass equal to an application of SSP (Rajan;1983). He used a volcanic ash soil of pH(water) 5.6 and a reactive Sechura PR. The experiments referred to above suggest that PR-So products, especially those prepared from reactive PRs, are effective for long term crops such as pastures. Extrapolating these results, one would expect similar results with other crops such as plantation crops where a low amount of P may be required over a long period of time. The effectiveness of PR-So on short-term crops is still unknown. Whereas Swaby ( 1 983) concluded that biosupers were inferior to SSP
PHOSPHATE ROCKS FOR DIRECT APPLICATION T O SOILS
145
for fast-growing crops, it is noteworthy that unreactive PRs were used in his studies. PR-So assemblages, especially those without inoculum, are attractive because (i) the production is not capital intensive, (ii) they enable flexible P:S ratios, (iii) they use low-grade PRs which may be unsuitable for making soluble fertilizers, and (iv) they behave as a controlled-release P and S fertilizer. However, the product has not been commercialized for want of a suitable granulation process. Recently, acceptable-size granules of PR-So have been prepared on a small scale in Thailand (D.E. Higgins, personal communication). Studies also need to be conducted on the agronomic effectiveness of PR-So granules on short-term crops. The product should be prepared using reactive PRs and applied preferably I month before crop sowing to ensure oxidation of 3’ and reaction of the H,SO, on the PR, and therefore a ready supply of soluble P for the plants. An alternative to advanced application is to preincubate PR with So and a Thiobacilli source (Ghani et a / ., 1994).
C. PARTIALLY ACIDULATED PHOSPHATE ROCKS Partially acidulated phosphate rocks (PAPRs) are PRs which have been acidulated usually with sulfuric or phosphoric acids with less than the stoichiometric quantities of acid needed for making SSP or TSP. The products usually contain a part of the P as monocalcium phosphate, the proportion of which depends on the level of acidulation, and the rest as unreacted PR. Small amounts of dicalcium phosphate may also be present. Products similar to PAPRs can also be prepared by cogranulating or compacting soluble P with PRs. Comprehensive reviews on PAPRs have been published recently by Hammond et a / . (1986), Stephen and Condron ( 1986), Bolan ef d.(1993), Hagin and Harrison, (1993) and Rajan and Marwaha (1993). In the field, partial acidulation has been found to enhance the dissolution of the PR component in PAPR, compared with that applied directly (Rajan and Watkinson, 1992). In this study the PAPR was prepared by partially acidulating a reactive North Carolina PR with phosphoric acid. The greater dissolution of PR was attributed primarily to increased root proliferation caused by the soluble P and the ensuing increase in exploitation of PR-P. Evidence was also presented of a greater rate of dissolution of PR component with increasing levels of partial acidulation. This was explained by the soil surface being initially “saturated” with P by the soluble P component, allowing the unreacted PR to dissolve at an effectively greater mean diffusion coefficient (Watkinson, 1994). The reviews referred to above show that PAPRs prepared using H,SO, directly by partial acidulation or by cogranulating PR with SSP, not only may not enhance PR dissolution, but may actually depress it. This has been attributed to CaSO,
146
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
present as coatings on the PR particles (Hammond et al., 1980). Recent studies using electron microscopic and energy-dispersive X-ray diffraction techniques have shown that the CaSO, coatings may delay disintegration of the PAPR granules and therefore dispersion of the PR component for maximum contact with soil acidity (SSS Rajan, personal communication).
X. CONCLUDING REMARKS Because of their controlled-release properties PRs are ideally suited for longterm crops such as permanent pastures and plantation crops. In very acidic soils they may be effective even on seasonal crops. On long-term crops there is also likely to be a time lag between the time of application of PR and its adequate availability to plants. The time lag can, however, be shortened by amending PRs; it also overcomes the need for greater soil acidity for effective use of PRs on seasonal crops. Application of PRs could be more widespread if the resulting yield increases are predictable and profitable. This would aid in the economic development of countries, especially those with indigenous PR resources, and also minimize pollution in industrialized countries. To achieve predictability of PR effectiveness a network of long-term experiments should be established at well characterized sites and the rate of PR dissolution and agronomic yield parameters should be measured. Further development and application of models similar to those described in this review should be able to relate soil and site properties with dissolution of PRs and their agronomic and economic effectiveness.
ACKNOWLEDGMENTS The authors are grateful to Dr.K.W. Perrott for useful comments on the manuscript.
REFERENCES Aguilar, A., and Van Diest, A. (1981). Rock-phosphate mobilization induced by the alkaline uptake pattern of legumes utilizing symbiotically fixed nitrogen. Plant Soil 61, 27-42. Alston, A. M.. and Chin, K. W. (1974). Response of subterranean clover to rock phosphates as affected by particle size and depth of mixing in the soil. Ausr. J . h p . Agrir. Anim. Husb. 14, 649-654. Amer, F., Bouldin. D. R., Black, C. A., and Duke. F. R. (1955).Characterization of soil phosphorus by anion exchange resin adsorption and 32Pequilibrium. Pluiir Soil 6, 391-408. Anderson, D. L., Kussow, W. R., and Corey, R. B. (1985). Phosphate rock dissolution in soil: Indications from plant grwoth studies. Soil Sri. Sor. Am. J . 49, 918-925.
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
147
Anderson, G. C., and Sale, P. W. G. (1993). Application of the Kirk and Nye phosphate rock dissolution model. Fert. Res. 35, 61-66. Archer, F. C. ( 1978). Comparison of different forms of phosphate fertilizers 11. Grassland. J. SoilSci. 29, 277-285. Attoe, 0.J., and Olson, R. A. (1966). Factors affecting rate of oxidation in soils of elemental sulphur and that added in rock phosphate-sulphur fusions. Soil Sci. 101, 317-324. Axelrod, S . , and Greidinger, D. (1979). Phosphate solubility tests-Interference of some accessory minerals. J. Soil Sci. 30, 153-157. Baifan, J., and Yichu. G. (1989). A suggested fractionation scheme of inorganic phosphorus in calcareous soils. Fert Res. 20, 159-165. Bangar, K. C., Yadav, K. S., and Mishra, M. M. (1985). Transformation of rock phosphate during cornposting and the effect of humic acid. Plant Soil 85, 259-266. Barber, S. A. (1984). “Soil Nutrient Bioavailability. I . A Mechanistic Approach.” Wiley, New York. Bardiya, M., and Gaur. A. (1972). Rock phosphate dissolution by bacteria. Indian J. Microbiol. 12, 269-27 I . Bardiya, M. C., and Gaur, A. C. (1974). Isolation and screening of microorganisms dissolving lowgrade rock phosphate. Folio Microbiol. 19, 386-389. Barnes. J. S.. and Kamprath, E. J. (1975). Availability of North Carolina rock phosphate applied to soils. Tech. Bull. 229, 3-23. Barrow, N. J. (1974). The slow reactions between soil and anions: I . Effects of time, temperature, and water content of a soil on the decrease in effectiveness of phosphate form plant growth. Soil Sci. 118, 380-386. Barrow. N. J. (1980). Evaluation and utilization of residual phosphorus in soils. In “The Role of Phosphorus in Agriculture” (F. E. Khasawneh, E. C. Sample, and E. J. Kamprath, Eds.), pp. 333-343. Am. Soc. Agron., Madison, WI. Barrow, N. J. (1985). Comparing the effectiveness of fertilizers. Fert. Res. 8, 85-90. Barrow, N. J . , and Campbell. N. A. (1972). Methods of measuring residual value of fertilizers. Aust. J. Exp. Agric. Anim. Husb. 12, 502-510. Barrow, N. J., and Carter, E. D. (1978). A modified model for evaluating residual phosphate in soil. Aust. J. Agric. Res. 29, 101 1-1021. Barrow, N. J., and Shaw, T. C. (1975).The slow reactions between soil and anions: 2. Effect of time and temperature on the decrease in phosphate concentration in the soil solution. Soil Sci. 119, 167- 177. Bascomb. C. (1964). Rapid method for the determination of cation-exchange capacity of calcareous and non-calcareous soils. J . Sci. Fd. Agric. 15, 821-823. Bationo, A., Baethgen, W. E., Christianson, C. B., and Mokwunye, A. U . (1991). Comparison of five soil testing methods to establish phosphorus sufficiency levels in soil fertilized with watersoluble and sparingly soluble-P sources. Ferr. Res. 28, 271-279. Bekele, T..Cino, B. J., Ehlert, P. A. I., Maas, V. D., and Diest, V. A. (1983). An evaluation of plant-borne factors promoting the solubilization of alkaline rock phosphates. Plant Soil 75, 361 -378. Bekele, T., and Hofner, W. (1993). Effects of different phosphate fertilizers on yield of barley and rape seed on reddish brown soils of the Ethopian highlands. Ferr. Res. 34, 243-250. Benbi, D. K.. and Gilkes. R. J. (1987). The movement into soil of P from superphosphate grains and its availability to plants. Fert. Res. 12, 21-36. Bhujbal, B. M.. and Mistry, K. B. (1985). Studies on the dissolution of Major Indian Phosphate Rocks in an acid soil. J. lndian Soc. Soil Sci. 33, 568-573. Black, C. A. (1993). “Soil Fertility Evaluation and Control.” Lewis, Boca Raton, FL. Black, C. A,, and Scott. C. 0. (1956). Fertilizer evaluation. 1. Fundamental principles. Soil Sci. Soc. Am. Proc. 20, 176- 179.
148
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
Bolan, N. S., Robson, A. D., and Barrow, N. J. (1983). Plant and soil factors including mycorrhizal infection causing sigmoidal response of plants to applied phosphorus. Plant Soil 73, 187-201. Bolan, N. S . , and Hedley, M. J. (1989). Dissolution of phosphate rocks in soils. 1. Evaluation of extraction methods for the measurement of phosphate rock dissolution. Ferf. Res. 19, 65-75. Bolan, N. S., and Hedley, M. J. (1990). Dissolution of phosphate rocks in soils. 11. Effect of pH on the dissolution and plant availability of phosphate rock in soil eith pH dependent charge. Fert. Res. 24, 125- 134. 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. Agric. 30, 297313. Bolland, M. D. A. (1985). Residual values of Australian rock phosphates. J. Agri. West Aust. Dept. Agric. 2, 52-56. Bolland, M. D. A. (1993). Summary of research on soil testing for rock phosphate fertilizers in Western Australia. Ferr. Res. 35, 83-91. Bolland, M. D. A. ( 1994). Effect of water supply on the response of wheat and tricale to applications of rock phosphate and superphosphate. Fert. Res. 39, 43-57. Bolland, M. D. A,, and Barrow, N . J. (1988). Effect of level of application on the relative effectiveness of rock phosphate. Fert. Res. 15, 181-192. Bolland, M. D. A,, and Gilkes, R. J. (1992). Evaluation of the Bray I , calcium acetate lactate (CAL), Truog and Cowell soil tests as predictors of triticale grain production on soil fertilized with superphosphate and rock phosphate. Fert. Res. 31, 363-372. Bolland, M. D. A,, Gilkes, R. J., and Allen, D. G. (1988a). The residual value of superphosphate and rock phosphates for lateritic soils and its evaluation using three soil phosphate tests. Fert. Res. 15, 253-280. Bolland, M. D. A,, Gilkes, R. J., and D’Antuono, M. (1988b). The effectiveness of rock phosphate fertilizers in Australian agriculture: A review. Aust. J. Exp. Agric. 28, 655-668. Bolland, M. D. A , , Weatherley, A. J., and Gilkes, R. J. (1988~).Residual effectiveness of superphosphate is greater than that of rock phosphate fertilizers for lateritic soils in south-western Australia. Ausf. J. Expl. Agric. 28, 83-90. Bolland, M. D. A., Kumar, V.. and Gilkes, R. J. (1994). A comparison of five soil phosphorus tests for crop species for soil previously fertilized with superphosphate and rock phosphate. Fert. Res. 37, 125-132. Bowman, R. and Savory, D. (1992). Phosphorus distribution in Calcareous soil profiles of the central plains. Soil Sci. Soc. Am. J. 56, 423-426. Bray. R . M.. and Kutrz. L. T. (1945). Determination of total, organic and available forms of phosphorus in soils. Soil Sci. 59, 39-45. Chang, S. C., and Jackson, M. L. (1957). Fractionation of soil phosphorus. SoilSci. SOC.Am. f r o c . 84, 133-144. Chaudhary, M. L . , and Mishra. B. (1980). Factors affecting transformation of rock phosphate in soils. J. Indian. Soc. Soil Sci. 28, 295-301. Chien, S . H. (1977). Thermodynamic considerations on the solubility of phosphate rock. Soil Sci. 123, 117-121. Chien, S. H. (1978). Interpretation of Bray I-extractable phosphorus from acid soil treated with phosphate rocks. Soil Sci. 126, 34-39. Chien, S. (1979). Dissolution of phosphate rock in acid soils as influenced by nitrogen and potassium fertilizers. Soil Sci. 127, 371-375. Chien, S. H. (1993). Solubility assessment for fertilizer containing phosphate rock. Fert. Res. 35, 93-99. Chien, S . H . , and Hammond, L. L. (1978). A comparison of various laboratory methods for predicting
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
149
the agronomic potential of phosphate rocks for direct application. Soil Sci. Soc. Am. J . 42,935939. Chien, S . H., Leon, L. A., and Tejeda, H. (1980). Dissolution of North Carolina phosphate rock in acid columbian soils as related to soil properties. Soil Sci. Soc. Am. J . 44, 1267-1271. Chien. S. H., Savant, N., and Mokwunye, U. (1982). Effect of temperature on phosphate sorption and desorption in two acid soils. Soil Sci. 133, 160-166. Chien, S. H., Adams. F., Khasawneh, F. E., and Henao, 0. (1987a). Effects of combinations of triple superphosphate and a reactive phosphate rock on yield and phosphorus uptake by corn. Soil Sci. Soc. Am. J . 51, 1656-1658. Chien. S. H., Hammond, L. L., and Leon, L. A. (l987b). Long-term reactions of phosphate rocks with an oxisol in Columbia. Soil Sci. 144, 257-265. Chien. S . H., Sompongse, D., Henao, J . , and Hellums, D. T. (1987~).Greenhouse evaluation of phosphorus availability from compacted phosphate rocks with urea or with urea and triple superphosphate. Fert. Res. 14, 245-256. Chien, S . H., Sale, P. W. G., and Hammond, L. L. (1990). Comparison of the effectiveness of phosphorus fertilizer products. I n “Phosphorus Requirements for Sustainable Agriculture in Asia and Oceania,” pp. 143- 156. International Rice Research Institute, Manila, Philippines. Christoffersen, J . , and Christoffersen, M. R . (1979). Kinetics of dissolution of calcium hydroxyapatite. 11. Dissolution in nonstoichiometric solutions at constant pH. J . Crys/. Growrh 47,671679. Christoffersen, J. , and Christoffersen, M. R. (1982). Kinetics of dissolution of calcium hydroxide. J . Crvst. Grow/h 57, 21-26. Chu, C. R.. Moschler, M. W., and Thomas, G. W. (1962). Rock phosphate transformations in acid soils. Soil Sci. Soc. Am. Proc. 26, 476-478. Colwell, J. D. (1963). The estimation of phosphorus fertilizer requirement in southern New South Wales by soil analysis. Ausr. J . E.rp. Agric. Artim. Hush. 3, 190-197. Colwell. J. D., and Goedert, W. J. (1988). Substitution rates as measure of the relative effectiveness of alternative phosphorus fertilizers. Fur. Rrs. 15, 163- 172. Cornforth, 1. S . , Smith, G . S . , and Fox, R. L. (1983). Phosphate extractability and availability to plants in phosphate rock-treated soils. N.Z.J . Exp. Agric. 11, 243-246. Dalal, R. C. (1985). Comparative prediction of yield response and phosphorus uptake from soil using cation-anion-exchange resins. Soil Sci. 139, 227-23 I . Dash, R. N . , Mohanty, S. K . , and Patnaik, S . (1988). Influence of reactivity of phosphate rocks on phosphorus utilization by dhaincha (Seshanio aculeatu). J . Indian. Soc. Soil. Sci. 36, 375378. De Swart. P. H.. and Van Diest, A. (1987). The rock-phosphate solubilizing capacity of Pueraria juvcmica as affected by soil pH, superphosphate priming effect and symbiotic N2 fixation. Plan/ Soil 100, 135-147. Di. H. J . , Harrison, R., and Campbell, A. S. (1994). Assessment of methods for studying the dissolution of phosphate fertilizers of differing solubility in soil. I. An isotopic method. Fert. Rrs. 38, 1-9. Dyer. B . (1894). Determination des matieres minerales assimilables par les plants. Ann Agron. 291298. Edmeades, D. C.. Wheeler, D. M.. and Clinton, 0. E. (1985). The chemical composition and ionic strength of soil solutions from New Zealand topsoils. Aust. J . Soil Res. 23, 15 I - 165. Edwards, G. H. A. ( 1956). Observations on the response of crops to rock phosphate. Inr. Congr. Soi/ Sci Trans. Paris Comm. 4 4, 13-20, Ellis, R.. Jr., Quader, M. A,, and Truog, E. (1955). Rock phosphate availability as influenced by soil pH. SoilSci. Soc. Am. Proc. 19, 484-487.
150
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
Engelstad, 0. P. (1978). Relative agronomic and economic effectiveness values for phosphate rocks. “Seminar on Phosphate Rock for Direct Application,” pp. 291-303. Haifa, Israel. Engelstad, 0. P., Jugsujinda. A,, and De Datta, S. K. (1974). Response by flooded rice to phosphate rocks varying in citrate solubility. Soil Sci. Soc. Am. Proc. 38, 524-529. Fardeau, J., Morel, C., and Jahiel, M. (1988). Does long contact with the soil improve the efficiency of rock phosphate? Results of isotopic studies. Ferr. Res. 17, 3-19. Fitts, J. W., and Hanway, J. J. (1971). Prescribing soil and crop nutrient needs. I n “Fertilizer Technology and Use“ (R. A. Olsen er a / . ,Eds.), pp. 57-79. Soil Sci. Soc. Am., Madison, WI. Flach, E. N., Quak, W., and Van Diest, A. (1987). A comparison of the rock phosphate-mobilising capacities of various crop species. Trop. Agric. 64, 347-352. Fox, R. L., Jones, R., Lower, R., and De La Pena, R. S . (1986a). Immediate and residual effects of phosphate fertilizers in relation to tropical soil mineralogy. Trans. Inr. Congr. Soil Sci. XIII, 4, 737-74 I . Fox, R. L., Saunders, W. M. H., and Rajan, S. S. S. (l986b). Phosphorus nutrition of pasture species: Phosphorus requirement and root saturation values. Soil Sci. Soc. Am. J. 50, 142-148. Fox, T. R., Comerford, N. B., and McFee, W. W. (1990). Phosphorus and aluminum release from a spodic horizon mediated by organic acids. SoilSci. Soc. Am. J. 54, 1763-1767. Fried, M., and MacKenzie, A. C. (1949). Rock phosphate studies with neutron irradiated rock phosphate. Soil Sci. SOC. Am. Proc. 14, 226-23 1. Frossard, E., Fardeau, J. C., Brossard, M., and Morel, J. L. (1994). Soil isotopically exchangeable phosphorus:A comparison between E and L values. Soil Sci. SOC. Am. J. 58, 846-851. Gahoonia, T. S., Claassen, N., and Jungk, A. (1992). Mobilization of phosphate in different soils by ryegrass supplied with ammonium or nitrate. Planr Soil 140, 241-248. Gahoonia, T.S., and Nielsen, N. E. (1992). The effects of root-induced pH changes on the depletion of inorganic and organic phosphorus in the rhizophere. Planr Soil 143, 185-191. Gahoonia, T. S., Raza, S., and Nielsen, N. E. (1994). Phosphorus depletion in the rhizophere as influenced by soil moisture. Planr Soil 159, 213-218. Germida, J. J., and Janzen, H. H. (1993). Factors affecting the oxidation of elemental sulfur in soils. Fert. Res. 35, 101-1 14. Ghani, A., Rajan, S . S. S., and Lee, A. (1994). Enhancement of phosphate rock solubility through biological processes. Soil Biol. Biochem. 26, 127- 136. Ghosh, A., and Gilkes, R. J. (1987). The initial and residual agronomic effectiveness of some Indian, USA and Australian rock phosphates. Ferr. Res. 12, 201-218. Gillman, G. P., and Bell, L. C. (1978). Soil solution studies on weathered soils from tropical North Queensland. Aust. J. Soil Res. 16, 67-77. Gregg, P. E. H., Mackay, A. D., Currie, L. D., and Syers, J. K. (1988). Application strategies for sechura phosphate rock use on permanent pasture. Ferr. Res. 17, 219-234. Grigg, J. L. (1980a). Relative efficiency of phosphatic fertilisers in pasture topdressing. I. Introduction and design of experiments. N.Z.J. Agric. Res. 23, 43-47. Grigg, J. L. (1980b). Relative efficiency of phosphatic fertilisers in pasture topdressing. 11. On a Kokotau silt loam. N.Z.J. Agric. Res. 23, 49-65. Hagin, J., and Hamson, R. (1993). Phosphate rocks and partially acidulated phosphate rocks as controlled release P fertilizers. Ferr. Res. 35, 25-32. Hagin ,J., Rajan, S. S. S . , Boyes, M. K., and Upsdell, M. P. (1990). Partially acidulated phosphate rocks: Phosphate release characteristics. Ferr. Res. 22, 109- 117. Halder, A. K., Mishra, A. K.. Bhattacharyya, P., and Chakrabartty, P. K. (1990). Solubilization of rock phosphate by Rhizobium and Bradyrhizobium. J. Gen. Appl. Microbiol. 36, 81-92. Hammond, L. L., and Leon, L. A. (1983). Agronomic effectiveness of natural and altered phosphate rocks from Latin America. Proc. Third Int. Congr. Phosphorus Compounds Brussels Belgium, 503-5 18.
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
15 1
Hammond, L. L., Chien, S. H., and Polo, J. R. (1980). Phosporus availability from partial acidulation of two phosphate rocks. Fert. Res. 1, 37-49. Hammond, L. L., Chien, S. H., and Easterwood, G. W. (1986a). Agronomic effectiveness of Bayovar phosphate rock in soil with induced phosphorus retention. Soil Sci. Soc. Am. J . 50, I60 1- 1606. Hammond, L. L., Chien, S. H., and Mokwunye, A. U . (1986b). Agronomic value of unacidulated and partially acidulated phosphate rocks indigenous to the tropics. Adv. Agron. 40, 89138. Haynes, R. J. (1983). Soil acidification induced by legurninus crops. Grass Forage Sci. 38, I - I I , Haynes, R. J. (1992). Relative ability of a range of crop species to use phosphate rock and monocalcuim phophate as P sources when grown in soil. J . Sci. Food Agric. 74, 1-7. Hedley, M. J., White, R. E., and Nye, P. H. (1982). Changes in L value, soil phosphate fractions and phosphatase activity. N . Phytol. 91, 45-56. Hedley, M. J., Nye, P. H., and White, R. E. (1983). Plant induced changes in the rhizosphere of rape (Brassica napus cv 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 plants. N e w Phvtol . 95, 69-82. Helling, C. S., Chesters, G., and Corey, R. B. (1964). Contribution of organic matter and clay to soil cation-exchange capacity as affected by the pH of the saturating solution. Soil Sci. Soc. Am. Proc. 28, 517-520. Hieltjes, A. H. M., and Lijklema, L. (1980). Fractionationation of inorganic phosphates in calcareous sediments. J. Environ. Qual. 9, 405-407. 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. 11. Local root exudation of organic acids as a response to P-starvation. Plant Soil 113, 161- 165. Holford, 1. C. R.. and Mattingly, G. E. G. (1975). The high and low-energy phosphate adsorbing surfaces in calcareous soils. J. Soil Sci. 26, 407-417. Hooker, M. L., Peterson, G. A,, Sander, D. H., and Daigger. L. A. (1980). Phosphate fractions in calcareous soils as altered by time and amounts of added phosphate. Soil Sci. Sor. Am. 1.44, 269-277. Huffman, E. (1962). Reacations of phosphate in soils: Recent research by TVA. Proc. Fert. Soc. London. 71, 5-48. Huffman, E., Cate, W., Deming, M., and Elmore, K. (1957). Rates of solution of calcium phosphates in phosphoric acid solutions. Agric. Food Chem. 5, 266-275. Hughes, J., and Gilkes, R. J. (1984). The effect of chemical extractant on the estimation of rock phosphate fertilizer dissolution. Aust. J . Soil Res. 22, 475-481. Hughes, J. and Gilkes, R. J. (1994). Rock phosphate dissolution and bicarbonate-soluble P in some soils from South-Western Australia. Aust. J. Soil Res. 32, 767-779. Johnston, H. W. (1952). The solubilization of phosphate. I. The action of various organic compounds on dicalcium and tricalcium phosphates. N . Z . J. Sci. Tech.B. 33, 436-446. Johnston, H. W. (1954a). The solubilization of “insoluble” phosphate. 11. A quantitative and comparative study of the action of selected aliphatic acids on tricalcium phosphate. N . Z . J. Sci. Tech.B. 36, 49-55. Johnston, H. W. ( 1954b). The solubilization of’insoluble” phosphate. 111. A quantitative and comparative study of the action of chosen aromatic acids on tricalcium phosphate. N . Z . J. Sci. Tech.B. 36, 28 1-284. Johnston, H. W. (1959). The solubilization of “insoluble” phosphate. V. The action of some organic acids on iron and aluminum phosphates. N . Z . J. Sci. 2, 215-218.
152
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
Johnstone, P., and Sinclair, A. G. (1991). Replication requirements in field experiments for comparing phosphatic fertilizers. Ferf. Rrs. 29, 329-333. Joret, G . , and Hebert, J., (1955). Contribution a la determination du besoin des sols en acide phosphorique. A m . Agron. 6, 233-299. Kanabo, I., and Gilkes, R. J. (1987a). The Influence of the addition of goethite to soil on the dissolution of North Carolina phosphate rock. Aust. J. Soil Res. 25, 313-322. Kanabo, I., and Gilkes, R . J. (1987b). A comparison between plant response and chemical measurements and the dissolution of reactive phosphate rock in soils of different pH and phosphorus retention. Ausf. J. Soil Rrs. 25, 451-460. Kanabo, I., and Gilkes, R. J. (1987~).The role of soil pH in the dissolution of phosphate rock fertilizers. Frrt. Res. 12, 165-174. Kanabo. I., and Gilkes, R. J. (1988a). The Effects of moisture regime and incubation period on the dissolution of North Carolina phosphate rock in soil. Ausf. J. Soil Res. 26, 153-163. Kanabo, I., and Gilkes, R. J. (l988b). A comparison of mixed and band applications on the dissolution of North Carolina phosphate rock and on bicarbonate-soluble soil phosphorus. Fert. Rrs. 15, 3-12. Kanabo, I., and Gilkes, R. J. (1988~). The effect of particle size on North Carolina phosphate rock on its dissolution in soil and on levels of bicarbonate-soluble phosphorus. Fert. Res. 15, 137- 145. Kanabo, I., and Gilkes. R. J. (1988d). The effect of the level of phosphate rock application on its dissolution in soil and on bicarbonate-soluble phosphorus. Fert. Res. 16, 67-85. Khanna, S. S., Tomar. N. K., and Gupta. A. P. (1983). Efficiency of incubated phosphate fertilizers varying in water solubility with organic matter to wheat. Proc. Third I n t . Congr. Phosphorus Compounds Brussels Belgium. 567- 5 80. Khasawneh, F. E., and Doll, E. C. (1978). The use of phosphate rock for direct application to soils. Ad): Agron. 30, 159-203. Kirk, G . J. D.,and Nye, P. H. (1985). The dissolution and dispersion of dicalcium phosphate dihydrate in soils. 11. Experimental evaluation of the model. J . Soil Sci. 36, 461-468. Kirk, G . J. D.,and Nye, P. H. (1986a). A simple model for predicting the rates of dissoltion of sparingly soluble calcium phosphates in soil. I. The basic model. J. Soil Sci. 37, 529-540. Kirk, G. J. D.,and Nye, P. H. (1986b). A simple model for predicting the rates of dissoltion of sparingly soluble calcium phosphates in soil. 11. Applications of the model. J. SoilSci. 37, 541553. Kittams, H., and Attoe, 0. (1965). Availability of phosphorus in rock phosphate-sulfur fusions. Agron. J. 57, 331-333. Kothandaraman, G . V.. Manickam, T. S., and Natarajan, K. (1987). “Rock Phosphate in Agriculture.’’ Tamil Nadu Agricultural University, Coimbatore, India. Kpomblekou. K., and Tatabai, M. A. (1994). Effect of organic acids on release of phosphorus from phosphate rocks. J. Soil Sci. 158, 442-453. Kucey, R., and Bole, 1. (1984). Availability of phosphorus from 17 rock phosphates in moderately and weakly acidic soils as determined by 32P dilution, a value, and total P uptake methods. Soil Sci. 138, 180-188. Kucey. R. M. N., Janzen, H. H., and Leggett, M. E. (1989). Microbially mediated increases in plant-available phosphorus. Adv. Agron. 42, 199-228. Kumar, V., Gilkes, R. J., and Bolland. M. D.A. (1991). Residual phosphate fertilizer compounds in soils. 11. Their influence on soil tests for available phosphate. Ferf. Rrs. 30, 31-38. Kumar, V., Gilkes, R. J., and Bolland, M.D.A. (1992a). Phospdhte fertilizer compounds in soils: Their influence on the relationship between plant yield and soil test value. Frrt. Res. 23, 14611477. Kumar, V., Gilkes, R. J., and Bolland, M. D. A. (l992b). A comparison of seven soil P tests for plant species with different external P requirements grown on soils containing rock phosphate and superphosphate residues. Frrt. Res. 33, 35-45.
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
153
Kurnar, V., Gilkes, R. J., and Bolland, M. D. A. (1994). Comparison of the Pi, Colwell. Bray I , calcium acetate lactate (CAL) and Truog soil phosphorus test for predicting growth of oats, barley, tricale and clover in the field in lateritic soils fertilised with superphosphate and rock phosphate. Fert. Res. 37, 115-124. Larsen, S. ( 1952). The use of P32 in studies on the uptake of phosphorus by plants. Plant S o i l 4 , 1-9. Larsen, S . , and Probert, M. E. (1968). A theoretical treatment of the maintenance of soil phosphorus status. Phos. Agric. Bull. Doc. 51, 1-6. Lee, A., Watkinson, J. H., Orbell, G., Bagyaraj, J., and Lauren, D. R . (1987). Factors intluencing dissolution of phosphate rock and oxidation of elemental sulphur in some New Zealand soils. N.Z.J. Agric. Res. 30, 373-385. Lehr, J. R., and McClellan, 0 . H. (1972) A revised laboratory reactivity scale for evaluating phosphate rocks for direct application. Bulletin Y-43, Tennessee Valley Authority, Muscle Shoals, AL. Leon. L. A., Fenster, W. E., and Hammond, L. L. (1986). Agronomic potential of eleven phosphate rocks from Brazil, Columbia, Peru, and Venezuela. Soil Sci. Soc. Am. J. 50, 798-802. Lin, T. H., Ho, S . B., and Houng, K. H. (1991). The use of iron oxide-impregnated filter paper for the extraction of available phosphorus from Taiwan soils. Plant Soil 133, 219-226. Lipman, J. G., and McLean, H. C. (1916). Sulfur oxidation in soils and its effect on availability of mineral phosphates. Soil Sci. 2, 499-538. Lipman, J. G.. and McLean, H. C. (1918). Experiments with sulfur-phosphate composts conducted under field conditions. Soil Sci. 5, 243-250. Loganathan, P., Hedley, M. J., and Bretherton. M. R. (1994). The agronomic value of co-granulated Christmas Island Grade C phosphate rock and elemental sulfur. Ferr. Res. 39, 229-237. Mackay, A. D., and Syers. J. K. (1986). Effect of phosphate. calcium, and pH on the dissolution of a phosphate rock in soil. Fert. Res. 10, 175- 184. Mackay, A. D., Gregg, P. E. H., and Syers, J. K . (1980). A comparison of dicalcium phosphate and single superphosphate as sources for pasture. N.Z.J. Exp. Agric. 8, 7- I I . Mackay, A. D., Gregg. P. E. H., and Syers, J. K . (1984a). Field evaluation of chatham rise phosphorite as a phosphatic fertiliser for pasture. N.Z.J. Agrie. Res. 27, 65-82. Mackay, A. D.. Syers, J. K., and Gregg, P. E. H. (1984b). Ability of chemical extraction procedures to assess the agronomic effectiveness of phosphate rock materials. N.Z.J. Agric. Res. 27, 219230. Mackay, A. D., Syers, J. K., and Gregg, P. E. H. (1984~).A glasshouse comparison of 6 phosphate fertilisers. N.Z.J. Exp. Agric. 12, 131-140. Mackay, A. D.. Syers, J. K . , Gregg, P. E. H., and Tillman, R. (l984d). A comparison of 3 soiltesting procedures for estimating the plant availability of phosphorus in soils receiving either superphosphate or phosphate rock. N.Z.J. Agric. Res. 27, 23 1-245. Mackay, A , , Syers, J . , Tillman. R., and Gregg, P. (1986). A simple model to describe the dissolution of phosphate rock in soils. Soil Sci. Soc. Am. J. 50, 291-296. Mackay, A. D., Syers, J. K., and Gregg, P. E. H. (1987). Potential losses of fertilizer phosphorus by animal transfer. Fert. Res. 14, 125-134. Magdoff, F. R.. and Bartlett. R . J. (1985). Soil pH buffer revisited. Soil Sci. Soc. Am. J. 49, 145148. Mahimairaja, S.. Bolan, N. S., and Hedley, M. J. (1995). Dissolution of phosphate rock during the cornposting of poultry manure: An incubation experiment. Fert. Res., in press. Mattingly, G. E. C., and Widdowson, F. W. (1963). Residual value of superphosphate and rock phosphate on an acid soil. I . Yields and phosphorus uptakes in the field. J. Agric. Sci. 60, 399407. McAuliffe, C. D., Hall, N. S., Dean, L. A,, and Hendricks, S . B. (1947). Exchange reactions between phosphates and soils: Hydroxylic surfaces of soil minerals. Soil Sci. Sor. Am. J. 12, 119-123.
154
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
McLaughlin, M. J., Lancaster, P. A,, Sale, P. G., Uren, N. C., and Peverill, K. 1. (1994). Comparison of catiodanion exchange resin methods for multi-element testing of acidic soils. Aust. J. Soil Res. 32, 229-240. McClellan, G. H., and Gremillion, L. R. (1980). Evaluation of phosphate raw materials. in “The Role of Phosphorus in Agriculture” (F. E. Khasawneh, E. C. Sample, and E. J. Kamprath, Fds.), pp. 43-80. Am. SOC.Agron.. Madison, WI. McClellan, G. H., and Van Kauwenbergh, S. J. (1992). Relationship of mineralogy to sedimentary phosphate rock reactivity. Pror. Workshop Phosphate Sources Acid Soil Kuala Lumpur Malaysia 1990. 1-17. Menon, R. G., and Chien, S . H. (1990). Phosphorus availability to maize from partially acidulated phosphate rocks and phosphate rocks compacted with triple superphosphate. Plant Soil 127, 123-128.
Menon, R. G, Chien, S. H., and Hammond, L. L. (1989a). Comparison of Bray I and Pi tests for evaluating plant-available phosphorus from soils treated with different partially acidulated phosphate rocks. Plant Soil 114, 21 1-216. Menon, R. G , Chien, S. H, Hammond, L. L., and Henao, J. (1989b). Modified techniques for preparing paper strips for the new Pi soil test for phosphorus. Fert. Res. 19, 85-91. Mey, P., Sayag, D., and Andre, L. (1986). Chemical or microbiological solubilisation of rock phosphate. Comp. Rend. Sean L’Acad. d’Agri. France 72, 81-89. Mishra., Kumar, P., and Dwivedi, G. (1985). Effectiveness of Indian rock phosphates in acid soils. Indian. Soc. Soil Sci. 3, 574-580. Mishra, M., Khurana, A.. Dudeja, S., and Kapoor, K . (1984). Effect of phosphorocompost on the yield and P uptake of red gram (Cajanus cajart (L.) mill sp.). Trop . Agric. 61, 174-176. Mishra, M. M., and Bangar, K. C. (1986). Rock phosphate composting: Transformation of phosphorus forms and mechanisms of solubilizdtion. Biol. Agric. Horr. 3, 331-340. Moghimi, A., and Tate, M. E. (1978). Does 2-ketogluconate chelate calcium in the pH range 2.4-6.4? Soil Biol. Biochem. 10, 289-292. Moorby, H., White, R., and Nye, P. (1988). The influence of phosphate nutrition on H ion efflux from the roots of young rape plants. Plant Soil 105, 247-256. Morel, C. H., and Fardeau, J. C. (1990). Agronomical evaluation of phosphate fertilizer as a nutrient source of phosphorus for crops: Isotopic procedure. Fert. Res. 24, 115-122. Morel, C. H., and Plenchette, C. (1994). Is the isotopically exchangeable phosphate of a loamy soil the plant available P? Plant Soil 158, 287-297. Morel, C., Tiessen, H., Moir, J. 0.. and Stewart, J. W. B. (1994). Phosphorus transformations and availability under cropping and fertilization assessed by isotopic exchange. Soil Sci. Soc. Am. J . 58, 1439-1445.
Morton, J. D., Waller, J. E., Ledgard, S. F., and Roberts, A. H. C. (1995). Effects of measurement techniques on responses to phosphorus fertilisers. In “Fertilizer Requirements of Grazed Pasture and Field Crops-Macro and Micro Nutrients” (L. D. Currey and P. Loganathan, Eds.), pp. 1220. Occasional Report No. 8, Fertilizer and Lime Research Center, Massey University, Palmerston North, New Zealand. Munns, D. N., and Fox, R. L. (1976). The slow reaction which continues after phosphate adsorption: Kinetics and equilibrium in some tropical soils. Soil Sci. Soc. Am J . 40, 46-51, Nimgade, N. M. (1968). Effect of mixing rock phosphate with molten sulphur on phosphorus availability. Trans. IX Iiit. Cong. Soil Sci. 2, 765-774. Nguyen, L., and Risk, W. H. (1992). Pasture production and soil phosphorus fractions resulting from six previous annual applications of triple superphosphate or Sechura phosphate rock. N.Z. J . Agric. Res. 35, 307-319. Nye, P. H. (1979). Diffusion of ions and uncharged solutes in soils and soil clays. Adv. Agron. 31, 225-272.
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
15s
Nye. P. H. (1981a). Diffusion of ions and uncharged solutes in soils and soil clays. Adv. Agron. 31, 225-272. Nye, P. H. (198 I b). Changes of pH across the rhizosphere induced by roots. Plant Soil 61, 7-26. Nye. P. H. (1986). Acid-base changes in the rhizosphere. Adv. Plant Nutr. 2. Nye, P. H. (1992). Towards the quantitative control of crop production and quality. 1. The role of computer models in soil and plant research. J. Plant Nutr. 15, 1131-1 150. Nye, P. H., and Tinker, P. B. (1977). “Solute Movement in the Soil-Root System.” University of California Press, Berkeley. Olsen, R. A. (1975). Rate of dissolution of phosphate from minerals and soils. Soil Sci. SOC. Am. Proc. 39, 634-639. Olsen, S. R., Cole, C. V., Watanabe, F. S . , and Dean, L. A. (1954). Estimation of available phosphorus in soil by extraction with sodium bicarbonate. US Dept. Agric. Circ. No. 939. Parfitt, R. L. (1989). Phosphate reactions with natural allophane, ferrihydrite and goethite. J. Soil Sci. 40,359-369. Partridge. I. (1980). The efficacy of biosupers made from different forms of phosphate on forage legumes in hill land in Fiji. Trop. Grass. 14, 87-93. Peaslee, D. E., Bums, G. R., and Black, C. A. (1962). Estimation of relative value of phosphate rock and superphosphate to plants on different soils. Soil Sci. Soc. Am. Proc., 566-570. Perrott. K. W. ( 1992) Effect of exchangeable calcium on fractionation of inorganic and organic soil phosphorus. Commun. Soil Sci. Plant Anal. 23, 827-840. Perrott, K. W.. and Kerr, B. E. (1994). Determination of phosphate rock residues in acid soils. 15rh I n t . Congr. Soil Sci. Acapulco Mexico 5a, 230-243. Perrott, K. W., and Wise, R. G. (1993). An evaluation of some aspects of the iron oxide-impregnated filter paper (Pi) test for available phosphorus with New Zealand soils. N . Z . f. Agric Res. 36, 157-162. Perrott. K. W.. and Wise, R. G. (1995). A routine test for quantifying RPR residues in soil. In “Fertilizer Requirements of Grazed Pasture and Field Crops” (L.D. Currie, Ed.). Occasional Report No. 8, Fertilizer and Lime Research Centre, Massey University, Palmerston North, New Zealand. Perrott. K. W., Roberts, A. H. C., Saggar, S . , Shannon, P. W., O’Connor, M. B., Nguyen, L., and Risk, W. H. (1992). Pasture prdouction and soil phosphorus fractions resulting from six previous annual applications of triple superphosphate or Sechura phosphate rock. N.Z. J. Agric. Res. 35, 307-319. Perrott. K . W., Saggar, S.. and Menon, R., G. (1993). Evaluation of soil phosphate status where phosphate rock based fertilizers have been used. Fert. Res. 35, 67-82. Peterson, G. W., and Corey, R. B. (1966). A modified Chang and Jackson procedure for routine fractionation of inorganic soil phosphates. Soil Sci. Soc. Am. Pror. 30, 563-565. Pohlman, A. A,. and McColl. J. G. (1986). Kinetics of metal dissolution from forest soils by soluble organic acids. J. Environ. Qual. 15, 86-92. Probert, M. E. (1985). A conceptual model for initial and residual responses to phosphorus fertilizers. Fen. Res. 6, 131-138. Purnomo, E., and Black, A. S. (1994). Wheat growth from phosphorus fertilizers as affected by time and method of application in soil with an acidic subsurface layer. Fert. Res. 39, 77-82. Quin, B . F. (1989). Phosphate rocks for direct application-Which, where and why. Comment on “The effectiveness of rock phosphate fertilisers in Australian agriculture: a review,” by Bolland, M. D. A,, Gilkes, R. J., and D’Antuono, M. F. (Aust. J. Exp. Agric. (1988) 28, 655-668). Aust. J . Expl. Agric. 29, 695-700. Rajan S. S. S. ( 1973). Phosphorus adsorption characteristics of Hawaiian soils and their relationships to equilibrium concentrations required for maximum growth of millet. Planr Soil 39, 5 19532.
156
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
Rajan, S. S. S. (1975). Adsorption of divalent phosphate on hydrous aluminum oxide. Nature 253, 434-436. Rajan, S. S . S. (1982). Influence of phosphate rock reactivity and granule size on the effectiveness of ‘biosuper.’ Fert. Res. 3, 3-12. Rajan, S. S . S . (1983). Effect of sulphur content of phosphate rock/sulphur granules on the availability of phosphate to plants. Fert. Res. 4, 287-296. Rajan, S . S. S. (1987a). Partially acidulated phosphate rock as fertiliser and dissolution in soil of the residual rock phosphate. N.Z. J . Exp. Agric. 15, 177-184. Rajan, S. S . S . (1987b). Phosphate rock and phosphate rock/sulphur granules as phosphate fertilizers and their dissolution in soil. Fert. Res. 11, 43-60. Rajan, S . S . S . . and Fox, R. L. (1975). Phosphate adsorption by soils. 11. Reactions in tropical acid soils. Soil Sci. Soc. Am. Proc. 39, 846-85 I . Rajan, S . S. S . . and Gillingham, A. G . (1986). Phosphate rocks and phosphate rocklsulphur granules as fertilisers for hill country pasture. N . Z . J . Exp. Agric. 14, 313-318. Rajan, S . S . S . , and Marwaha, B. C. (1993). Use of partially acidulated phosphate rocks as phosphate fertilisers. Fert. Res. 35, 47-59. Rajan, S . S. S., and Watkinson, J. H. (1992). Unacidulated and partially acidulated phosphate rock: Agronomic effectiveness and the rates of dissolution of phosphate rock. Fert . Res. 33, 267277. Rajan, S . S . S . , Fox, R. L., Saunders, W. M . H.. andupsdell, M . P. (1991a). InfluenceofpH, time and rate of application on phosphate rock dissolution and availability to pastures. 1. Agronomic benefits. Fert . Res. 28, 85-93. Rajan, S . S. S . . Fox, R . L., Saunders, W. M . H.. and Upsdell, M. P. (1991b). influence of pH, time and rate of application on phosphate rock dissolution and availability to pastures. 11. Soil Chemical Studies. Ferr , Res. 28, 95-101. Rajan, S . S . S . , Brown, M . W., Boyes, M. K., and Upsdell. M. P. (1992). Extractable phosphorus to predict agronomic effectiveness of ground and unground phosphate rocks. Fert. Res. 32, 291302. Rajan, S . S . S . , O’Connor, M . B.. and Sinclair, A. G. (1994). Partially acidulated phosphate rocks: Controlled release phosphorus fertilizers for more sustainable agriculture. Fert . Res. 37,69-78. Reinhorn. T.,and Hagin, J. (1978). Reactions of a phosphate rock with soil. In “Seminar on Phosphate Rock for Direct Application,” pp. 2 I 1-221. International Fertilizer Development Center, Alabama. Rennie, D. A., and McKercher. R. B. (1959). Adsorption of phosphorus by four Saskatchewan soils. J . Soil Sci. 39, 64-75. Robinson, J. S . , and Syers, J. K . (1991). Effects of solution calcium concentration and calcium sink size on the dissolution of Gafsa phosphate rock in soils. J . Soil Sci. 42, 389-397. Robinson, J. S . , Syers, J. K.. and Bolan, N. S. (1992a). Influence of calcium carbonate on the dissolution of sechurd phosphate rock in soils. Fert. Res. 32, 91-99. Robinson, J. S . , Syers, J. K., and Bolan, N. S . (199’2b). Importance of proton supply and calciumsink size in the dissolution of phosphate rock materials of different reactivity in soil. J . Soil Sci. 43, 447-459. Robinson, J. S . , Syers, J. K.. and Bolan, N. A. (1994). A simple conceptual model for predicting the dissolution of phosphate rock in soils. J . Sci. Food Agric. 64, 397-403. Saggar, S . , Hedley, M. J.. and White, R. E. (1990). A simplified membrane technique for extracting phosphorus from soils. Fert. Res. 24, 173-180. Saggar, S . , Hedley, M. J . , and White, R. E. (1992a). Development and evaluation of an improved soil test for phosphorus: I . The influence of phosphorus fertilizer solubility and soil properties on the extractability of soil P. Ferr . Res. 33, 81-91. Saggar, S . , Hedley, M. J., White, R. E., Gregg, P. E. H., Perrott, K. W., and Comforth, 1. S.
PHOSPHATE ROCKS FOR DIRECT APPLICATION T O SOILS
157
( 1992b). Development and evaluation of an improved soil test for phosphorus. 2. Comparison of the olsen and mixed cation-anion exchange resin tests for predicting the yield of ryegrass grown in pots. Fert. Res. 33, 135-144. Sale. P. W. G., and Mokwunye, A. U. (1993). Use of phosphate rocks in the tropics. Fert. Res. 35, 33-45. Sample, E. C . , Soper, R. J . , and Racz, G . J. (1980). Reactions of phosphate fertilizers in soils. I n “The Role of Phosphorus in Agriculture” (F. E. Khasawneh, E. C. Sample, and E. J. Kamprath, Eds.), pp. 263-310. Am. Soc. Agron.. Madison. WI. Sanchez, P. A. (1976). “Properties and Management of Soils in the Tropics.” Wiley, New York. Saroa. G. S . , and Biswas, C. R. (1989). A semi-descriptive model for predicting residual-P from fertilizer P applications. Fen. Res. 19, 121- 126. Saunders. W. M. H. (1964). Extraction of soil phosphate by anion-exchange mbrane. N . Z . 1. Agric. Res. 7 , 427-43 I . Saunders, W. M. H. (1965). Phosphate retention by New Zealand soils and its relationship to free sesquioxides, organic matter, and other soil properties. N . Z . J . Agric. Res. 8 , 30-57. Schafer, H. N . S. (1963). Application of ion exchange to analysis of phosphate rocks. 35, 53-56. Schoenau, J. J. and Huang. W. Z. (1991) Anion exchange membrane, water and sodium bicarbonate extractions as soil tests for phosphorus. Cornmun. Soil Sci. Plant Anal. 22, 465-492. Schuller. H. ( 1969). Die CAL-methode, eine neue methode zur bestimming des pflanzenverfugbaren phosphates in Boden. Z fflanzenvernuehr Bodenkd. 123, 48-63. Schultz, J. J. (1992). An examination of the environmental issues facing the phosphate fertilizer production sector-indicated cost of environmental compliance. I n “Phosphate Fertilizers and the Environment” (J. J. Schultz, Ed.), pp. 283-3 10. Scott, R. S . , and Cullen, N. A. (1965). Some residual effects of phosphatic fertilzers. N . Z . J . Agric. Res. 8 , 652-666. Sharpley, A. N. ( 1991). Soil phosphorus extracted by iron-aluminum-oxide-impregnated filter paper. SozlSci. Soc. Am. J . 55, 1038-1041. Sibbesen. E. (1978). An investigation of the anion-exchange resin method for soil phosphate extraction. Plant Soil 50, 305-321. Sibbesen. E. (1983). Phosphate soil tests and their suitability to assess the phosphate status of soil. J . Sci. Food Agric. 34, 1368-1374. Sidhu, S. S . (1978). A framework to evaluate phosphate rock as an alternative in phosphate fertilization. I n “Proceedings, Seminar on Phosphate Rock for Direct Application,” pp. 265-290. Haifa, Israel. Silverman, S. R . , Fuyat, R. K . , and Weiser. J. D. (1951). Quantitive determination of calcite associated with carbonate -bearing apatites. U.S.Geological Survey, 2 I 1-222. Sinclair, A. G , and Johnstone. P. D. (1995) Ability ofa dynamic model of phosphorus to account for observed patterns of response in a series of phosphate fertilizer trials on pasture. Fert. Res. 40, 21-29. Sinclair, A. G . , Bowditch, M . . Perrott, K. W., and Watkinson, J. H. (1990a). A new approach to the economic evaluation of reactive phosphate rock as a replacement for single superphosphate. I n “The Challenges for Fertilizer Research in the 1990s.” pp. 296-319. Proc. N . Z . Fertilizer
Manufacturers’ Res . A .ssoc. A ucklund New Zeulund.
Sinclair, A. G . , Shannon. P. W.. and Risk, W. H. (l990b). Sechura phosphate rock supplies plantavailable molybdenum for pastures. N . Z . J . Agric. RPS.33, 499- 502. Sinclair, A. G . . Rajan, S . S . S . , and h’dtkinson, J. H. (1992). Standardization of phosphate rocks. Proc. Workshop Phosphate Sources Acid Soil Kuula Lumpur Mula?.siu 1990. 59-72. Sinclair, A. G , Johnstone, P. D., Smith, L. C., O’Connor, M. B., and Nguyen, L. (1993a). Agronomy, modelling and economics of reactive phosphate rocks as slow-release phosphate fertilizers for grasslands. Fert. Res. 36, 229-238.
158
S. S. S. RAJAN, J. H. WATKINSON, AND A. G. SINCLAIR
Sinclair, A. G., Johnstone, P. D., Smith, L. C., and Risk, W. H. (l993b). Effect of reactive phosphate rock on the pH of soil under pasture. N.Z.J. Agric. Res. 36, 381-384. Sinclair, A. G., Johnstone, P. D., Smith, L. C., Risk, W. H., and Morton, J. D. (1994). Patterns of, and a model for, dry matter yield response in grass/clover pastures to annual applications of triple superphosphate fertiliser. N.Z.J. Agrir. Res. 37, 239-253. Singaram, P., Rajan, S. S. S., and Kothandaraman, G. V. (1995). Phosphate rock and a phosphate rock/superphosphate mixture as fertilizers for crops grown on a calcareous soil. Commn. Soil Sci. Planf Anal. 26, 157I - 1593. Singh, C. P., and Amberger, A. (1991). Solubilization and availability of phosphorus during decomposition of rock phosphate enriched straw and urine. Biol. Agric. Hort. 7 , 261-269. Sissingh, H. A. (1983). “Estimation of Plant Available-Phosphate in Tropical Soils: A New Analytical Technique.” Note 235, Institute for Soil Fertility, Harm, Netherlands. Smith, A. N., Posner, A. M., and Quirk, J. P. (1977). A model describing the kinetics of dissolution of hydroxyapatite. J. Colloid Interface Sci. 62, 475-494. Smyth, T. J., and Sanchez, P. A. (1982). Phosphate rock dissolution and availability in Cerrado soils as affected by phosphorus sorption capacity. Soil Sci. SOC.Am. J . 46, 339-345. Starkey, R. L. (1966). Oxidation and reduction of sulfur compounds in soils. Soil Sci. 101,297-306. Stephen, R. C., and Condron, L. M. (1986). An assessment of the agronomic efficiency of partially acidulated phosphate rock fertilizers. Ferf. Res. 10, 269-282. Sverdrup, H., and Bjerle, 1. (1982). Dissolution of calcite and other related minerals in acidic aqueous solutions in a pH-stat. Fatfan 38, 59-73. Swaby, R. J. (1975). BioSuper-Biological superphosphate. In “Sulfur in Australasian Agriculture” (K. D. McLachlan, Ed.), pp. 213-220. Sydney University Press, Sydney. Swaby, R. J. (1983). Production and uses of biological superphosphate. In “Soils: An Australian Viewpoint,” pp. 817-823. CSIRO, Melbourne/Academic Press, London. Syers, J. K., Smillie, G. W., and Williams, J. D. H. (1972). Calcium fluoride formation during extraction of calcareous soils with fluoride: 1. Implications to inorganic P fractionation schemes. Soil Sci. Soc. Am. Proc. 36, 20-25. Syers, 1. K., and Mackay, A. D. (1986). Reactions of Sechura phosphate rock and single superphosphate in soil. Soil Sci. Sor. Am. J . 50, 480-485. Tambunan, D., Hedley, M. J., Bolan, N. S.. and Turner, M. A. (1993). Comparison of sequential extraction procedures for measuring phosphate rock residues in soils. Ferf. Res. 35, 183-191. Tandon, H. L. S. (1987). Status of fertilizer products underdevelopment in India. Ferf. Res. 13, 181189. Tandon, H. L. S. (1987).“Phosphorus Research and Agricultural Production in India.” Fertiliser Development and Consultation Organization, New Delhi. Tinker, P. B. (1980). Role of Rhizosphere microorganisms in phosphorus uptake by plants. In “The Role of Phosphorus in Agriculture” (F. E. Khasawneh, E. C. Sample, and E. J. Kamprath, Eds.), pp. 617-654. Am. SOC.Agron., Madison, WI. Truog, E. (1930). The determination of the readily available phosphorus of soils. J. Am. Sor. Agron. 22, 874-882. Van Raij, B., Quaggio, J. A,, and Da Silva. N. M. (1986). Extraction of phosphorus, potassium, calcium and magnesium from soils by an ion-exchange resin procedure. Commun. Soil Sci.Planr. Anal. 17, 547-566. Van Raij, B., and Van Diest, A. (1980). Phosphate supplying power of rock phosphate in an oxisol. Fert. Res. 26, 1-10, Van Ray, B., and Van Diest, A. (1979). Utilization of phosphate from different sources by six plant species. Plant Soil. 51, 577-589. Waidyanatha, U. P. S., Yogaratnam, N., and Ariyante, W. A. (1979). Mycorrhizal infection of growth and nitrogen fixation of Pueraria and Stylosanthes and uptake of phosphorus from two rock phosphates. N. P h y o l . 82, 147-152.
PHOSPHATE ROCKS FOR DIRECT APPLICATION TO SOILS
159
Watkinson, J. H . , and Perrott, K. W. (1993). Estimating fertilizer substitution values from the dissolution rates of reactive phosphate rocks in New Zealand pastoral soils. Proc. Int. Grassland Congr. XVII New Zealand, 1460-1461. Watkinson, J. H. (1994a). Modelling the dissolution rate of reactive phosphate rock in New Zealand pastoral soils. Ausr. J. Soil Res. 32, 739-753. Watkinson, J. H. (1994b). Dissolution rate of phosphate rock particles having a wide range of sizes. Aust. J. Soil Res. 32, 1009-1014. Watkinson, J. H. (1994~).A test for phosphate rock reactivity in which solubility and size are combinet in a dissolution rate function. Ferr. Res. 39, 205-215. wdtkinson. J. H. (1995). Modelling RPR dissolution in soil. In “Fertilizer Requirements of Grazed Pasture and Field Crops” (L. D. Currie, Ed.). Occasional Report No. 8, Fertilizer and Lime Research Centre, Massey University, Palmerston North, New Zealand. Weil, S., Gregg, P. E. H.. and Bolan, N. S. (1994). Influence of soil moisture on the dissolution of reactive phosphate rocks. In “The Efficient Use of Fertilizers in a Changing Environment: Reconciling Productivity and Sustainability” (L. D. Currie and P. Loganathan, Eds.), pp. 7581. Occasional Report No. 7. Fertilizer and Lime Research Centre, Massey University, Palmerston North. Williams, C. ( 1980). Soil acidification under clover pasture. A m . J. Exp. Agric. Anim. Husb. 20, 561-567. Williams. R. (1937). Solubility of soil phosphorus and other phosphorus compounds in sodium hydroxide solution. J. Agric. Sci. 27, 259-270. Williams, J. D. H., Syers, J, K., and Walker, T. W. (1967). Fractionation of Soil Inorganic Phosphate by a Modification of Chang and Jackson’s Procedure. Soil Sci. SOC. Am. Proc. 31, 736-739. Wilson, M. A,, and Ellis, B. G. (1984). Influence of calcium solution activity and surface area on the solubility of selected rock phosphates. Soil Sci. 138, 354-354. Wright, R. J.. Baligar, V. C., and Belesky, D. P. (1992). Dissolution of North Carolina phosphate rock in soils of the Appalachian region. Soil Sci. 153, 25-36. Yeates, J. S., and Clarke, M. F. (1993). Developing alternatives to phosphate fertilizers of high water solubility. Ferr. Res. 36, 141-150. Yost, R. S., Kamprath, E. J., Naderman, G., and Lobato, E. (1981). Residual effects of phosphorus applications on a high phosphorus absorbing oxisol of central Brazil. Soil Sci. SOC. Am. J . 45, 540-543. Yost. R. S.. Naderman, G. C., Kamprath, E. I., and Lobato, E. (1982). Availability of rock phosphate as measured by an acid tolerant pasture grass and extractable phosphorus. Agron. J. 74, 462-468. Zapata. F., Casanova, E.. Salas, A. M., and Pino, I. (1994). Dynamics of phosphorus in soils and phosphate fertilizer management in different cropping systems through the use of isotopic techniques. Inr. Cong. Soil Sci. X V , Acapulco, Mexico. 5a, 451-466.
This Page Intentionally Left Blank
BREEDINGAND IMPROVEMENT OF FORAGE SORGHUMS FOR THE TROPICS R. R. Duncan University of Georgia, Griffin, Georgia 30223
I. Introduction 11. Genetic Parameters A. Nutritional Quality 111. Breeding A. Basic Hybrid Development B. Selection Techniques W. Germplasm V. Conclusions A. Biotechnology Potential B. Summary References
I. INTRODUCTION Sorghum belongs to the Tribe Andropogonae of the Graminae family Poaceae. It is a coarse, perennial grass that is treated as an annual in temperate or subtropical climes where cold temperature limits growth and development. In tropical regions, however, it can be harvested many times. Forage sorghums can be grouped into ( 1) greenchop-haylage-pasture grazing types that generally include sudangrass (subrace sudanense), sudangrass X sudangrass hybrids, or sorghum (bicoforgrain type) X sudangrass hybrids, (2) dual-purpose sorghums, usually bicolor grain types, tall 3-dwarf or normal 2-dwarf in height, that are characterized by high grain:stover ratios, (3) sweet sorghum cultivars or hybrids involving grain x sweet hybrids fed as greenchop, silage, or hay, (4) silage types, usually 0-, I - , or 2-dwarf types, that are ensiled, or ( 5 ) stubble residue (Reed et al., 1988) after grain harvest that is supplemented with high-protein legume forages and/or grain concentrates as a good temporary roughage source. All types, except the sudangrasses, produce some grain that adds to the feeding quality and enhances subsequent utilization. Forage sorghums in the tropics are 161 Admmm in A p n m y , Volume 17
Copyright 6 1996 by Academic Press, Inc. All rights of reproduction in any form resewed.
162
R. R. DUNCAN
used predominately as a temporary alternative to perennial forages when the latter are in a low biomass production cycle or their quality is not adequate to sustain animals. In general, temperate grasses are more digestible than tropical ones (Hoveland and Monson, 1980). High temperatures, low nutrient inputs, pathogen pressure, acid soil infertility and stresses, and genetic variability all contribute to this low digestibility problem in the tropics. This chapter will concentrate on the genetic improvement of forage sorghums. Environmental and management factors, as well as animal performance, will not be considered in this review.
11. GENETIC PARAMETERS Forage feeding value (FFV) involves a number of criteria, affecting both plants and animals. Several plant criteria are listed in Table I. Forage sorghums differ widely in chemical composition and nutritive value, both of which are genetically controlled (Hoveland and Monson, 1980; Gourley and Lusk, 1978). Genetic variability has been found for in vitro dry matter digestibility (IVDMD), acid detergent fiber (ADF), acid detergent lignin (ADL), cell wall components, and crude protein (CP) (Hoveland and Monson, 1980). Table I Forage Feeding Value Criteria to Be Considered in a Breeding Improvement Program" 1 . Nutritional quality (value)
Adequate content and balance of essential nutrients; high digestible energy and reasonable protein 2. Biomass yield Dry matter per unit land area; grain-to-stover ratio 3. Digestibility High intake and high nutritive value per unit intake 4. Physiological-chemical characteristics conducive to rapid comminution
High mmen outflow rate that promotes higher intake and greater net microbial yield per unit intake 5 . Nonstructural carbohydrate content Provides energy to facilitate high yield of microbial protein; promotes proprionate production. Negatively correlated with cmde protein content 6 . Lipid and condensed tannin content Lipids are used efficiently in ruminate weight gain, but high lipid content is inversely correlated with dry matter production. Tannins reduce rate and extent of protein degradation in the rumen; excess tannins reduce intake and decrease feed utilization efficiency a Adapted from Hoveland and Monson (1980), Wheeler and Corbett (1989), Gourley and Lusk (1978), and Kalton (1988).
FORAGE SORGHUMS FOR THE TROPICS
163
A. NUTRITIONAL Q U A L ~ Good forage quality can be associated with a high leafstem ratio (Hoveland and Monson, 1980). This ratio can be manipulated by the introduction (usually through backcrossing) of one or more dwarfing genes, which shortens the stem internodes without otherwise altering the plant (Burton er al., 1969). Plant height and DM yields are reduced, but leaf percentage, IVDMD, and CP are increased (Burton et al., 1969). The concentration of energy and protein in the forage is significantly increased (Hoveland and Monson, 1980). The reduced yields can be compensated for somewhat by developing hybrids and by taking advantage of hybrid vigor. Increasing stem diameter and nonlignified cellulose (more pith parenchyma) may also be used to increase IVDMD of sorghum (Bums et al., 1970). Microanatomical features of leaves and stems affect digestion rates in ruminants. Cultivars within species vary in leaf and stem vascular bundles, mesophyll, and cutinized and lignified cell walls (Hanna er al., 1973, Schertz and Rosenow, 1977), and the variability can be used in a selection program. The most rapid method of improving forage sorghum quality is to improve IVDMD (Pedersen et al., 1982). When IVDMD is increased, winter hardiness is decreased, prussic acid glycosides may be increased, and maturity (photoperiod sensitivity) is extended (Hoveland and Monson, 1980). General combining ability (GCA) effects are important for DM, IVDMD, and Brix (total soluble carbohydrates in squeezed-stem exudate) (Pedersen er al., 1982, 1983). Sugars and starches are the principal energy-storage compounds in the sorghum plant. One major recessive gene affects soluble carbohydrate accumulation in plants, but modifier genes condition the magnitude of this accumulation (Gourley and Lusk, 1978). High nonstructural carbohydrates are negatively correlated with CP content and increased concentrations could reduce fiber digestion in the rumen because by-products of fermentation reduce pH (Wheeler and Corbett, 1989). A 3: 1 ratio of acetate:propionate in rumenal fluid can be used as an indicator of desirable forage carbohydrate concentrations (Corbett, 1976). Since nonstructural carbohydrates are beneficial for rumen microbial growth, selecting for a lower ratio of structural to nonstructural carbohydrates and reducing forage lignin contents might be more effective in the selection program rather than directly increasing carbohydrate concentrations, as long as the structural integrity (standability) of the plant is not sacrificed (Wheeler and Corbett, 1989). 1. Protein
Protein content is often used as a measure of forage quality. Forage protein levels lower than 7 to 8% may have adverse effects on IVDMD and intake (Milford and Minson, 1965). Temperate grasses are generally higher in protein than are tropical grasses, but management and environment can contribute to this
164
R. R. DUNCAN
difference more than inherent genetic capacity (Hovelands and Monson, 1980). Protein digestibility is positively related to plant protein concentration, but the digestibility component is rarely addressed directly in plant breeding programs. An average of 20% of the crude protein of dried sorghum forage may be unavailable to ruminants (Gourley and Lusk, 1978). This undigestible crude protein is positively correlated ( r = 0.62) with tannin content. Tannins can reduce protein digestibility by rumen microflora. High CP is associated with plant leafiness and with a higher proportion of chloro-plastic protein, lipids, and sulfur-amino acids (Wheeler and Corbett, 1989). Methionine, the essential sulfur-amino acid, might be increased more effectively by genetically engineering rumen bacteria rather than by modifying plants (Wheeler and Corbett, 1989).
2. Antiquality Factors Secondary metabolites (tannins, cyanogenic glycosides) under genetic control in forage plants may affect animal response (palatability, intake potential, digestibility) directly or indirectly (Barnes and Gustine, 1973). Comprehensive discussions on forage phenolics, lignins, flavonoids, and tannins are published (Mole, 1989; Fahey and Jung, 1989; Grisebach, 1981). a. Forage Tannins Tannins are a complex group of astringent, phenolic compounds that have been associated with lowered digestibility and intake (Barnes and Gustine, 1973). A beneficial effect of tannins is as a protein-binding agent that protects the protein from rumen microbial degradation and thus increases N utilization (Hoveland and Monson, 1980). Moderate levels can reduce or prevent bloat and control fungicidal activity in ruminants. Sorghum forages generally do not contain condensed tannins, but flavonoid monomers such as leucoanthocyanidins (flavan-3,4-diols), which are responsible for the false-positive tannin reactions (Walton et al., 1983). All tannins are water-soluble polyhydroxy phenols that precipitate proteins from solution (Gourley and Lusk, 1978). The flavonoids do not precipitate proteins and are therefore not a tannin. Tannins vary among organs of a plant, with monomers being synthesized continuously in younger leaves and in the developing seed during early grain filling (Gourley and Lusk, 1978). The condensing of the monomeric units of low molecular weight (flavanols) during early grain fill causes astringency. As the leaf or seed ages, the monomers are polymerized and astringency decreases (Swain, 1965). The principal flavonoids in sorghum leaves are monomeric leucoanthocyanidins (LAC) (Swain 1969, which are different from the polymeric proanthocyanidins (linamarin and lotaustralin, which are cyanogenic glycosides) found in Lotus corniculatus (Seigler, 1976; Ross and Jones, 1983; Watterson and Butler, 1983). Sorghum leaves contain about 74% of
FORAGE SORGHUMS FOR THE TROPICS
165
the total plant flavonoids, even though the leaves contribute only about 2 1 % to the total plant dry matter (Gourley and Lusk, 1978). LAC content is governed by a single, incompletely dominant allelic gene pair (Haskins and Gorz, 1986a). Variability as great as 25 X in LAC content between cultivars has been found (Haskins and Gorz, 1986b). Whole-plant tannin levels are negatively correlated with CP and IVDMD, and positively correlated with fibrous constituents (Montgomery et al., 1988). Whole plant ADF is higher for low-"tannin" forage hybrids, as opposed to high tannin types. ADF and NDF decrease, while IVDMD increases from boot to hard dough stages (Montgomery et al., 1988). Tannins inhibit the f3-glucosidase-mediated hydrolysis of cyanogenic glucosides in vitro (Goldstein and Spencer, 1985). Red pigments (apigeninidin plus one of its derivatives and two luteolinidin derivatives) are negatively correlated with in vitro fiber digestion (Reed et al., 1987; MuellerHarvey and Reed, 1992). Phenolic acids characterized from sorghum forages include p-hydroxybenzaldehyde (PHBAL), vanillic acid (VA), caffeic acid (CA), vanillin (VAN), trans-p-coumaric acid (PCA), trans-ferulic acid (FA), sinapic acid (SA), p-hydroxybenzoic acid (PHBA), and syringic acid (SYA) (Cherney et af., 1989). Total and aqueous ethanol-soluble phenolics in sorghum leaves are negatively correlated with digestibility of NDF (Reed et al., 1987). Both soluble and alkalilabile phenolic compounds are involved in inhibition of structural carbohydrate digestion (Cherney et al.,1989; Reed et al., 1988). Alkali-labile PCA and FA are higher in stem than in leaf tissues and their solubility in neutral detergent (ND) is generally <30%, probably because they are esterified to the cell wall lignin (Cherney et al., 1991). The solubility of the other minor alkali-labile phenolics is >50% because they accumulate in soluble form with less cell wall esterification. Sorghum plants produce large amounts of very diverse phenolic compounds (Butler, 1989). Several derivatives of cinnamic acid (particularly p-coumaric acid) and of monomeric and oligomeric flavonoids occur in sorghum tissues (Ring et al., 1988). In leaf blades, flavone 7-0-glucosides (apigenin and luteolin 7-0-glucosides) are present (Mueller-Harvey and Reed, 1992). In leaf sheaths, flavones (apigenin and luteolin) and 3-desoxyanthocyanidins (apigenenidin aglycone, apigenindin, and luteolinidin derivatives) are found in low-tannin cultivars. The luteolinidin derivative and apigeninidin occur in 90% of high-tannin leaf sheaths, while the apigeninidin derivative occurs in 38% and a second luteolinidin derivative in only 10% of the samples (Nicholson et al., 1988; Mueller-Harvey and Reed, 1992). Flavones (five luteolin derivatives in leaf blades, one apigenin derivative in leaf blades, and one unknown flavone derivative in leaf blades and leaf sheaths) are negatively correlated (r = -0.24--0.47 with in vitro digestibilities) (Mueller-Harvey and Reed, 1992). Five flavanones/dihydroflavonols in leaf sheaths are also negatively correlated ( r =
166
R. R. DUNCAN
-0.27--0.57) with digestibility. One flavonone (butin) is highly (P < 0.001) negatively correlated ( r = -0.54--0.57) with digestibility, suggesting that ortho-di-OH groupings in the flavonoid B-ring (butin, luteolin) are associated with lower digestibilities (Mueller-Harvey and Reed, 1992; Julian et al., 1971). Negative correlations ( r = -0.22--0.5 1) with cinnamic acid derivatives are found in leaf sheaths (Mueller-Harvey and Reed, 1992). Butin is strongly correlated with red pigments (Reed et al., 1987) (soluble A550:r = 0.81, P < 0.001) and insoluble 3-desoxyanthocyanidins (1A:r = 0.91, P < 0.001) (MuellerHarvey and Reed, 1992). Flavone aglycones and glycosides are present in leaf blade phenolics of high-tannin and low-tannin cultivars and in leaf sheath phenolics of low-tannin cultivars (Mueller-Harvey and Reed, 1992). Sorghum plants containing the single recessive gene for tan plant color inherently have a low (8%) tannin content (Gourley and Lusk, 1978). Purple plant types, in contrast, range from 10 to 18% leaf tannins (18 to 31 mg/g total plant). Three dominant genes are required to produce the brown testa in high-tannin grain types (Gourley and Lusk, 1978). b. Cyanogenic Glycosides Sorghum leaves contain the cyanogenic glucoside dhurrin (p-hydroxy-(S)mandelonitrile-P-D-glucoside).Dhumn and its catabolic enzymes are compartmentalized in young tissue of green seedlings. Glycosides are stored in the vacuole of epidermal cells (Saunders and Conn, 1978; Saunders et al., 1977), and the catabolic enzymes are localized in the mesophyll cells (Thayer and Conn, 1981). Degradation of dhurrin yields equimolar amounts of prussic acid or hydrocyanic acid (HCN), glucose, and p-hydroxybenzaldehyde (p-HB) (Haskins et al., 1988; Halkier and Moller, 1991; Cutler et al., 1981; Conn, 1980). Large amounts of dhurrin may be produced rapidly when plants are environmentally stressed (drought, frost) (Barnes and Gustine, 1973) and when leaf tissues are disrupted. HCN is readily absorbed into the bloodstream of grazing ruminants, causing cellular asphyxiation and eventual death (Hoveland and Monson, 1980). Dhurrin turns over rapidly in sorghum seedlings (Adewusi 1990). The rate of synthesis in the shoot has been calculated at 17.4 nmol/h, while the breakdown rate is 4.8 nmol/h. In roots, the synthesis rate is 4.1 nmol/h, while the breakdown rate is 1.4 nmol/h. Consequently, the breakdown rate can range from 27 to 34% of the synthesizing capacity (Adewusi, 1990). Cyanogenic glycosides serve a protective function since more dhurrin is accumulated than is catabolized, resulting in the storage of excess dhurrin for repelling predators (Jones, 1979). A primary metabolic function of dhurrin is to provide carbon atoms for the formation of P-cyanoalanine and asparagine (Jones, 1979), or ubiquinone via in vitro oxidation of p-hydroxybenzaldehyde to p-hydroxybenzoic acid (Moller and Conn, 1979). Hydrocyanic potential (HCN-p) inheritance studies have revealed a single
FORAGE SORGHUMS FOR THE TROPICS
167
major gene intermediate in dominance (Gorz et al., 1986a,b), quantitative inheritance with additive genetic effects (Lamb et al., 1987), and multiple loci affecting this trait (Kalton, 1988). Low HCN-p is partially dominant to high H C N p (Lamb er al., 1987). LAC and dhurrin levels are negatively associated ( r = -0.68) (Haskins and Gorz, 1986b). They are also inherited independently and are not linked (Haskins and Gorz, 1988). Since both are aromatic compounds, this negative association may result from competition for intermediates or products of the aromatic biosynthetic pathway or from a linkage between genes governing LAC and HCNp levels (Haskins and Gorz, 1986b). Nevertheless, neither trait would be a reliable indicator of the other in a selection program, and high LAC forage should not be assumed to be free of cyanide poisoning. Sorghum forage with 1750 p@g HCN is generally not detrimental to ruminant absorption via grazing (Elder and Denman, 1966; Bennett et al., 1990), although forage hybrids harvested on 60-day cycles seldom exceed the “danger threshold” of 200 Fg/g HCN in Puerto Rico (Torres-Cardona et al., 1983), unless severe stress (drought, temperature) is involved.
3. Lignin and Brown Midrib Mutants Lignin is a complex, amorphous biopolymer matrix formed in vascular plants to maintain structural integrity of cell wall tissues. Monomers of this complex constituent are (4-hydroxyphenyl) propenes that are synthesized by diaminating tyrosine and/or phenylalanine (Bucholtz er al., 1980). Lignin is a major factor limiting the extent of digestibility of cell wall polysaccharides by animals (Jung and Fahey, 1983; Grisebach, 1981). The indigestible neutral detergent fiber (INDF) is the component limiting intake of less digestible forages, and a ratio of 7 digestible organic matter: 1 INDF is generally desirable (Lippke, 1986). The usefulness of forage in ruminant diets is limited by the quantity of lignin; consequently, an effective method for improving energy availability is to reduce or alter lignin content. Normal sorghum genotypes vary by 2 X in cell wall lignin concentration (Cherney et al., 1991a,b). Brown-midrib (bmr) mutations are one way to modify lignin quantity and quality in forages (Cherney, 1990). Chemical mutagen treatments (Porter et al., 1975, 1978) have been used to induce reddishbrown pigmentation in the leaf midrib that is associated with lignin, and the pigmentation persists in the cell wall residue after hemicellulose or cellulose are removed from the cell wall (Cherney ef al., 1991a,b). The pigmentation is most conspicuous in the leaf midrib on the abaxial side and on the stem covered by the leaf sheath (Porter et al., 1975). The mutant lignin contains -3 X more aldehyde groups than normal lignin, and there is less lignin in the mutant than in the normal (Bucholtz et al., 1980). The mutagenesis causes the accumulation of aldehyde lignin intermediates for incorporation into the lignin polymer. The bmr
168
R. R. DUNCAN
mutant increases polysaccharide digestion by ruminant animals (Porter et al., 1978). bmrd mutants have 16% lower lignin concentration and substantially lower PCA concentrations that result in higher IVDMD than normal types (Cherney et al., 1991a,b). bmr-12 and bmr-18 lines have 8 and 1 1 % higher IVDMD and 30 and 32% lower lignin contents than normal lines (Hanna el al., 1981), respectively. The bmr- 12 mutation has improved microbial degradation of marginally digestible tissue (leaf sclerenchyma, parenchyma bundle sheath, midrib parenchyma), which is part of the poorly degraded INDF portion of forages (Akin et al., 1986). p-Coumaric acid concentration and the linear xylan content are important factors limiting the rate and extent of cell wall digestion (Akin et al., 1986). Brown midrib genotypes have a significantly lowerp-coumaric acidferulic acid ratio than normal genotypes. Phenolic-carbohydrate complexes with low molar PCA:FA ratios and high xy1ose:arabinose ratios inhibit in vitro rumenal fiber digestion (Cherney et al., 1992). The negative effect on digestion occurred when FA content was higher than PCA. The rate of NDF digestion does not differ between bmr and normal types, but the extent of NDF digestion was greater for bmr types (Fritz et al., 1990). Increased p-coumaric acid concentration and linear xylan content are associated with decreased rate and extent of NDF digestion (Fritz et al., 1990). Hemicellulosic xylans form covalent complexes with lignin (Morrison, 1974). Linear xylan polymers have a higher phenolic acid content (especially p-coumaric acid) than branched xylans (Bittner, 1983). Noncovalent binding between arabinoxylans and cellulose fibers is greatly influenced by frequency of arabinosyl substitutions on the xylose backbone (McNeil et a!. , 1975). These arabinose sidechains disrupt binding by causing termination of binding sites (McNeil et al., 1975). Removal of arabinosyl sidechains increases the extent of degradation (digestibility) of grass hemicellulose (Brice and Morrison, 1982). Consequently, the degree of lignification is an important factor in determining the degradability of lignin-hemicellulose complexes. The amount of linear xylan, as compared to branched xylan, also influences the rate and extent of cell wall digestion (Fritz et al., 1990). Linear xylans have a greater propensity to form physicaUchemica1 associations with additional cell wall components (lignins) and this complex lignification decreases cell wall digestibility (Fritz et al., 1990). Cell wall polysaccharides of sorghum increase in proportion of dry matter until physiological grain maturity (Goto et al., 1991). Even though the nonstarch polysaccharides increase, the composition of the cell-wall structural polysaccharide components (5% arabinose, 28% xylose, 63% glucose) remain constant during growth and maturation. Cellulose and arabinoxylan increase as a proportion of dry matter during immature stages. With growth, xylose residues free of glycosidic side chains and of arabinose residue-carrying alkali-labile substituents increase proportionally. The extent of alkali-labile linked substitution of arabinose residues is highly correlated (r = 0.90) with polyphenolic material, suggesting the lignin-hemicellulose complex was formed at least partly through
FORAGE SORGHUMS FOR THE TROPICS
169
arabinose residues (Goto et al., 1991). Ferulic acid remains static during growth, but decreases as a proportion of total phenolic acids as the plant ages. It is associated with the primary cell wall and makes a larger contribution to the total phenolic acid content of young plants. In contrast, p-coumaric acid is associated with secondary cell wall formation, and is characteristic of lignified tissue. It acts as a marker for the degree of lignification and for the extent of ester bonding between lignin and cell-wall polysaccharides (Chesson, 1988). Cellulose degradability is more highly correlated with the polyphenolic content (r = -0.98) than withp-coumaric acid content (r = -0.76) (Goto et al., 1991). Cellulose degradability is also correlated with the proportion of arabinose residues carrying alkalilabile substituents (r = -0.92). Consequently, the lignin-hemicellulose complex containing ester linkages formed between arabinose and phenolic residues is present. Decreased sorghum degradability during growth and maturation is ascribed to an overall increase in lignin content and an accompanying increase in lignin-hemicellulose bonding as indicated by the increase in alkali-labile substitution of arabinose residues (Goto et al., 1991). This core lignin (non-alkalilabile polyphenolic complex) exerts control over incorporation of phenolics into the lignin polymer (Chesson, 1988). Cellulose degradability decreases from 83% at the seedling stage to 37% at the milk-dough stage in sorghum (Goto et al., 1991). Degradabilities of arabinose and uronic acid residues are consistently higher than those of xylose and glucose, the main monosaccharide components of structural carbohydrates contributing to cell wall polysaccharides. Total nonstarch polysaccharides increase from 3 1% at the seedling to 45% at the boot stage, but change little after that (Goto et al., 1991). Cellulose content increases in similar fashion, from 14 to 22% of dry matter. Nineteen bmr phenotypes have been identified (Porter et al., 1978), but bmr-6, bmr-12, and bmr-18 have been used to backcross the trait into sudangrass and other grain sorghums (Fritz et al., 1981). Allelism tests of bmr mutants indicate that more than one locus with possible modifying genes is involved in the expression of this trait (Bittinger et al., 1981). Intragenic complementation may be involved since bmr-12 and bmr-18 are allelic, while bmrd apparently is located on a different portion of the genome (Bittinger et al., 1981). Four nonallelic genes function in com (Zea mays L.), and double mutants possess less lignin than single mutants (Kuc and Nelson, 1964). These bmr genes appear to be simple recessives (Bittinger et al., 1981). As many as three recessive brnr genes may influence low lignin concentration in sorghum (Gourley and Lusk, 1978).
4. Lipids: Bloom vs Bloomless Epicuticular waxes are deposited on the external surfaces of leaf laminae, leaf sheaths, stems, and grain (Cannon and Kummerow, 1957). The primary fatty acid (FA) of the leaf and stem wax is the unsaturated linolenic, but sorghum
170
R. R. DUNCAN
leaves and stems are generally lower in this FA than saturated FAs such as palmitic (Burnett et al., 1958; Burnett and Lohmar, 1959). Plant waxes are produced throughout growth and development, with leaves containing 0.300.33% and stems 0.60% total wax at maturity (Pedersen et al., 1983). The epicuticular wax exudate is commonly referred to as “bloom” or glaucousness and is characterized by a white, powdery appearance on the leaves and stems. The amount of wax present varies from a heavy, uniform covering (bloom) to a light covering at critical areas (sparse-bloom), to no exudate present (bloomless or nonglaucous) (Peterson et al., 1982). Bloomless types have 22% higher IVDMD than bloom types, and forage quality can be improved by selecting for the bloomless characteristic (Cummins and Dobson, 1972; Cummins and Sudweeks, 1975). Two genes control the expression of bloom in normal sorghums (Ayyangar et al., 1937; Ayyangar and Poinnaiya, 1941). Bm (bloom) is completely dominant over bm (bloomless) and H (bloom) is completely dominant over h (sparsebloom) in the presence of BmBm or Bmbm, but with no visible expression in the presence of bmbm. Homozygous recessive alleles at two different loci condition the expression of bloomlessness, with gene designations of bm, and bm, (Peterson et al., 1982). Expression of sparse-bloom is governed by homozygous recessive alleles at a minimum of three loci, with gene designations of h, , h,, and h, (Peterson et al., 1982). The bloomless and sparse-bloom genes are not allelic and segregate independently. Sorghum seedlings can be morphologically nonglossy (normal seedings with dark green leaves) and glossy (seedlings with light yellow-green color and shiny leaf surfaces) (Maiti et al., 1984). Only 2% of the world collection of sorghums is classified as glossy. Durra, durra-bicolor, durra-caudatum, durra-kafir, and durra-guinea types account for about 96% of the glossy types (Maiti et al., 1984). The glossy trait is correlated with a reduction or absence of wax deposits on leaf surfaces (Traore et al., 1989). Inheritance studies show that glossy is a simple recessive to nonglossy (Tarumoto, 1980). Glossiness is clearly manifested during the seedling stage and gradually disappears as the seedling grows (Maiti et al., 1984). This trait would be important where ruminants are grazing the forage.
5. Stem Sweetness Forage sorghum dry matter increases with maturity of the plant, and -50% of the nonstructural carbohydrates (NSC) is partitioned to the stem, whereas 15% is allocated to the leaves (Eilrich et al., 1964). The cell wall is a composite matrix of cellulose microfibrils embedded in an amorphous gel containing a relatively wide range of polysaccharides. Regions may be enriched with specific types of polysaccharides, but without clear delineation of pectic and hemicellulosic fractions. The interactions among polysaccharides and satellite components are more
FORAGE SORGHUMS FOR THE TROPICS
171
important to function and use of the cell wall than the absolute classification of individual polysaccharides (Eilrich et al., 1964). The three groups of chemically fractionated structural polysaccharides include ( 1 ) pectic polysaccharides that are rich in galacturonic acid residues, (2) hemicellulosic polysaccharides that are rich in xylose, and (3) the cellulose fraction (Eilrich et al., 1964). The NSC and cellulose fractions are higher in the stem than in the leaf (Stallcup et al., 1964; McBee and Miller, 1990). Hemicellulose contents are higher in blades than stems (McBee and Miller, 1990). Larger average leaf areas per plant and larger CO, exchange rates are associated with greater concentration of NSC (Vietor and Miller, 1990). Inheritance of stem sweetness in sorghum is qualitative, with nonsweet being monogenically dominant over sweet (Bangarwa ef al., 1987). Sweetness is controlled by a single recessive gene.
III. BREEDING
A. BASICHYBRID DEVELOPMENT Parents used in producing sorghum hybrids are designated as male-sterile or “A”-lines, sterility maintainers or “B”-lines, and restorer or “R”-lines (House, 1985; Doggett, 1988). Most sorghum hybrids involve three-way crosses. The male-sterile parent is developed by crossing with pollen from a maintainer line: A-line
X
B-line
sterile A line.
The maintainer line is perpetuated by selfing. The A and B lines are phenotypically isogenic. The interaction of “kafir” nuclear genes with “milo” cytoplasm causes sterility, and is governed by ms, genes. Hybrid seed is produced by crossing the male-sterile line by pollen from a restorer line: A-line
X
R-line + male-fertile F, hybrid.
Hybrid forage sorghums can be developed using several combinations. (1) Male-sterile grain-type X male-fertile grain type + dual-purpose or silage hybrids (2) Male-sterile grain-type X male-fertile sudangrass + sorghum-sudangrass hybrid (3) Male-sterile sudangrass (Alam and Sandal, 1967) X male-fertile sudangrass += hybrid sudangrass Other options include (a) using photoperiod-sensitive-producing restorer lines to develop hybrids that do not produce panicles (remain vegetative) in temperate
172
R. R. DUNCAN
environments, and (b) using sterility-inducing sudangrass pollinator lines such that hybrids are completely sterile and will not outcross with wild progenitors to produce weedy production problems. Forage sorghum production in the Caribbean has potential for 25 T/ha of dry forage in 180 days with the best hybrids (Sotomayor-Rios and Santiago, 1981; Sotomayor-Rios and Telek, 1977; Sotomayor-Rios and Torres-Cardona, 1984a,b, 1990). Crude protein content up to 15% can be achieved (Sotomayor-Rios and Santiago, 1981). A three-way hybrid of [(ATX624 X B “Rhodesian” sudan) X common sudan] produced the highest dry forage and crude protein yield, compared to single cross hybrids (ATX623 X “Greenleaf” sudan) and (ATX624 X “Common” sudan) (Sotomayor-Rios et af., 1985). The sweet sorghum “Rio” produced low dry forage yields, but high IVDMD (Torres-Cardona et af., 1986). “Florida 357” was superior in having low HCNp among male parents, while Lahoma had the highest yield potential of male parents, but had an intermediate HCNp level (Sotomayor-Rios and Torres-Cardona, 1984a). A Rhodesian is a line developed from Rhodesian sudangrass (PI 156549) and CK-60 (Craigmiles, 1961). This particular male-sterile line has shown excellent forage potential with various male parent combinations in Puerto Rico (Sotomayor-Rios and Santiago, 1981). A Rhodesian X Common sudangrass produced 20 T/ha dry forage in 140 days with 14% CP at 30-day cutting intervals. Sorghum male-steriles by sudangrass and hegari have potential for high forage production in Japan (Tarumoto, 1975, 1978). “Millo blanco” is a cultivar well-adapted to the tropics of the Caribbean and is highly tolerant to drought and soil stress problems indigenous to the region (Sotomayor-Rios and Telek, 1977). Millo blanco hybrids are potentially good forage producers, producing 17 T/ha dry forage with 18% CP in 210 days. Cytoplasmic male-sterile X adapted photoperiod sensitive sorghum offer tremendous potential for producing high-quality, high-yielding dry forage in the tropics based on extended vegetative growth during long days ( 1 2+ h) (Sotomayor-Rios et af., 1986). Traits receiving the most attention for commercial sorghum silage breeding improvement include (Kalton, 1988): total biomass production, standability, disease resistance, insect resistance, male-sterility, seedling cold tolerance, and drought tolerance. Improved nutritional quality traits include: better grain-toforage ratio, leafiness, green leaf retention (stay-green), juicy sweet stalks, lower lignin concentration, tan plant color, non-testa grain, and brown midrib. Sudangrass improvement objectives include (Kalton, 1988): total biomass production, leafiness, ratoonability (regrowth potential) (Duncan et al., 1980), disease resistance (Duncan, 1984), low HCN-p, insect resistance, tillerability, seedling cold tolerance, green leaf retention, later maturity, tan plant color, finer stems, male-sterility, seed production, and brown midrib. Breeders must also screen their lines for combining ability effects and, in the
FORAGE SORGHUMS FOR THE TROPICS
173
absence of specific combining ability effects, produce hybrids among various parental combinations that combine the best grain and forage traits (Ross et al., 1980). High-yielding hybrids with high-forage-quality traits are difficult to achieve. Sterilization of locally adapted photoperiod sensitive lines [e.g., rnillo blanco (SC0971) is being sterilized using 10 CMS sources] will provide excellent CMS sources for use in hybrid combinations with desirable forage or sudangrass sources (Sotornayor-Rios et a l . , 1986; Miller, 1986; Weibel et al., 1984). This approach is essential since most of the quality traits are quantitative and recessive, requiring both parents in hybrid combinations to possess genes for the trait. Three-way crosses [(S. bicolor CMS x S . arundinaceum) x desirable sudangrass line] offer the potential for utilization of wild species for improved forage production in the tropics (Sotomayor-Rios et al., 1986). Concurrent to these approaches, parental line sources must be identified that have improved disease resistance, and improved soil stress (acid soil problems, drought) tolerance in combination with better agronomic potential and enhanced forage quality.
.
B SELECTION TECHNIQUES Recurrent restricted phenotypic selection (RRPS) has effectively increased (16% per cycle) forage yields of Pensacola bahiagrass (Paspalum notatum var. saure Parodi) without reducing IVDMD (Burton, 1982). Use of grid selection with a t-score adjustment (for variation among block means and for differences among blocks in their phenotypic variation) increased the efficiency of phenotypic recurrent selection for low NDF concentration in smooth bromegrass (Bromus inermis Leyss.) (Casler, 1992).
W. GERMPLASM A collaborative program between the Genetic Resources Unit of ICRISAT and the National Bureau of Plant Genetic Resources, Indian Council of Agricultural Research, has resulted in a comprehensive evaluation of forage sorghums in the World Collection (Mathur et a l . , 1991, 1992). Several random-mating populations have been developed in Puerto Rico (Webster, 1976; Sotomayor-Rios et al., 1984) that are sources for photoperiod-sensitive and -insensitive breeding lines (Table 11). Low dhurrin sudangrasses are available for improvement programs (Sotomayor-Rios et al., 1984; Gorz et al., 1984, 1986,a,b, 1990a,b,c,f; Haskins et al., 1990a,b). Brown midrib sources are available (Gorz et al., 1990d,e,f). Bloomless and
R. R. DUNCAN
174
Table 11 Germplasm Sources with Potential Adaptation to Tropical Environments for Use in Breeding Programs -
Name
Description
Designation
PRlBR PR2BR PR3BR PR4BR PR5BR
Population Population Population Population Population
ms7
NP22
Population
NP23,24
Populations
NP25
Population
NP28 NP29
Composite Population
NP30
Population
NP3 1.32
Populations
NP33-35
Populations
NP36,37
Populations
N97 N98-100 N10 I- I03 N 104- I07
Inbred R line R lines B/R lines B lines
N108- I 1 I N112- 121
B lines B lines
OK GPI 1-16 OK GP17-26 PL162- 169
B lines R lines A/B lines
GA 337 Greenleaf
R line R line
Sudangrass, low dhurrin, ms3, from Piper and NP2B Sudangrass, low dhumn, ms,. sweetjuicy culms Sudangrass, low dhurrin, ms,, dry stemmed Sudangrass, low dhurrin, ms, Sudangrass, low dhunin, germination/ seedling cold tolerance Sudangrass, low dhunin, good regrowth potential Sudangrass, low dhurrin, ms3-NP31, B-lines(NP32) Sudangrass, ms,(NP34,35), low dhurrin(NP35) Grain sorghum, low dhurrin, bmr-6(NP37), B line sources, ms3 Low dhurrin Sweet sorghum forages Low dhurrin, bmr-6 (N103) Bloom, green midribs(N104) Bloom, brown midribs(Nl08) Bloomless, green midribs(Nl06) Bloomless, brown midribs(N 107) Sweet sorghums Sudangrass, low dhumn, bmr-6 (N120, 121) Bloomless, bm, Sparse-bloom, h, or h, Bloomless, biotype C and E greenbug resistance Sudangrass Sudangrass
ms7, 34 cytoplasms Source from TP4R Insect/disease resistance KP5 x millo blanco, forage
Reference Webster (1976) unpublished unpublished unpublished Sotomayor-Rios et al. (1984) Gorz et al. (1984) Gorz et al. (1986) Haskins et al. (1986) C o n et al. (1990a) Haskins et al. (1990a) Gorz et a!. (1 990b)
Haskins et al. (1990b) Gorz et al. (1990~) Gorz et al. (199Od)
Gorz et al. (199Oe) Gorz er al. (199Oe) Gorz et al. (199Oe) Gorz et al. (199Oe)
Gorz et al. (1990e) Gorz er al. (19900 Weibel (1986a) Weibel (1986a) Weibel (1986b) Burton ( 1964) Karper (1955)
175
FORAGE SORGHUMS FOR THE TROPICS Table I1 Name
Designation
Lahoma Piper Suhi-1 GPlR GPT2RB GP140-142 GPTM3BR 87BL2598 GPP4BR GTPP7R
R line R line Hybrid Population Population R lines Population R line Population Population
GPPSBR GAC102
Population R line
GC103, 104
R lines
(continued)
Description
Reference
Sudangrass Sudangrass Sudangrass Sorghum, acid soil tolerance, ms3 Anthracnose resistance, ms3 Acid soil/disease resistance Fusarium resistance, ms3 Disease resistance Acid soil tolerance White seeded, tan plant, disease resistance, ms3 Anthracnose resistance, ms3 Hegari forage line with acid soil tolerance, tissue culture regenerant Tx430 tissue-culture regenerated, acid soil tolerant
Kramer (1960) Smith et al. (1973) Craigmiles (1963) Duncan (1981) Duncan et al. (1982) Duncan (1984) Duncan et al. (1987) Duncan er al. (1988) Duncan er al. (1989) Duncan er al. (1990) Duncan er al. (1991a) Duncan et al. (1991b) Duncan et al. (1992)
sparse-bloom genetic sources have been released (Weibel 1986a,b). Acid soil stress-tolerant (Duncan, 1981, 1984; Duncan et al., 1989, 1991b, 1992), disease-resistant (Duncan et al., 1982, f987, 1988, 1990, 1991a), and tan plant (low tannin) (Duncan et al., 1990) germplasm sources have been developed. A Hegari line with acid soil tolerance could be useful in sorghum forage breeding programs in the tropics (Duncan et al., 1991).
V. CONCLUSIONS Selection and breeding of improved nutritious forages is paramount to enhanced animal performance. The emphasis must be on improvement and maintenance of forages high in digestible energy (Hoveland and Monson, 1980). Digestible energy affects intake of forages as well as subsequent ruminant performance. Since hybrids are rarely as good as the best parental line in IVDMD, highly digestible parents must be identified. A high leafstem ratio is associated with good forage quality (Hoveland and Monson, 1980). By introducing dwarf genes, the concentration of energy and protein and ultimate nutritive value can be significantly increased by emphasizing leafiness and fineness of stem (Burton ef
176
R. R. DUNCAN
al., 1969). Internode length, plant height, and dry matter yields are decreased, but IVDMD, CP, and leaf percentage are increased. Leaf area and height are positively correlated with dry and green forage (0.47 and 0.62, respectively) and CP (0.31, 0.50) (Sotomayor-Rios and Torres-Cardona, 1984b). In sorghum silage types, the grain component is an additional source of energy. Four brachytic dwarfing genes control height (Quinby and Karper, 1954), with tallness being partially dominant to shortness. As each recessive gene for dwarfness is replaced by a dominant gene for height, the plant grows taller, the percentage of stalk increases, and the percentage of grain decreases. The dilemma for the forage breeder is to balance stem biomass and grain production in a favorable grain-to-stover ratio that supplies high energy feed to ruminants. This favorable ratio and energy balance is usually accomplished with 2-dwarf dual purpose types (Gourley and Lusk, 1978). Long growth duration is significantly associated with high grain yield (Dalton, 1967). Four genes control sorghum maturity (Quinby, 1967). Late maturity is dominant to early maturity. Photoperiod sensitivity can be used advantageously in hybrid combinations for effective tropical forage production (Sotomayor-Rios and Torres-Cardona, 1984a). Plant fiber is a major constituent of forage crops, comprising 30 to 80% of the dry matter (Cherney et al., 1991). The usefulness of forages is primarily limited by the degradability of the plant fiber. The major constraint to plant cell wall digestion is lignin. The complexity of physiological-chemical properties of individual polysaccharides accounts for some diversity (< 400 g kg-1 in stubble residues to 900 g kg-1 utilization in immature plants) in cell wall degradation (Hatfield, 1989). Individual polysaccharide properties, however, are more important in controlling the composition and component interactions of the total cell wall matrix rather than their direct influence on degradation. Interactions within the cell wall matrix govern the speed and extensiveness of degradation (Hatfield, 1989; Jung, 1989). Selection of cultivars that partition more photosynthates to NSC, cellulose, and hemicellulose and less to lignin would be desirable for ruminant digestion (McBee and Miller, 1990). One major recessive gene causes accumulation of soluble carbohydrates in sorghum, but modifying genes condition the magnitude of accumulation (Gourley and Lusk, 1978). Dry-stem genotypes (identified by white leaf midrib) have about 8% less moisture than juicy-stem types (light green leaf midrib) in the stover at physiological grain maturity (Gourley and Lusk, 1978). The dry-stem trait (pithy cortex) is controlled by one single dominant gene (Gourley and Lusk, 1978). Dominant genes for low carbohydrate and dry stems could be helpful during the ensiling process. Forage protein content is positively correlated with phosphorus content (r = 0.75); height is negatively correlated with nitrogen (-0.73) and P (-0.74); Brix is negatively correlated with potassium (-0.71), and NDF is negatively correlated with P (-0.74) (Gorz et al., 1987). Consequently, mineral nutrient concen-
FORAGE SORGHUMS FOR THE TROPICS
177
tration and forage quality factor interactions should not be overlooked in breeding programs. Nitrate toxicity in sorghum forages generally occurs when nitrogen fertilization is excessive and moisture stress causes continued root uptake and concentration of nitrates in the lower stem portion with conversion to protein precursors terminated (Bennett et al., 1990). Photoperiod-sensitive hybrids can occasionally be a problem because of no sink for grain production. Nitrate-N contents less than 1000 pg/g dry w are safe, while levels exceeding 1500 pg/g can result in poisoning (Bennett et al., 1990). Proper management plus balanced fertilization can minimize the problem (Spoelstra, 1985).
A. BIOTECHNOLOGY POTENTIAL Modification of plant proteins by genetic engineering could be effective in improving this forage quality component (Wheeler and Corbett, 1989). An essential sulfur-amino acid, such as methionine, would be a quality-enhancing target either from the plant manipulation side or from the rumen bacteria modification side (Wheeler and Corbett, 1989). Transposon tagging could be used for molecular isolation of genes associated with a phenotype, but not affiliated with a specific gene product (Bennetzen et al., 1987; Fitzmaurice et al., 1992). Modification of the processes governing lignin biosynthesis would be one example. Genes controlling the bmr trait could be isolated using a controlling element system for transposon mutagenesis and tagging (Cherney et al., 1991a,b). The mutable “candy stripe” system in sorghum caryopses could also be used as a marker to isolate brnr genes. The candy stripe locus specifies a somatically and germinally unstable synthesis of anthocyanin pigments in the seed pericarp (Cherney et al., 1991a,b). Identification of the protein product from the bmr gene followed by DNA cloning would be a second method for isolating the gene (Cherney et al., 1991a,b). Identification of the single peptide differing between the mutant isoline and the normal line would suggest that the peptide is related to the mutant gene. Isolation of bmr genes, LAC genes, low HCN-p genes, and low arabinose residue-controlling genes and understanding their patterns of expression would offer tremendous potential in modifying quantitatively controlled traits for enhanced degradation and utilization of tropical forages.
B. SUMMARY Forage sorghum hybrids carrying the following genes or traits are necessary to improve quality: (1) brown midrib (trigenic, recessive inheritance) to reduce lignin content in leaves and stems, (2) bloomless (digenic, recessive) to aid
178
R. R. DUNCAN
rumen degradation, (3) glossiness (monogenic, recessive) to reduce wax load on leaf surfaces, (4) sweet stem (monogenic, recessive) to supply more energy, (5) nonlignified cellulose (more pith parenchyma in the stem) to facilitate rapid degradation, (6) desirable carbohydrate levels as indicated by a 3:l ratio of acetate:propionate in rumenal fluid, (7) low tannin content (multigenic) or low LAC-producer (single allelic gene pair) to reduce interference with protein digestion, (8) low contents of luteolin and its derivatives, butin, and other flavanones/dihydroflavonolsas well as high concentrations of apigenin derivatives (Mueller-Harvey and Reed, 1992), (9) tan plant (recessive, monogenic) to minimize tannin content, (10) low HCNp (quantitative, additive) to minimize prussic acid poisoning, (1 1) low arabinosyl-linked hemicellulose to enhance cell wall degradation, (12) high grain-to-stover ratio, (13) dry stem (monogenic, dominant) with increased pithy cortex and easier digestibility, (14) green leaf retention (Wanous et al., 1991), (15) ratoonability for multiple cuttings (Duncan et al., 1980), (16) tropical adaptation (Duncan et al., 1980, 1981; Mann et al., 1985; Gerik and Miller, 1984), and (17) acid soil tolerance (Duncan, 1991).
REFERENCES Adewusi, S. R. A. (1990). lhrnover of dhurrin in green sorghum seedlings. P h i Physiol. 94, 12191224. Akin, D. E., Hanna, W. W., and Rigsby, L. L. (1986). Normal-I2 and brown midrib-I2 sorghum. 1. Variations in tissue digestibility. Agmn. J. 78, 827-832. Alam, S., and Sandal, P. C. (1967). Inheritance of cytoplasmic male-sterility in sudangrass. Crop Sci. 7 , 668-669. Ayyangar, G. N. R., Rao, V. P., Nambiar, A. K., and Poinnaiya, B. W. X. (1937). The occurrence and inheritance of waxy bloom on sorghum. Proc. Indian Acad. Sci., 5B, 4-15. Ayyangar, G. N. R., and Poinnaiya, B. W. X. (1941). The occurrence and inheritance of a bloomless sorghum. Curr. Sci. 10, 408-409. Banganva, K. S., Grewal, R. P. S., and Lodhi, G. P. (1987). A single recessive gene for stem sweetness in sorghum. Euphyrica 36, 225-226. Barnes, R. F., and Gustine, D. L. (1973). Allelochemistry and forage crops. In “Antiquality Components of Forages” (A. G . Matches, Ed.), pp. 1-13. Spec. Pub. 4, Crop Sci. SOC. Amer., Madison, WI. Bennett, W. F., Ihcker, B. B., and Maunder, A. B. (1990). “Modern Grain Sorghum Production,” pp. 108. Iowa St. Univ. Press, Ames, I A . Bennetzen, J. L., Cresse, A,, Brown, W. E., and Lee, L. (1987). Molecular cloning of maize genes by transposon tagging with mutator. In “Plant Gene Systems and Their Biology” (J. L. Key and L. McIntosh, Eds.), pp. 183-204. A. R. Liss, New York. Bittinger, T. S., Cantrell, R. P., and Axtell, J. D. (1981). Allelism tests of the brown-midrib mutants of sorghum. J. Hered. 72, 147-148. Bittner, A. S. (1983). Alkali soluble phenolic acids from an xylan fraction isolated from reed canarygrass. J. Dairy Sci. 66, 920-923. Brice, R. E., and Morrison, I. M. (1982). The degradation of isolated hemicelluloses and lignin cellulose complexes by cell free, mmen hemicelluloses. Curbohydr. Res. 101, 93- 100.
FORAGE SORGHUMS FOR THE TROPICS
I79
Bucholtz, D. L., Cantrell, R. P., Axtell, J. D., and Lechtenberg, V. L. (1980). Lignin biochemistry of normal and brown midrib mutant sorghum. J. Agric. Food Chem. 28, 1239- 1241. Burnett, M. C., Lohmar, R. L.. and Dutton, H. J. (1958). Countercurrent distribution of sorghum lipids in leaf and stem extract. J. Agric. Food Chem. 6, 374-377. Burnett, M. C., and Lohmar, R. L. (1959). Fatty acids of sorghum leaf and stem. J. Agric. Food Chem. 7, 436-437. Burns, J. C., Barnes, R. F., Wedin, W. F., Rhykerd, C. L., and Noller, C. H. (1970). Nutritional characteristics of forage sorghum and sudangrass after frost. Agron. J. 62, 348-350. Burton, G. W. (1964). Georgia 337 sudangrass. Crop Sci. 4, 666. Burton, G. W., Monson, W. G., Johnson, J. C., Lowrey, R. S., Chapman, H. D., and Marchant, W. H. (1969). Effect of the d, dwarf gene on the forage yield and quality of pearl millet. Agron. J. 61, 607-612. Burton, G. W. ( 1 982). Improved recurrent restricted phenotypic selection increases bahiagrass forage yields. Crop Sci. 22, 1058-1061. Butler, L. G. (1989). Sorghum polyphenols. I n “Toxicants of Plant Origin. IV. Phenolics” (P. R. Cheeke, Ed.), pp. 95-121. CRC Press, Boca Raton, FL. Cannon, C., and Kummerow, F. A. (1957). A comparison of plant and grain wax from two varieties of sorghum. J. Am. Oil Chem. Soc. 34, 519-520. Casler, M. D. (1992). Usefulness of the grid system in phenotypic selection for smooth-bromegrass fiber concentration. Euphyrica 63, 239-243. Cherney, J. H. (1990). Normal and brown midrib mutations in relation to improved lignocellulose utilization. I n “Microbial and Plant Opportunities to Improve Lignocellulose Utilization by Ruminants” (D. E. Akin and L. G. Lungdahl, Eds.), pp. 205-214. Elsevier, Amsterdam. Cherney, J. H., Anliker, K. S., Albrecht, K. A , , and Wood, K. V. (1989). Soluble phenolic monomers in forage crops. J. Agric. Food Chem. 37, 345-350. Cherney, D. J. R., Patterson, J. A., Cherney, J. H., and Axtell, J. D. (1991a). Fiber and soluble phenolic monomer composition of morphological components of sorghum stover. J. Sci. Food Agric. 54, 645-649. Cherney, I. H., Cherney, D. I. R., Akin, D. E., and Axtell, J. D. (1991b). Potential of brown-midrib, low-lignin mutants for improving forage quality. Adv. Agron. 46, 157- 198. Cherney, D. J. R., Cherney, J. H., Patterson, J. A., and Axtell, J. D. (1992). In vitro ruminal fiber digestion as influenced by phenolic-carbohydrate complexes released from sorghum cell walls. Animal Feed Sci. Tech. 39, 79-93. Chesson, A. (1988). Lignin-polysaccharide complexes of the plant cell wall and their effect on microbial degradation in the rumen. Anim. Feed Sci. Tech. 21, 219-228. Conn, E. E. (1980). Cyanogenic glycosides. I n “Encyclopedia of Plant Physiology, New Series, Secondary Plant Products,” Vol. 8 (E. A. Bell, and B. V. Charlwood, Eds.), pp. 461-492. Springer-Verlag, Berlin. Corbett, J. L. (1976). Nutritional aspects of the growth of grazing animals. froc. Austr. SOC. Anim. Prod. 11, 281-288. Craigmiles, J. P. (1961). The development, maintenance, and utilization of cytoplasmic male sterility for hybrid sudangrass seed production. Crop Sci. 1, 150-152. Craigmiles, J. P. (1963). Registration of varieties of sudangrass. Suhi-I (Rhodesian grass-sudangrass hybrid). Crop Sci. 3, 367. Cummins, D. G . , and Dobson, J. W., Jr. (1972). Digestibility of bloom and bloomless sorghum leaves as determined by a modified in vitro technique. Agron. J. 64, 682-683. Cummins, D. G . , and Sudweeks, F. W. (1975). I n vivo performance of bloom and bloomless sorghum. Agron. J. 68, 735-737. Cutler, A. J., Hosel, W., Sternberg, M., and Conn, E. E. (1981). The in vitro biosynthesis of taxiphyllin and the channeling of intermediates in Triglochin mariima. J. Eiol. Chem. 256, 4253-4256.
180
R. R. DUNCAN
Dalton, L. G. (1967). A positive regression of yield on maturity in sorghum. Crop Sci. 7, 271. Doggett, H. (1988). “Sorghum.” Longmen Scientific, UK. Duncan, R. R. (1981). Registration of GPlR acid soil tolerant sorghum germplasm population (Reg. No. GP73). Crop Sci. 21, 637. Duncan, R. R. (1984a). Registration of acid soil tolerant sorghum germplasm (Reg. Nos. GP140-142). Crop Sci. 24, 1006. Duncan, R. R. (1984b). The association of plant senescence with root and stalk diseases of sorghum. In “Proceedings Consultative Group Discussion on Research & Strategies for Control of Sorghum Root and Stalk Rot Diseases, Bellagio, Italy,” pp. 99-1 10. ICRISAT, Patancheru, India. Duncan, R. R. (1991). Acid soil tolerance breeding in sorghum. Adv. Agron. (India) 1, 71-79. Duncan, R. R., Miller, F. R., and Bockholt, A. J. (1980). Inheritance of tiller regrowth in ratooned sorghum. Can. J . Plant Sci. 60, 473-478. Duncan, R. R., Bockholt, A. J., and Miller, F. R. (1981). Descriptive comparison of senescent and nonsenescent sorghum genotypes. Agron. J . 73, 849-853. Duncan, R. R., Rosenow, D. T., Sotomayor-Rios, A., and Frederiksen, R. A. (1982). Registration of GPT2RB anthracnose resistant sorghum germplasm population (Reg. No. GP130). Crop Sci. 22, 1274- 1275. Duncan, R. R., Sotomayor-Rios, A,, Hepperly, P. R., Rosenow, D. T., Miller, F. R., Narro S., J., Forbes, G . A,, and Frederiksen, R. A. (1987). Registration of GFTM3BR(H)C4 fusarium head blight/stalk rot resistant sorghum population. Crop Sci. 27, 1321-1322. Duncan, R. R., Rosenow, D. T., and Frederiksen, R. A. (1988). Registration of sorghum diseaseresistant fertility restorer germplasm line, 87BL2597. Crop Sci. 28, 1037. Duncan, R. R., Torres-Cardona, S . , Goenaga, R., Sotomayor-Rios, A., and Shuman, L. M. (1989). Registration of GPP4BR(H)C5 acid-soil-tolerant sorghum germplasm population. Crop Sci. 29, 1581-1582. Duncan, R. R., Miller, F. R., Rosenow, D. T., Sotomayor-Rios, A., and Torres-Cardona, S. (1990). Registration of GTPP7R(H)C5 white-seeded, tan plant, and foliar disease resistant sorghum germplasm population. Crop Sci. 30, 239-240. Duncan, R. R., Hepperly, P. R., Torres-Cardona, S., and Sotomayor-Rios, A. (1991a). Registration of GPP5BR(M/H/F)C3 anthracnose-resistant sorghum germplasm population. Crop Sci. 31, 24 1-242. Duncan, R. R., Waskom, R. M., Miller, D. R., Voigt, R. L., Hanning, G. E., Timm, D. A,, and Nabors, M. W. (1991 b). Registration of GACIO2 acid-soil-tolerant Hegari regenerate. Crop Sci. 31, 1396-1397. Duncan, R. R., Waskom, R. M., Miller, D. R., Hanning, G . E., Timm, D. A,, and Nabors, M. W. (1992). Registration of GC103 and GC104 acid-soil-tolerant Tx430 sorghum regenerants. Crop Sci. 32, 1076-1077. Eilrich, G. L., Long, R. C., Stickler, F. C., and Pauli, A. W. (1964). Stage of maturity, plant population, and row widths as factors affecting yield and chemical composition of Atlas forage sorghum, Kansas Agric. Exp. Stn. Tech. Bull. 138. Elder, W. C., and Denman, C. E. (1966). Sudangrass and sudangrass hybrid research. Prog. Rept. Processed Ser. P-543, Okla. St. Expt. Stn. Fahey, G. C., Jr., and Jung, H-J. G. (1989). Phenolic compounds in forages and fibrous feedstuffs. In “Toxicants of Plant Origin. IV. Phenolics” (P. R. Cheeke, Ed.), pp. 123-190. CRC Press, Boca Raton, FL. Fitzmaurice, W. P., Lehman, L. J., Nguyen, L. V., Thompson, W. F., Wemsman, E. A., and Conkling, M. A. (1992). Development and characterization of a generalized gene tagging system for higher plants using an engineered maize transposon Ac. Plant Mol. Biol. 20, 177198.
FORAGE SORGHUMS FOR THE TROPICS
181
Fritz, J. O., Cantrell, R. P., Lechtenberg, V. L., Axtell, J. D., and Hertel, J. M. (1981). Brown midrib mutants in sudangrass and grain sorghum. Crop Sci. 21, 706-709. Fritz, J. 0.. Moore, K. J., and Jaster, E. H. (1990). Digestion kinetics and cell wall composition of brown midrib sorghum X sudangrass morphological components. Crop Sci. 30, 213-2 19. Gerik, T. J . , and Miller, F. R. (1984). Photoperiod and temperature effects on tropically- and temperately-adapted sorghum. Field Crops Res. 9 , 29-40. Goldstein, W. S . , and Spencer, K. C. (1985). Inhibition of cyanogenesis by tannins. J. Chem. Ecol. 11, 847-849. Gorz, H. J., Haskins, F. A,, Kindler, S. D., and Sotomayor-Rios, A. (1984). Registration of NP22 sudangrass germplasm. Crop Sci. 24, 391. Gorz, H. J., Haskins, F. A., and Vogel, K. P. (1986a). Inheritance of dhumn content in mature sorghum leaves. Crop Sci. 26, 65-67. Gorz, H. J., Haskins, F. A., and Vogel, K. P. (1986b). Registration of NP23 and NP24 sudangrass germplasm. Crop Sci. 26, 212. Gorz, H. J., Haskins, F. A., Pedersen, J. F., and Ross, W. M. (1987). Combining ability effects for mineral elements in forage sorghum hybrids. Crop Sci. 27, 216-219. Gorz, H. J., Haskins, F. A., and Vogel, K. P. (1990a). Registration of NP28 sudangrass germplasm, a composite of 90 low-dhunin self-pollinated lines. Crop Sci. 30, 758. Gorz, H. J., Haskins. F. A., and Vogel, K. P. (1990b). Registration of NP30, a low-dhurrin sudangrass population selected for regrowth potential. Crop Sci. 30, 759. Gorz, H. J., Haskins, F. A., and Vogel, K. P. (1990~).Registration of NP33, NP34, and NP35, three broadly based random-mating populations of sudangrass. Crop Sci. 30, 760. Gorz, H. J., Haskins, F. A,, Johnson, B. E., and Sotomayor-Rios, A. (1990d). Registration of NP36 and NP37, two random-mating grain sorghum populations selected for reduced dhurrin content. Crop Sci. 30, 761. Gorz, H. J., Haskins, F. A,, and Johnson, B. E. (1990e). Registration of 15 germplasm lines of grain sorghum and sweet sorghum. Crop Sci. 30, 762. Gorz, H. J., Haskins, F. A,, and Vogel, K. P. (1990f). Registration of 10 sudangrass inbred lines. Crop Sci. 30, 763. Goto, M., Gordon, A. H., and Chesson, A. (1991). Changes in cell-wall composition and degradability of sorghum during growth and maturation. J . Sci. Food Agric. 54, 47-60. Gourley, L. M., and Lusk, J. W. (1978). Genetic parameters related to sorghum silage quality. J. Dairy Sci. 61, 1821-1827. Grisebach, H. (1981). Lignins. In “The Biochemistry of Plants” (P. K. Stumpf and E. E. Conn, Eds.), Vol. 7. “A Comprehensive Treatise,” pp. 457-478. Academic Press, New York. Halkier, B. A , , and Moller, B. L. (1991). Involvement of cytochrome P-450 in the biosynthesis of dhunin in Sorghum bicolor (L.) Moench. PIunr Physiol. 96, 10-17. Hanna, W. W., Monson, W. G., and Burton, G. W. (1973). Histological examination of fresh forage leaves after in vitro digestion. Crop Sci. 13, 98-102. Hanna, W. W., Monson, W. G., and Gaines, T. P. (1981). IVDMD, total sugars, and lignin measurements of normal and brown midrib (bmr) sorghums at various stages of development. Agron. J . 73, 1050-1052. Haskins, F. A., and Gorz, H. J. (1986a). Inheritance of leucoanthocyanidin content in sorghum leaves. Crop Sci. 26, 286-288. Haskins, F. A,, and Gorz, H. J. (1986b). Relationship between contents of leucoanthocyanidin and dhunin in sorghum leaves. Theor. Appl. Genet. 73, 2-3. Haskins, F. A,, and Gorz, H. J. (1988). Independent inheritance of genes for dhunin and leucoanthocyanidin in a sorghum cross. Crop Sci. 28, 864-865. Haskins, F. A,, Gorz, H. J., Kindler, S. D., Jensen, S. G., and Sotomayor-Rios, A. (1986). Registration of NP25 low-dhurrin sudangrass germplasm. Crop Sci. 26, 213.
182
R. R. DUNCAN
Haskins, F. A,, Gorz, H. J . , and Hill, R. M. (1988). Colorimetric determination of cyanide in enzyme-hydrolyzed extracts of dried sorghum leaves J. Agric. Food Chem. 36, 775-778. Haskins, F. A,, Gorz, H. J., and Vogel, K. P. (1990a). Registration of NP29, a low-dhurrin sudangrass population selected for tolerance to early spring seeding. Crop Sci. 30, 758. Haskins, F. A , , Gorz, H. J., and Vogel, K. P. (1990b). Registration of NP31 and NP32, two populations of sudangrass selected for low dhurrin content. Crop Sci. 30, 759-760. Hatfield, R. D. (1989). Structural polysaccharides in forages and their degradability. Agron. J. 81, 39-46. House, L. R. (1985). “A Guide to Sorghum Breeding,” 2nd ed. ICRISAT, India. Hoveland, C. S., and Monson, W. G. (1980). Genetic and environmental effects on forage quality. In “Crop Quality, Storage, and Utilization,” pp. 139-168. Amer. Soc. Agron. and Crop Sci. SOC. Amer., Madison, WI. Jones, D. A. (1979). Chemical defense. Primary or secondary functions? Am. Nururulisr 113, 445451. Julian, E. A,, Johnson, G., Johnson, G. K., and Donnelly, B. J. (1971). The glycoflavonoid pigments of wheat, Triricurn uestivum. leaves. Phytochemisrry 10, 3185-3193. Jung, H. G. (1989). Forage lignins and their effects on fiber digestibility. Agron. J. 81, 33-38. Jung, H. G., and Fahey, G. C., Jr. (1983). Nutritional implications of phenolic monomers and lignin: A review. J. Anim. Sci. 57, 206-219. Kalton, R. R. (1988). Overview of the forage sorghums. In “Proceedings, 43rd Ann. Corn & Sorghum Indus. Res. Conf.” (Wilkinson, D., Ed.), pp. 1-12. Amer. Seed Trade Assoc., Washington, DC. Karper, R. E. (1955). Registration of sorghum varieties. VIII. Greenleaf and Tracy. Agron. J. 47, 540. Kramer, N. W. (1960). Registration of sorghum varieties. IX. Wiley, Lahoma, and Brawley. Agron. J. 52, 666. Kuc, J., and Nelson, 0. E. (1964). The abnormal lignins produced by the brown midrib mutants of maize. I. The brown midrib-I mutants. Arch. Biochem. Biophys. 105, 103-113. Lamb, J. F. S . , Haskins, F. A., Gorz, H. J., and Vogel, K. P. (1987). Inheritance of seedling hydrocyanic acid potential and seed weight in sorghum-sudangrass crosses. Crop Sci. 27, 522525. Lippke, H. (1986). Regulation of voluntary intake of ryegrass and sorghum forages in cattle by indigestible neutral detergent fiber. J . Anim. Sci. 63, 1459- 1468. Maiti, R. K., Rao, K. E. P., Raju, P. S., and House, L. R. (1984). The glossy trait in sorghum: Its characteristics and significance in crop improvement. Field Crops Res. 9, 279-289. Mann, I. A., Gbur, E. E., and Miller, F. R. (1985). A screening index for adaptation in sorghum cultivars. Crop Sci. 25, 593-598. Mathur, P. N., Rao, K. E. P., Thomas, T. A,, Mengesha, M. H., Sapra, R. L., and Rana, R. S. (1991). “Evaluation of Forage Sorghum Germplasm,” Part 1. NBPGR-ICRISAT, New Delhi, India. Mathur, P. N., Rao, K. E. P., Singh, I. P., Agrawal, R. C., Mengesha, M. H., and Rana, R. S. (1992). “Evaluation of Forage Sorghum Germplasm,” Part 2. NBPGR-ICRISAT, New Delhi, India. McBee, G. G., and Miller, F. R. (1990). Carbohydrate and lignin partitioning in sorghum stems and blades. Agron. J. 82, 687-690. McNeil, M., Albersheim, P., Taiz, L., and Jones, R. L. (1975). The structure of plant cell walls. VII. Barley aleurone cells. Plant Physiol. 55, 64-68. Milford, R., and Minson, D. J. (1963). Intake of tropical pasture species. In “Proceedings, IX International Grassland Congress, Sao Paola, Brazil,” pp. 8 15-822. Miller, F. R. (1986). Registration of seven sorghum A- and B-line inbreds. Crop Sci. 26, 216.
FORAGE SORGHUMS FOR THE TROPICS
I83
Mole, S. (1989). Polyphenolics and the nutritional ecology of herbivores. In “Toxicants of Plant Origin. IV. Phenolics” (P. R. Cheeke, Ed.), pp. 191-223. CRC Press, Boca Raton, FL. Moller, B. L., and Conn, E. E. (1979). The biosynthesis of cyanogenic glucosides in higher plants. N-Hydroxytyrosine as an intermediate in the biosynthesis of dhunin in Sorghum bicolor (Linn) Moench. J. B i d . Chem. 254, 8575-8583. Montgomery, C. R., Nelson, B. D., Joost, R., and Mason, L. F. (1988). Tannin concentration and quality changes in sorghum as affected by maturity and sorghum type. Crop Sci. 26, 372-375. Morrison, I. M. (1974). Structural investigations on the lignin carbohydrate complexes of Lolium perenne. Biochem. J. 139, 197-204. Mueller-Harvey, I . , and Reed, J. D. (1992). Identification of phenolic compounds and their relationships to in-vim digestibility of sorghum leaves from bird-resistant and non-bird-resistant varieties. J. Sci. Food Agric. 60, 179-196. Nicholson, R. L., Jamil, F. F., Snyder, B. A,, Lue, W. L., and Hipskind, J. (1988). Phytoalexin synthesis in the juvenile sorghum leaf. Physiol. Mol. Plant Purhol. 33, 271-278. Pedersen, J. F., Gorz, H. J., Haskins, F. A,, and Ross, W. M. (1982). Variability for quality and agronomic traits in forage sorghum hybrids. Crop Sci. 22, 853-856. Pedersen, J. F., Haskins, F. A,, Con, H. J., and Britton, R. (1983). Variability for traits used to estimate silage quality in forage sorghum hybrids. Crop Sci. 23, 376-379. Peterson, G. C . , Suksayretrup, K., and Weibel, D. E. (1982). Inheritance of some bloomless and sparse-bloom mutants in sorghum. Crop Sci. 22, 63-67. Porter, K. S., Axtell, J. D., Lechtenberg, V. L., and Colenbrander, V. F. (1975). The brown midrib gene (bmr) in sorghum and its effect on forage quality. In “Proceedings, International Sorghum Workshop, Mayaguez, P.R.,” pp. 471-482. Porter, K. S., Axtell, J. R., Lechtenberg, V. L., and Colenbrander, V. F. (1978). Phenotype, fiber composition, and in vitro dry matter disappearance of chemically induced brown midrib (bmr) mutants of sorghum. Crop Sci. 18, 205-208. Quinby, J. R. (1967). The maturity genes of sorghum. Adv. Agron. 19, 267-305. Quinby, 1. R., and Karper, R. E. (1954). Inheritance of height in sorghum. Agron. J. 46, 21 1-216. Reed, 1. D., Tedla, A , , and Kebede, Y. (1987). Phenolics, fiber and fiber digestibility in the crop residue from bird resistant and non-bird resistant sorghum varieties. J. Sci. Food Agric. 39, 113-121. Reed, J. D., Kebede, Y., and Fussell, L. K. (1988). Factors affecting the nutritive value of sorghum and millet crop residues. In “Plant Breeding and The Nutritive Value of Crop Residues, Proc. Workshop” (J. D. Reed, B. S. Capper, and P. J. H. Neate, Eds.), pp. 233-251. ILCA, Addis Ababa, Ethiopia. Ring, A. S., Waniska, R. D., and Rooney, L. W. (1988). Phenolic compounds in different sorghum tissues during maturation. Biomass 17, 39-49. Ross, M. D., and Jones, W. T. (1983). A genetic polymorphism for tannin production in Lo/us corniculatu.s and its relationship to cyanide polymorphism. Theor. Appl. Genet. 64, 263268. Ross, W. M., Gorz, H. J., and Haskins, F. A. (1980). A preliminary investigation of forage quality characters in hybrid grain sorghum residues. Maydicu XXV, 41-53. Saunders, J. A,, and Conn, E. E. (1978). Presence of the cyanogenic glucoside-dhurrin in isolated vacuoles of sorghum. Plant Physiol. 61, 154-157. Saunders, J. A,, Conn, E. E . , Linn, C. H., and Stoking, C. R. (1977). Subcellular localization of the cyanogenic glucoside of sorghum by autoradiography. Plum Physiol. 59, 647-652. Schertz, K. F., and Rosenow, D. T. (1977). Anatomical variation in the stalk internodes of sorghum. Crop Sci. 17, 628-63 I . Seigler, D. (1976). Plants of the northeastem U.S. that produce cyanogenic compounds. Econ. Bot. 30, 395-407.
184
R. R. DUNCAN
Smith, D. C., Ahlgren, H. L., Sund, J. M., Hogg, P. G., and Goodloe, H. F. (1973). Registration of Piper-sudangrass. Crop Sci. 13, 584. Sotomayor-Rios, A,, and Santiago, A. (1981). Performance of F, hybrids from crosses of three sudangrasses and six forage sorghums with A Rhodesian sudangrass. J. Agric. Univ. P.R. 65, 142- 146. Sotomayor-Rios, A,, and Telek, L. (1977). Forage yield and protein content of millo blanco (Sorghum bicolor) and two F, hybrids. J . Agric. Univ. P.R. 61, 300-304. Sotomayor-Rios, A , , and Torres-Cardona, S. ( 1984a). Agronomic performance, hydrocyanic acid potential (HCN-p) and heterosis in forage sorghum hybrids. J . Agric. Univ. P.R. 68, 131-141. Sotomayor-Rios, A,, and Torres-Cardona, S. (1 984b). Agronomic comparison, heterosis, and hydrocyanic acid potential (HCN-p) of sudangrass-sorghum and sudangrass-sudangrass hybrids and their parents. J. Agric. Univ. P.R. 68, 143-155. Sotomayor-Rios, A,, and Torres-Cardona, S. (1990). Breeding and agronomic studies with sorghum in Puerto Rico. Proc. Caribbean Food Crops SOC. 20, 289-292. Sotomayor-Rios, A,, Hepperly, P. R., and Torres-Cardona, S. (1984). Registration of PRSBR sorghum germplasm population. Crop Sci. 24, 627. Sotomayor-Rios, A,, Torres-Cardona, S., and Quiles-Belen, A. (1985). Forage sorghum response to N fertilization and harvest intervals. J . Agric. Univ. P.R. 69, 341-355. Sotomayor-Rios, A,, Torres-Cardona, S., and Hepperly, P. R. (1986). Sorghum germplasm research in Puerto Rico and introduction in St. Croix. In “Proc. 41st Ann. Corn & Sorghum Ind. Res.Conf.” (D. Wilkinson, Ed.), pp. 40-47. Amer. Seed Trade Assoc., Washington, DC. Spoelstra, S. F. (1985). Nitrate in silage. Grass Forag. Sci. 40, 1-1 1. Stallcup, 0. T., Davis, G. W., and Ward, D. A. (1964). Factors influencing the nutritive value of forages utilized by cattle. Ark. Agric. Exp. Sfn. Bull. 684. Swain, T. (1965). The tannins. In “Plant Biochemistry” (J. Bonner and J. E. Varner, Eds.), pp. 552580. Academic Press, New York. Tarumoto, I. (1975). Breeding method of hybrid forage sorghum by using male-sterile lines. Jpn. Agric. Res. Quart. 8, 242-250. Tarumoto, I. (1978). Forage sorghum breeding by using male-sterile lines. Trop. Agric. Res. Ser. 11, 23-32. Tarumoto, I. (1980). Inheritance of glossiness of leaf blades in sorghum. Sorghum bicolor (L.) Moench. Jpn. J. Breed. 30, 237-240. Thayer, S. S., and Conn, E. E. (1981). Subcellular localization of dhurrin P-glucosidase and hydroxynitrile lyase in the mesophyll cells of sorghum leaf blades. Plant Physiol. 67,617-622. Torres-Cardona, S . , Sotomayor-Rios, A,, and Telek. L. (1983). Agronomic performance and hydrocyanic acid potential (HCN-p) of single and three-way sorghum forage hybrids and DeKalb hybrid SX-17. J . Agric. Univ. P.R. 67, 39-49. Torres-Cardona, S . , Sotomayor-Rios, A., and Miller, F. (1986). Agronomic comparison and in virro dry matter digestibility of eight sorghums at two locations in Puerto Rico. J. Agric. Univ. P.R. 70, 37-44. Traore, M., Sullivan, C. Y., Rosowski, J. R., and Lee, K. W. (1989). Comparative leaf surface morphology and the glossy characteristic of sorghum, maize, and pearl millet. Ann. Bor. 64, 447-453. Vietor, D. M., and Miller, F. R. (1990). Assimilation, partitioning, and nonstructural carbohydrates in sweet compared with grain sorghum. Crop Sci. 30, 1109- I 1 15. Walton, M. F., Haskins, F. A., and Gorz, H. J. (1983). False positive results in the vanillin-HCI assay of tannins in sorghum forage. Crop Sci. 23, 197-200. Wanous, M. K . , Miller, F. R., and Rosenow, D. T. (1991). Evaluation of visual rating scales for green leaf retention in sorghum. Crop Sci. 31, 1691-1694.
FORAGE SORGHUMS FOR T H E TROPICS
185
Watterson, J. J., and Butler, L. G. (1983). Occurrence of an unusual leucoanthocyanidin and absence of proanthocyanidin in sorghum leaves. J . Agric. Food Chem. 31, 41-45. Webster, 0. J. (1976). Registration of PBlBR sorghum germplasm. Crop Sci. 16, 447. Weibel, D. E. (1986a). Registration of 12 bloomless and four sparse-bloom lines of sorghum germplasm. Crop Sci. 26, 840. Weibel, D. E. (1986b). Registration of eight bloomless parental lines of sorghum. Crop Sci. 26, 842. Weibel, D. E., Sieglinger, J. B., and Davies, F. F. (1984). Registration of fourteen sorghum parental lines. Crop Sci. 24, 628. Wheeler, J. L., and Corbett, J. L. (1989). Criteria for breeding forages of improved feeding value: Results of a 146.DeIphi survey. Grass Forag. Sci. 44, 77-83.
This Page Intentionally Left Blank
NITROGEN MINERALIZATION IN TEMPERATE AGRICULTURAL SOILS: PROCESSES AND MEASUREMENT Stephen C. Jamis,' Elizabeth A. Stockdale,2 Mark A. Shepherd,3 and David S. Powlson* 'Institute of Grassland and Environmental Research, North Wyke, Okehampton, Devon, EX20 ZSB, United Kingdom 2Institute of Arable Crops Research, Rothamsted, Harpenden, Hertfordshire, ALS ZJQ, United Kingdom 3ADAS, Gleadthorpe Research Centre, Meden Vale, Mansfield, Nottinghamshire, NGZO 9PF, United Kingdom
I. Introduction 11. Pools and Processes A. Soil Organic Matter B. Mineralization/Immobilization C. Nitrification D. Soil Biomass E. Effects of Organic Matter Additions 111. Process Controls A. Resource Quality B. Environmental Controls C. Effects of Cultivation D. Microsites, Diffusional Constraints and Soil Architecture IV.Measurement and Prediction of Mineralization A. Background B. Mineralization from N Balance/Cropping Data C. Laboratory Determination of Potential Mineralization Indices D. Field Measurements E. Measurement of Gross Mineralization F. The Role of Models in Predicting Mineralization V. The Impact of Mineralization VI. Conclusions and Future Progress References
187 Advances m A p n m y , Volume 17
Copyright 0 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
188
S. C. JARVIS ETAL.
I. INTRODUCTION The soil nitrogen (N) cycle is an integral part of an overall, global recycling system (Fig. 1). Soils form a major repository of N within both natural and agricultural terrestrial ecosystems, containing, on a global basis, an estimated 2.4 x 10” tons of N (Stevenson, 1982a). The soil receives N inputs through fertilizer additions and from the atmosphere in precipitation and dry deposition or via biological fixation: inputs are also made in plant and animal residues. Nitrogen is removed in the harvested crop and is lost by leaching and surface run-off of soluble forms, by gaseous transfer as nitrogen gas and nitrogen oxides (during nitrification and denitrification processes), and by ammonia volatilization. In some circumstances, erosion may also be important. In addition to these interactions with the total ecosystem, internal cycles also operate within the soil, so that even if gains and losses are in balance, then N still continues to cycle in the soil. Nitrogen is continuously assimilated into organic forms in the soil, i.e., “immobilized,” and released as ammonium ions (NH;) from organic matter, i.e., “mineralized.” The processes of mineralization and immobilization are therefore central to the control of the flows of N within agricultural cycles and mineralization has been recognized as an important soil process since the early years of this century (Lohnis, 1910) because N in mineral forms is essential for plant growth and development.
Figure 1 The “global” nitrogen cycle.
NITROGEN MINERALIZATION
189
Increasingly, questions are being raised about the efficiency of N use within managed systems. Recent research has demonstrated clearly that there are substantial losses of N either to waters or to the atmosphere with potential for environmental impact. In order to improve efficiency and reduce emissions while sustaining production targets, a thorough knowledge of the N flows from all sources, especially those from organic materials, is required. Although studies of mineralization/immobilization have been undertaken over many decades, our general understanding remains fragmentary. New approaches, concepts, and information have become available which will enable a more detailed understanding of the impact of these processes on, for example, (i) improvement of fertilizer recommendations, (ii) leaching of NO, (and NO,) into ground or surface waters, (iii) recycling of N from crop and animal residues, (iv) production of N,O (an important greenhouse gas) and other gaseous oxides of N, (v) changes in soil quality through changes in the nature of soil organic matter, and (vi) effective and accurate inputs to computer models to simulate N cycling. This review therefore describes the current understanding of the conceptual basis of the processes involved in mineralization, relationships between the processes and other factors, and also how their effects can be determined practically. The aim is to present this in a way that is relevant to current and future agricultural development and to environmental issues. While we use, in the main, information that has been collected or derived from temperate agricultural systems, the general principles that we describe are applicable across other climatic zones and ecosystem types.
11. POOLS AND PROCESSES
A. SOILORGANIC MATTER In undisturbed systems, soil organic matter (SOM) can be considered to attain a steady state level governed by the soil forming factors and their interaction, i.e., climate, topography, parent material, vegetation, soil flora and fauna, and time (Jenny, 1941). Where soils are disturbed and used for agricultural production, the various equilibria involved are not maintained and SOM content tends to decline. The rate of the decline and the establishment of a new steady state during cultivation will depend on soil type (and its textural, structural, and drainage status), crop rotations, and management of both residues and soil, including the intensity and mode of tillage. All of these influence the balance between inputs of residues and their breakdown by mineralization. Cropping rotations which return few residues to the soil, and do not include additions of other organic materials such as manures, show the greatest rates of decline in
190
S. C. JARVIS ETAL.
organic N (Janzen, 1987), whereas the continued long-term use of N fertilizer has been shown to slow the decline of soil organic N, largely by increasing the returns of N in residues (Glendining and Powlson, 1995). However, the net retention of N from fertilizer is small; in one long-term experiment the increase in soil organic N was equivalent to only 3% of the fertilizer N applied over a period of more than a century. In many circumstances organic matter accumulates or declines only slowly. Disturbing the soils by cultivation enhances mineralization of SOM and levels decline more rapidly. Uncultivated soils with natural vegetation tend to have most organic matter, and long-term grassland generally has more organic N than long-term arable soils. Agricultural soils therefore hold substantial but variable amounts of SOM and organic N, as shown with data taken from 106 arable fields in England (Table I). The organic N pool present in soils is always much larger than the mineral N pool; for example, mineral N measured in autumn, averaged across all of the sites in Table I for 1989-92, was only 76 kg N ha-1 compared with 7 t ha-1 of organic N . SOM contains a large reservoir of nutrients and is commonly divided into a number of pools, into which materials behaving similarly are grouped. These are linked by a number of interacting, competing, and sometimes antagonistic processes. In fact, soil organic matter is composed of a continuum of materials stabilized against mineralization to varying degrees by molecular recalcitrance, physical separation from the soil microbial biomass (SMB), and/or direct asso-
Table I The Average Topsoil Organic N Contents of 106 Arable Fields in England Sampled to 25 cm as Part of a Long-Term Soil Mineral Nitrogen Monitoring Scheme (ADAS, Unpublished Data) Organic N (t ha-' to 25 cm depth) Topsoil texture
No. of samples
Mean
Range
Clay Clay loam Sandy clay loam Silty clay loam Silt loam Sandy loam Loamy sand
7 34 6 17 5 31 6
11.7 7.5 6.4 5.8 8.6 4.7 4.3
8.0-13.6 5.0-12.0 4.3-9.0 4.0-8.6 6.3-12.6 2.7-7.6 3.7-4.9
~
NITROGEN MINERALIZATION
191
ciation with inorganic ions and clay surfaces. Chemical fractionation techniques have been used to define the chemical structures of SOM and have shown that a wide range of functional groups is present (Stevenson, 1982b). The relationships between these fractions and the soil N cycle have not been clearly identified or defined, and no chemical technique has yet separated SOM into biologically meaningful fractions (Paul, 1984). Four pools of SOM which made differing contributions to mineralization were identified by Paul and Juma (1981) using I5N. The active pools of biomass, active nonbiomass, and metabolites were small but had rapid turnover, whereas old organic N had a much larger pool but slow turnover. A very stable pool of SOM has also been identified with a turnover rate of more than 1000 years (Hsieh, 1992), which does not take part in seasonal nutrient cycling. Physical fractionation techniques have also been used to separate SOM pools into sand, silt, and clay-sized fractions (Skjemstad et al., 1988). Organic matter separated into sand-sized or light fractions declined most rapidly when soil was brought into cultivation (Dalal and Mayer, 1986), while the rate of loss of C from the clay-sized fraction was slower than that from the soil as a whole (Hassink, 1992). Physical fractionation techniques may provide a practical way to divide the SOM into fractions which can be related to the soil N cycle but this has not been explored to any great extent as yet. Many models of SOM turnover have been developed (Jenkinson et al., 1987; Parton et al., 1987; Jenkinson, 1990). Bosatta and Agren (1985) introduced the idea that decomposition is a continuum, with organic matter decreasing in quality as a nutrient resource as it decays. The mathematics of this approach are complex and it is therefore not surprising that most modellers have retained the concept of a variable number of discreet pools. Contributions to those pools will be made from organic materials added to soils in crop residues after harvest or by senescence during the growing season, by incorporation of green manures or cover crops before the next planting, in waste materials (which can range from farm manures, municipal wastes, and sewage sludges to by-products from industrial processes), roots and their exudates, and from the turnover of SMB. The simplest description of these materials is as fresh organic matter (e.g., crop residues, recent manure additions), which decomposes rapidly and is eventually stabilized to join the old/native organic matter pool, which releases N slowly (Jenkinson, 1984). The simple separation of SOM into active and passive pools has been used to model the contribution of N from recently added residues and stabilized pools (Matus and Rodriguez, 1994). More complex models separate more pools and usually include the SMB explicitly as one or more of the defined pools (van Veen and Frissel, 1981; Parton et al., 1987; Hansen et al., 1991; Rijetema and Kroes, 1991; Bradbury et al., 1993). However, the pools defined in models are often conceptual and few can yet be measured by physical, chemical, or biological methods.
192
S. C. JARVIS ET AL.
B. MINERALI~ATION/~[MMOBILIZATION Decomposition and mineralization are the means by which nutrients held in organic materials in the soil are released into the soil as inorganic (often referred to as mineral) forms. Inorganic N released in this way is then available either for subsequent recycling and utilization by plants or micro-organisms or to be lost from the system. Mineralization is the transformation process whereby ammonium (NHT) or ammonia (NH,) is released by soil micro-organisms as they utilize organic N compounds as an energy source (Jansson and Persson, 1982; Royal Society, 1983). The process is complex and depends upon the activities of nonspecific heterotrophic soil micro-organisms under both aerobic and anaerobic conditions. Mineralization occurs, to differing extents, with both newly added residues and existing, already degraded organic materials of varying ages and degrees of recalcitrance (Fig 2). Mineralization is always coupled with immobilization (Fig. 2); the two processes are intimately connected and dependent. Much of any NH;, NO,, or
1
Atmosphere volatilization
denifnfication N~ fixation
. I 1 d
I Plant
Soil fauna
i Soil organic matter
+ t.
immobrbafion
I leaching
Figure 2 The soil nitrogen cycle.
NITROGEN MINERALIZATION
193
simple organic N compounds that are released is assimilated rapidly by the SMB population and transformed into the organic N constituents of their cells during the oxidation of suitable C substrates, i.e., it is “immobilized.” However, immobilized N is likely to be available subsequently for mineralization as the microbial population turns over. Degradation of microbial tissues is of great importance in terms of the final release of N originally bound in organic residues, and biomass N contributes, over the short term, substantial amounts of N to the pools of mobile N. Concurrent with release from the SMB will also be direct release from both fresh residues and “native” soil organic materials of various ages. Gross mineralization is the total release of NH,+ through microbial activities, i.e., before any immobilization back into the soil biomass. The difference between gross rates of mineralization and immobilization is net mineralization or, in some circumstances, net immobilization. Net rates are the integration of a number of soil N processes and a number of effects, which act on other interacting processes as well as directly on mineralization/immobilization: extrapolation of information to other circumstances and sites is therefore difficult. Immobilization has been shown to occur predominantly from the NH,+ pool (Jansson, 1958; Recous et al., 1988). However, where NH,+ is not available, NO, is assimilated by SMB in the presence of readily available C (Azam et al., 1986; Recous et al., 1988). There has been no evidence of any difference between the subsequent rates of release of immobilized NH,+ or NO, (Bjamason, 1987). The process of continuous transfer of mineralized N into organic materials in SMB and the release of immobilized N back into inorganic pools is known as “Mineralization-Immobilization-Turnover” or MIT (Jansson and Persson, 1982). A basic assumption of MIT is that all immobilization occurs from the inorganic pool (NH,+ or NO,). It has also been proposed that at a microsite scale there may be direct immobilization of small organic compounds such as amino acids (Hadas et al., 1987; Drury et al., 1991), known as the “Direct Hypothesis.” Although it has been demonstrated that SMB can utilize amino acids in this way (Barak ef al., 1990), MIT generally describes overall mineralization more accurately (Barak ef al., 1990; Hadas et al., 1992). However, since both mechanisms are not mutually exclusive and both have been demonstrated, both may occur concurrently, at least in some circumstances. The way that the mineralization/immobilizationprocesses operate is of importance to the turnover, recycling and fate of released N and many questions remain. For example, we do not know how much competition occurs between plant roots and loss processes and immobilization for the products of gross mineralization or whether SMB absorbs all the N that it requires so that the remainder is available for plant uptake or loss. Similarly, the microsite distribution of net mineralized N and how this relates to interpretation of measurements of soil mineral N measurements is not known. In a practical sense, it is the balance between mineralization and immobilization (i.e., net mineralization)
194
S. C. JARVIS E T A .
which seems to be important in influencing N available for crop uptake or loss to waters or the atmosphere, and it is this which is most often measured or estimated. However, gross mineralization is the more fundamental process which reflects the properties of the substrate and its interaction with the ambient environment. An understanding of gross mineralization will provide the sounder basis on which to base mechanistic models or allow confident extrapolation. Gross mineralization is more difficult to measure but new techniques and approaches have become available over recent years (see Section 1V.E.). Immobilization is involved with all mineralization activities as part of an intricate cycle of organic matter degradation and renewal, and the assimilation of mineral nutrients to provide the basis for the multiplication and maintenance of the SMB. In theory, the balance between mineralization and immobilization can fluctuate from positive to negative according to ambient soil and environmental conditions and the quantity and quality of available organic substrates. In practice, in most situations, there is usually net mineralization during an annual cycle, although over the shorter term immobilization may dominate, for example when organic materials such as cereal straw are returned to the soil (Ocio et al., 1991). Long-term accumulation of organic N in soils may also be referred to as “immobilization.” This is not a direct result of SMB activities, but represents the N returned to the soil in plant shoot and root residues. It is quite usual and reasonable for a soil to have net positive mineralization and to release mineral N, but to accumulate organic N during the year, provided that the N input was greater than any offtake in harvested crops plus that accumulated in organic matter. This is particularly the case in undisturbed grassland soils where there is usually long-term accumulation of organic N through this means.
C. NITRIFICATION As well as the linkage between mineralization and immobilization, it is also important to consider interactions with nitrification. In most circumstances nitrification is the oxidation of reduced N compounds, primarily NHZ, by two groups (in the main) of autotrophic bacteria (Nitrosomonasand Nitrobacter) according to the following pathways.
NHZ
+ l+ 0,
6 e-
NO;
Nitrosomonas
NO;
+ 4 0,
2 e-
NO,
Nitrobacter
+ 2H+ + H,
NITROGEN MINERALIZATION
195
The transformation from a relatively immobile (NHJ) to a highly mobile state (NO,) via nitrite (NO,), provides a key step in the soil N cycle often leading to an excess of NO, and subsequent N loss by leaching or denitrification. In many circumstances the major source of NH,+ is through release by mineralization; exceptions are addition of NH,+or urea-based fertilizers, or where animal excreta are being returned. Despite its importance as a rate-limiting process controlling the availability and loss of NO,, and despite a reasonable knowledge of the ecology of the bacteria involved (Prosser, 1986), nitrification is poorly defined in many soils. In arable and other tillage systems, it is usually assumed that nitrification is not a limiting process. Conditions are such that rapid oxidation takes place and it is unusual to find accumulations of NHJ, i.e., nitrification rate exceeds that of mineralization. However, the situation differs for grassland soils especially where swards are grazed or farm wastes are applied and there are large returns of ammoniacal N. Under these circumstances there may be significant quantities of NHJ in the soil profile at various stages through the year (Jarvis and Barraclough, 1991). Nitrification is dependent on soil aerobicity (and thus soil texture, structure, and water contents), on pH (i.e., inhibited by high pH caused by liquid or anhydrous ammonia fertilizer, for example), and on substrate (NHZ) availability and appropriate populations of micro-organisms. Nitrification interacts strongly with ambient local soil conditions and there is a high degree of spatial compartmentalization of NHJ production and consumption sites. This, coupled with diffusional constraints between microsites, controls the rate at which nitrification proceeds (Bramley and White, 1991). In grassland soils there are known to be latent nitrification potentials which have developed because of previous conditions and managements and which can be displayed over both the relatively short (Jarvis and Barraclough, 1991) and the longer term (Willison and Anderson, 1991). Nitrification rates are also strongly influenced by additions of mineral N as fertilizers. At low inputs of NHJ to grassland soils it has been suggested that competition between plant uptake and nitrification reduces nitrification rates (Barraclough and Smith, 1987). Where there was considerable excess of input over removal by plants, nitrification rates were substantially higher. It is worth emphasizing that nitrification is central to the flows, losses, or utilization of N through the conversion of NH,+ into labile NO,. Its greater understanding is an important further step in maximizing N efficiency and reducing losses, particularly in grassland, but also as a key interactive process linked by flows of substrates and spatial distributions in all ecosystems.
D. SOILBIOMASS Biomass is at the center of the internal soil N cycle (Jenkinson, 1990) and is important as (i) an agent of change, decomposition, and release of N (and other
196
S. C. JARVIS ET AL.
nutrients) from fresh organic residues and native SOM into more labile forms, (ii) a major sink for “active” soil N, and (iii) a potential source of labile N. Biomass N is in a constant state of turnover (Jenkinson and Ladd, 1981) and represents a significant proportion of the total soil N which remains relatively constant throughout the year (Holmes, 1994). For example, SMB N alone accounted for 0.5-15.3% (3-108 kg N ha-’) of the total N in the surface 12.5 cm of the soil (Anderson and Domsch, 1990), and Jenkinson and Ladd (1981) showed that in an unmanured wheat field (to 23 cm depth) SMB contained 95 kg N ha-’. The annual N flux through the SMB can be greater than the offtake in harvested grass (Brookes et af., 1984). Bristow and Jarvis (1991) have shown that, in cut and rotationally grazed pastures with a range of N inputs, the average N content of the SMB (0-10 cm depth) ranged between 138 and 240 kg N ha-’. The turnover time for N derived from microbial biomass has been estimated to be from 5 (Marumoto et al., 1982a) to 10 (Smith and Paul, 1990) times faster than that in SOM or returned in plant residues. This rate of transformation is assumed to be very important in those production systems with low or no annual inputs from fertilizer or other immediately available sources. However, the exact nature of the participating populations, the community structure, and the prediction of their ability to react to environmental and other changes requires characterization and definition. Understanding these issues will be a key to future successful management of low-input systems. The soil biomass comprises almost every class and order of invertebrates, as well as a wide range of fungal and bacterial species and genotypes. The community structure and size have been shown to be related to soil type and management (Chaussod et d., 1988). Thus, mesofauna activity decreased with intensity of management (Anderson, 1988) and cultivation has resulted in high earthworm mortality (Curry and Byrne, 1992). Heterotrophic microflora are the primary decomposers and contribute more than 90% of the energy flux in soil (Ladd and Foster, 1988). They comprise diverse assemblies of organisms which are often divided into two groups, i.e., autochthonous species which are able to maintain low and relatively constant activities in utilizing the more resistant SOM, and zymogenous species which react rapidly to inputs of readily metabolized substrates but then return to dormancy (Ladd and Foster, 1988). The existence of these two types has been used to explain changing C and N dynamics during decomposition of straw (Cochran et al., 1988) and the occurrence of a greater proportion of active biomass in the upper layers of grassland soils (Hassink, 1993). Although management has been shown to have an effect on SMB activities and possibly on community structures (Lovell et af., 1995), substantial differences in management of grassland soils have had little or no effect on overall biomass size (Bristow and Jarvis, 1991; Lovell et al., 1995). Resolution of SMB into labile and resistant components has allowed the development of a recent model to simulate changes in active vs quiescent biomass fractions (Grant et af., 1993). The funga1:bacteria biomass ratio has been shown to be an impor-
NITROGEN MINERALIZATION
197
tant factor regulating the relationship between activity and SMB size (Sakamoto and Oba, 1994). Techniques involving antibiotics have been developed to separate effects of these two groups of organisms (Landi et al., 1993). Mineralization should not be considered totally as a microbial activity (Woods et al., 1982). Invertebrate fauna make an important contribution by (i) redistributing organic materials over a range of spatial scales, (ii) enhancing the rate of cycling through chemical change during metabolism, and (iii) having an effect on microbial populations themselves by creating or removing appropriate conditions for their various activities. Soil invertebrates thus contribute to N fluxes by changing microsite environments and controlling populations of other organisms, and through trophic transfers in food webs and turnover of tissues (Anderson, 1988). Although there is no clear consensus about the net effects of feeding and other activities of micro-, meso-, or macrofauna, it is, however, clear that they have substantial impact. Thus earthworms have increased CO, evolution, decreased SMB, and increased mineralization (Ruz Jerez et al., 1988), mineral N levels were higher in casts than in surrounding soil (Scheu, 1987) and denitrification has been increased (Elliott et al., 1990). The main impact of earthworms, however, was to bring plant residues and soil materials into close contact (Curry and Byrne, 1992). The role of microbivorous fauna (e.g., protozoa, nematodes) has been extensively studied (Bouwman et d., 1994) and there have been suggestions that predation stimulated N turnover since more N was mineralized when protozoa were present (Woods et al., 1982; Kuikman and van Veen, 1989). Recent studies (Yeates et al., 1993) have shown the importance of microfaunal grazing on microfloral populations and nutrient cycling. However, Hassink et al. (1993) did not find any effect of grazing on microbial activity. It is also thought that bacterial predators can enhance efficient recycling of N in root exudates (Griffiths and Robinson, 1992; Griffiths, 1994). It is clear that interactions within the soil biomass and between this biological component of the soil and the net effects of mineralization and other N cycling processes are complex and poorly understood but of much importance since key functions in many biologically based soil processes are involved (Powlson, 1994; Patra, 1994). The influences on net release of mineral N are profound but it is not yet possible to manipulate N flows through these complex food webs to the advantage of management and environmental impact. Interactions of this nature may be more significant in relatively undisturbed systems such as grassland or in low-input and organic managements, but an understanding is essential to aid the efficiency of N utilization in all systems.
E. EFFECTS OF ORGANIC MATTERADDITIONS Recent additions of organic materials have the potential to mineralize at the greatest rates. Shen et al. (1989) showed that fresh residues were about seven
198
S. C. JARVIS ET AL.
times more decomposable than SOM in arable soils. Fresh materials may be returned to the soil either gradually in small amounts throughout the growing season, e.g., by leaf-drop or root exudation, or in larger pulses, e.g., after incorporation of a cover crop or harvest residue, or application of slurry or manure. Roots and root exudates offer a source of organic matter for mineralization during the growing season for both arable and grass crops. A larger pulse is made available when the soil is cultivated. Below ground, plant biomass holds substantial amounts of N (Saffigna, 1988), the rate of breakdown of which will depend on composition. It has been estimated that between 10 and 40% of the increase in root dry matter is exuded from the zone immediately behind the root tip (P. Darrah,personal communication), which provides substrates for hotspots of microbiological activity in the rhizosphere. The effects of this activity may be beneficial, harmful, or neutral for crop growth depending on whether net mineralization or net immobilization occurs. Robinson et al. (1989) showed that the potential increase in N availability to the roots was greater where exudates had a high C:N ratio. However, Griffiths and Robinson (1992) showed that the main effect of root exudation was to allow recycling of N in the exudates in forms available to plants, rather than causing any extra release from SOM. The amounts of N returned to the soil following crop harvest can vary widely depending both on crop quality and yield and on the residue management strategy (Table 11). Large amounts of N, in excess of 200 kg N ha-1, can result after the harvest of vegetable crops (Rahn et al., 1992). The use of cover crops or green manures may return from 10 to 150 kg N ha-1 (Christian et al., 1992) when residues are incorporated as young and fresh materials shortly before drilling the next crop. Some cover crops, e.g., mustard, are affected by frost so that residues resulting from defoliation occur earlier (Shepherd, 1992). Other crops are grazed and large amounts of N are returned in feces and urine. The effects of all these returns are considerable and have been reviewed recently (Shepherd er al., 1996). Even in grassland after either cutting or grazing, much plant material remains in the stubble and roots and there is a continuous turnover of N through the leaf litter and roots (Parsons el al., 1991). The quantities of roots and macro-organic matter in the soil increase with the age of the sward (Garwood, 1967). Where swards are cultivated and resown or brought into arable production, the macroorganic fraction has been shown to be relatively labile (Warren and Whitehead, 1988) and large amounts of N may be released for a number of years. Potential mineralization has also been related to the “light” SOM fraction in a sandy soil (Gaiser and Stahr, 1994). Francis et al. (1992) measured 230 kg N ha-1 accumulated in the stubble and roots of a 3-year ryegrass/clover sward. Whitehead et al. (1990) measured 536 and 602 kg N ha-1 accumulated in stubble, leaf litter, roots, and macro-organic matter in an 8- and 15-year-old sward, respectively.
199
NITROGEN MINERALIZATION Table II
Typical Balances between N Uptake (or Fixation) and N Removed in the Harvested Produce of Arable Crops in the United Kingdom on >lOO,OOO ha (Sylvester-Bradley, 1993) Crop
Yield (t ha-')
N content (kg t-1)
N uptakea (kg ha-')
N offtakeb (kg ha-')
Winter wheat (milling) Winter barley (feed) Spring barley (malting) Winter oilseed rape sugar beet Maincrop potatoes Linseed Winter field beans Spring oats
7.5 6.5 4.5 3.2 42.0 40.0 1.8 3.5 4.7
19 17 14 32 1.7 2.5 38 42 17
190
140 110 65 100 70 100 70 145 80
N leftC (kg ha-') ~____
150 80 250 200 200 90 225 115
~
50 40 25 145 130 100 20 80 35
Yield X N content of whole above-ground crop. N removed in harvested crop. N remaining in crop residue after harvest.
Young ( 1986) calculated that N mineralization increased after ploughing from approximately 100 kg N ha-' for a I-year ley to 280-380 kg N ha-1 for leys older than 4 years. It has been shown that grassland management, e.g., N inputs and grazing, influences N accumulation and, hence, mineralization after ploughing (Whitehead et al., 1990). Shepherd (1993) found that soil mineral N contents were of no value in predicting response of wheat to N after ploughing grass swards because subsequent immobilization/mineralization processes were too variable. Biologically fixed N may also increase the N content of specific crop residues and, thus, potential mineralization rates. Further, there has been some evidence that white clover (Trifolium repens L.) has an effect in increasing mineralization perhaps because of high N contents in roots and nodules, but also because soil structure may be improved in the presence of the clover which could increase mineralization (Mytton et al., 1993). The effect of clover is uncertain and the clover content of sward and its contribution to N cycling have only rarely been well defined. Animal manures are often added in large quantities to soils and are a mixture of NHJ, urea, uric acid (especially in poultry manures), and more complex materials such as proteins, either undigested or partially digested during transit through the gut (Chescheir et al., 1986). Manures vary in both total N content and the forms of N present which are affected by many factors, e.g., animal species, diet, amount and type of litter, and dilution by water. The main source of carbon in animal slurries or layer poultry manures will be that which has passed
2 00
S. C . JARVIS ETAL.
through the animal. However, bedding will introduce other carbon sources and the interaction of manure and bedding, either during storage or after application to the land, will have important effects on the release of N from organic forms and subsequent nitrification (Jensen, 193l), but this is poorly defined. Dung and urine directly voided to pastures by grazing animals behave differently to stored manures (but have enormous effects on N transformations) (Jarvis et al., 1995). Storage of manures before application to land results in some losses of N and other nutrients, and changes in the pools of residual N. Composting of manure caused a large reduction in volume and produced a material with an increased proportion of stable organic materials with a higher C:N ratio (Lampkin, 1990). Anaerobic storage of manures or slumes results in materials with high NH,+ contents. Thus, on average, 23% of the N in dairy cattle manure was lost as NH, during composting (Lampkin, 1990), compared to 10-15% during anaerobic storage. However, losses after application to land are much increased in manures which have been stored anaerobically, since they contain a much greater proportion of NH,+ (Pain et al., 1993). The losses of N by volatilization during composting have been reduced by increasing the initial C:N of the compost (Kirchmann, 1985) and where compost heaps are not covered, then substantial losses of potassium (K) and N as NO, may occur through leaching (Russell, 1973). Since urea and uric acid hydrolyze rapidly to NH,+ after application, these compounds are usually included as part of the inorganic N pool of manures and the hydrolysis is not considered strictly as mineralization. The amount of N released in available forms from manures has often been calculated assuming that the inorganic N (NH,+, urea, and uric acid) and organic N pools supply N independently, e.g., N supply = E , N , E,MNo, where Ni = inorganic N content, No = organic N content, E , and E , are efficiency factors for N which allow for loss processes, and M is the proportion of organic N mineralized (Sims, 1986). However, stimulation of biological activity in soils due to the readily available C added in manures can lead to rapid and significant immobilization of the manure inorganic N pool (Flowers and Arnold, 1983; Bernal and Kirchmann, 1992). The interactions between the N pools in manure and soil biomass and the inherent variability of manures make the prediction of the timing and amount of N release from manures difficult. There is a need for better characterization of the forms and amounts of N in animal manures and better assessment of the availability and interactions between these pools under field conditions. Effects of cattle slurry on N release have been found to at least 0.5 m depth when applied to grassland (Kandeler et al., 1994). Sewage sludges may be applied to agricultural land when they do not contain significant amounts of metals or other toxic chemicals and can provide a significant input of organic materials. As with animal manures, sewage sludges contain both inorganic and organic N fractions. The relative proportions of inorganic and organic N in sewage sludge and the composition of the organic fraction (i.e.. its
+
NITROGEN MINERALIZATION
201
recalcitrance) depend on the sewage production process and effects are therefore variable (Smith et al., 1992).
111. PROCESS CONTROLS Mineralization is a highly variable process and although there have been many studies over several decades, predictive capability is poor. The rate at which mineralization occurs is the result of complex interactions between biological, chemical, and physical components of the soil and is subject to many external influences that have been reviewed extensively (e.g., Haynes, 1986). Basic knowledge was established about mineralization kinetics in the early 1970s, mostly derived from experiments under controlled environmental conditions and a first assumption (Stanford and Smith, 1972) was that mineralization depended upon first-order kinetics with only one pool of degradable SOM that was available for utilization and N release. Because the concept of multiple organic-N pools (Section 1I.A) with different decomposition rates has been developed recently (Bonde and Lindberg, 1988; Hassink et al., 1990; van Veen et al., 1985; Warren, 1985), it seems likely that mixed-order kinetics will provide a better description. More usually, multiple pools are considered in models, each having a different rate constant for decomposition, but with first-order kinetics.
A. RESOURCEQUALITY Soil organic matter is a very variable material and, in the first instance, is influenced by the nature of the returned organic materials which has a substantial effect in determining mineralization. Russell (1973) described crop residues as comprising three separate groups of materials, (i) cell wall and structural materials, consisting of the skeletal framework (i.e., cellulose) and cementing/ encrusting materials (with carbohydrates predominating in young shoots and lignin accumulating in older tissues), (ii) reserve substrates including starches, fats, and proteins, and (iii) cell contents (i.e., proteins, sugars, unassimilated NO,, and traces of NO, and NHJ). Mengel and Kirkby (1978) estimated that the relative proportions of N components in residues were as follows: inorganic fraction ( N O , NHJ, NO,) generally <2%; soluble amino compounds (acids, amides, and amines) approximately 5%; and protein and nucleic acids 90-95%. All three fractions are influenced by crop growth and nutrition, especially the supply of N. The N concentration in plants declines with age because a greater proportion of resource is diverted to the production of N-free material required for structural support (Raymond ef al., 1960). Increasing the N supply increases
2 02
S. C. JARVIS ET AL.
all fractions, but not to the same extent, and the largest effect is on soluble amino compounds. Once in the soil, the simpler N compounds mineralize more quickly than complex materials (Rubins and Bear, 1942). Lignin, for example, is resilient to attack and it may also protect cellulose by encrustation, i.e., an example of physical protection and chemical composition influencing mineralization. The relative proportions of the different N forms in crop residues have a marked effect on the rate at which breakdown occurs (Waksman and Tenney, 1927). The nature of the returned organic materials is therefore of some importance; C:N characteristics of materials provide an indication of the likely balance between mineralization and immobilization processes when residues are added to soil. However, there is little evidence to suggest that critical C:N ratios for either SOM or added materials can be defined which are generally applicable to aid prediction of mineralization rates (Haynes, 1986). In many respects this would be useful in providing an easily determined characteristic, but the wide range of other constituents (e.g., lignin and polyphenol contents) and their behavior makes dependence on interpretation of C:N too simplistic. C:N status does, however, provide an important guide to whether net N mineralization or immobilization will occur when organic materials are added to soil. In general, as substrate C:N increases, mineralization decreases until some critical point is reached (Marstorp and Kirchmann, 1991). This relationship has not always held (Douglas and Magdoff, 1991) and has been criticized since not all the C and N is readily available to micro-organisms (Frankenberger and Abdelmagid, 1985; Reinersten et al., 1984). The chemical composition, competitive activities, and community structure of the SMB also vary between different systems and therefore influence the efficiency of C use and N demand (Ladd and Foster, 1988). This in turn influences the reliability of simple ratios as an index of mineralization. Attempts to provide a more mechanistic approach by considering C either alone (Douglas and Magdoff, 1991) or in conjunction with specific components such as lignin (Frankenberger and Abdelmagid, 1985) have met with only limited success. Jenkinson (1984) suggested that the C:N ratio was a good guide to the amount of N released from a residue in a year in temperate arable soils with typical crop residues. However, to make progress it will be important to define differences in the short-term release patterns for different residues with widely different characteristics as illustrated in Table 111. The C:N of manures has also been suggested to have a major influence on subsequent mineralization in soils. Salter and Schollenberger (1939) found that yield depressive effects were not serious unless the manure C:N ratio exceeded 20, whereas Kirchmann (1985) suggested that a C:N of 15 was critical, with N immobilization occurring above and mineralization below this value. This critical ratio of 15 has further been corroborated by Castellanos and Pratt (1981) and Beauchamp (1986). Analysis of 36 cattle farmyard manure samples found an average ratio of 14, with 14 samples having a ratio > 15, i.e., with a potential for
203
NITROGEN MINERALIZATION Table III Examples of C:N Ratios of Arable Crop Residues and N Return (kg N ha-') (Source: ADAS, Unpublished Data, Unless Otherwise Stated)
C:N ratio
No. of Crop
Residue
samples
Mean
S.D.
Range
N return in residues
Sugar beet Potatoes Dried peas Field beans Oilseed rape
Tops Haulma Haulma Haulma Haulma
35
17.1 22.7 36.9 43.6 55.0
2.35 5.41 12.70 6.87 -
12.1-25.0 15.5-30.3 19.5-62.5 32.8-56.3 22-74b
118 41 48 45 87
a
10
20 14 -
Sampled at harvest, and therefore does not include material senescing before harvest. From Holmes (1980).
net immobilization (ADAS, unpublished data). Castellanos and Pratt (198 1) reported C:N values of 6.5 for chicken manure and 9.9-15.9 for cattle and pig manures. In a UK survey (ADAS, unpublished data), average values for cattle and pig slurry were 5 and 2, respectively. However, as large proportions of the N in all the above materials were NHZ, it is more meaningful to look at the solid fraction alone where typical ratios were 13 and 7 for cattle and pig slurries, respectively.
B. ENVIRONMENTAL CONTROLS Availability of appropriate sites with appropriate conditions for mineralization, and any other transformation, is controlled by the relationships between the soil and the external environment, especially those factors which influence the degree of soil aerobicity. Aeration status is dependent upon soil texture, structure and moisture, the cultural management of the system and the returns of organic materials. Soil temperature and moisture then further interact by influencing microbial activity. Physical location of organic materials at a microscale is therefore of some importance: distribution in locations with access to 0, increases the potential for N release. Significant proportions of SOM are associated intimately with clay and other inorganic components of the soil matrix which restricts the potential for mineralization. The vertical distribution of organic materials in the soil profile also has effects. Thus, although the concentration of mineralizable N has been shown to decrease exponentially with depth (Power, 1980), laboratory incubations of soils from different depths have indicated that although the majority of mineralization occurs in the top 20 cm, 20-30% of total N released would
2 04
S. C. JARVIS ET AL.
be mineralized below 60 cm (Cassman and Munns, 1980; Hadas et al., 1986). Mineralization in the subsoil may therefore contribute substantially to the amount of N mineralized, and this has important implications for leaching losses. Mineralization is influenced profoundly by the temperature changes that are normally encountered under field conditions, with a lower limit close to freezing. The majority of soil micro-organisms are mesophyllic and prefer moderate temperatures within optimum activity between 25 and 37°C and a base temperature of 5°C. Freezinghhawing changes may have comparable effects to those of wetting/drying and result in the release of soluble materials and/or disruption of soil aggregates. As well as mesophiles, psychrophiles, which are active at lower temperatures, are also common and have been shown to play an important role in British soils (Dickinson, 1974). This has implications, therefore, for release of N over winter during periods of maximum drainage. Recent studies in long-term grassland soils have indicated that between 21 and 38% of a total net annual mineralization occurred during the period November-February (Gill et al., 1995): this again has implications for leaching losses. In general, net mineralization rates increase with temperature and tend to become less variable at higher temperatures (Stanford et al., 1973). At optimum soil water content, an Arrhenius function with a Q,,of approximately 2 described the relationship between temperature and net N mineralization (Kladivko and Keeney, 1987). Changes in the availability of soil water have a number of effects: (i) deficiency/stress limits biological activities and hence mineralization; (ii) excess reduces aerobicity and therefore alters the activities of different microbial populations in different microsites, i.e., reduces mineralization; (iii) soil water content controls solute diffusion and mass distribution of the products of microbial activity; and (iv) cycles of wetting/drying increase the availability of substrates (Cabrera, 1993). Drainage of permanent grassland soils increased the amounts of NO, leached because ‘it was thought’ of an enhanced mineralization capacity where aerobicity was increased (Scholefield et al., 1993). Some bacteria can operate under anaerobic conditions but their rate of activity is much lower than under aerobic conditions. The mechanisms involved during wetting and drying are not clear, but may involve a release of energy supplies after physical disruption and/or increased accessibility of degradable organic materials and, in the first instance, a significant flush of mineral N through a release of readily mineralized materials from freshly killed biomass cells (Marumoto et al., 1982b). Moisture distribution in the soil appears to be more important than the absolute amounts present (Cassman and Munns, 1980). Optimum net mineralization rates have been achieved at between -0.33 and -0.1 bar where water occupied 8090% of the pore space; the rates fell as soil moisture potential fell below -0.33 bar (Stanford and Epstein, 1974). At very high soil moisture contents the mechanisms which control mineralization are unclear and both increased and decreased rates of mineralization have been observed. Although moisture/temperature re-
NITROGEN MINERALIZATION
20s
gimes are likely to be the most important external controls over mineralization rates, there is contradictory evidence for a significant interaction between the two variables (Cassman and Munns, 1980; Kladivko and Keeney, 1987). It has been suggested that separate equations to describe moisture and temperature effects can be used to predict mineralization rates without introducing significant errors (Kladivko and Keeney, 1987). However, whereas net mineralization rates in grassland soils throughout the year were not significantly influenced by variation in soil moisture, between 31 and 41% of the variance was accounted for by soil temperature (Gill et a l . , 1995). There have been suggestions that adding immediately available N, in fertilizers for example, stimulates mineralization, a so-called priming effect. However, it has been shown by both modelling (Jenkinson el al., 1985) and experimental studies (Hart et al., 1986) that this is usually an artifact of procedure and unlikely to be significant in most circumstances. Soil texture exerts an important control over mineralization by (i) influencing aeration/moisture status, (ii) affecting the physical distribution of organic materials and hence their potential for degradation, and (iii) conferring some degree of “protection” through an association of organic materials with clay particles (Hassink et al., 1993). Dickinson (1974) suggested that adsorption of microorganisms on the surface of clays and the creation of protected sites was an important controlling factor whose impact increased as soil clay content increased. Other studies have shown (Gregorich et al., 1991) that there was a close interaction between SMB and their decay products in clay soils which allowed a more efficient transfer of nutrients between subsequent microbial generations. Major differences in mineralization rates in soils of different textures have been found and the fraction of small pores, < 1.2 pm, explained more than 50% of the variation in N mineralization rate between soils (Hassink, 1992). Over 70% of the variation was explained by the soil particle size fraction <50 pm (i.e., silt plus clay content). Thus it seems probable that a major proportion of SOM that is vulnerable to mineralization may be located in pores or on surfaces which cannot be reached by micro-organisms or do not have the requisite degree of aerobicity and therefore can be regarded as protected. This therefore occurs to a greater extent in finer than in coarser textured soils resulting in lower net mineralization rates (Verberne et al., 1990). The degree of structure development and the relationship between particles and pores are therefore extremely important and have implications for understanding the effects of cultivation on mineralization (Hassink, 1994). Physical disruption of the soil increases both aeration potential and microbial access to readily mineralizable SOM (Cabrera and Kissel, 1988) and provides at least a partial explanation for the flush of mineral N that is released when grassland is ploughed (see Section 1II.C). Other soil factors are also important and include pH [decomposition is inhibited under strongly acidic conditions; Jenkinson ( 1981)] and contamination of
2 06
S. C . JARVIS ET AL.
soils by, for example, heavy metals which have immediate impact in reducing mineralization rates (Chang and Broadbent, 1982). The presence of plants can be expected to influence mineralization rates through an increase in the availability of substrates. The importance of this depends on the C:N ratio of the materials either exuded from the roots or returned to the soil during senescence (Section 1II.A). After five species of grasses were grown in the same soil for 3 years (Wedin and Tilman, 1990) there were 10-fold differences in net mineralization which corresponded to species differences in tissue N concentration, belowground biomass (roots), and lignin contents.
C. EFFECTS OF CULTIVATION Cultivation, its timing and methodology, is an important tool for modifying mineralization of old organic matter as well as recent additions. Mechanical disruption of the soil structure makes previously protected SOM available for degradation and increased rates of mineralization have been observed in disturbed soils (Ballesdent et al., 1990). The cultivation method, type of machinery used, and the energy input into a soil all have effects on the disruption of soil structure and therefore can be expected to affect the amount of N mineralized. Ploughing has caused more mineralization than establishing crops by direct drilling. Goss et al. (1993) estimated that extra net mineralization of approximately 20 kg N ha-1 occurred with ploughing as compared with direct drilling, although other studies have shown smaller differences (e.g., Powlson, 1980). Where residues were left on the soil surface, mineralization rates were reduced and immobilization rates enhanced, compared with an incorporation management (Power et d . , 1984). Richter et al. (1989) showed that deeper ploughing slowed mineralization, but Francis et al. (1992) in contrast found no differences between mouldboard and chisel ploughing a grass ley. Similarly, Webb et al. (1991) and Lloyd (1992) found only small or no differences between ploughing compared with direct drilling or minimum cultivations following a grass ley. Timing of cultivation also affects mineralization patterns and, therefore, availability of NO, for crop uptake, leaching, or other processes. Vinten et al. (1992) and Stokes et al. (1992) have shown that delaying cultivations in autumn significantly reduced NO, release. It is also important to consider the effect of adverse soil conditions on mineralization. Compaction, for example caused by heavy machinery or cultivating the soil at the wrong time, reduces soil porosity, which modifies and reduces net mineralization. Furthermore, the age and dynamics of SOM in relation to the soil structural hierarchy (Dexter, 1988), which will be determined in large part by cultural procedures, are thought to be important in separating SOM into pools which are biologically significant. Correlations have also been established be-
NITROGEN MINERALIZATION
207
tween the bacterial component of SMB and a specific pore size fraction (Hassink et al., 1993).
D. MICROSITES, DIFFUSIONAL CONSTRAINTS, AND SOIL ARCHITECTURE When inorganic N is extracted from soil it represents the sum of that in many microsites. Although the bulk concentration of NHJ is not thought to control microbial processes such as immobilization and nitrification, their rates may vary widely even when the bulk soil NHJ is constant (Davidson and Hackler, 1994) through microsite heterogeneity. There are substantial diffusional constraints to the movement of ions in soils and these result in a diversity of micro-organisms within close spatial and temporal proximity which otherwise which would not occur (Focht, 1992) and which influence further transformation and transfer of N. Restricted diffusion of NH,+ may result in preferential immobilization by SMB (Davidson et al., 1990) and only the N in excess of microbial requirements enters the available NH$ pool (Drury et al., 1991). Where organic N resources are low, little NH,+ will be released and any mineral N diffusing to a particular site will be immobilized. Recent studies of soil N cycling, and mineralization in particular, have considered the architecture of the soil in relation to the internal N cycle and the distribution of soil organisms (Drury et al., 1991; Killham et al., 1993). Any such consideration has to be made in association with a knowledge of the diffusion constraints to NHJ and NO, movement between microsites, i.e., in different physical pools (Darrah et al., 1983). On this basis, a mechanistic understanding of the controls over mineralization at a fundamental level can be developed and it should then be possible to integrate effects sequentially into a series of larger scales, taking into account the unique spatial and temporal properties of each one. A model of soil N cycling based on more realistic, diffusionlimited activities should provide a more fundamental understanding of mineralization and perhaps allow, with other appropriate data, direct interpretation of soil mineral N contents. The concepts of microsites and different “pools” of potentially mineralizable N and mineral N are implicit in this approach. Much is also known qualitatively about many of the soil factors affecting mineralization (Section 111). While there is quantitative information on temperature and moisture, this is restricted and does not address interactions with other factors and therefore does not, at present, allow a practical interpretation of the impact of mineralization and recycling other than in a few specific circumstances. Further, the current information does not allow the development of effective broad-brush predictive models. Progress requires better definition not only of the SOM, but also its location and interaction with soil biology and environmental status. The relationships between pore size (resulting from an
208
S. C.JARVIS ET AL.
interaction between soil texture and structure) and mineralization, and whether “protected organic matter” is of practical relevance, need to be determined.
IV. MEASUREMENT AND PREDICTION OF MINERALIZATION
A. BACKGROUND Methods used to measure mineralization and predict its relevance for plant uptake, losses to the environment, and overall N cycles have, for the most part, relied on determination of net mineralization, i.e., the balance over immobilization. Many methods to assess net mineralization have been described, but because it has been seen to be of only limited immediate practical use, and because it has been difficult to measure, gross rates of mineralization have rarely been determined. The recent development of stable isotope methodology has meant that investigation of gross mineralization is now more easily achieved and is an important development to aid further understanding. It is possible to partition the available methods to assess mineralization in various groupings according to different criteria. In this review we identify three main groupings, i.e., those which (i) rely on either the plant as a means of estimation or a knowledge of N budgets for the system, (ii) attempt to provide indices of potentially available N by chemical extraction or by incubation under controlled conditions, or (iii) aim to determine, either directly in the soil profile or under conditions as close as possible to those in the field, the products of the processes involved, i.e., NH,+ and NO,. The aim of all these methods has been to provide a more precise basis for evaluating the capacity of soil to supply N. We discuss the more important techniques that are currently used and indicate their potential for further use as well as their shortcomings.
B. MINERALIZATION FROM N BALANCE/~ROPPING DATA One of the simplest methods to provide information on the supplies of N from native SOM sources is to use the crop as a sink for any N released, i.e., by growing plants under natural conditions in the absence of external inputs (fertilizers, excreta, fixation, and, ideally but not often easily achieved, atmospheric deposition). Uptake and removal into the crop are then assumed to equate with mineralized N; such information is an essential component of many fertilizer response trials and thus many data sets exist from which mineralization can be estimated. While this can provide useful information it does require careful interpretation in order to be translated into net mineralization rates because (i) N released and partitioned into nonharvested components, i.e., roots, stubble, senesced leaves, etc., is not accounted for, (ii) there may be significant losses
NITROGEN MINERALIZATION
2 09
through leaching or denitrification, (iii) there may be significant influence of the plant itself through the release of exudates and their impact on soil microbial populations and their activities, and (iv) soil mineral N contents are required at the beginning and end of measurements. The information obtained in this way is also retrospective and applicable only to a specific soil/crop/environment combination. Nevertheless, because of its extent and, in contrast to data from other methods, availability, it may be of value in providing a guide to the ranges of mineralization likely to be encountered under a wide range of conditions. The interpretation of N in crop removal is therefore complex. While such data integrate crop growth and soil N dynamics for any crop-climate system (Meisinger, 1982), they do not, for example, allow for any residual effects from the previous season. Further, removal into the crop is season-specific and different crops have different uptake patterns and total demand for N (Stockdale er al., 1992). The below-ground plant biomass contains substantial amounts of N (Saffigna, 1988) and with perennial crops there will be a reservoir of root N which does not relate to the current season’s input (Huntington er al., 1985). Preharvest N loss in leaf fall and/or by volatilization from leaves is also important (Huntington er al., 1985) and difficult to quantify. Nitrogen balance studies have also been used to determine mineralization either directly or by difference (Powlson et al., 1986; Vinten et al., 1992), and combine measurements of crop uptake, mineral N in the soil profile, N in soil organic pools, and estimates or measurements of losses. This is a very laborintensive method and is reliant on a large range and number of measurements and analyses. At present, information from crop N uptake studies provides the most extensive data sets for estimating mineralization, with information from fertilizer response trials for many soils, crops, and previous managements. Interpretation of these data could provide a valuable means of estimating flows of N from mineralization for a wide range of systems, but careful calibration will be required in order to estimate accurate rates of release of N.
C. LABORATORY DETERMINATION OF POTENTIAL MINERALIZATION INDICES 1. Incubation Methods
One of the most widely used approaches has been incubation of a fixed mass or volume of soil under standard conditions over a defined period (typically up to 34 weeks) and determination of the increases in NO, and NH$ concentrations. The protocols for doing this are many and varied. The objective has been to estimate the organic N pool that is available for mineralization, i.e., it is assumed that there is a pool which can be mineralized by micro-organisms in a finite time under optimal conditions. The advantage of this approach is that conditions can
210
S. C . JARVIS ET AL.
be defined and held constant and some of the problems of field spatial and temporal variability can be overcome, but rates are usually higher than those measured in the field where optimum conditions rarely prevail (Adams and Attiwill, 1986). Thus, although good correlations often exist between N released during incubation and uptake in pot studies, relationships are much poorer in field studies. Many methods rely on aerobic conditions and are usually some variation of the method proposed by Stanford and Smith (1972). There are a number of modifications of this basic method. Anaerobic systems have sometimes been preferred (Keeney, 1982; Lober and Reeder, 1993) as only NH$ has to be determined; because nitrification is restricted, there is no need to optimize water contents and higher temperatures can be used so that more N can be released in a given period than under aerobic conditions. Most of the methods evaluate mineralization in soils which are much modified by mixing, sieving, drying, and/or rewetting. The possible combinations of methods/pretreatments/storage/incubation are numerous and all have impact on mineralization rates. Thus soils have been sieved (Nordmeyer and Richter, 1985), air dried (Stanford and Smith, 1972), field moist (Skjemstad et al., 1988), or frozen (Beauchamp et al., 1986) prior to incubation under a variety of conditions. The disruption of aggregates and disturbance of anaerobic or aerobic microsite activity can have profound impact on mineralization potential. Some methods try to reduce these problems by using intact cores. The use of structurally undisturbed cores is also difficult, in large part because the spatial variability in the field demands much replication (Macduff and White, 1985), but also because they contain excised roots which could affect mineralization kinetics (Ross ef al., 1985). It is clear that N mineralization is dependent on sample pretreatment and incubation conditions as well as the inherent properties of the soil and the chemical constitution of the organic materials (Nordemeyer and Richter, 1985). The conditions chosen in any one study usually reflect the particular aims of that study. Further, this approach provides a “snapshot” picture only, dependent on conditions prevailing at the time of sampling. Contents of more labile forms of organic materials change through the year, as will populations of micro-organisms. There are therefore difficulties in extrapolating results from incubation system to field scale where many factors other than those held to be important in the laboratory will be operating and/or interacting. Assumptions therefore have to be made in terms of the models used to estimate rates in the field. The use of data from laboratory incubations to predict field mineralization rates has had variable success. Where results from incubations of disturbed samples were used to predict those in the field, overestimates of 67-343% were obtained (Cabrera and Kissel, 1988). On other occasions, reasonable agreement between prediction from incubation and that measured in lysimeters has been found, but not when there were flushes of mineralization during wetting/drying sequences (Campbell et al., 1988). There may also be problems where crop residues are concerned because of an unrealistic degree of mixing in incubation
NITROGEN MINERALIZATION
211
systems. If the objective is to represent field conditions, then structurally undisturbed cores with much replication are more appropriate. With all experimental approaches, models have to be developed to describe and extrapolate the information. Although used successfully in some instances, not all mineralization patterns have been described by first-order kinetics (Matus and Rodriquez, 1994; Stanford and Smith, 1972), for example zero-order kinetics were observed in fresh, undried and unamended soils (Addiscott, 1983) and in meadow soils (Simard and N’Dayegamiye, 1993). Other more complex multiphase patterns have been described for both meadow soils (Simard and ”Dayegamiye, 1993) and plant and animal residues (Chae and Tabatabai, 1986). To determine potentially mineralizable N precisely requires that measurements are continued until the rate declines to a small constant level (Bonde and Linberg, 1988). In most instances, time dictates that mineralizable N and its associated rate factor have to be determined by fitting models to the data. It was assumed that a one-pool model (Stanford and Smith, 1972) would adequately describe the process (Section II.A), but recently two-pool models have been used with both having either first-order (Deans et al., 1986) or first- and zero-order (Beauchamp et al., 1986) kinetics. In fact, in the field it is likely that an exponential model with an infinite number of pools is most representative, although from a practical point of view the number of pools to be included would depend upon the objectives and desired accuracy (Sierra, 1990). Other more complex relationships have been proposed, including parabolic equations (Broadbent, 1986) and the use of the van Bertalanfly function and Weibull distributions (White and Marinakis, 1991). All of these depend on nonlinear least squares fitting of equations and produce a predicted mineralization rate as a mathematically defined quantity, and depend upon there being a homogeneous, discrete presence of organic N in the soil. The use of incubation methods has told us much about the mechanics of the mineralization process and allowed comparison between soil types under controlled conditions. To be of further value for practical applications, this approach requires standardization, the development of a rapid method to estimate potentially mineralizable N, and the means to extrapolate the information to different cropping systems. This presents a considerable challenge but is required so that results can be used to aid advice and to add to the general understanding of the supplies of available N under field conditions.
2. Chemical Methods a. Extractants Chemical extractants have been used as rapid and convenient means to “quantify” N released from labile SOM. The severity of extraction varies widely from strong hydrolysis to relatively mild extraction with hot water or hot salt solutions; although the latter have shown the most promise (Stanford, 1982), acidified potassium permanganate and dichromate have been used with some success
212
S. C . JARVIS ET AL.
(Wilson et al., 1994). The amount of N extracted from the soil is assumed to represent, or to be related to, the N available to plants during the following growing season or to leach during winter if the extractions are undertaken in the spring or autumn, respectively. The use of such a method is therefore dependent upon correlation with crop yield, N offtake, or N leached. Some of the more recently successful methods have used KC1 solutions at differing concentrations and temperatures over various times (Whitehead, 1981; Gianello and Bremner, 1986). Nitrogen released with hot 2 M KCl correlated well with N uptake by malting barley on a restricted range of Scottish soils (McTaggart and Smith, 1993). Whitehead (1981), using a similar extraction, found a good relationship with N uptake by perennial ryegrass in a pot experiment. However, the method has not been tested widely and, while it might provide an index of the basal rate of mineralization (Stockdale and Rees, 1994), it has not predicted the changes resulting from recent additions of plant or animal residues (McTaggart and Smith, 1993). None of the proposed methods or derived indexes has been adequately tested under a broad range of field conditions or put to general use. b. Labile N Another recent method has been the use of electro-ultrafiltration in which anions and cations in soil suspensions are separated in an electrical field. The solutions recovered from this treatment contain not only NO, and NH$ but also soluble organic N. They are therefore assumed to contain readily mineralized organic N, i.e., all forms of N that are likely to be available for plant uptake or transformation processes over the shorter term. The method has been used in some European studies to provide better prediction for fertilizer requirements (Mengel, 1991) and has in some instances provided good correlation with N uptake and N mineralized from soils in pot experiments with rape (Appel and Mengel, 1990). Other relationships have been developed for the supply of N to barley in field trials and EUF extracts (and also with UV absorption at 205 nm) (Linden et al., 1993). In other studies, the technique has provided poor correlation with N uptake by maize from unfertilized plots (Saint Fort et al., 1990) and also where fresh residues have been incorporated (Appel and Mengel, 1993). While there may be a useful role for this technique to improve short-term prediction of N supplies and early season utilization of N, it is unlikely to be relevant to a whole season or to have general applicability across a wide range of soils/sites. c. Soil Organic Matter/Total N It has been suggested (Hadas et al., 1986) that a good general approach is to use total soil organic matter N content as an index of the potential to supply N. On theoretical grounds this would seem unlikely given the nonuniform nature of the substrates and the wide range of soil and environmental conditions to which they are subjected. There may be some value in using this approach to examine extremes, for example highly organic versus mineral soils, but the need to do this
NITROGEN MINERALIZATION
213
is restricted. There have been only a few recent attempts to use this approach and often there has been little or no relationship between total soil N with either uptake or estimated mineralization rates (Appel and Mengel, 1990). Freytag et al. (1989) showed that although there were no simple relationships, by grouping the soils by site and/or by some soil characteristics, a degree of success in estimating mineralization could be achieved. To provide sufficient confidence that the method could be used for a wide range of cropping/soils/environment conditions would require substantial input. d. Extractions for Residues and Organic Wastes There are many reports of methods to assess the likely immediate impact of manures and plant residues on mineral N release. Parker and Sommers (1983) found that the use of chemical indices oversimplified the complex process of N release from manures. Various methods have been used, e.g., acidified and alkaline permanganate in aerobic incubations (Castellanos and Pratt, 1981; Parker and Sommers, 1983), autoclaving (Parker and Sommers, 1983; Douglas and Magdoff, 1991), 6 M HCl (Serna and Pomares, 1991), pepsin digestion (Castellanos and Pratt, 198I), Walkley-Black digestion (Douglas and Magdoff, 1991), and CO, respiration during incubation (Castellanos and Pratt, 1981), with varying degrees of success. C:N ratio, %N, and phenolic content of residues have been identified as characteristics which influence mineralization, but there is little consistency in their relative importance (Section 1II.A). Whereas in some studies (Frankenberger and Abdelmagid, 1985) there was a strong correlation between residue N content and mineralized N, in others (Silvlapan et al., 1985) soluble N content was found to be more important. On their own, lignin or phenolic contents have not always good been predictors of N mineralized but have helped when integrated with some means of assessing N content (Fox et al., 1990; Silvlapan et al., 1985). While C:N values are often available, these give variable results when used as a sole factor to explain mineralization rates. While much is to be learned from the use of many of the methods noted above, the general applicability of the information that each achieves is likely to be restricted. Where residues are involved there is a need to define the various interactions between their characteristics and soil/environment conditions. There is no suitable database for either SOM or residue characteristics to permit this definition.
D. FIELDMEASUREMENTS 1. Changes in Mineral N Various studies have attempted to deduce some measure of mineralization by following the changes in soil contents and/or losses of NO, and NH,+ over time.
2 14
S. C.JARVIS ETAL.
Although useful, this is a simplistic approach and is not likely to provide a realistic assessment because of the internal recycling, transformation, and external inputs and losses that may occur. Mineralization rates calculated from observations of change in size of the mineral N pool should measure all of the processes or aim to exclude them. It is also worth noting that the size of the pool does not necessarily reflect its importance. Thus, although the NO, pool is often very small at any one time, it can turn over in 1-2 days and maintain available supplies (Davidson et al., 1990). Good correlations between mineral N measurements made in the topsoil at sowing of spring barley and N uptake of the crop (McTaggart and Smith, 1993) have been shown. It was suggested that, by the time the soil samples were taken, some mineralization had already occurred and this gave a good indication of the site differences in mineralization which would have persisted throughout the season. A major problem in using changes of mineral N to measure mineralization rates is the degree of spatial variability that exists in soils and the difficulty of making comparisons over time with spatially separated samples. Even under uniform management mineral N is spatially highly variable; if this is increased by residue addition or excretal returns under grazing, the problem is confounded. It has been estimated that, depending on the area involved, between 24 and 40 individual samples per field are required for analysis (Goovaerts and Chiang, 1993; Stockdale ef al., 1994). A further complication is that mineral N contents are often log-normally distributed (Macduff and White, 1984; Stockdale et al., 1994). Grassland soils provide a further complication because (i) they support a perennial crop with potential for removal of mineral N throughout the year whenever environmental conditions are appropriate and (ii) they may accumulate large amounts of mineral N through recycling in excreta (Jarvis, 1992) either at grazing or after application of farm wastes.
2. Field Incubations A range of field incubation techniques has been proposed with varying degrees of sophistication. There has been a recent resurgence of interest in these techniques because of the current need to understand all inputs and internal cycles of N under practical conditions. Techniques have included using a soil sample sealed into a polythene bag and repositioned back into the soil under ambient conditions, i.e., at the same temperature, with minimal change to soil moisture status and no risk of interaction with plants (Gordon et al., 1987). In other measurements intact cores have been used (Alves et al., 1993) to reduce the changes due to disturbance either in polythene bags (Nadelhoffer ef al., 1985) or in capped tubes (Raison et al., 1987). The latter system has been refined to include exchange resins to trap any NO, leaving and, on occasion, entering the columns so that changes in soil NO, and NH,+ could be determined more accurately (DiStefano and Gholz, 1986).
NITROGEN MINERALIZATION
215
Another recent method has been developed to help reduce the errors incurred because denitrification removed NO,. Intact cores were incubated in sealed containers in the field at ambient soil conditions with acetylene added to inhibit nitrification and thus reduce the opportunity for denitrification (Hatch et al., 1990). This has been used on a regular, routine basis (Hatch er al., 1991; Gill e l al., 1995) to provide estimates for net mineralization in grassland soils. A further degree of refinement was introduced by Blantern (1991) who incubated intact cores sheathed in plastic to minimize effects on the core surface through changes in gaseous composition and in an atmosphere of 20% 0, in helium after purging N, from the soil atmosphere so that full account could be taken of denitrification effects.
3. Applicability of Field Methods Field methods have the advantage that they allow measurement under ambient conditions over long periods, are relatively simple to use, make some attempt at reducing soil variability, and take steps to reduce or stop losses. On the other hand, there is a requirement for frequent measurements and a need for some assurance that, where it is used, acetylene is effective and that atmospheric conditions within the vessel do not deviate too far from those present originally. Furthermore, little attempt is made to control water status which is only representative of what is present at the start of a measurement period. Ideally, large numbers of replicates are needed (Raison et al., 1987). Few cross comparisons between methods have been made, but Redman et al. (1989) compared net mineralization rates determined with field core incubations with those derived from a budget study. They found that while there were no differences during winter, there were significantly higher rates from the incubation methods between April and July. Undisturbed cores in plastic bags compared with exposed cores with ion-exchange resins gave similar estimates (Zou et al., 1992). All methods, even where cores are left intact, are affected by sampling and incubation procedures which will have effects on net N release, including that derived from damaged roots. It seems unlikely that any one method could ever fulfill all needs and that a range of approaches will have to be utilized.
4. Use of ‘SN-Labeled Fertilizer By labeling inorganic N fertilizer with ‘5N it is possible to distinguish between N derived from fertilizer and that derived from other sources which, in most cases, will be through mineralization. Thus, in principle, a measurement of unlabeled N in the crop at the time of harvest plus the net change in unlabeled inorganic N in the soil should provide a measure of net mineralization. This is an extension of the “zero N plot” method already discussed and it is
2 16
S. C. JARVIS ET AL.
arguable whether the data derived from a 15N experiment in this way are superior to those from the zero N plot method. The values obtained from the two methods very often differ, with the uptake of unlabeled N in 15N experiments often exceeding the total N uptake by a crop given no N fertilizer under the same environmental conditions (e.g., Powlson er al., 1986, 1992). This has been described as a priming effect, with the implication that the addition of fertilizer N stimulated the mineralization of native organic N. However, in most cases, the difference is likely to arise through pool substitution, in which labeled inorganic N substitutes for a part of the unlabeled inorganic N that would otherwise have been immobilized or lost through denitrification (Jenkinson er al., 1985). Although this results in a larger amount of unlabeled N being available for crop uptake, it is an artifact of the experimental procedure as discussed in detail by Hart et al. (1986). If 15N labeling is used to assess soil N supply through mineralization, the data must be interpreted with great caution. Data showing crop uptake of unlabeled N are usually obtained as a by-product of '5N experiments with other objectives, e.g., following the fate of fertilizer N in a specific system (e.g., Hart er al., 1986), to construct N balances in order to assess overall loss of fertilizer N (e.g., Powlson et al., 1986, 1992), or to obtain estimates of gross mineralization as described below (Section 1V.E).
E. MEASUREMENT OF GROSS MINERALIZATION The importance of understanding gross mineralization has already been indicated (Section 1.B). Technology has provided the ability to distinguish between stable isotopes of elements such as N with relative ease and has opened up opportunities to examine soil N processes with considerable precision. Monitoring changes in enriched sources of N added to soils allows a good deal of sensitivity, precision, and positive identification of labeled forms of N as they enter, are transformed in, or leave the system under study. There are limitations due to the biological internal recycling occurring in soil systems, and interpretation of 15N data thus needs considerable care. Two general approaches have been used. The first, noted above (Section 1V.D) involves N balance studies with enriched N added with fertilizer, for example, and with budgets for all the appropriate inputs and outputs. The degree of dilution of the enrichment within the system provides some estimate of input from other sources, i.e., through mineralization (Nason and Myrold, 1991). Second, 15N labeling has been used to determine specific rates of transformation (Myrold and Tiedje, 1986; Barraclough and Smith, 1987). Thus the rate that added enriched I5NH,+is diluted as unlabeled organic N is mineralized and is added to the soil NH,+ pool provides, over the short term, a very accurate measure of gross mineralization rate.
NITROGEN MINERALIZATION
217
By labeling with I5NH,+or 15NO, and examining soils after a relatively short period, rates of mineralization as indicated by the decline in 15N enrichment of NH,+, of nitrification by the decline in I5N enrichment of NO,, and of consumption/removal or immobilization of NH,+ and NO, by the disappearance of 15N label have all been determined on a limited range of soils (Barraclough, 1991; Barraclough and Smith, 1987; Myrold and Tiedje, 1986). However, there are a number of assumptions implicit in the use of these methods, i.e., (i) the processes occur at constant rates, or vary in a known way, between sampling times, (ii) there is no rapid internal cycling, and (iii) the processes exploit 15N and 14N in all available pools in proportion to their relative amounts. Development of stable isotope methodology for N, especially if linked to similar studies with carbon, can contribute much to our understanding of the complexity of soil mineralization processes. The problems of execution and interpretation are numerous. As well as the assumptions above, it is necessary to assume that I4N and 15N are not discriminated by soil processes and that any added 15N equilibrates rapidly with the pool to which it has been added. It is known that some microbial processes do discriminate between I4N and '5N (Heaton, 1986). The heterogeneity of soils, especially the spatial variability of NH,+ and NO,, presents a serious problem in obtaining, first, a uniform distribution (Barraclough, 1991) and, second, the development of in situ methods of isotope dilution, since rate estimates are calculated by differences which amplify errors (Myrold and Tiedje, 1986). 15NH,+injected or mixed with soils is assumed to be uniformly distributed but will inevitably enter macropores and be spatially separated from much of the background soil solution, exchangeable and fixed NHJ. Added '5NH,+ and native l4NH,f may then be subjected to different consumption rates leading to errors in calculation of gross mineralization. The use of pool dilution techniques, therefore, is not always straightforward. Continued, sensible use and reassessment of the procedures involved should, however, lead to an increased understanding of the supplies of mineral N from native soil sources. Increased definition of gross mineralization rates of recently returned organic materials as compared with "old" native SOM is an essential prerequisite to manipulating existing systems or devising novel management systems to control and utilize N from mineralization. Most information can be obtained where pool dilution experiments are coupled with direct tracer methods, where the fate of added 15N is determined at the end of the experiment (Powlson and Barraclough, 1993).
F. THEROLEOF MODELS IN PREDICTING MINERALIZATION Most models of N flows in cropping systems incorporate some consideration of mineralization but generally lack a mechanistic basis. Model design depends
218
S. C . JARVIS ET AL.
on intended use, how much detail is used to describe the subsystems, and the way in which they are influenced by environmental factors. Choice of an appropriate time step is also important, and controls the amount of data needed to run the model and the nature of the output. Specific processes are often excluded because there are insufficient reliable data available either to derive or to validate the relationships; the release of N from manures is a good example of this. Models intended for widespread use as management tools may be restricted in their design due to limited available data inputs or they may have to incorporate default data sets.
1. Practical Models One of the main problems is to decide at what level of detail to model the soil system: this will be dependent on the intended use of the model. Management models to be used in policy development or by farmers and advisors have to use inputs which are relatively easily available and which are likely to be on a field or hectare basis. The intricacies of the soil N cycle at the microsite scale then seem to be relatively unimportant, but most microsite processes, e.g., denitrification, which is dependent on local 0, and soluble C and NO, concentrations, are important at field scales (Section 1II.D). It is important to be able to estimate the effects of field scale management either by studying the literature or by running a model which operates at an appropriate scale to derive a simple relationship. The amount of detail included in models designed to be used as management tools varies enormously. However, the prediction of the amount of N released by mineralization by more complicated models has not given better results than simpler ones (de Willigen, 1991). Nitrogen management models are used to give immediate fertilizer recommendations, to determine the likely impact of management changes on farm profitability or environmental impact, and to allow development of good agricultural practice to minimize losses. These models should have mineralization as an important subsystem to determine the release of N from crop residues, manures, and SOM. The detail at which mineralization, and the other subsystems with which it interacts, is described will vary between models.
2. Research Models Research models try to integrate new understanding about soil processes and their interactions. These models may be very different from each other and may use entirely new approaches but have an important role in increasing our understanding of mineralization and its relationship with other components of the soil N cycle. Such models enable both synthesis of new information from experimental data and confirmation or rejection of hypotheses about how the system oper-
NITROGEN MINERALIZATION
2 19
ates, and highlight gaps in knowledge. Research models seem certain to become larger and larger as the importance of more and more subsystems is recognized. The development of a mathematical representation of soil porosity is crucial to allow the modeling of the interrelationships between soil structure, soil biota, and nutrient cycling (Juma, 1993). Changes in the output of models caused by the introduction of new subsystems may be very small: often the effect is to alter the pathways and partitioning of N within the SOM pools or to add new pools. However, future changes in the perception of microsite activities may influence the way the whole system responds to environmental perturbations and so should not be disregarded when a management model is being designed or reevaluated. If management models are to be accurate and scientifically credible they must take note of new concepts, derived from both modeling and experimental approaches, at the research level. New, and often complicating, factors can then be included explicitly or implicitly in a simplified way or excluded on the basis of a rational judgement. A different recent approach has been to model N mineralization from the trophic interactions among the groups of organisms constituting the soil food web which produces results close to in situ measurements (Ruiter et al., 1993).
V. THE IMPACT OF MINERALIZATION In this review we have shown that the effects of mineralization are condiderable and widespread. Current issues related to excesses of N in agricultural management being transferred to other systems demand a knowledge of, and an ability to predict, the flows of N from all sources. There is little doubt that supplies from native sources are substantial, but direct measurements of mineralization for most conditions are few and far between. It is essential to the development of useful models that the current database is extended to a wider range of soil types and agricultural managements. This could then be coupled with the existing information for soil and climate and used to provide more field-specific advice. The rates of transfer in grassland can be particularly large. There have been only a few studies which have measured either gross or net mineralization directly under field conditions. Geens et al. (1991) measured gross mineralization and immobilization rates under long term grass on a clay loam soil. In April, gross mineralization rates were 2-3 kg N ha-1 day-1 and gross immobilization rates were 1-5 kg N ha-1 day-', while in June mineralization rates had increased to 1-5 kg N ha-1 day-1 and immobilization rates were 1-3.5 kg N ha-1 day-'. Recently, three studies have determined net mineralization in grazed pastures through extended periods. Average daily rates of net release ranged from 0.2 to
S. C. JARVIS ET AL.
220
2.3 kg N ha-', depending on background sward management (Table IV). As well as the direct effects of the very large rates of N released and available for plant uptake or loss, it is of interest to note that in the study of Gill el al. (1995) an average of 85 kg N ha-1 was released during the period November to February, i.e., when opportunities to be leached were at their greatest in the UK. Substantial winter mineralization rates have also been observed in grassland soils in the Netherlands (Olff et al., 1994). Disturbance of grassland can have substantial impact. Older swards generally mineralize more N after disturbance than short-term leys, because of greater N accumulation, but there is always a declining release of N which continues for decades. Young (1986) calculated that N mineralization increased from about
Table IV Net Mineralization Rates in Grassland Soils Fertilizer N management (kg ha-I) Site
Soil type
Past
Present
N mineralization Measurement period
kg ha-'
kg ha-' day-'
415 321 310
2.3 1.8 1.7
314
0.9
SE Englanda
Sandy loam
0 420 420
0 0 420
SW Englandb
Silty clay loam, drained Silty clay loam, drained Silty clay loam, undrained Silty clay loam, undrained Silty clay loam, drained Silty clay loam, undrained Silty clay loam, undrained Silty clay loam, undrained Silty clay loam, undrained
400
400
April-Oct April-Oct April-Oct 12 months
0
0
12 months
162
0.4
400
400
12 months
173
0.5
0
0
12 months
65
0.2
200
200
12 months
376
1 .o
200
200
12 months
317
0.9
200
0
12 months
292
0.8
0
0
12 months
135
0.4
0
200
12 months
270
0.7
SW EnglandC
a Hatch et al. (1991).
Blantern (1991). Gill et al. (1995).
NITROGEN MINERALIZATION
22 1
100 kg N ha-' after ploughing a I-year ley to 280-380 kg N ha-1 for leys older than 4 years. Whitehead et al. (1990) measured N accumulation in stubble, leaf litter, roots, and soil macro-organic matter, which was 536 kg N ha-1 for an 8-year ryegrass sward and 602 kg N ha-' for a 15-year-old sward. They also demonstrated that management, in particular N inputs and grazing, influenced N accumulation and, hence, mineralization after ploughing. Williams and Clement (1965) had previously shown this with incubation studies. Shepherd (1993) found that the use of soil mineral N sampling to predict N response of wheat after ploughing grass swards was of no value because subsequent immobilization/mineralization processes were too variable. Manure application appeared to increase N mineralization potential on ploughing out grass (Withers, 1988). Increased stocking rates, stimulated by increased fertilizer use, had a large effect on potential N mineralization (Lloyd, 1992). N mineralization from fields cut for silage has been reported as being negligible (e.g., Clement, 1975; Withers, 1988) but depended very much on the previous N balance of the sward. Large amounts of manure or slurry are often applied during the winter before ploughing out in the UK and this will further affect mineralization. Clover-based swards may be a special case (Section 1I.E) and have a specific effect in increasing mineralization on ploughing out grassland. This may become more important if increased symbiotic N fixation within farming systems results from the search for increased sustainability. In general, although the amount and timing of N release on ploughing grassland are known qualitatively, the relationships need further quantification. In arable systems, uptake of N from soil and plant residue sources forms a significant part of total crop N uptake. In experiments using 15N fertilizers to measure fertilizer use efficiency, unlabeled N uptake assumed to be that from the soil has been shown to be significantly different between soils, even where fertilizer N uptakes are similar (McTaggart and Smith, 1993). This may be due to the effect of previous crop residues: increased uptake of soil N occurred where the residues of the brussel sprout crop were involved (Table V). However, at other sites where cereals were the previous crop, significant differences in soil N uptake were seen. In most, but not all, cases, uptake of soil N increased as the SOM content increased. Understanding rates of supply from N catch crops is an important issue in relation to potential leaching (Thorup-Kristensen, 1993) but little is known about about long-term effects of this. Understanding N mineralization from manures is particularly important to provide better information about short-term N supply, but there are important longer term implications because of contributions to fertilizer value in the following years [discussed elsewhere by Shepherd et al. (1996)l and the need to avoid increased leaching risk with time. Thus the mineralizable N pool can be increased over the long term by frequent additions. For example, N uptake by plots receiving no mineral fertilizer has been shown to vary annually because of
222
S. C. JARVIS ET AL. Table V Nitrogen Uptake (kg ha-1) of Spring Barley Grown on Six Sites in Southern Scotland in 1990 Divided into That Derived from Fertilizer and That from the Soil (McTaggart and Smith, 1993) Nitrogen uptake (kg N ha-’) Previous crop
Soil texture
OM 9%
From fertilizer
From soil
Cereals Brussels sprouts Cereals Cereals Cereals Cereals
Sandy loam Loamy sand Sandy clay loam Sandy loam Clay loam Sandy loam
2.4 2.8 3.3 4.7 5.1 5.7
63.9 73.3 51.0 70.0 46.4 58.5
47.2 122.8 49.1 71.3 18.6 41.1
weather differences, but the variation was greater where the amount of mineralizable substrate was larger, in plots treated annually with farmyard manure. Similarly, the application of fertilizer N for many years increased mineralizable soil N in many long- and medium-term experiments (Glendining and Powlson, 1995) by increasing the amount of crop residues returned to the soil. The increases were generally substantial, i.e., 20% or more compared with the control plots and tended to be greater than any increases in total soil N. Residues returned and manures added to the soil are particularly important by acting as a source of generally readily mineralized N for a following crop, and also as a major source of N lost from the system. Optimization of when and how a residue or manure is returned to the soil is important for efficient use of the resource. Although there is much qualitative information about these processes, this will require a transformation of this knowledge and of the controlling factors to quantitative principles which can be used to improve the construction of predictive models and, thus, advice. The current, major practical objective of obtaining information on mineralization is to provide the basis on which to achieve increased efficiency/reduced environmental impact in agricultural systems. Inevitably, attempts to do this have used methods which integrate the effects of many interactive processes and controlling factors by measuring, for example, changes in mineral N. The detail with which the N cycle processes are viewed decrease as the hierarchy of scale increases from microsite through soil profile, field to catchment, regional and global levels, with the primary difference between scales being the complicating effect of spatial variability (Fig.3). Spatial variability of the processes and their products occurs both laterally and with depth and makes it difficult to calculate an appropriate field “average” from data collected at individual points (Biggar,
NITROGEN MINERALIZATION
223
Figure 3 Mineralization can be considered at a range of scales: shown here are catchment, soil profile, and microsite. The detail at which the processes are seen increases as the hierarchy descends from landscape to farm, to soil profile toward the microsite scale.
1978). This problem arises whenever there is a need to transfer information between any two levels in the hierarchy. Relationships between different levels are difficult, but important, to establish, so that the results of research can be put into practical applications. From the outset, the success of transposing information from one scale to another will depend upon (i) better process definition and
224
S. C. JARVIS ET AL.
(ii) improved methods of making estimates of their effects. Predictive or advisory tools that aim to involve mineralization will be of more functional use if based on an improved understanding of the distributions of biological activities in relation to microsites and soil mineral N pools is obtained (Ladd et al., 1993).
VI. CONCLUSIONS AND FUTURE PROGRESS The overall effect of all the processes involved in the net release of available N has implications for all ecosystems, whether managed or natural, and at a range of scales from microsite to global. It is clear from this review that while much information exists about the process generally known as mineralization, we are a long way from being able to either maximize its potential or reduce its impact. The very basis of being able to do this will be an increased understanding at the process level and an improved ability to measure the results of those processes. The following points summarize the current status of the problems involved and indicate the way in which we consider that progress can be made. Better quantification of the role of mineralization would (a) help minimize losses of N to the environment, (b) enable better fertilizer advice, and (c) improve management of N within the crop-soil system. While there is some basis for this, detailed information is restricted to a limited number of situations. At global, national, regional, and farm scales, changes in policies and agricultural management for which there is no current advice will demand greater knowledge and predictive capacity for mineralization. Increased efficiency of N use will demand that a full account is taken of the contribution of mineralization and immobilization to N budgets at field, farm and catchment scales. This will require not only a synthesis of current available knowledge, but in many instances the development of new databases for both short- and long-term effects. There will be an urgent need for more data from new managements such as set-aside, for new crops and ley-arable farming systems as practices respond to political, environmental, and socioeconomic pressures. There is a need to (a) examine the impact of management practices, e.g., cultivation and residue distribution, on soil properties at the microsite level, (b) determine the influence of these changes on biological activity, and (c) determine the extent to which variability at the microsite level influences the flows of N at a field scale. All of these will depend upon an appreciation of the new concepts of the way in which soil architecture interacts with spatial distributions of biological activities and specific N pools and a knowledge of how diffusional constraints affect the processes in widely heterogeneous microsites. The new techniques
NITROGEN MINERALIZATION
225
available for measuring gross rates of N mineralization will provide a foundation on which to base the principles for more applied considerations. Mineralization controls much of the NO, losses by leaching. However, the contribution of mineralization to other losses, e.g., denitrification, need to be quantified (Seitzinger, 1994). Other specific areas requiring attention include the role of nitrification as a limitation to leaching and denitrification, and the contribution of mineralization of SOM and leached organic N at depth to NO, leaching into subsurface aquifers. It should be possible to reexamine crop N uptake in past N response trials to provide immediate information on N supplies from native sources for the most common crops. This could then be modified by any future improvements in direct methodology or approaches to mineralization estimates. Detailed experiments are required to measure gross and net mineralization for representative sites across a range of crops, climates, and soils. A comprehensive understanding of N flows would enable the principles developed to be applied with appropriate models to a full range of cropslsoils. The identification of pools within both SOM and added residues which can be readily separated from soil and analyzed and which have biological relevance will be important. The quantification of C and N in these pools and their interactions will enable better understanding of mineralization/immobilizationand more precise definition of resource quality in relation to the amounts and timing of N released. It will also provide more appropriate definition of pools within models and greater possibilities for testing model simulations. At the same time, the development of field methods for measuring net mineralization is important and requires the selection of the most appropriate technique to produce robust information. In the first instance, there will be a need to reexamine whether a simple test (e.g., soil extraction or incubation) is a feasible way to predict N mineralization. Even if this approach is not universally applicable it may still be of value in a significant number of situations. Data will also be required which define the factors controlling soil N supply including the effect of season and long-term management history. Finally, the long-term nature of the process and its effects should be emphasized. The full impact on mineralization of, for example, a change in agricultural practice may be apparent only after some years and the longer-term results may be very different from those obtained immediately.
ACKNOWLEDGMENTS We are grateful to the Ministry of Agriculture, Fisheries and Food, London, for providing funding for the preparation of this review.
226
S. C . JARVIS ET AL.
REFERENCES Adams, M. A., and Attiwill, P. M. (1986). Nutrient cycling and nitrogen mineralization in eucalypt forests of south-eastem Australia. 11. Indices of nitrogen mineralization. Plant Soil 92, 341362. Addiscott, T. M. ( I 983). Kinetics and temperature relationships of mineralization and nitrification in Rothamsted soils with differing histories. J . Soil Sci. 34, 343-353. Alves, B. J. R., Vrquiaga, S., Cadisch, G., Souto, C. M., and Boddey, R. M. (1993). In situ estimation of soil nitrogen mineralization. In “Soil Organic Matter Dynamics and the Sustainability of Tropical Agriculture” (K. Mulongoy and R. Merckx, Eds.), pp. 173-180. Wiley, Chichester. Anderson and Domsch (1990). Carbon link between microbial biomass and soil organic matter. In “Perspectives in Microbial Ecology Proceedings IV,” F. Megusar and M. Gantar, Eds.), pp. 467-47 1. Slovenic Society for Microbiology, Ljubljana. Anderson, J. M. (1988). The role of soil fauna in agricultural systems. In “Advances in Nitrogen Cycling in Agricultural Ecosytems” (J. R. Wilson, Ed.), pp. 89- I 12. CAB International, Wallingford. Appel, T., and Mengel, K. (1990). Importance of organic nitrogen fractions in sandy soils, obtained by electro-ultrafiltration or CaCI, extraction for nitrogen mineralization and nitrogen uptake of rape. Biol. Ferril. Soils 10, 97-101. Appel, T., and Mengel, K. (1993). Nitrogen fractions in sandy soils in relation to plant nitrogen uptake and organic matter incorporation. Soil Biol. Biochem. 25, 685-691. Azam, F., Malik, K. A,, and Hussain, F. (1986). Microbial biomass and mineralizationimmobilization of nitrogen in some agricultural soils. Biol. Fertil. Soils 2, 157-163. Ballesdent, J., Maniotti, A,, and Boisgontier, D. (1990). Effect of tillage on soil organic carbon mineralization estimated from I3C abundance in maize fields. J. Soil Sci. 41, 587-596. Barak, P., Molina, J. A. E.. Hadas, A,, and Clapp, C. E. (1990). Mineralization of amino acids and evidence of direct assimilation of organic nitrogen. Soil Sci. SOC. Am. J. 54, 769-774. Barraclough, D. (1991). The use of mean pool abundances to interpret I5N tracer experiments I. Theory. Plant Soil 131, 89-96. Barraclough, D., and Smith, M. J. (1987). The estimation of mineralization, immobilization and nitrification in nitrogen-I5 field experiments using computer simulation. J . Soil Sci. 38, 519530. Beauchamp, E. G. (1986). Availability of nitrogen from three manures to corn in the field. Cannd. J. Soil Sci. 66, 713-720. Beauchamp, E. G., Reynolds, W. D., Brasche-Villeneuve, D., and Kirby, K. (1986). Nitrogen mineralization kinetics with different soil pretreatments and cropping histories. Soil Sci. SOC. Am. J. 50, 1478-1483. Bemal, P., and Kirchmann, H. (1992). Carbon and nitrogen mineralization and ammonia volatilization from fresh, anaerobically treated pig manure during incubation with soil. Biol. Fertil. Soils 13, 135-141. Biggar, J. W. (1978). Spatial variability of nitrogen in soils. In “Nitrogen in the Environment” Vol. 1 (D. R. Nielsen and J. G. MacDonald, Eds.), pp. 201-211. Academic Press, New York. Bjamason, S. (1987). Immobilization and remineralization of ammonium and nitrate after addition of different energy sources to soil. Plant Soil 97, 381-389. Blantem, P. (1991). Factors affecting nitrogen transformations in grazed grassland soils with specific reference to the effects of artificial land drainage and N fertilizer. PhD Thesis, University of Exeter. Bonde, T. A., and Lindberg, T. (1988). Nitrogen mineralization kinetics in soil during long-term aerobic laboratory incubations: A case study. J. Environ. Qual. 17, 414-417.
NITROGEN MINERALIZATION
227
Bosatta, E., and Agren, G. I. (1985). Theoretical analysis of decomposition of heterogeneous substrates. Soil Biol. Biochem. 17, 601-610. Bouwman, L. A., Bloem, J., Boogert, van den P. H. J. F., Bremer, F., Hoenderboom, G. H. J., and Ruiter, de P. C. (1994). Short-term and long-term effects of bacteriavorous nematodes and nematophagus fungi on carbon and nitrogen mineralization in microcosms. Biol. Fertil. Soils 17, 249-256. Bradbury, N. J., Whitmore, A. P., Hart, P. B. S., and Jenkinson, D. S . (1993). Modelling the fate of nitrogen in crop and soil in the years following the application of 15N-labelled fertilizer to winter wheat. J. Agric. Sci. 121, 363-379. Bramley, R. G. V., and White, R. E. (1991). An analysis of variability in the activity of nitrifiers in a soil under pasture. 1. Spatially dependent variability and optimum sampling strategy. Aust. J. Soil Res. 29, 95-108. Bristow A. W., and Jarvis S. C. (1991). Effects of grazing and nitrogen fertilizer on the soil microbial biomass under permanent pasture. J. Sci. Food Agric. 54, 9-2 I , Broadbent, F. E. (1986). Empirical modelling of soil nitrogen mineralization. Soil Sci. 141, 208213. Brookes, P .C., Powlson, D. S., and Jenkinson D. S. (1984). The microbial biomass in soil. In “Ecological Interactions in Soil, Plants, Microbes and Animals” (A. H. Fitter, Ed.), pp. 123125. Publication No. 4, British Ecological Society. Cabrera, M. L. (1993). Modelling the flush of nitrogen mineralization caused by drying out and rewetting soils. Soil Sci. Soc. Am. J . 57, 63-66. Cabrera, M. L., and Kissel, D. E. (1988). Evaluation of a method to predict nitrogen mineralized from soil organic matter under field conditions. Soil Sci. SOC. Am. J . 52, 1027-1031. Campbell, C. A., Jame, Y.W., and DeJong, R. (1988). Predicting net nitrogen mineralization over a growing season: Model verification. Canad. J. Soil Sci. 68, 537-552. Cassman, K. G . , and Munns, D. N. (1980). Nitrogen mineralization as affected by soil moisture, temperature and depth. Soil Sci. Soc. Am. J. 44, 1233-1237. Castellanos, J. Z., and Pratt, P. F. (1981). Mineralization of manure nitrogen-Correlation with laboratory indexes. Soil Sri. SOC.A m . J . 45, 354-357. Chae, Y. M., and Tabatabai, M. A. (1986). Mineralization of soils in soils amended with organic wastes. J. Enwiron. Qual. 15, 193- 198. Chang, F. H., and Broadbent, F. E. (1982). Influence of trace metals on some soil nitrogen transformations. J. Enwiron. Qual. 11, 1-4. Chaussod, R., Houot, S., Guiraud, G., and Hetier, J. M. (1988). Size and turnover of the microbial biomass in agricultural soils: Laboratory and field measurements. In “Nitrogen Efficiency in Agricultural Soils” (D. S. Jenkinson and K. A . Smith, Eds.), pp. 312-326. Elsevier, London. Chescheir, G. M. 111, Westermann, P. W., and Safley L. M . , Jr. (1986). Laboratory methods for estimating available nitrogen in manures and sludges. Agric. Wastes 18, 175- 195. Christian, D. G., Goodlass, G., and Powlson, D. S. (1992). Nitrogen uptake by cover crops. Aspects Appl. Biol. 30, 291-300. Clement, C. R. (1975). Grassland management, soil conditions and crop yield: The ley fertility experiments at Hurley. In “Soil Physical Conditions and Crop Production,” pp. 388-405. MAFF Technical Bulletin 29. HMSO, London. Cochran, V. L., Horton, K. A,, and Cole, C. V. (1988). An estimation of microbial death rate and limitations of N or C during wheat straw decomposition. Soil Biol. Biochem. 20, 293-298. Curry, J. P., and Byrne, D. (1992). The role of earthworms in straw decomposition and nitrogen turnover in arable land in Ireland. Soil Biol. Biochern. 24, 1409-1412. Dalal, R. C., and Mayer, R. J. (1986). Long-term trends in fertility of soils under continuous cultivation and cereal cropping in Southern Queensland. 111. Distribution and kinetics of soil organic carbon in particle-size fractions. Aust. J. Soil Res. 24, 293-300.
228
S. C. JARVIS ET AL.
Darrah, P. R., Nye, P. H., and White, R. E. (1983). Diffusion of NHZ and NO, mineralized from organic N in soil. J. Soil Sci. 34, 693-707. Davidson, E. A,, Stark, J. M., and Firestone, M. K. (1990). Microbial production and consumption of nitrate in an annual grassland. Ecology 71, 1968-1975. Davidson, E. A,, and Hackler, J. L. (1994). Soil heterogeneity can mask the effects of ammonium availability on nitrification. Soil Eiol. Eiochem. 26, 1449- 1453. Deans, J. R., Molina, J. A. E., and Clapp, C. E. (1986). Models for predicting potentially mineralizable N and decomposition rate constants. Soil Sci. SOC. Am. J . 50, 323-326. Dexter, A. R. (1988). Advances in the characterisation of soil structure. Soil Tillage Res. 11, 199238. Dickinson, C. H. (1974). Decomposition of litter in soil. In “Biology of Plant Litter Decompsition,” Vol. I1 (C.H. Dickinson and G. J. F. Pugh, Eds.), pp. 633-658. Academic Press, London. DiStefano, J. F., and Gholz, H. L. (1986). A proposed use of ion exchange resins to measure nitrogen mineralization and nitrification in intact soil cores. Commun.Soil Sci. Plant Anal. 17, 989-998. Douglas, B. F., and Magdoff, F. R. (1991). An evaluation of nitrogen mineralization indices for organic residues. J. Environ. Quol. 20, 368-372. Drury, C. F., Voroney, R. P., and Beauchamp, E. G. (1991). Availability of NHt-N to microorganisms and the soil internal N cycle. Soil Eiol. Eiochern. 23, 165-169. Elliot, P. W., Knight, D., and Anderson, J. M. (1990). Denitrification in earthworm casts and soil under different fertilizer and drainage regimes. Soil Eiol. Eiochem. 22, 601-605. Flowers, T. H., and Arnold P. W. (1983). Immobilization and mineralization of nitrogen in soils incubated with pig slurry or ammonium sulphate. Soil Eiol. Eiochem. 15, 329-335. Focht, D. D. (1992). Diffusional constraints on microbial processes in soil. Soil Sci. 154, 300-307. Fox, R. H., Myers, R. J. K., and Vallis, I. (1990). The nitrogen mineralization rate of legume residues in soil as influenced by their polyphenol, lignin and nitrogen contents. Plant Soil 129, 25 1-259. Francis, G. S., Haynes, R. J., Sparling, G. P., Ross, D. J., and Williams, P. H. (1992). Nitrogen mineralization, nitrate leaching and crop growth following cultivation of a temporary leguminous pasture in autumn and winter. Fertilizer Res. 33, 59-70. Frankenberger, W. T., and Abdelmagid, H. M. (1985). Kinetic parameters of nitrogen mineralization rates of leguminous crops incorporated into soil. Planr Soil 87, 257-271. Freytag, H. E., Ramsch, H., and Liittich, M. (1989). Determination of nitrogen mineralization potentials from nitrogen contents of soils. Archiv. Acker Pflanzenbau Eodenkunde 33,741-746. Gaiser, T., and Stahr, K. (1994). The significance of the light fraction of soil organic matter for nitrogen mineralization and the yield potential of a ferralic-haplic Acrisol. Mitreilungen Deurschen Eodenkundlichen Gesellschaft 71, 243-246. Ganvood, E. A. (1967). “Studies on the Roots of Grasses,” pp. 72-79. Annual Report 1966, Grassland Research Institute, Hurley. Geens, E. L., Davies, G. P.,Maggs, J. M., and Barraclough, D. (1991). The use of mean pool abundances to interpret 15N tracer experiments. I1 Application. Plant Soil 131, 97-105. Gianello, C., and Bremner, J. M. (1986). A simple chemical extraction method for assessing potentially available organic nitrogen in soil. Commun. Soil Sci. Plant Anal. 17, 195-214. Gill, K., Jarvis, S. C., and Hatch, D. J. (1995). Mineralization of nitrogen in long-term pasture soils: Effects of management. Plant Soil,A2, 153-162. Glendining, M. J., and Powlson, D. S. (1995). The effects of long continued applications of inorganic nitrogen fertilizer on soil organic nitrogen-A review. Adv. in Soil Sci. Soil Management: Experimental Basis for Sustainability and Environmental Quality. pp.385-446. IACR. Rothamsted. Goovaerts, P.,and Chiang, C. N. (1993). Temporal persistence of spatial patterns for mineralizable nitrogen and selected soil properties. Soil Sci. Soc. Am. J . 57, 372-381.
NITROGEN MINERALIZATION
229
Gordon, A. M., Tallas, M., and van Cleve, K. (1987). Soil incubations in polyethylene bags: Effect of bag thickness and temperature on nitrogen transformations and CO, permeability. Canad. J . Soil Sci. 67, 65-76. Goss, M. K., Howse, K. R., Lane, P. W., Christian, D. G., and Harris, G. L., (1993). Losses of nitrate nitrogen in water draining from under autumn-sown crops established by direct drilling or mouldboard ploughing. J. Soil Sci. 44, 35-48. Grant, R. F., Juma, N. G., and McGill, W. B. (1993). Simulation ofcarbon and nitrogen transformations in soil: Microbial biomass and metabolic products. Soil Biol. Biochem. 25, 1331-1338. Gregorich, E. G., Voroney, R. P., and Kachanoski, R. G. (1991). ’Ibrnover of carbon through the microbial biomass in soils with different textures. Soil Biol. Biochem. 23, 799-805. Griffiths, B. (1994). Microbial-feeding nematodes and protozoa in soil: Their effects on microbial activity and nitrogen mineralization in decomposition hotspots and the rhizosphere. Planr Soil 164, 25-34. Griffiths, B., and Robinson, D. (1992). Root-induced nitrogen mineralization: A nitrogen balance model. Plant Soil 139, 253-263. Hadas, A., Feigenbaum, S., Feigin, A,, and Portnoy, R. (1986). Distribution of nitrogen forms and availability indices in profiles of differently managed soil types. Soil Sci. SOC. Am. J . 50, 308313. Hadas, A., Molina, J. A. E., Feigenbaum, S., and Clapp, C. E. (1987). Simulation of nitrogen-I5 immobilization by the model NCSOIL. Soil Sci. SOC. Am. J . 51, 102-106. Hadas, A., Sofer, M., Molina, J. A. E., Barak, P., and Clapp, C. E. (1992). Assimilation of nitrogen by the soil microbial population: NH, versus organic N. Soil Biol. Biochem. 24, 137- 143. Hansen, S., Jensen, H. E., Nielsen, N. E., and Svendsen, H. (1991). Simulation of the nitrogen dynamics and biomass production in winter wheat using the Danish simulation model DAISY. Fertilizer Res. 27, 245-259. Hart, P. B. S., Rayner, J. H., and Jenkinson, D. S. (1986). Influence of pool substitution on the interpretation of fertilizer experiments with ‘5N. J. Soil Sci. 37, 389-403. Hassink, J. (1992). Effects of soil texture and structure on carbon and nitrogen mineralization in grassland soils. Biol. Fertil. Soils 14, 126-134. Hassink, J. (1993). Relationship between the amount and activity of the microbial biomass in Dutch grassland soils: Comparison of the fumigation-incubation method and the substrate-induced respiration method. Soil Biol. Biochem. 25, 533-538. Hassink, J. (1994). Effects of soil texture and grassland management on soil organic C and N and rates of C and N mineralization. Soil Biol. Biochem. 26, 1221-1231. Hassink, J., Scholefield, D., and Blantern, P. J. (1990). Nitrogen mineralization in grassland soils. In “Proceedings of the 13th Meeting of the European Grassland Federation,” Vol. 2, pp. 25-32. Banska Bystrica, Czechoslovakia. Hassink, J., Bouwman, L. A., Zwart, K. B., and Brussaard, L. (1993). Relationships between habitable pore space, soil biota and mineralization rates in grassland soils. Soil Biol. Biochem. 25, 47-55. Hatch, D. J., Jarvis, S. C., and Philipps, L. (1990). Field measurement of nitrogen mineralization using soil core incubation and acetylene inhibition of nitrification. Plant Soil 124, 97-107. Hatch, D. J., Jarvis, S. C., and Reynolds, S. E. (1991). An assessment of the contribution of net mineralization to N cycling in grass swards using a field incubation method. Plant Soil 138, 23-32. Haynes, R. J. (1986). The decomposition process: Mineralization, immobilization, humus production and degradation. In “Mineral Nitrogen in the Soil-Plant System” (R. J. Haynes, Ed.), pp. 52-186. Academic Press, London. Heaton, T. H. E. (1986). Isotopic studies of nitrogen pollution in the hydrosphere and the atmosphere: A review. Chem. Geol. 59, 87-102.
230
S. C. JARVIS ET AL.
Holmes, W. E. (1994). Soil microbial biomass dynamics and net mineralization in northern hardwood ecosystems. Soil Sci. Soc. Am. J. 58, 238-243. Hsieh, Y-P. (1992). Pool size and mean age of stable soil organic carbon in cropland. Soil Sci. SOC. Am. J . 56,460-464. Huntington, T. G., Grove, J. H., and Frye, W. W. (1985). Release and recovery of nitrogen from winter annual cover crops in no-till corn production. Commun. Soil Sci. PIanr Anal. 16, 193211.
Jansson, S. L. (1958). Tracer studies on nitrogen transformations in soils with special attention to mineralization-immobilization relationships. Kongliga Landbrakshogskolans Annaler 24, 101361.
Jansson, S. L., and Persson, J. (1982). Mineralization and immobilization of soil nitrogen. In “Nitrogen in Agricultural Soils” (F. J. Stevenson, Ed.), pp. 229-252. ASA, CSSA, SSSA, Madison. Janzen, H. H. (1987). Soil organic matter characteristics after long-term cropping to various spring wheat rotations. Canad. J . Soil Sci. 67, 845-856. Jarvis, S. C. (1992). Grazed grassland management and nitrogen losses: An overview. Aspects Appl. Biol. 30, 207-214. Jarvis, S. C., Scholefield, D., and Pain, B. F. (1995). Nitrogen cycling in grazing systems. In “Nitrogen Fertilization in the Environment” (P. E. Bacon, Ed.), pp. 381-419. Marcel Dekker, New York. Jarvis, S. C., and Barraclough, D. (1991). Variation in mineral nitrogen content under grazed grassland swards. Plant Soil 138, 177-188. Jenkinson, D. S . (1981). The fate of plant and animal residues in soils. In “The Chemistry of Soil Processes” (D. J. Greenland and M. H. B. Hayes, Eds.), pp. 505-562. Wiley, Chichester. Jenkinson, D. S. (1984). The supply of nitrogen from the soil. In “The Nitrogen Requirement of Cereals,” pp. 79-93. Reference Book No. 385. ADAS, Ministry of Agriculture, Fisheries and Food, London. Jenkinson, D. S. (1990). The turnover of organic carbon and nitrogen in soil. Philos. Trans. R. Soc. London B 329, 361-368. Jenkinson, D. S., and Ladd, J. N. (1981). Microbial biomass in soil: Measurement and turnover. In ‘‘Soil Biochemistry” (E. A. Paul and J. N. Ladd, Eds.). Marcel Dekker, New York. Jenkinson, D. S., Fox, R. H., and Rayner, J. H. (1985). Interactions between fertilizer nitrogen and soil nitrogen-The so-called “priming” effect. J. Soil Sci. 36, 425-444. Jenkinson, D. S., Hart, P. B. S., Rayner, J. H., and Parry, L. C. (1987). Modelling the turnover of organic matter in long-term experiments at Rothamsted. INTECOL Bull. 15, 1-8. Jenny, H. (1941). “Factors of Soil Formation.” McGraw-Hill, New York. Jensen, H. L. (1931). The microbiology of farmyard manure decomposition insoil. I Changes in the microflora and their relation to nitrification. J. Agric. Sci. Camb. 21, 38-80. Juma, N. G. (1993). Interrelationships between soil structure/texture, soil biota/soil organic matter and crop production. Geoderma 57, 3-30. Kandeler, E., Eder, G., and Sobotik, M. (1994). Microbial biomass, N mineralization and the activities of various enzymes in relation to nitrite leaching and root distribution in a slurry amended grassland. Biol. Fertil. Soils 18, 7- 12. Keeney, D. R. (1982). Nitrogen availability indices. In “Methods of Soil Analysis,” Part 2 (A. L. Page, R. H. Miller, and D. R. Keeney, Eds.), pp. 71 1-734. ASA, Madison. Killham, K., Amato, M., and Ladd, J. N. (1993). Effect of substrate location in soil and soil porewater regime on carbon turnover. Soil B i d . Biochem. 25, 57-62. Kirchmann, H. (1985). Losses, plant uptake and utilisation of manure nitrogen during a production cycle. Acta Agric. Scand. Suppl. 24.
NITROGEN MINERALIZATION
23 1
Kladivko, E. J., and Keeney, D. R. (1987). Soil nitrogen mineralization as affected by water and temperature interactions. Biol. Fertil. Soils 5, 248-252. Kuikman, P. J., and van Veen, J. A. (1989). The impact of protozoa on the availability of bacterial nitrogen to plants. Biol. Fertil. Soils 8, 13-18. Ladd, J. N., and Foster, R. C. (1988). Role of soil microflora in nitrogen turnover. In “Advances in Nitrogen Cycling in Agricultural Ecosystems” (J. R. Wilson, Ed.), pp. 113-133. CAB International, Wallingford. Ladd, J . N., Foster, R. C., and Skjemstad, J. 0. (1993). Soil structure: Carbon and nitrogen metabolism. Geoderma 56, 401-434. Lampkin, N. (1990).“0rganic Farming,” pp. 89- 106. Farming Press, Ipswich. Landi, L., Badalucco, L., Pomare, F., and Nannipieri, P. (1993). Effectiveness of antibiotics to distinguish the contributions of fungi and bacteria to net nitrogen mineralization, nitrification and respiration. Soil Biol. Biochem. 25, 1771-1778. Linden, B., Lyngstrad, I . , Sippola, J., Nielsen, J.D., Soegaard, K., and Kjellerup, V. (1993). Evaluation of the ability of three laboratory methods to estimate net nitrogen mineralization during the growing season. Swedish J. Agric. Res. 23, 161-170. Lloyd, A. (1992). Nitrate leaching following the break-up of grassland for arable cropping. Aspects Appl. Biol. 30, 243-248. Lober, R. W., and Reeder, J. D. (1993). Modified waterlogged incubation method for assessing mineralization in soils and soil aggregates. Soil Sci. Soc. Am. J. 57, 400-403. Lohnis, F. (1910). “Handbuch der landwirtschaftlichen Bakteriologie.” Gebruder Borntraeger, Berlin. Lovell, R. D., Jarvis, S. C., and Bardgett, R. D. (1995). Soil microbial biomass and activity in longterm grassland: Effects of management changes. Soil Biol. Biochem. 27, 969-975. Macduff, J. H., and White, R. E. (1984). Components of the nitrogen cycle measured for cropped and grassland soil-plant systems. Plant Soil 76, 35-47. Macduff, J. H., and White, R. E. (1985). Net mineralization and nitrification rates in a clay soil measured and predicted in permanent grassland from soil temperature and moisture content. Plant Soil 86, I5 1- 172. Marstorp, H., and Kirchmann, H. (1991). Carbon and nitrogen mineralization and crop uptake of nitrogen from six green manure legumes decomposing in soil. Acta Agric. Scand. 41, 243252. Marumoto, T., Anderson, J. P. E., and Domsch, K. H. (1982a). Decomposition of I4C and 15Nlabelled microbial cells in soil. Soil Biol. Biochem. 14, 461-467. Marumoto, T., Anderson, I. P. E., and Domsch, K. H. (l982b). Mineralization of nutrients from soil microbial biomass. Soil Biol. Biochem. 14, 469-475. Matus, F. H., and Rodriquez, J. (1994). A simple model for estimating the contribution of nitrogen mineralization to the nitrogen supply of crops from a stabilized pool of soil organic matter and recent organic input. P lunt Soil 162, 259-27 1. McTaggart, I. P., and Smith, K. A. (1993). Estimation of potentially mineralizable nitrogen in soil by KCI extraction 11. Comparison with plant uptake in the field. Plant Soil 157, 175-184. Meisinger, J. J. (1982). Evaluating plant available nitrogen in soil-crop systems. In “Nitrogen in Crop Production” (R. D. Hauck, Ed.), pp. 391-416. ASA, Madison. Mengel, K., and Kirkby, E. A. (1978). “Principles of Plant Nutrition.” International Potash Institute, Switzerland. Mengel, K. (1991). Available nitrogen in soils and its determination by the ‘N-Min’ method and by electroultrafiltration (EUF). Fertilizer Res. 28, 251-262. Myrold, D. D., and Tiedje, J. M. (1986). Simultaneous estimation of several nitrogen cycle rates using ISN: Theory and application. Soil Biol. Biochem. 18, 559-569.
S. C. JARVIS ET AL.
232
Mytton, L. R., Cresswell, A , , and Colbourn, P. (1993). Improvement in soil structure associated with white clover. Grass Forage Sci. 48, 84-90. Nadelhoffer, K. J., Aber, J. D., and Melillo, J. M. (1985). Fine roots, net primary production and soil nitrogen availability: A new hypothesis. Ecology 66, 1377- 1390. Nason, G. E., and Myrold, D. D. (1991). l5N in soil research: Appropriate application of rate estimation procedures. Agric. Ecosystems Environ. 34, 427-44 I . Nordmeyer, H., and Richter, J. (1985). Incubation experiments on nitrogen mineralization in loess and sandy soils. Plant Soil 83, 433-445. Ocio, J. A,, Martinez, I., and Brooks, P. C. (1991). Contribution of straw derived N following incorporation of cereal straw to soil. Soil Biol. Biochem. 23, 655-659. Olff, H., Berendse, F., and de Vissier, W. (1994). Changes in the impact in nitrogen mineralization, tissue nutrient concentrations and biomass compartmentation after cessation of fertilizer application to mown grassland. J . Ecol. 82, 61 1-620. Pain, B. F., Misselbrook, T.A,, and Rees, Y. T. (1993). Effect of nitrification inhibitor and acid addition on nitrogen losses and herbage yield. Grass Forage Sci. 49, 209-215. Parker, C. F., and Sommers, L. E. (1983). Mineralization of nitrogen in sewage sludges. J . Environ. Qual. 12, 150-156. Parsons, A. J., Orr, R. J., Penning, P. D., and Lockyer, D. R. (1991). Uptake, cycling and fate of nitrogen in grass clover swards continuously grazed by sheep. J. Agric. Sci. Camb. 116,47-61. Parton, W. J., Schimel, D. S., Cole, C. V., and Ojima, D. S. (1987). Analysis of factors controlling soil organic matter levels in Great Plains grasslands. Soil Sci. SOC. Am. J. 51, 1173-1 179. Patra, D. D. (1994). Soil microbial biomass. In “Ecology and Biology of Soil Organisms,” (S. C. Bhandari and L. L. Somani, Eds.), pp. 81-104. Agrotech Publishing Academy, Vdaipur. Paul, E. A. (1984). Dynamics of organic matter in soils. Plant Soil 76, 275-284. Paul, E. A,. and Juma, N. G. (1981). Mineralization and immobilization of soil nitrogen by microorganisms. 111 “Terrestrial Nitrogen Cycles. Processes, Ecosystems, Strategies and Management Impacts,“ pp. 179-204. Ecological Bulletin 33 (F. E. Clark and T. Rosswall. Eds.), Stockholm. Power, R. F. (1980). Mineralizable soil nitrogen as an index of nitrogen availability to forest trees. Soil Sci. SOC. Am. J . 44, 1314-1320. Powlson, D. S. (1980). Effect of cultivation on the mineralization of nitrogen in soil. Plant Soil 57, 151-153.
Powlson, D. S. (1994). The soil microbial biomass: Before, beyond and back. In “Beyond the Biomass, Compositional and Functional Analysis of Soil Microbial Communities’’ (K.Ritz, J. Dighton, and K. Giller, Eds.), pp. 3-20. Wiley, Chichester. Powlson, D. S., and Barraclough, D. (1993). Mineralization and assimilation in soil plant systems. In “Nitrogen Isotope Techniques in Plant, Soil and Aquatic Biology” (R. Knowles and T. H. Blackbum, Eds.), pp. 209-242. Academic Press, San Diego. Powlson, D. S., Pruden, G., Johnston, A. E., and Jenkinson, D. S. (1986). The nitrogen cycle in the Broadbalk Wheat Experiment: Recovery and losses of ‘sN-labelled fertilizer applied in spring and inputs of nitrogen from the atmosphere. J . Agric. Sci. 107, 591-609. Powlson, D. S . , Hart,P. B. S., Poulton, P. R., Johnston, A. E., and Jenkinson, D. S. (1992). Influence of soil type, crop management and weather on the recovery of 15N-labelledfertilizer applied to winter wheat in spring. J. Agric. Sci. 118, 83-100. Prosser, J. (1986). Nitrification (Special Publication ofSociety ofceneral Microbiology). IRL Press, Oxford. Rahn, C. R., Vaidyanathan, L. V., and Paterson, C. D. (1992). Nitrogen residues from brassica crops. Aspects Appl. Biol. 30, 263-270. Raison, R. J., Connel, M. J., and Khanna, P. K. (1987). Methodology for studying fluxes of soil mineral-N in situ. Soil Biol. Biochem. 19, 521-530.
NITROGEN MINERALIZATION
233
Raymond, W. F., Tilley, J. M . , Derias, R. E., and Minson, D. J. (1960). In “Production and Use of Grass,” pp. 181-190. SCI, London. Recous, S., Machet, J. M., and Mary, B. (1988). The fate of 15N urea and ammonium nitrate applied to a winter wheat crop. 11. Plant uptake and N efficiency. Plant Soil 112, 215-224. Redman, M. H., Wigglesworth, S. A., and Vinten, A. J. A. (1989). Nitrogen dynamics of a leguminous green manure. In “Nitrogen in Organic Wastes Applied to Soils” (J. A. Hansen and K. Henricksen, Eds.), pp. 98-1 12. Academic Press, London. Reinertsen, S. A., Elliott, L. F., Cochran, L. V., and Campbell, G. S. (1984). Role of available carbon and nitrogen in determining the rate of wheat straw decomposition. Soil Biol. Biochem. 16, 459-464. Richter, G. M., Hoffman, A , , Nieder, R . , and Richter, J. (1989). Nitrogen mineralization in loamy arable soils after increasing the ploughing depth and ploughing grasslands. Soil Use Munagement 5, 169-173. Rijtema, P. E., and Kroes, J. G. (1991). Some results of nitrogen simulations with the model ANIMO. Fertilizer Res. 27, 189- 198. Robinson, D., Griffiths, B., Ritz, K., and Wheatley, R., (1989). Root-induced nitrogen mineralization: A theoretical analysis. P lunt Soil 117, 185- 193. Ross, D. I., Speir, T. W., Tate, K. R., and Orchard, V. A. (1985). Effects of sieving on estimations of microbial biomass and carbon and nitrogen mineralization under pasture. Aust. J. Soil Res. 23, 319-324. Royal Society (1983). “The Nitrogen Cycle of the United Kingdom.” (Report of a Royal Society Study Group). The Royal Society, London. Rubins. E. J., and Bear, F. E. (1942). Carbon:nitrogen of fertilizer materials in relation to the availability of their nitrogen. Soil Sci. 54, 41 1-423. Ruiter de, P. C., van Veen, J. A,, Moore, J. C., Brussard, L., and Hunt, H. W. (1993). Calculation of nitrogen mineralization in soil food webs. Plant Soil 157, 263-273. Russell, W. E. (1973). “Soil Conditions and Plant Growth,’’ 10th ed. Longman, London. Ruz Jerez, B. E., Ball, P. R., and Tillman, R. W. (1988). Role of earthworms in N release from herbage residues. In “Nitrogen Efficiency in Agricultural Soils” (D. S. Jenkinson and K. A. Smith, Eds.), pp. 355-370. Elsevier, London. Saffigna, P.G. (1988). I5N methodology in the field. In “Advances in Nitrogen Cycling in Terrestrial Ecosystems” (J. R. Wilson, Ed.), pp. 433-451. CAB International, Wallingford. Saint Fort, R., Frank, K. D., and Schepers, J. S . (1990). Role of nitrogen mineralization in fertilizer recommendations. Commun. Soil Sci. Plant Anal. 21, 1945- 1958. Sakamoto, K., and Oba, Y. (1994). Effect of fungal to bacterial biomass ratio on the relationship between CO, evolution and total soil microbial biomass. Biol. Fertil. Soils 17, 39-44. Salter, R. M., and Schollenberger, C. J. (1939). “Farm Manure.” Bulletin 605, Ohio Agricultural Experiment Station. Scheu, S. (1987). Microbial activity and nutrient dynamics in earthworm casts (Lurnbricidae). Biol. Fertil. Soils 5, 230-254. Scholefield, D., Tyson, K . C., Garwood, E. A , , Armstrong, A. C., Hawkins, J., and Stone, A. C. ( 1993). Nitrate leaching from grazed grassland lysimeters: Effects of fertilizer input, field drainage age of sward and patterns of weather. J. Soil Sci. 44, 601-613. Seitzinger, S . P. ( 1994). Linkages between organic matter mineralization and denitrification in eight riparian wetlands. Biogeochernisrry 25, 19-39. Serna, M. D., and Pomares, F. (1991). Comparison of biological and chemical methods to predict nitrogen mineralization in organic wastes. B i d . Fertil. Soils 12, 89-94. Shen, S . M., Hart, P. B. S., Powlson, D. S., and Jenkinson, D. S. (1989). The nitrogen cycle in the Broadbalk wheat experiment: ‘SN-labelled fertilizer residues in the soil and in the soil microbial biomass. Soil B i d . Biochem. 21, 529-533.
2 34
S. C . JARVIS ET AL.
Shepherd, M. A. (1992). Sugar beet. Leaching of fertilizer nitrogen and nitrogen requirements. SBREC Comm. Paper No. 2527. Shepherd, M. A. (1993). Measurement of soil mineral nitrogen to predict the response of winter wheat to fertilizer nitrogen after applications of organic manures or after ploughed-out grass. J. A @ . Sci. 121, 223-231. Shepherd, M. A., Jarvis, S. C., Powlson, D. S . , and Stockdale, E. A. (1996). Nitrogen management in UK agricultural systems as influenced by mineralization. Soil Use Munugemenr, submitted for publication. Sierra, J. (1990). Analysis of soil nitrogen mineralization as estimated by exponential models. Soil Biol. Biochem. 22, 1151-1153. Silvlapan, K., Fernando, V., and Thenabadu, M. W. (1985). N mineralization in polyphenol-rich plant residues and their effect on nitrification of applied ammonium sulphate. Soil Biol. Biochem. 17, 547-55 I . Simard, R. R., and N’Dayegamiye, A. (1993). Nitrogen-mineralization potential of meadow soils. Cunad. J. Soil Sci. 73, 27-38. Sims, J. T. (1986). Nitrogen transformations in a poultry manure amended soil: Temperature and moisture effects. J. Environ. Quul. 15, 59-63. Skjemstad, J. O., Vallis, I . , and Myers, R. J. K. (1988). Decomposition of soil organic nitrogen. In “Advances in Nitrogen Cycling in Terrestrial Ecosystems” (J. R. Wilson, Ed.), pp. 134-144. CAB International, Wallingford. Smith, J. L., and Paul, E. A. (1990). The significance of soil microbial biomass estimates. In “Soil Biochemistry,” Vol. 6. (J. M. Bollag and H. Stotzheyl, Eds.), pp. 357-396. Marcel Dekker, New York. Smith, S. R., Tibbett, M., and Evans, T. D. (1992). Nitrate accumulation potential of sewage sludges applied to soil. Aspecrs Appl. Biol. 30, 157-162. Stanford, G. (1982). Assessment of nitrogen availability. In “Nitrogen in Agricultural Soils” (F. J. Stevenson, Ed.), pp. 651-688. ASA, CSSA, SSSA, Madison. Stanford, G., and Smith, S. J. (1972). Nitrogen mineralization potentials of soils. Soil Sci. SOC. Am. Proc. 109, 190-196. Stanford, G., and Epstein, E. (1974). Nitrogen mineralization-water relations in soils. Soil Sci. SOC. Am. Proc. 38, 103-107. Stanford, G., Frere, M.H., and Schwaninger, D. H. (1973). Temperature coefficient of soil nitrogen mineralization. Soil Sci. 115, 321-323. Stevenson, F. J. (1982a). Origin and distribution of nitrogen in soil. In “Nitrogen in Agricultural Soils” (F. J. Stevenson, Ed.), pp. 1-42. ASA, CSSA, SSSA, Madison. Stevenson, F. J. (1982b). Organic forms of nitrogen. In “Nitrogen in Agricultural Soils” (F. J. Stevenson, Ed.), pp. 67-122. ASA, CSSA, SSSA, Madison. Stockdale, E. A., McKinlay, R. G., and Rees, R. M. (1992). Soil nitrogen management and interaction with crop pests and diseases in organic farming. Aspecrs Appl. Biol. 30, 387-392. Stockdale, E. A,, Davies, M. G., Koch, M., Rees, R. M., and Vinten, A. J. A. (1994). Assessment of the problems associated with measuring gross mineralization rates in soil cores. In “Proceedings of the Symposium: N mineralization in agricultural soils” (J.J. Neeteson and J. Hassink, Eds.), pp. 95-1 10. AB-DLO, WageningedHaren, 1993. Stockdale, E. A., and Rees, R. M. (1994). Relationships between biomass nitrogen and nitrogen extracted by other nitrogen availability indices. Soil B i d . Biochem. 26, 1213-1220. Stokes, D. T.,Scott, R. K., Tilston, C. H., Cowie, G., and Sylvester-Bradley, R. (1992). The effect of time of disturbance on nitrate mineralization. Aspects Appl. Biol. 30, 279-282. Sylvester-Bradley, R. (1993). “Nitrate Research.” MAFF Review, MAFF, London. Thorup-Kristensen, K. (1993). The effect of nitrogen catch crops on the nitrogen nutrition of a succeeding crop. Effects through mineralization and pre-emptive competition. Acru Agric. S c u d . B 43, 74-81.
NITROGEN MINERALIZATION
235
Van Veen, J. A., and Frissel, M. J. (1981 ). Simulation model of the behaviour of nitrogen in soil. In “Simulation of Nitrogen Behaviour of Soil-Plant Systems” (M. J. Frissel and J. A. Van Veen, Eds.). PUDOC, Wageningen. Van Veen, J. A., Ladd, J. N., and Amato, M. (1985). Turnover of carbon and nitrogen through the microbial biomass in a sandy loam and a clay soil incubated with [I4C(U) glucose] and 15N-(NH4),S04under different moisture regimes. Soil Biol. Biochem. 19, 559-565. Verberne, E. L. J., Hassink, J., de Willigen, P., Groot, J. J. R., and Van Veen, J. A. (1990). Modelling organic matter dynamics in different soils. Netherlands J . Agric. Sci. 38, 221-238. Vinten, A. J. A., Vivian, B. J., and Howard, R. S . (1992). The effect of nitrogen fertilizer on the nitrogen cycle of two upland arable soils of contrasting textures. froc. Fertilizer Soc. 329. Waksman, S . A., and Tenney, F. G. (1927). The composition of natural organic materials and their decomposition in the soil. Soil Sci. 24, 317-333. Warren, G. P. (1985). Available nitrogen in relation to fractions of soil nitrogen in grassland soils. PhD Thesis, University of Reading. Warren, G. P., and Whitehead, D. C. (1988). Available soil nitrogen in relation to fractions of soil nitrogen and other soil properties. Plant Soil 112, 155- 165. Webb, J., Froment, M., and Sylvester-Bradley, R. (1991). Effects of cultivation and nitrification inhibitor on soil nitrogen availability after a grass ley and the response of the following cereal crop to fertilizer nitrogen. J. Agric. Sci. 117, 9-22. Wedin, D. A,, and Tilman, D. (1990). Species effects on nitrogen cycling: A test with perennial grasses. Oecologia 84, 433-441. White, C. S., and Marinakis, Y. D. (1991). A flexible model for quantitative comparisons of nitrogen mineralization patterns. Biol. Fertil. Soils 11, 239-244. Whitehead, D. C. (1981). An improved chemical extraction method for predicting the supply of available nitrogen. J . Sci. Food Agric. 32, 359-365. Whitehead, D. C . , Bristow, A. W., and Lockyer, D. R. (1990). Organic matter and nitrogen in the unharvested factions of grass swords in relation to the potential nitrate leaching after ploughing. Plant Soil 123, 39-49. Williams, T. E., and Clement, C. R. (1965). Accumulation and availability of nitrogen in soils under leys. Proc. First Gen. Meeting Eur. Grassland Fed.. 39-45. Willigen, P. de (1991). Nitrogen turnover in soil-crop system; Comparison of fourteen simulation models. Fertilizer Res. 27, 141-149. Willison, T., and Anderson, J. M. (1991). Spatial patterns and controls of denitrification in a Norway spruce plantation. Forest Ecol. Management 44, 69-76. Wilson, C. E., Norman, R. J., and Wells, B. R. ( I 994). Chemical estimations of nitrogen mineralization in paddy rice soils: 1. Comparison to laboratory indices. Commun. SoilSci. Plant Anal. 25, 573-590. Withers, P. J. (1988). Nitrogen prediction after grass-Preliminary review. ADAS Internal Report. Woods, L. E., Cole, C. V., Elliott, E. T., Anderson, R. V., and Coleman, D. C. (1982). Nitrogen transformations in soil as affected by bacterial-microfaunal interactions. Soil Biol. Biochem. 14, 93-98. Yeates, G.W., Wardle, D. A., and Watson, R. N. (1993). Relationships between nematodes, soil microbial biomass and weed management strategies in maize and asparagus cropping systems. Soil Biol. Biochem. 25, 869-876. Young, C. P. (1986). Nitrate in ground water and the effects of ploughing on release of nitrate. I n “Effects of Land Use on Fresh Waters” (J. F. Solbe, Ed.), pp. 221-237. WRC/Ellis Horwood, Chichester, UK. Zou, X., Valentine, D. W., Sanford, R. L., Jr., and Binkley, D. (1992). Resin-core and buried bag estimates of nitrogen transformations in Costa Rican lowland rainforests. f /ant Soil 139, 275283.
This Page Intentionally Left Blank
THEBUFFERINGPOWER OF PLANT NUTRIENTS AND EFFECTS ON AVAILABILITY K. P. Prabhakaran Nair Institute of Plant Nutrition, University of Hohenheim, 70599 Stuttgart, Federal Republic of Germany
I. Introduction 11. Efficient Plant Nutrient Management - The Ley Factor in Sustainable Soil Management A. Soil Tests and Nutrient “Availabilitv” B. Rating Soil Tests To Define Nutrieit Availability and a Fertility Index 111. T h e Buffer Power and Effect on Nutrient Availability A. Basic Concepts B. Measuring the Nutrient Buffer Power and Its Importance in Affecting Nutrient Concentrations on Root Surfaces IV. Quantifying the Buffer Power of Soils and Testing Its Effect on Nutrient Availability A. Phosphorus B. Potassium V. T h e Role of Electro-ultrafiltration in Measuring P and K Intensity for the Construction of Buffer Power Curves VI. Quantifylng the Buffer Power for Precise Availability Prediction - Heavy Metals A. Zinc B. Quantifylng Zn Buffer Power C. Other Heavy Metals D. Molybdenum E. Iron F. Manganese G. Boron WI. Influence of Heavy Metal Contamination on Buffering of Major Elements VIII. Possible Buffering Effect on Plant Acquisition of Heavy Metals IX. Concluding Comments and Future Imperatives References
237 Advances m A p n m y , volump 57 Copyright 0 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
238
K. P. PRABHAKARAN NAIR
I. INTRODUCTION Many years ago, in one of the early editions of Advances in Agronomy, Roy W. Simonson writing a chapter entitled “Concept of Soil,” noted “Someone has said that the fabric of human life is woven on earthem looms-it everywhere smells of the clay.” More than three decades later, we Agronomists and Soil Scientists have come very far in our understanding of “the fabric of human life” which “everywhere smells of the clay.” That “the fabric of human life” which is so very intimately linked to soil has dramatically changed is beyond dispute. Yet, there is no denying the fact that this “fabric of human life” will always be linked to the soil which is “the pragmatic, the reality, the entity that dictates much of what societies can do” (Boul, 1994). Soil, in my opinion, is that invaluable gift of God to life on planet Earth and can aptly be termed “The Soul of Infinite Life.” Though the basic concept of soil, since its early description as a “thin mantle over the land surface” has vastly changed over the years, this thin mantle has always been the focal point since it is the medium of plant growth. For early man it was nothing more than a physical support for his predation. Quite likely, some areas were known to provide better footing than others, and some were to be avoided if possible. It is amazing that even after decades of research in soil science which has provided such invaluable information on this “thin mantle over the land surface” so crucial to the existence of life, both human and animal, on the planet earth, this basic instinct of predation has remained unchanged. How else can we explain the disdain and callousness so often witnessed in modem societies, propelled by an insatiable greed to acquire unlimited wealth, which leads to the abuse of soil, this invaluable gift of God to man? Undoubtedly, the earliest shift in attitude toward soils must have originated at a time when man began to grow food, rather than gather it as his ancestors did. In many ways, this shift in attitude was the precursor to modem day soil science. Though this shift must have occurred in pre-Christian times, about 9000 years ago (Simonson, 1968), and focused on the inevitability of a proportionally smaller land surface supporting a larger human population, it is only in recent times that we have witnessed the magnitude of the impact of this shift in attitude on human existence. Much land has degraded and become unsuitable for agriculture since a century ago. The 1992-93 World Resources report (Stammer, 1992) from the United Nations on the status of the world soils contains very alarming conclusions. For example, nearly 10 million ha of the best farm lands of the world have been so ruined by human activity since World War I1 that it is impossible to reclaim them. Over 1.2 billion ha that have been seriously damaged can be restored only at a great cost. This loss in soil capability could mean that there will be enormous food shortages in the next 20-30 years and, as is usual, the people of disadvantaged nations will suffer. Two-thirds of the seriously eroded land is in Asia and Africa. About 25% of the cropped land in Central America is moderately to severely damaged. In North America, this is a small percentage-only 4.4%. Since the time of the
BUFFERING POWER OF PLANT NUTRIENTS
239
“green revolution,” food production has declined dramatically in 80 developing countries in the past decade. Soil degradation is the major factor. Nearly 40% of the world’s farming is done on very small parcels of I ha or less (Robison et al., 1981). Ignorance and poverty characterize this situation. Yet, emphasis on agriculture has been confined mostly to large-scale farming. Large-scale farming, grand projects at huge costs and huge profits, have been the order of the day for many decades. In a lighter vein, it can be said that even the “lebens raum” concept of Adolf Hitler had an echo in the inevitability of this modem day fact. What else can justify the ruthless conquest of vast territories of land by this master strategist who set out to conquer the world-or, more appropriately, the world’s soils? Despite the complexity of soil science and the emergent soil management practices, the basic concept of soil as a medium of plant growth can be expected to persist for an indefinite length of time. But it is becoming increasingly clear that the earlier views on soil as merely the “supportive medium” for plant growth is giving place to newer ones on “managerial concepts” of this supportive medium. This is amply illustrated by the shift in focus from the green revolution phase of the 1960s to mid-1970s where application of increasing quantities of soil inputs such as fertilizers and pesticides was emphasized, to the “sustainable agriculture” phase from the early 1980s to the present (probably to continue?); sustainable agriculture places more reliance on biological processes by adopting genotypes to adverse soil conditions, enhancing soil biological activity and optimizing nutrient cycling to minimize external inputs, such as fertilizers, and maximize their efficiency of use. In fact, the paradigm of the earlier phase has given way to the emergent new paradigm (Sanchez, 1994) and this is clearly reflected in the dialogue of the world leaders during the Earth Summit in 199 1 in Rio de Janeiro, Brazil, where Agenda 21 has incorporated six chapters on soil management issues (Keating, 1993). The focus of this review will be on the second paradigm inasmuch as prescriptive soil management is concerned with regard to understanding soil nutrient availability and its efficient management in crop production.
11. EFFICIENT PLANT NUTRIENT MANAGEMENTTHE KEY FACTOR IN SUSTAINABLE SOIL MANAGEMENT Agricultural systems differ from natural systems in one fundamental aspect: while there is a net outflow of nutrients by crop harvests from soils in the first, there is no such thing in the second (Sanchez, 1994). This is because nutrient losses due to physical effects of soil and water erosion are continually replenished by weathering of primary minerals or atmospheric deposition. Hence,
240
K. P. PWHAKARAN NAIR
the crucial element of sustainability of crop production is the nutrient factor. But, of all the factors, the nutrient factor is the least resilient (Fresco and. Kroonenberg, 1992). The thrust of high input technology, the hallmark of the “green revolution,” in retrospect, or the moderation by low input technology, the foundation stone of sustainable agriculture, in prospect, both dwell on this least-resilient nutrient factor. If the pool of nutrients in the soil, both native and added, could be considered as the “capital,” efficient nutrient management might be analogous to raising the “interest” accrued from this capital in such a way that there is no great danger of the erosion of this capital. Hence, sound prescriptive soil management should aim at understanding the actual link between the “capital” and the “interest” so that meaningful management practices can be prescribed.
A. SOILTESTS AND NUTRIENT “AVAILABILITY” It is universal knowledge that soil tests are the basis for predicting nutrient “availability.” There are, perhaps, as many soil tests for each nutrient as there are nutrients. This review will not dwell on the merits or demerits of any single soil test or group of them. Suffice to say that fertilizer recommendations traditionally are made at the point where marginal revenues equal marginal costs which involve some positive synergism (DeWit, 1992). The most common result of this approach is the vast build-up in the soil nutrient pool in intensively cultivated soils (Whitmore and van Noordwijk, 1994). Data in Table 1 indicate positive balances (in kg/ha/year) for N(61), P(23), and K(37) in intensive crop production systems (Frissel, 1978). Over several decades, such positive balances can lead to a huge build-up of the Table 1 Nutrient Balance (kg/ha/yr) in Intensively Managed Arable Soils (after Frissel, 1978)
Inputs Fertilizers Other Total outputs Harvest Removal Other Total
N
P
K
156 32 188
39 39
119 9 128
103 24
16 -
91 -
127
16
91
Balance: 61
23
31
BUFFERING POWER OF PLANT NUTRIENTS
241
nutrient capital, especially in the case of high-input, intensive agricultural systems as in the case of many European, North American, and Scandinavian countries. A dangerous consequence of such huge soil build-up is nutrient contamination of groundwater to such extremes that “environmental soil tests” become necessary to assess critical limits of nutrient pollution (Sharpley et a l . , 1993). Nitrogen is a prime candidate for this scenario especially in the temperate zone. At the other end of the spectrum are the marginal areas of the tropical zone where inadequate replenishment of nutrient removal by crop growth and also nutrient loss by soil and water erosion has left that capital “in the red.” Initially fertile Alfisols of much of Africa with subsequent severe depletion of N and P (Yates and Kiss, 1992) are an example of this nutrient “bankruptcy.” Either way, contemporary soil tests are the basis on which prescriptive management practices are formulated.
B. RATINGSOILTESTSTo DEFINENUTRIENT AVAILABILITY AND A FERTILITY INDEX Most soil test laboratories around the world use some kind of “rating system” to evaluate soil test values. These rating systems invariably use qualitative terms such as “low,” “medium,” or “high” to describe the availability of a specific plant nutrient. Admittedly, these terms denote different meanings in the context of availability of a particular plant nutrient and, at best, are empirical terminologies. This problem has been recognized by researchers over the years. Morgan (1935) suggested a scale of 1- 10 with 8 equal to the point of no response. Bray and Kurtz (1945) used relative yield or percentage sufficiency to describe the degree of deficiency, with 100 defined as the point of no response. The index below 100 follows the curvilinear relationship between soil-test values and yield without addition of the element. Above 100, the index displays a straight-line relationship indicating the relative margin of adequacy or the proximity to an excessive level. To eliminate the need for a percent sign, the values are referred to as “Fertility Indices” and they are reported to the nearest multiple of 10 from 0 to 9990 (Cope and Evans, 1985). In addition to ratings, most laboratories use some method of reporting results more precisely, mainly for use by farmers in record keeping and monitoring soil fertility. Some report kg/ha, lb/a, or ppm extracted, but these would be confusing to farmers, because each element has a different level for a specific degree of adequacy (Cope and Evans, 1985). For instance, the adequate or critical level for one soil may be 25 ppm P, 120 ppm K, 200 ppm Ca, and 30 ppm Mg. Adequate levels in other soils and from other extracting procedures would be different for each element (Cope and Evans, 1985). Despite the fact that a number of soil tests and others such as Diagnosis and Recommendation Integrated Systems (DRIS) are in vogue, to predict nutrient
2 42
K. P. PRABHAKARAN NAIR
availability, it must be said that a universal picture is yet to emerge in this field with regard to precise availability prediction. This is primarily because a soil test in the laboratory can never simulate plant root absorption of a nutrient in a field soil, though most of the time the assumption is that it does. In the final analysis it is the plant and plant alone which will decide whether or not the nutrient is available. This review will examine the question of whether a better and more reliable alternative exists.
111. THE BUFFER POWER AND EFFECT ON NUTRIENT AVAILABILITY A. BASICCONCEPTS In any nutrient management approach that is sound and reproducible, one must start with a basic understanding of the chemical environment of plant roots. When we consider this, the first term that we come across is the “soil solution,” because the plant root is bathed in it and is most affected by its chemical properties. The Soil Science Society of America (1965) defines soil solution as “the aqueous liquid phase of the soil and its solutes consisting of ions dissociated from the surfaces of the soil particles and of other soluble materials.” Adams (1974) has given a simple definition: “The soil solution is the aqueous component of a soil at field-moisture contents.” Perhaps it is important to emphasize here that much of contemporary soil testing has considered a soil extract as synonymous with the soil solution. Since soil extraction is supposed to simulate plant root extraction, it is pertinent to consider the chemical environment of the root, though briefly, from this angle. It is worth noting that the chemical environment of roots in natural soil systems is so obviously complex that both soil scientists and plant physiologists have been unable to provide a precise definition. If this complex chemical system is to be accurately quantified, thermodynamic principles will need to be used to evaluate experimental data. Even then, the limitations are obvious, as in the case of K where the thermodynamic investigations are quite often inapplicable under field conditions. This is because, although a quasi-equilibriumin K exchange can be achieved in the laboratory, these conditions are seldom, if ever, attained under field conditions (Sparks, 1987). Agricultural soils are, for the most part, in a state of disequilibrium owing to both fertilizer input and nutrient uptake by plant root. It thus appears that a universal and accurate definition of a root’s chemical environment awaits the proper application of thermodynamics for the root’s ambient solution (Adams, 1974) or even kinetics, as in the case of K (Sparks, 1987), where thermodynamics has been found inadequate.
BUFFERING POWER OF PLANT NUTRIENTS
2 43
Soil extractions with different extractants provide a second approach in defining the root’s chemical environment. This approach has been particularly successful in understanding cases like P insolubility, soil acidity, and K fixation. However, this approach also fails to define precisely the root’s chemical environment. Though this approach also suffers from deficiencies, such as the extractants removing arbitrary and undetermined amounts of solid-phase electrolytes and ions (or the extractants causing precipitation of salts or ions from the soil solution) and the soil-plant interrelationship defined in terms of the solid phasecomponent of the soil, even though the solid phase is essentially inert except as it maintains thermodynamic equilibria with the solution phase (Adams, 1974), the latter part could be researched more to understand how the solid phase-solution phase equilibria can be interpreted to give a newer meaning to quantifying nutrient availability. It is in this context that the role of the plant nutrients’ “buffer power” assumes crucial importance. The close, almost linear, relationship in a low concentration range of <0.5 mM for NOT-N, NHZ-N, K+ , H,PO;, and HPOj- , which has been established by numerous solution culture experiments, can be quantitatively described by the equation U = 2nr aC,,
(1)
where U = uptake of a I-m root segment, r = root radius, C, = concentration of the ion at the root surface and, a = root absorbing power (Mengel, 1985). The metabolic rate of the root determines its absorbing power. A high root absorbing power would imply that a relatively high proportion of nutrient ions coming in contact with the root surface is absorbed and vice versa. The nutrient ion concentration at the root surface (C,) depends on a since a high root absorbing power tends to decrease C,; it also depends on the rate of movement from bulk soil toward the root (Mengel, 1985). Diffusion and/or mass flow controls this movement. But it is now established that nearly 95% of this movement for nutrients such as P, K , and Zn (among heavy metals) and, possibly NHJ, is by way of diffusion. When root uptake of an ion species is less than its movement toward it, accumulation of the ion species on the root surface is bound to occur, as has been shown to be the case with Ca2+ where mass flow contributes to this accumulation (Barber, 1974). The diffusive path for ions, such as P and K, which plant roots take up at high rates but which are in low concentration in the soil solution near the root, is the concentration gradient. In a sense, the effective diffusion coefficient which quantifies the diffusive path and the buffer power are analogous because the diffusive flux across the root surface is integrally related to the nutrient buffer power. This has been shown to be true in the case of P where a highly significant positive correlation between the two was found to exist in 33 soil samples obtained from experimental sites in the U.S.A. and Canada (Kovar and Barber, 1988). However, in a routine laboratory set-up, it is far easier to
244
K. P. PRABHAKARAN NAIR
measure the buffer power than the effective diffusion coefficient and this review will further focus on the question of how buffer power can be quantified without recourse to cumbersome analytical techniques and how its integration into routine soil test data will considerably improve predictability of nutrient uptake.
B. MEASURING THE NUTRIENT BUFFERPOWER ITS IMPORTANCE IN AFFECTINGNUTRIENT CONCENTRATIONS ON ROOT SURFACES
AND
The ability to predict the mobility of dissolved chemicals, such as fertilizers, in the soil is of considerable value in managing fertilizer applications. Soil testing, in its essence, aims to achieve this. While modeling transport and retention of ions from thermodynamic (Selim, 1992), kinetic (Sparks, 1989), and mechanistic (Barber, 1984) angles could be very informative, the importance of translating this information into practically feasible procedures in crop production calls for an understanding not only of the basic concepts but of their intelligent application as well. In a dynamic state of plant growth, the concentration of any nutrient on the root surface is nearly impossible to measure since both the nutrient in the plant tissue and the root absorbing power, which directly affects it, change quickly due to root metabolic processes. The inability of even mechanical mathematical models to accurately predict nutrient influx rates has recently been highlighted (Lu and Miller, 1994). Hence, if an effective soil testing procedure is to be devised for a nutrient, which is an alternative to defining the plant root’s chemical environment, one must resolve the problem of quantifying the nutrient concentration on the root surface indirectly, even if it is impossible to resolve it directly, for the reasons mentioned above. Using Fick’s first law,
F = -D (dcldx),
(2)
where F = the flux, dC/dx = concentration gradient across a particular section, and D = the diffusion coefficient, Nye (1979) has suggested that the formula can be applied to both ions and molecules. The negative sign for D implies net movement from high to low concentration. Although for molecules in simple systems like dilute solutions D may be nearly constant over a range of concentrations, for ions in complex systems like soils and clays D will usually depend on the concentration of the ion, and on that of other ions as well (Nye, 1979). Nye (1979) has further suggested that though Fick’s first law may be derived from thermodynamic principles in ideal systems, in a complex medium such as the soil
BUFFERING POWER OF PLANT NUTRIENTS
2 45
the above equation may be regarded as giving an operational definition of the diffusion coefficient. Thus, Nye (1979) defines the diffusion coefficient as D = D, 8f,(dC,ldC) + DE,
(3)
where D ,= diffusion coefficient of the solute in free solution, 8 = the fraction of the soil volume occupied by solution and gives the cross section for diffusion, fi = an impedance factor, C , = concentration of solute in the soil solution, DE = an excess term which is zero when the ions or molecules on the solid have no surface mobility, but represents their extra contribution to the diffusion coefficient when they are mobile. D, can generally be neglected since only in rare instances will it play any role in diffusion of plant nutrient ions in soil (Mengel, 1985). From the point of view of nutrient availability, dC,ldC, which represents the concentration gradient, assumes crucial importance, as we shall see below. The term dC,/dC, where C , = concentration of the nutrient ion in the soil solution and C = concentration of the same ion species in the entire soil mass, assumes considerable significance in lending a practical meaning to nutrient availability. If we ascribe the term “capacity” or “quantity” to C and “intensity” to C , , we have in this term an integral relationship between two parameters that may crucially affect nutrient availability. Since the concentration gradient of the depletion profile of the nutrient in the zone of nutrient uptake depends on the concentration of the ion species in the entire soil mass (represented by “capacity” or “quantity”) in relation to the rate at which this is lowered on the plant root surface by uptake (represented by “intensity”), it could be argued that a quantitative relationship between the two should represent the rate at which nutrient depletion and/or replenishment in the rooting zone should occur (Nair, 1984). This relationship has been functionally quantified by Nair and Mengel (1984) for P in eight widely differing central European soils (Table 11) and the term dC,ldC has been referred to as the “nutrient buffer power.” Nair and Mengel (1984) used electro-ultrafiltration to quantify C , while using an incubation and extraction technique to quantify C . For P, C was found to closely approximate isotopically exchangeable P (Keerthisinghe and Mengel, 1979), but in the experiments conducted by Nair and Mengel (1984), it was estimated by the extraction of incubated soil with an extractant which was a mixture consisting of 0.1 M Ca lactate + 0.1 M Ca acetate + 0 . 3 M acetic acid at pH 4.1. The extractant exchanges adsorbed phosphate and dissolves Ca phosphates except apatites; the method known as the “CAL-method,” developed by Schuller (1969), is now widely used in central Europe. In the case of K+ and NHJ-N, C denotes the concentration of exchangeable, and to some extent nonexchangeable, fractions (Mengel, 1985). Since very low concentrations in the range of 2.0 pM may be attained on the root surface for both P and K (Claassen and Barber, 1976, Claassen et af., 1981; Hendriks et a l . , 1981), Nair and Mengel (1984) had to use electro-ultrafiltration
2 46
K. P. PRAl3HAKARAN NAIR Table I1 Comparison of P Buffer Power of Eight Widely Differing Central European Soils (Determined by Two Different Techniques) r
Regression Soil Benzheimer Hof Hungen Oldenburg B6 Wolfersheim Obertshausen Oldenburg B3 Klein-Linden GNningen
(2)
(1)
Y Y Y Y Y
+
1 8 . 8 ~ 7.94 3 8 . 2 ~- 1.03 4 9 . 8 ~+ 0.52 7 0 . 3 ~+ 0.03 7 0 . 5 ~+ 2.66 Y = 7 3 . 6 ~+ 2.07 Y = 7 5 . 0 ~+ 0.38 Y = 7 5 . 4 ~+ 0.89 = = = = =
Y = 0 . 2 3 ~+ 8.98 Y = 4.32 + 0 . 2 5 ~ Y = 0.72 + 0.261 Y = 0.11 0 . 2 7 ~ Y = 2.89 0.301 Y = 0.61 + 0 . 3 1 ~ Y = 1.81 0 . 3 2 ~ Y = 3.62 + 0.361
+ + +
(1)
(2)
0.912 0.967 0.994 0.998 0.966 0.994 0.999
0.995 0.997 0.999 0.983 0.998 0.997 0.991 0.996
1.OOO
Note. The b values in the regression functions represent the P buffer power of each soil. In regression function (1) (after Nair and Mengel, 1984) Y = CAL-P (Schuller’s method) and in regression function (2) (after Nair, 1992) Y = the author’s method. x in both refers to electroultrafiltrable P. Note the very high r values in all the cases. The soils are arranged in their sequential increase in P buffer power.
to quantify C , . Thus, the nutrient depletion around the roots which is caused by the diffusive flux of the nutrients toward the root surface is related to both the quantity and the intensity parameters, and a quantifiable relationship between both represents the buffer power specific to the nutrient and the soil. A growing root will at first encounter a relatively high concentration of P which is in the range of the concentration of the bulk soil solution (Nair and Mengel, 1984). As uptake continues, a depletion will occur at the root surface. This depletion profile gets flatter with enhanced nutrient uptake (Lewis and Quirk, 1967; Claassen et al., 1981; Hendriks et al., 1981). But it is the capacity of the soil to replenish this depletion which ensures a supply of nutrient ions to the plant root without greatly depressing its average concentration on the root surface. It is the nutrient’s buffer power that decides these depletion and/or replenishment rates. A soil with a high P buffer power implies that the P absorbed from the soil solution is rapidly replenished. In such a case, P concentration at the root surface decreases only slowly and mean P concentration at the root surface remains relatively high. In soils with a low P buffer power, the reverse is true, and P concentration at the root surface is rapidly diminished and remains relatively low. This has been proved experimentally for P (Nair and Mengel, 1984, Nair, 1992). This phenomenon holds true for Zn2+ (Nair, 1984), K (Nair et al., 1994), and NHJ-N as well (Mengel, 1985).
BUFFERING POWER OF PLANT NUTRIENTS
2 47
IV. QUANTIFYING THE BUFFER POWER OF SOILS AND TESTING ITS EFFECT ON NUTRIENT AVAILABILITY A. PHOSPHORUS When the exchangeable or desorbable nutrient ions are plotted on the y axis and the corresponding concentration of the same ions in the equilibrated solution on the x axis, we obtain a slope, the steepness of which represents the buffer power for that ion. The P buffer powers of eight central European soils widely differing in their physicochemical properties are reported in Table 11. The “quantity” factor ( C ) in the Eq. (3) was determined by two methods: (i) by an incubation and extraction technique (Nair and Mengel, 1984) and (ii) by an absorption-desorption equilibrium technique (Nair, 1992) while the “intensity” factor (C,), was determined by electro-ultrafiltration (Nemeth, 1979). It is highly remarkable that the correlation coefficients, in the case of all the soils investigated, between C (represented by y) and C , (represented by x ) are very highly significant as indicated by the r values, which in almost all cases were close to unity [including one soil (Gruningen) in which it was unity]. This clearly establishes the fact that the techniques employed will accurately quantify the concentration gradient for P in the soils investigated when plants are grown in those soils. If this hypothesis is true, the impact of the P buffer power will clearly be reflected in the P uptake by the test plant since the P buffer power becomes the most crucial factor in predicting P availability. This can be tested by correlating P uptake with the routine soil test data without integrating the P buffer power and also by integrating the P buffer power. Nair and Mengel (1984) used the multiple linear regression equation Y = a + bx + cz, where Y = P uptake by the test plant, x = routine soil test data, z = P buffer power, and a and b = constants, to test the hypothesis. The impact of the P buffer power on P uptake vis a vis P availability is clearly shown by the data in Table 111. In both cases of routine soil test data, the predictability of uptake was remarkably increased by integrating the P buffer power. The coefficient of determination, which is a measure of the precision of these computations, increased from 15 to 79% in the case of EUP-P and from 5 1 to 89% in the case of CAL-P. This in effect means that by integrating the P buffer power, the precision of P uptake predictability increased by 427% in the case of EUF-P and 74% in the case of CAL-P. Obviously, the increase has been very remarkable in the case of EUF-P. This also shows that the quantification of either the intensity factor alone (as in the case of EUP soil test data) or the quantity factor alone (as in the case of CAL soil test data) is insufficient to precisely predict P availability. A substantial proportion of the variance in P uptake is contributed by the P buffer power. From the above arguments, it follows that soils with a low buffer power for a particular
248
K. P. PRABHAKARAN NAIR Table In Correlation Coefficients ( r ) for the Relationship of P Uptake vs Routine P Soil Tests without (A) and with (B) Integration of the P Buffer Power (after Nair and Mengel, 1984)
P uptake vs EUF-P P uptake vs CAL-P
A
B
0.393* 0.714**
0.8870* * 0.9415 * *
Nore. Electro-ultrafiltratable P (EUF-P) and Ca lacacetic acid mixture extratable P tate + Ca actate (CAL-P) (Schiiller, 1969). respectively, used routinely to characterize P availability in central European soils. *,** Significant at 5 and 1% confidence levels, respectively.
+
nutrient requires a high initial concentration of the ions and vice versa. Mengel and Busch (1982) provided experimental evidence to substantiate this in the case of K+ uptake of rye grass growing in nine soils of widely differing physicochemical properties. This would also point to the fact that the relationship between buffer power for the nutrient and the “critical” or “cut off” limit where plant yield is perceptibly reduced have an inverse relationship-a soil with a low buffer power for the nutrient resulting in a higher critical concentration in the plant tissue as compared to a soil with a relatively high buffer power having a lower critical concentration in the plant tissue for the same nutrient. This has been found in the case of K (Mengel and Busch, 1982) and P (Nair, 1988 unpublished data). Nair (1988, unpublished data) found a very highly significant negative correlation between the plant top critical concentration for P and the P replenishment rate in the soil substrate which the P buffer power represents in summer rye (Secale cereale L.). In the case of dry matter yield this relationship was significant, but positive. This suggests that the P buffer power controls the P depletion and/or replenishment in the rooting zone and its accurate measurement would be the key to a precise quantification of P availability. Remarkably, the P buffer power of the soils had no direct relationship to clay or humus content, or soil pH (Nair and Mengel, 1984).
1. P Buffer Power Measurement vs Soil Test P Data for Dependability of Availability Prediction The important question on which the choice between P buffer power and/or contemporary or routine soil tests has to be made depends on what one aims to
BUFFERING POWER OF PLANT NUTRIENTS
2 49
achieve inasmuch as prescriptive soil management practices are concerned. Contemporary soil tests are based on philosophies and procedures which were developed several decades ago, with essentially no change in the general approach. In order to be accurate, a soil test must measure both solution concentration of nutrients and their rates of movement through the soil. The “buffer power approach” seems to accompolish this. Virtually no soil test provides information about the most important factors which govern phosphate availability, which are phosphate concentration in the soil solution and the phosphate buffer power. In the widely used bicarbonate method (Olsen et al., 1954) or using water extraction (van der Paauw, 1971), the P quantities recovered may be 10-100 times more than that found in the soil solution (Mengel, 1985). Thus, all the methods which use an acid extraction extract much more phosphate from the “labile pool” (Larsen, 1967). Acid extractions, in certain instances, can also extract phosphate that under normal crop growth will be found unavailable to plant roots. The discrepancy between contemporary soil tests and crop response to P application has recently been highlighted (Nair, 1992). In an extended study comparing eight central European soils with extremely differing physico-chemical properties, Nair (1992) found that the response to a wide range of P application to summer rye grass (Secale cereale L.) has no relevance at all to routine soil test data. Three routinely used soil tests, namely, the 0.01 M CaCI, test (Mattingly et al., 1963), the CAL-P test (Shciiller, 1969), and the EUF-P test (Nemeth, 1979), were considered. The fluctuations in shoot P concentration with regard to P application were not at all reflected by the soil test values. Two of the soils which were at either extremes of “availability” (as characterized by the soil tests) showed totally divergent responses to P application, while the response was found to be very consistent considering the soil’s P buffer power. This proves that the accuracy of routine soil tests can be vastly enhanced by integrating the corresponding P buffer power of the soil. Buffer curves are extremely important in determining phosphate availability. The equilibration of a soluble fertilizer incorporated into a soil is not merely a question of adsorption and/or precipitation, but also of diffusion. Even if the fertilizer source is soluble and well mixed with the soil, the phosphate distribution is not homogenous, resulting in areas of high and low phosphate concentration. Since phosphate diffusion is a slow process, an even distribution brought about by diffusion takes time. The contact period between phosphate fertilizer and the soil has an impact on P availability, which has practical importance, and this can be depicted by buffer curves as shown in Fig. 1 (Barezkai, 1984). This figure represents on the y axis P quantity (Ca lactate soluble) and on the x axis P intensity (electro-ultrafiltrable) and shows that the solubility of the applied P fertilizer decreased considerably during the 6 weeks of incubation as compared to the time immediately after application. In fact these curves show the P aging process and reflect the phosphate concentration of the bulk soil solution. The steepness of the curve at a given phosphate concentration reflects the buffer
K. P. PRABHAKARAN NAIR
250
=
/ -
7 1 2 weeks
4
8
12
16
20
24
28
32
Soil aolutlon P, mgL-' Figure 1 Phosphate buffer curves of an acid soil obtained immediately after phosphate application and 6 and 12 weeks after phosphate application. After Barekzai (1984).
power and this steepness for eight widely differing central European soils was quantified by computing a regression function between P quantity (Ca lactate soluble P) and P intensity (electro-ultrafiltrable P); the b values in the regression functions were calculated to represent the P buffer power (Nair and Mengel, 1984). Considerable differences can be found among soils in their P buffer power; acid tropical soils invariably have a high buffer power, meaning that at a given quantity of available P, the P concentration of the soil solution is relatively low (Pagel and Huay, 1976). With soils widely differing in their P buffer power, at an identical quantity of available P, the P concentration of the soil solution can differ by a factor of 100 (Barezkai, 1984).
2. P Buffer Power and Q/Z Relationship Almost half a century ago Schofield (1955), while introducing the Q/I (Quantity/Intensity) concept to soil P, argued that the amount of P available to plants is not necessarily dependent on the size of the labile soil P pool and suggested that the work required to remove P from the labile pool might be crucial in determining the uptake of this element by plant roots. In fact, the P buffer power might be an indirect measure of the work required to remove P from the labile pool to affect plant uptake. There has been research on both sorption and desorption Q l l relationships (Raven and Hossner, 1994) and their theoretical representations are
BUFFERING POWER OF PLANT NUTRIENTS
251
shown in Figs. 2 and 3. As soil P release is a key feature in P supply to plants, the study of P desorption Qll characteristics might be more adequate than P sorption Qll characteristics when evaluating plant available P and P buffer power of soils. P buffer power quantification by the sorption method would only be valid if both sorption and desorption curves overlap and if the proportion of solution P in the total desorbed P is negligible and all of the sorbed P is also desorbed (Raven and Hossner, 1994). The success of the adsorption-desorption equilibrium approach in quantifying exactly the P buffer power of eight widely varying central European soils (Nair, 1992) lies in the above fact. However, P sorption and desorption QII curves generally do not coincide and sorption-desorption hysterisis is commonly observed (Fox and Kamprath, 1970; Okajima et a / . , 1983; Barrow, 1983). However, P buffer power values derived from sorption curves appear to be adequate capacity indexes in terms of P uptake by plants (Bowman and Olsen, 1985; Holford, 1988). The latter author, using results from 15 soils of diverse P characteristics, reported that the buffer capacities derived from P adsorption isotherms are satisfactory for characterizing P desorption, as well as adsorption, in the context of P diffusion in the soil and P uptake by plants. He further stated
Qs,
.
...
1st Solutlon P Concentration (Intenrlty) Flgure 2 Theoretical soil P sorption curve and related parameters (Qs,, amount of P sorbed by the soil in situation i; Is,, solution P concentration in situation i; and Bps,, P buffering power at IS,). After Raven and Hossner (1994). 0
2 52
K. P. PWHAKARAN NAIR
0 0
Solution P Concentratlon (Intensity) Figure 3 Theoretical soil P desorption Q / f curve and related parameters (Q,,,, maximum desorbed soil P, labile soil P, or resin-extractable P at a 4/3 soil/resin weight ratio; f,, solution P concentration when no P is desorbed or Q = 0; and BP,, P buffering power at fo). After Raven and Hossner (1994).
that during the later stages of P uptake, buffer capacities were always greater than indicated by adsorption measurements, suggesting that nonlabile P was making a significant contribution to the replenishment of the soil solution. Recent investigations (Raven and Hossner, 1994) indicate that P buffer power values derived from sorption Qlt curves differ from those of desorption Qlt curves. They have reported that test plant P uptake (in this case a single corn plant in a short-term container culture) was better related to the quantity parameter derived from the desorption Q / t relationship (Qmax)as shown in Fig. 4. Fox and Kamprath (1970) developed a standard procedure to determine sorption curves that allow computation of the quantity of P fertilizer required to establish a desired P concentration in the soil solution. Though Beckwith (1965) tentatively suggested that adjustment to 0.2 mg P liter-' would be adequate for successful growth of most plants, these critical values would depend on plant species and nutritional factors (Fox, 1981). Further, recourse to merely a critical solution P concentration value would neglect P buffering effect. Procedures evaluating buffering capacities of soils are
BUFFERING POWER OF PLANT NUTRIENTS
253
1
loo
80
aI Y
0 c.
n
3
0
JP
60
n
40
I Psarnrnenttc Paleudult Typic Calctustoll
0 Udic Pellustert 0 Vertic Albaqualf
20
A y= -4 1
Artdic Calciustoll
+ 2 4~ - 0 0 1 4 ~ ‘R2 =
0 82
0 0
20
40
60
80
100
Qm, (mg Pkg-1) Figure 4 Relationship between the relative shoot P uptake of corn plants and the maximum desorbed soil P (QmaX).After Raven and Hossner (1994).
often time consuming, but the question as to the need for precise documentation of soil parameters quantifying P uptake with precision for field crops, where other important factors may come into play, cannot be underestimated. Empirical correlations based on Mitcherlich-type equations have been fairly successful in extension work, but these equations are restricted to areas with similar soil, climate, and management. Since these equations cannot accommodate soils of divergent phosphate buffering capacities, the need to include a numerical value for P buffering capacity can hardly be overlooked (Bowman and Olsen, 1985b). The usefulness, for field application, of Qll relationships obtained in the laboratory is often questioned. Q / l relationships can give an idea of the relative ease with which the “readily extractable P ’ will be released (buffered)-something that other conventional soil testing procedures do not give (Bowman and Olsen, 1985b). Estimates can be made of the number of years the soil can supply P before deficiency levels set in. Hence, the interpretation of soil tests for soils of identical Q but with a higher P buffer power will need to be different. Another important question is whether it is critical to employ laboratory desorption isotherm procedures merely because the procedure more realistically simulates plant uptake (Bowman and Olsen, 1985b). The scientific understanding of the uptake process seems more appropriately explained by the desorption isotherm process since plant uptake is basically a desorption process. However, Bowman and Olsen (1985b) suggest that where field responses are concerned, adsorption isotherms are more appropriate to standardize P buffer power of soils. The work
254
K. P. PRABHAKAFUW NAIR
of Nair and Mengel (1984) and Nair (1992) suggests that both adsorptive and desorptive procedures are appropriate, the crucial question being how accurately these processes are determined in the laboratory and by plant culture techniques.
As discussed earlier with regard to P, the dynamics of K availability follows the same pattern as that of P, especially in the range of low concentrations. Beckett (1964) has used the activity ratio for K+ and Ca2+ to determine K availability. Since interlayer K would play an important role in K availability, it would be more logical to consider K buffer power in determining K availability. Routinely, it is the ammonium acetate extractant that is widely used to characterize K availability. The reason that this may not be suitable to characterize exchangeable K is that in the routine extraction, only the top layer is extracted, while interlayer K from which deep-rooted plants can feed is ignored (Nair er al., 1994). There is extensive evidence to substantiate this (Mengel, 1985). The importance of interlayer K in the nutrition of deep rooted and perennial crops, such as cardamom, (Elettaria cardamomum M.), the world’s most valuable spice crop, has been highlighted recently by Nair et al. (1994).
1. The Importance of K Buffer Power Determination in Predicting K Availability to Perennial Crops As in the case of P, the K buffer power assumes great importance in predicting K availability, especially with regard to deep-rooted and perennial crops. With annual or biennial crops, K availability has been studied with reference to the exchangeable K. However, with perennial and deep-rooted crops nonexchangeable and interlayer K play a crucial role in K availability. Three soil parameters that control the rate of K supply to plant roots which have been used for predicting K absorption by plants are the K intensity in the soil solution, the K buffer power, and the effective diffusion coefficient (Beckett, 1971; Mengel and Kirby, 1980; Claassen er al., 1986). K buffer power can be directly obtained from the K quantity-intensity relationship. The effective diffusion coefficient depends on, among other factors, the buffer power (Nye and Tinker, 1977). Plants feed not only from exchangeable K but also from nonexchangeable K + , which mainly consists of K+ trapped in the interlayers of nonexpanded 2:l clay minerals (Hoagland and Martin, 1933; Schachtschabel, 1937). Exchangeable K comprises that which can be exchanged with the NHZ ion, and is primarily planar K + , the interlayer K + of expanded 2: 1 clay minerals and some of the K+ at the interlayer edges of nonexpanded 2:l clay minerals. Interlayer K+ of nonexpanded clay minerals such as illites and interlayer and
BUFFERING POWER OF PLANT NUTRIENTS
255
lattice K + micas (present in feldspars) constitutes the nonexchangeable K + . The interlayer K + is of particular importance in the nutrition of deep-rooted and perennial crops such as cardamom, as demonstrated by the recent work of Nair et al. (1994), and also for annual crops such as rye grass (Lolium perenne. cv. Taptoe) (Mengel and Uhlenbecker, 1993). In most of the soil tests for K+ availability, nonexchangeable K is not considered. For cereals such as wheat (Triticum aestivum) 80% of the K+ extracted by the crop came from the non-exchangeable K pool (Jimenez and Para, 1991). This is one of the most important reasons for the poor soil test crop response relationship with regard to K fertilizer application based on such tests (Kuhlmann and Wehrmann, 1984). The contribution of nonexchangeable K to plant availability was assessed by 1 M HCI extraction by Schachtschabel(1961), similar to the 1 M HNO, extraction proposed by Pratt (1985) and McLean and Watson (1985). However, the efficiency of 1 M HCl extraction to quantify plant available K from the nonexchangeable pool has been disputed (Boguslawski and Lach, 1971; Grimme, 1974; Kuhlmann and Wehrmann, 1984). Soils containing primarily 2: 1 clay minerals such as vermiculite and illite have interlayer K in excess of crop demand. However, the availability of interlayer K of nonexpanded minerals is independent of the quantity of interlayer K as such, but dependent on its release rate depending on the type of K+ bearing minerals (Sparks, 1987). Release of K from interlayer positions is an exchange and diffusion process (von Reichenbach, 1972). While exchange depends on the cation species and their concentration near the surface of the mineral, diffusion depends largely on the expansion of the mineral and therefore on soil moisture. Net release of K will only occur if the K concentration of the adjacent solution is low (Mengel, 1985). Martin and Sparks (1983), while studying the release of nonexchangeable K+ from sandy loam and loamy sand extracted with a H+ charged ion exchanger resin, found large release of K + with a K+ concentration of about 1-2 pM in the contact solution. This concentration may approximate the rhizosphere concentration level. Under submerged conditions as in rice, there can be a depletion zone for K in the rhizosphere (Xu and Liu, 1983). Plant roots act as a sink for K and maintain the K solution concentration at very low levels (Kuchenbuch and Jungk, 1984); this would cause further release of interlayer K (Mengel, 1985). These considerations point to the important fact that a precise quantification of K availability, where nonexchangeable interlayer K is concerned, hinges primarily on its release rate, which the K buffer power attempts to quantify, as we shall see in the following discussion. Nair et al. (1994) selected cardamom (E. cardumomum M.) to demonstrate the importance of nonexchangeable and interlayer K on K availability vis-a-vis the K buffer power. The K buffer power curves were constructed by a two-step extraction in which 1 N HNO, was used to determine the quantity of K (Wood and De Turk, 1941) and 1 N NH,OAc was used to determine K intensity. The NH,OAc
256
K. P. PRABHAKARAN NAIR
extractant is universally used to determine exchangeable K. The contribution of nonexchangeable K to plant availability has been assessed both by extraction with 1 M HCl (Schachtschabel, 1961) and 1 M HNO, (Pratt, 1965; McLean and Watson, 1985). Nair et al. (1994) regressed 1 N HNO, extractable K (y) over 1 N NH,OAc extractable K ( x ) to obtain the K buffer power (Table IV). Data in Table IV clearly indicate that the Coorg soils, which had a much higher K buffer power, produced cardamom yield which was twice that obtained in the Idukki soils. The higher K buffer power of the Coorg soils was clearly reflected in the cardamom yield. By comparison, the 1 N NH,OAc extractable K had no significant relationship with leaf K (Table V) and further the integration of K buffer power with the routine soil test K data (NH,OAc extractable K) remarkably improved this relationship. Cardamom is a heavy feeder of K (Sadanandan et al., 1993), and in India, which grows most of this valuable spice crop, and in other countries on the Asian and African continents where this crop is grown, its K fertilizer needs are almost always based on the exchangeable K determined by 1 N NH,OAc. Data in Table V unequivocally show the inability of this extraction to precisely predict K availability; further, data in Table VI show how the situation is remarkably improved by the integration of the K buffer power into the computations. A substantial variation (302.7%) in leaf K is attributable to the K buffer power. These results have been obtained from a very extensive area (more than 20,000 ha), which demonstrates their significance. The K buffer power in this instance integrates both exchangeable (NH,OAc extraction) and nonexchangeable and/or interlayer K (HNO, extraction) and this
Table IV K Buffer Power of Cardamom-GrowingSoils from Two Regions of Southern India Extensively Growing this Crop (after Nair ef al., 1994) Region
Regression function (Y = a + bx)
Coorg Idukki
142.38 + 1.4443~ 592.46 + 0.91721
r 0.8561**
0.5799*
Crop yield (kg ha-l) 155 80
Nore. b values in the regression function represent the K buffer power of the soil. The K buffer power refers to pooled values of soil samples obtained from 94 locations covering an area of more than 20,000 ha in the two cardamom-growing regions of southern India, namely, Coorg and Idukki. Yield data refer to the same locations. *.** Significant at a confidence level of 5 and 1%, respectively.
BUFFERING POWER OF PLANT NUTRIENTS
257
Table V Correlation Coefficients and Regression Functions for the Relationship Leaf K ( Y ) vs Exchangeable K (x, NH,OAc Extraction) (after Nair et al., 1994) Regression function (Y = a + bx)
Details
Coorg 1.2701 0.0004
+
Leaf vs exchangeable K
Y
Leaf vs exchangeable K
Idukki Y = 1.6448 + 0.000006
=
Correlation coefficient (r)
0.2064 -0.006
Note. The correlation coefficients refer to the leaf samples from 94 locations from which soil samples were also obtained to calculate the K buffer power. In cardamom the fifth pair of leaves from the top of each panicle bearing tillers are sampled for K analysis (after Sadanandan et al., 1993).
gives an accurate measurement of K depletion around the plant roots. In a recent study (Mengel and Uhlenbecker, 1993) on K availability from interlayer K to rye grass (L. perenne L.cv. Taptoe) it was observed that the rate constant ( b values) obtained by correlating K released (from the interlayers of clay minerals) and time periods by a modified electro-ultrafiltration (EUF) technique was closely related to K uptake and represented the K availability index from nonexchangeable K . These rate constants, according to the authors, are of the utmost importance because they provide information on the availability of nonexchangeable K in attaining maximum yield; and a set of “critical b” values toward attaining this
Table VI Correlation Coefficients ( r )for the Relationship between Leaf K ( Y ) and Exchangeable K (x, NH,OAc Extraction) for the Pooled Data (94 Locations) from Two Regions (Coorg and Idukki) without (A) and with (B) K Buffer Power Integration (after Nair et ul., 1994) Correlation coefficient Details
A
B
0.2510
0.4367**
~
Leaf vs exchangeable K
** Significant at
1% confidence level.
258
K. P. PWHAKARAN NAIR
objective have been reported. It appears that the rate constants of Mengel and Uhlenbecker (1993) are analogous to the K buffer power values reported by Nair et al. (1994), because, though the techniques differ in their details, they have accomplished the same objective of precisely predicting K availability from the nonexchangeable pool and/or the interlayer K. The capability of tapping interlayer K varies among plant species. For instance, Steffens and Mengel (1979) found that rye grass (L. perenne) could feed from interlayer K+ for a longer period without yield depression, while red clover (Trifolium pratense) could not. These authors reported that since L. perenne had a longer and deeper root system compared to T. pratense, the former could grow satisfactorily while relatively low K+ concentration at which the latter would already suffer from K deficiency (Steffens and Mengel, 1981). The differences in root mass, root length and root morphology between monocots and dicots explain the better K+ feeding capacity from interlayer K + of the former compared to the latter (Mengel, 1985). Cotton, Gossypium hirsutum L., is another deep-rooted long-duration crop, on which the K buffer power exerts considerable influence on K acquisition. Brouder and Cassman (1994) evaluated K uptake by cotton in a vermiculitic soil using mechanistic models and observed that initial model output produced both substantial under- and overpredictions of whole-plant K accumulation. Model predictions were greatly enhanced by estimating the K buffer power. They further concluded that the contribution of the fixed K pool to the plant available K pool was likely to be substantial and that this influence must be captured in estimates of the soil K buffer power. These studies were conducted after observing in a San Joaquin Valley cotton field in California that cotton exhibited late season K deficiency while other crop species remained unaffected. In such cases, the precise estimation of K buffer power will lead to far more dependable K fertilizer recommendations than estimations by the routine NH,OAc extraction. Though it has long been recognized (Schachtschabel, 1937) that the soil K fraction which is not exchangeable by NH, ions (nonexchangeable K) may be important for the supply of K to plants, it is only of late that researchers have paid more attention to this aspect. The work of Sparks and Huang (1985) has very carefully examined the release mechanism from the nonexchangeable source and the factors controlling it. Considerable portions of initially nonexchangeable K can be utilized by plant roots even within a few days (Kuchenbuch and Jungk, 1984). The depletion zone, however, extends into the ambient soil for 2 mm only. Hinsinger et al. (1992) embedded phlogopite in agar and observed that the interlayer K of this mineral was entirely lost in the close vicinity of ryegrass roots within 4 days. Since the process limiting the rate of K uptake in the rhizosphere may be K transport through the soil rather than the release from minerals as such, some researchers have focused their attention on this aspect. One such example is the mechanistic mathematical model of Claassen and Barber (1976). Claassen el al. (1986) and Claassen (1990) have successfully applied the model referred to
BUFFERING POWER OF PLANT NUTRIENTS
2 59
above to predict K depletion profiles in soil around plant roots. Quite recently, Meyer and Jungk (1993) have used these models to predict K uptake by test plants from exchangeable and nonexchangeable K sources. They reported that 64-79% of the K taken up by wheat (T. aestivum L.) and sugar beet (Beta vulgaris L.) was derived from the rapidly released exchangeable and 21-36% from the nonexchangeable or less mobile soil K fraction. The buffer power describes the relationship between adsorbed K and the K concentration of the ambient solution. In simulation models it is assumed that this relationship is linear and hence independent of soil solution concentration. However, in desorption studies with soil a sharply curved buffer relationship could be found and Meyer and Jungk (1993) have referred to it. Very near the plant roots the soil can be subjected to a curved buffer function since plant roots strongly reduce the soil solution concentration. An important aspect to be considered in the utilization of nonexchangeable K is the role of plant roots. Plant species differ in their ability to utilize nonexchangeable K and this has been attributed to the differences in root length (Mengel and Steffens, 1985). Radial distance between two single roots decreases, consequently increasing root density and this would result in the overlapping of the depleted soil volumes between these roots. This would also lead to a decrease in rate of K uptake per unit root. In the case of the rapidly diffusing K fraction, which has a higher mobility, the competition effect between roots could be very intense. There is evidence to support this view, as shown by the work of Mitsios and Rowell (1987), who observed that the contribution of nonexchangeable K increased with a corresponding increase in root density. Additionally, the differences in root hair length and density among plant species (Fohse et a l . , 1991) affect their ability to acquire soil K . Accordingly, the work of Meyer and Jungk (1993) has shown that K uptake was higher when they included root hairs as well in their model calculations. Since root hairs contribute to an increase in root absorbing surface, a reduction in the distance of diffusion from the site of K release to the site of K uptake, and an increase in the K concentration gradient, they can be expected to exert a pronounced effect on K availability from the less mobile K fraction.
2. The Commercial Significance of K Buffer Power Determination in K Fertilizer Management for Perennial Crops The commercial significance of K buffer power determination for dependable K fertilizer recommendations assumes great importance in those countries which are faced with importing these fertilizers at a huge cost to the national exchequer. India is a case in point. The recent decontrol of prices of both phosphate and potassium fertilizers by the government of India resulted in an overnight escalation of their market prices. In a situation like that, the farmers become extremely
260
K. P. PRABHAKARAN NAIR
wary of their field use and unless the fertilizer application is cost effective, faith in their use, especially those mentioned above, would be shattered. The K fertilizer recommendation for cardamom has been based exclusively on NH,OAc extraction. The investigation of Nair et al. (1994) showed its ineffectiveness. Though the importance of K buffer power in predicting K availability has been reported earlier, these research reports related mainly to annual crops such as white clover (During and Duganzich, 1979) and rye grass (Mengel and Busch, 1982); the work of Nair et al. (1994) was the first of its kind in perennial crops.
3. The Role of the NH$ on K+ Availability One of the frequent assumptions made in predicting K availability in soils is that results from binary (two-ion) exchange systems can be extrapolated to ternary (three-ion) systems by using appropriate equations. The K-Ca exchange reactions in soils are often investigated in laboratory studies. Most of the research carried out on soil clay minerals and soils as exchanger surfaces (Vanselow, 1932; Gapon, 1933; Argersinger e t a l . , 1950; Sposito, 1981a,b; Sposito et al., 1981, 1983; Jardine and Sparks, 1984) are binary exchange systems. However, field soils are at least ternary systems (Adams, 1971; Curtin and Smillie, 1983). The evaluation of soils as binary systems implies that these reactions can be used to predict results in ternary systems such as field soils. For this assumption to be valid, the binary exchange selectivity coefficients need to be independent of exchanger-phase composition (Lumbanraja and Evangelou, 1992). But, the work of Shu-Yan and Sposito (1981) showed that it is impossible to predict exchange phase-solution phase interactions in a ternary system such as the field soil from a binary system such as the laboratory sample. This focuses the importance of the ternary systems. As far as K availability is concerned, it would be important to include the NH,+ ion as well. The work of Lumbanraja and Evangelou (1992) has shown that K+ adsorption to soil surfaces is suppressed in the presence of added NH;, while the adsorption of NH; to the same surface is enhanced in the presence of added K+. These observations point to the influence of added NH,+ on the desorption potential (chemical potential) of adsorbed K or vice versa (Lumbanraja and Evangelou, 1992) and would be relevant to the determination of K buffer power especially when agents containing NH; ions, such as NH,OAc, are used in determining K buffer power (Nair et al., 1994). The work of Lumbanraja and Evangelou (1992), though, clearly demonstrates the effect of the NH,+ ion on K+ desorption, with an increase in K+ desorption in the presence of added NH,+. In its absence, it might be safe to conclude that the shape of the K buffer power curve will not appreciably change even if larger quantities of K are removed due to cropping and therefore can be considered as a relatively constant property of soils. There is evidence to support this view
BUFFERING POWER OF PLANT NUTRIENTS
261
(Jimenez and Para, 1991). These authors, while investigating the Q/I relationship on K uptake by wheat (T. aestivum) in calcareous vertisols and inceptisols of southwestern Spain, found that 80% of the K extracted came from the nonexchangeable K pool. These observations, coupled with the ones discussed earlier, suggest that the precision of predicting K availability can be substantially enhanced by first quantifying the K buffer power of the soil in which the crop is intended to be grown. Admittedly, the rate-limiting steps involved in K dynamics are not entirely understood (Sparks, 1987). Notwithstanding this limitation, if one must move forward in devising better management of K fertilizers in crop production, a starting point has to be made with regard to precisely quantifying K availability. Quantifying the K buffer power of soils and basing K fertilizer recommendations on this seems to be the best starting point.
4. K Buffer Power Measurement vs Contemporary Soil Test K Data for Dependability of K Availability Prediction The Context of MYR (Maximum Yield Research) and MEY (Maximum Economic Yield) Approaches
-
The importance of K buffer power determination in devising effective K fertilizer recommendations must be viewed against the twin objectives of maximizing fertilizer use efficiency to attain high crop productivity-the crucial concern of developing countries-and minimizing nutrient loss to the environment-the crucial concern of industrialized countries. Potassium losses are primarily due to leaching and occur on coarse-textured soils under humid conditions (Mengel, 1985). Where such leaching losses are acute, split application of K fertilizer adjusted to the crop demands and to the fluctuating weather conditions are considered a useful measure to enhance efficiency of K use by crops. Despite the fact that predictive kinetic models of K dynamics are absent and the K-rate limiting steps are still inadequately understood (Sparks, 1987) (a better understanding would doubtless result in better prediction of the fate of applied K and, consequently, assist in enhancing crop yields and K uptake efficiency, through more effective K fertilizer recommendations), the K fertilizer recommendations still continue to revolve around quantifying available K, mostly by standard extractions with NH,OAc or acids such as 1 Nor 1 M HN03 or 1 M HCI. A case in point is the MYR (Maximum Yield Research) or MEY (Maximum Economic Yield) approach. The recent findings of Narang et al. (1994) in India indicate that in a rice-wheat rotation of five crop seasons, the decline in available K was 12.3 pg g- * at 120-30-30/ 120-60-30 (N-P-K rates) and 15.6 pg g-' at 180-3030/ 180-60-30 (N-P-K rates) over their respective initial levels, pointing out the anomalous and inconsistent fact that at identical K application rates the reduction in available K is variable. Obviously, such a conclusion is subject to
262
K. P. PRABHAKARAN NAIR
scrutiny from the viewpoint both of quantifying available K (in this case the standard NH,OAc extraction was employed) and of the varying N and P rates where interaction between the three nutrients (N, P, and K) is bound to play a role in their individual availability. However, one cannot ignore the inadequacy of such approaches, because these contemporary soil tests fail to detect precisely the differences in soil K availability. Contradictory results have been obtained where higher K fertilizer application resulted in a corresponding higher available K (Singh and Nambiar, 1986; Narang and Bhandari, 1990; Roy et al., 1990) while the same practice resulted in a decline of extractable K (Swarup and Singh, 1987; Hegde and Dwivedi, 1992). Obviously, these inconsistent results cast a deep shadow of doubt on the dependability of contemporary soil tests for K availability in soils. A similar situation prevails in the case of rice production for MEY in Costa Rica. Using a modified Olsen’s extractant, a mixture of NaHCO, EDTA, Cordero and Espinosa (1994) classified soil K critical levels as low (0.2 cmol, liter-I). The K critical level was set at 0.1 cmol, liter-’ and the K fertilizer recommendations varied from 30 to 60 kg K,O ha-I . But responses to K fertilizer up to 80 kg K,O ha-1 were obtained even in soils with a K critical level of 0.15 cmol, liter-’ in Vertisols and Mollisols with high Ca + Mg contents (Cordero and Espinosa, 1994) indicating that the K critical level based on the soil extraction is not a dependable index of availability and the situation can vary from one soil to another. The above situation would call for different K fertilizer recommendations for different soils based on extractable K and the Ca Mg/K ratio. This disallows K fertilizer recommendations in a universal manner. These are not the limitations which the K buffer power approach suffer from. Since the K buffer power quantifies the average K concentration on the plant’s root surface, though indirectly (Nair, 1984), it would provide greater precision in K availability prediction and, hence, result in more dependable and uniform K fertilizer recommendations. The supportive evidence provided by During and Duganzich (1979), Mengel and Busch (1982), Mengel and Uhlenbecker (1993), and Nair et al. (1994) reinforces this assumption. The MYR approach focuses on the BMP (Best Management Practice) and aims to maximize crop yields wherever possible. Yet some of the key inputs, such as decisions on fertilizer rates, are still based on contemporary soil tests based on philosophies and procedures developed several decades ago with essentially no change in the general approach. In an interesting paper entitled “Maximum Yield Research: Friend of the Environment,” presented at the 15th World Congress of Soil Science in Acapulco, Mexico, Ludwick (1994) makes a strong case for this approach in maximizing crop production. He cites Norman Borlaug (1990) the Nobel Peace Prize Winner, and Borlaug and Dowswell (1993) who said,
+
+
“The only way for agriculture to produce sufficient food to keep pace with population and to alleviate the hunger of the world’s poor is to increase the
BUFFERING POWER OF PLANT NUTRIENTS
263
intensity of agricultural production under those ecological conditions which lend themselves to intensification while decreasing the intensity of production in the more fragile ecologies. Most of the yield increase in food production needed over the next several generations must be achieved through yield increases on land now under cultivation. Moreover, these yield increases must be achieved through the application of technology already available or well advanced in the research pipeline. This will not only lead to economic development but it will also do much to solve the serious environmental problems that come as a consequence of trying to cultivate lands that are not suited for crop production. Fortunately many of the more-favoured agricultural lands currently under cultivation are still producing food at yield levels far below their potential.” He further cites Cassman (1990) who mentions that the greatest challenge to global food output is to understand how soil fertility management affects soil quality, and to apply this knowledge to integrate all aspects of crop management to optimize yield and input utilization efficiency over the long term. Soil fertility management governs plant nutrition, which in turn influences susceptibility to disease, plant growth and yield.
In a nutshell, the MYR approach lays great stress on nutrition, yet the nutrient management aspects move along the beaten track. Unless imaginative attempts are made to revise the basic concepts of quantifying nutrient availability, even the MYR approach is bound to falter as has been clearly shown in the case of K recommendations for rice-wheat rotations in India and rice production in Costa Rica. These are but a few examples which have been reported and are reviewed here, and yet there could be many more which are not reported and which are outside the purview of this review. These facts call for a new direction in soil testing for nutrient availability. The buffer power concept appears to provide this. The field experience of this new approach is detailed in the book “The Spirit of Enterprise-The 1993 Rolex Awards” which summarizes the work of Nair (1992) in the farmers’ fields in the Republic of Cameroon.
5. K Buffer Power and Q/Z Relationship The use of the QlI relationship to describe the K status of soils is based on Woodruff’s ( 1955a,b) observations that K availability to plants can be characterized by the free energy of K-Ca exchange with regard to solid-solution reactions. Beckett (1971) listed a number of assumptions as prerequisites for Woodruff’s observation to be valid. An important assumption is that the rate of K uptake by plant root must be regulated by “the difference in the equivalent free energies of K and Ca as offered to the uptake sites.” Maas (1969) has provided
K. P. PRABHAKARAN NAIR
2 64
experimental evidence to substantiate this from uptake data of excised maize roots. These data demonstrate that Ca*+ uptake in the presence of K + is a competitive process. The rate of K uptake was maximized in the AG range predicted by Woodruff (1955b). Such experimental evidence of K uptake by excised maize roots provides supportive proof to the validity of Woodruff’s observation that K availability could be described by the free energy of K-Ca exchange as long as all other soil factors related to plant growth are not limiting and are held constant (Beckett, 1972). The use of Qlt in predicting K availability to plants in soils has been widely tested (Beckett, 1972; Bertsch and Thomas, 1985). However, it has been shown that the Qlt relationship does not enjoy universal acceptance since a single relationship for all soils between K uptake by a given crop and ARK does not exist, perhaps due to the nature of the soil components regulating ARK (Evangelou er al., 1994) (Fig. 5). The reader is advised to read the excellent review of Evangelou er al. (1994) for a better appreciation of the new developments and perspectives on soil K Qlt relationships. The reasons for the failure of the Qlt concept to precisely predict soil K availability in all soils are varied and an important aspect that has to be recognized is that Q l t concept does not deal with physiological effects, but appears to
2.0 1.5
-2
1 .o
0 5 0.5
I
0 Binary (K-Ca) A Ternary (K-NH,-Ca)
E”
x-
a
-
Predicted Ternarv fK-NH.-Ca\
0
4
-0.5
-1 .o
EXK (1 M NH,C,H,O,) -1.5 0
0.005
0.010
0.015
0.020
AR, (mol L-’)l’*
Rgure 5 Potassium quantitylintensity ( Q / l ) plots of Eden soil in the absence (binary) and presence (ternary) of NH, [ternary prediction is from experimental binary Vanselow exchange selectivity coefficients for K-Ca and NH,-Ca exchange, and from experimental ternary activity ratios (AR) for K and NH,; ExK is exchangeable K]. After Lumbanraja and Evangelou (1990).
BUFFERING POWER OF PLANT NUTRIENTS
265
be influenced by them (Evangelou et al., 1994). Mengel and Busch (1982) conducted an interesting investigation dealing with some of these physiological effects. They demonstrated that a very close correlation (r2 = 0.908), reflected by a parabola ( y = c/x; Figs. 6 and 7), existed between K buffer power and the critical K + concentration of the soil solution. They estimated the K buffer power by correlating exchangeable K (0.2 N BaCl, exchangeable) as y (K, quantity) with soil solution K (electro-ultrafiltrable) as x (K, intensity), and the b values in the regression functions were taken as a measure of the K buffer power. Nine central European soils of extremely differing natures were tested for K availability to a test crop of L . multijlorum (cv. Avance). These authors reported that as the K buffer power increased, the critical K concentration (the critical K concentration defined as that level below which a yield increase is obtained when K fertilizer is applied) decreased. Though the Q/I concept does not deal with physiological effects, the latter influences it. For a better appreciation of the Q / l concept vis-a-vis the K buffering effect on prediction of K uptake, one should be quite familiar with plant nutrition concepts. Even if nutritional effects are assumed to play only a limited role in K uptake under a well-defined experimental set-up, the ability of a soil to replenish K+ in the soil solution must be considered a highly complex process Mg) exchange alone (Evangelou et al., and not solely dependent on K-(Ca 1994). K availability is also controlled by additional factors, such as hysteresis or
+
t
o t , 0
I 1
2
'
4
1
'
6
1
~
I
I 1
/ ~
I 1
I '
I 1
8 1 0 1 2 l b
. '
K+ buffer power x10
Figure 6 Relationship between the critical K + concentration of the soil solution and the K + buffer power: y = critical K + concentration; x = K+ buffer power. After Mengel and Busch (1982).
2 66
K. P. PRABHAKARAN NAIR A
120-
A
110100-
;I: Q)
5i
so--
A
y = b.x + a r* = 0,543
(0
0)
c (0 c
8070--
b = 5,106 a = 42,7
A
Q)
60-A A
50-A
40-
A A I
I
I
.
I
1
1
.
1
.
K+ buffer power x10 =
Figure 7 Relationship between the critical level of exchangeable K + and the K + buffer power: y critical level of exchangeable K + ; x = K + buffer power. After Mengel and Busch (1982).
exchange irreversibility effects, anion effects, multication effects, potentialdetermining ion effects, and kinetic effects (Evangelou et al., 1994). To predict K availability with precision, these factors need to be further scrutinized. But, most importantly, it is the soil/soil solution-plant root surface interactions which merit closest scrutiny in predicting K availability. The importance of K buffer power has to be examined in this context. Since the ultimate objective in perfecting prescriptive soil management practices with regard to K fertilizer application, vis-a-vis soil testing for plant available K, hinges on the soil/soil solution-plant root interactive effects, and also since plant root surface K concentration during the dynamic state of plant growth is nearly impossible to measure in a field soil, Nair (1984) has argued that a precise quantification of K buffer is an important prerequisite to accurately predict K availability and must be considered as such as an indirect measure of the plant root surface K concentration. The experimental evidence provided by Mengel and Busch (1982) and Nair et al. (1994) substantiates this view.
BUFFERING POWER OF PLANT NUTRIENTS
267
V. THE ROLE OF ELECTRO-ULTRAFILTRATION IN MEASURING P AND K INTENSITY FOR THE CONSTRUCTION OF BUFFER POWER CURVES In the construction of the buffer power curves it is imperative that both the quantity and the intensity factors are measured accurately. While the former can be quantified by an incubation and extraction technique (Nair and Mengel, 1984) or an adsorption-desorption equilibrium technique (Nair, 1992), the precise determination of the latter poses greater problems. Here, the role of electroultrafiltration in quantifying the intensity factor is briefly discussed. Electroultrafiltration is a flow and batch technique combined to determine the kinetics of K availability (Nemeth, 1971; Nemeth, 1979). It combines electrodialysis and ultrafiltration. Originally electro-ultrafiltration was used to purify colloidal suspensions (Grimme, 1979) but it was later modified by Kottgen and Diehl (1929) for soil testing purposes. These authors attempted to extract the labile fraction of the plant nutrients without a chemical treatment such as chemical extraction. The EUF method eliminates accumulation of reaction products in the electrode compartments, and also charge on membranes separating the compartments as well as the effect of electroosmosis (Nemeth, 1972; Grimme, 1979). Nemeth (1971, 1972) modified the procedure to extract soil at three field strengths instead of one as done originally. Thus, an extraction (desorption) time of 10 rnin at 50V and 20°C was followed by two sequential extractions of 10 min each at 200V and 20°C and five sequential extractions of 5 rnin each at 400V and 80°C. The extraction period can be modified for different time sequences. For instance Nair and Mengel (1984) collected three fractions, EUF-I + EUF-11, extracted at 200V and 20°C for 30 min and EUF-Ill extracted at 400V and 80°C between 30 and 35 min, in the case of P, while Mengel and Uhlenbecker (1993) collected EUF-I extracted at 200V and 20°C for 30 min followed by EUF-I1 extracted at 400V and 80°C for another 30 min in the case of K (Fig. 8). Each extraction period was 5 min. In the case of P, Nair and Mengel (1 984) considered all the three fractions as a measure of P intensity, while in the case of K, the first two fractions (5-10 min) were considered a measure of K intensity (Sparks, 1987). Since the stepwise extractions will provide an assessment of the changes in the mobility of the nutrient, the slopes of the desorption curves will indicate the rate at which the available quantity of the nutrient will be exhausted. For K , the ratio of the seventh and second fraction has been taken as a measure of K buffer power of the soil in question (Sparks, 1987). Nair and Mengel (1984) and Nair (1992) correlated the EUF fractions with CAL-extractable P (Schuller, 1969) and desorbable P (Nair, 1992), respectively, and the b values in the regression functions were used as a measure of P buffer power. The very high r values obtained in
2 68
K. P. PRABHAKARAN NAIR cmol K+ kg-I
0
I
W Vertical
6
4
-
Entisol
A
Alfisol2 Alfisol 1
-
10
20
I 2 0 0 V / 2 0 OC
30
40
50
400V/80 OC
60mn -I
FEgure 8 Cumulative K + extracted by electro-ultrafiltration(EUF). Extraction at 200 V, 20"C, 30 min (EUF,-,, fraction); extraction at 400 V, 80"C, 30 min (EUF,,-,, fraction). After Mengel and Uhlenbecker (1993).
both of the cases discussed earlier indicate that these b values will indeed provide a very dependable measure of P replenishment rates in the rooting zone which the P buffer power, in fact, attempts to measure. While EUF can provide dependable information on both intensity and quantity factors of P and K (Simrad and Tran, 1993; Simrad er al., 1991), it must be pointed out that this facility may not be universally available and is, of course, quite expensive to provide. Thus, in routine soil testing procedures, there is a need to examine alternative procedures to estimate the P and K intensity factors. Nair (1994, unpublished data) has used extractions with 0.1 M CaCl, in tropical African and Asian soils to quantity P and K intensity. Thus, it is relevant to mention that wherever an EUF facility exists, it could be used to quantify the quantity and intensity factors, and where it is not available, alternate procedures such as extraction with 0.1 M CaCl, can be used to measure K and P intensity factors. But in both cases, the essential point is that quantifying the P and K buffer power of soils and integrating the buffer power into the computations would provide a much better picture of the soil availability of these nutrients than contemporary soil tests alone. Additionally, EUF has the possibility to simultaneously provide information on quantity and intensity factors of other plant nutrients, besides P and K, at the same extraction cycles, as for Zn (Nair, 1984). In a recent study (Steffens, 1994), EUF has been used to study P kinetics. He quantified P kinetics by the Elovich b values [similar to that used by Mengel and Uhlenbecker (1993) in the case of K] by stepwise extraction using electro-
BUFFERING POWER OF PLANT NUTRIENTS
2 69
ultrafiltration. In all, I 1 fractions were used to study the P kinetics. The first two fractions were obtained, respectively, at 20°C, 200 V, and <15 mA during an extraction period of 30 min and subsequently at 80°C, 400 V, and <150 mA during an extraction period of 5 min, that is, a total of 30-35 min. This procedure was similar to that adopted by Nair and Mengel (1984) and Nair (1992). After these two fractions were extracted, nine other fractions were extracted at 5-min intervals at 80°C, 400 V, and <150 mA. The Elovich b values were computed using mg P kg-1 soil extracted vs time. P uptake was studied during a 4-year period using soils which were treated with basic slag phosphate, super phosphate, partially acidulated phosphate rock, and phosphate rock (Fig. 9). The test crops which were successively grown were rye grass (L. perenne L., cv. Taptoe) in the first year, rape (Brassica nupus L., cv. Akella) in the second year, corn (cv. Badischer Landmais) and summer wheat (T. aestivum L., cv. Selpek) in the third year, and summer rye (Secale cereale L., cv. Sorom) in the fourth year. The Elovich b values were more closely correlated to cumulative plant P uptake than water soluble P (van der Paauw, 1971), CAL-Calcium acetate + calcium lactate + acetic acid mixture extractable P (Schiiller, 1969), or Mehlich 3 extractable P (Mehlich, 1984). The cumulative plant P uptake from the 15 test soils was very well predicted ( r = 0.99) when the Elovich b values were also inte-
180
-
X
PR
r = 0.9ar**
y
-
r* = 0.97 n = 15 27.8 t (O.~*.CALP)+(O.W'**.b)
'Predicted" P, mg k g '
Figure 9 Relationship of measured and predicted plant P uptake from pot experiments during a 4-year period (where C A L P denotes the amount of calcium acetate-calcium lactate-acetic acidextractable P, and b denotes the slope term of the electro-ultrafiltrationextraction curves as described by the Elovich equation). Po, without P fertilization (control); PR, phosphate rock; PAPR, partially acidulated phosphate rock; BSP, basic slag phosphate; SP, superphosphate. *,***Significant at the 0.05 and 0.001 probability levels, respectively. After Steffens (1994).
270
K. P. PWHAKARAN NAIR
+
grated into the multiple regression model y = a bx + cz used earlier by Nair and Mengel (1984) and Nair (1992), where y denotes the plant P uptake, x denotes the CALP, and z denotes the Elovich b value. This indicates that considering the Elovich b value as a parameter for P release and the quantity of CAL extractable P for determining P availability following different P fertilizer applications provides a much more reliable estimation of plant P uptake than consideration of only a single parameter of P availability such as the water extractable P, CALP, or Mehlich P. The Elovich b values used in this investigation are analogous to the P buffer power values of Nair and Mengel (1984) for P, who, for the first time, showed that precise prediction of P availability can only be done when both the labile P and the P intensity are integrated within one framework. Larsen (1967) had postulated such a view much earlier, but experimental evidence utilizing the EUF facilities to measure P intensity as shown by the work of Nair and Mengel (1984) and Nair (1992) for P and by Nair (1984) for Zn came much later. All of these studies underscore the great importance of EUF in modem-day soil testing methods.
VI. QUANTIFYING THE BUFFER POWER FOR PRECISE AVAILABILITY PREDICTION -HEAW METALS A. ZINC Since Zn, among the heavy metals, has become the most problematic on a global scale, the scope of this review is confined to this very important plant nutrient. There is a great paucity of published material on the effect of buffer power on availability of heavy metals. Plants obtain most of their fertilizer Zn from reaction products and not applied sources as such, implying that any source of Zn added to soil has to necessarily conform to a chain reaction involving adsorptive, desorptive, and resorptive processes that govern the maintenance of an equilibrium between adequate Zn concentration in the soil solution nearest to the zone of Zn depletion on the one hand and plant uptake on the other. The Zn buffer power defines this. As Zn concentration in soil solution is normally very low, the supply to plant roots by mass flow can only account for a very small fraction of plant demand. For instance, with a transpiration coefficient of 300 liters kg-I dry matter and a corresponding Zn concentration of 10-7 M in the soil solution, approximately 2 mg of Zn can be supplied by mass flow against a demand of 10-30 mg Zn kg-1 dry w of plant tissue. In calcareous soils, as the Zn concentration is of a much lower order of approximately lo-* M, the supply by mass flow could be 'very much lower (Marschner, 1994) indicating that mass flow can only contribute
BUFFERING POWER OF PLANT NUTRIENTS
271
very negligibly to meet plant needs of Zn. Hence, Zn movement to the plant root surface is principally by diffusion and is essentially confined to a zone around the plant root which hardly extends beyond the root hair cylinder (Marschner, 1994). Though published papers on the relevance of Zn buffer power in soils to Zn availability are very limited, there is no dearth on other aspects. Genotypic differences to Zn nutrition in a very important crop such as maize, with regard to its hybrids, inbreds, and composites, was reported almost two decades ago in India (Nair and Prabbat, 1977); this study focused on the importance of genetic engineering for tailoring maize varieties for better efficiency of Zn utilization, while the complicated mechanism of Zn-P-Fe interaction was researched even prior to this (Nair and Babu, 1975). In a recent review on the mechanism of Zn uptake, Marschner (1994) indicated that flow culture experiments with various species showed adequate ranges of Zn concentration in the range of 6 x 10-8 to 8 x 10-6M which are concentrations greater than those that would be expected in the soil solution of most soils. He further pointed out that although work using chelate-buffered solutions has indicated adequate Zn concentrations between 10-10 and 10-11 M, extremely low adequate Zn concentrations required a concomitant excess of about 100 pM Zn-chelate as buffer at the plasma membrane of the root cells. This implies a need for an unlimited Zn pool for replenishment of Zn*+ at the plasma membrane. When plants grow in soil, it is impossible to expect a Zn buffer of this size to exist, and free Zn2+ and chelated-Zn concentrations will be at least threefold lower. Hence, critical deficiency or sufficiency concentrations obtained through research employing chelate-buffered solutions cannot be applied to soil-grown plants. Most of the work on Zn availability to plants in soil is based on chemical extractions among, which DTPA is the most frequently used. The DTPA extraction quantifies a labile fraction of soil Zn comprising water soluble, exchangeable, adsorbed, chelated, and some occluded Zn. The critical soil level of DTPA extractable Zn can vary from 0.3 to 1.4 mg kg-' soil which equates to about 900-4200 g ha-1 of Zn in heavy soils and about 600-2800 g ha-' Zn in light soils in the plough layer (0-20 cm). The crop requirements, on the other hand, are quite small, in the range of 100-300 g ha-' for a total dry matter production of about 10 t ha-' (Marschner, 1994). The inadequacy of DTPA extraction to reflect plant Zn demand shows that other important factors, such as replenishment of soil solution Zn (Nair, 1984), mobility, and transport to the root surface (Wilkinson et a l . , 1968; Nair et al., 1984), and also the activity of the roots themselves (Wilkinson et al., 1968; Marschner, 1994), are involved. Since the Zn buffer power is intricately involved in all three factors, the focus of this review is mainly on that attribute. As early as three decades ago it was suggested that colloidal Zn was released by some specific process associated with root activity (Wilkinson et al., 1968). Conditions in the rhizosphere and particularly root-induced changes markedly
2 72
K. P. PRABHAKARAN NAIR
affect Zn availability. A difference in rhizosphere pH of as much as 2, higher or lower compared to bulk soil, can be expected to occur as a result of imbalance in ionic uptake. For instance, any acidifying fertilizer such (NH,),SO, can result in a net excretion of H+ ions and others such as NH,NO, can result in a net excretion of HCO, or OH- ions. Additionally, secretion of organic acids and enhanced CO, production, as well, will affect rhizosphere pH, and all of the above-mentioned changes will markedly affect Zn availability. However, the scope of this review is confined to the kineticldynamic aspects of the changes occurring in the rooting zone mirrored in the Zn buffer power rather than changes in soil reaction in the rhizosphere per se on Zn availability. The distribution of Zn between the solid and solution phases can be described by the buffer power. The availability of soil Zn to the plant depends on the initial Zn concentration, Zn buffer power and effective diffusion coefficient (Barber, 1984).The Langmuir equation gives the relation between B and C, as
where C, = Zn concentration in the soil solution, xlm = the amount of Zn adsorbed per unit of soil, B = adsorption maximum, and a = a constant related to the soil’s bonding energy for Zn. A straight line is obtained when C,l(xlm)is plotted against C , with a slope of 1lB and intercept of 1laB. The inverse of C , / ( x l m ) = b, the Zn buffer power, where C, and xlm are both expressed in volume units (Barber, 1984).Using this approach, Shuman (1975)estimated the buffer power values varying from 5 to 100 for four soils representing different major physiographic regions of Georgia. Based on the diffusion model of Drew et al. (1969),Nair (1984)has argued that the E in the equation U = 27~01act (Drew et al., 1969) where U = quantity of Zn absorbed per centimeter root length, a = root radius in cm, 01 = root absorbing power, c = average Zn concentration on the root surface, and t = duration of the absorption period, in fact represents an indirect measure of the Zn buffer power. As we already know, the bulk of Zn uptake is by diffusion (Wilkinson et al., 1968;Elgawhary et al., 1970;Barber, 1984).This diffusive process will maintain a concentration gradient in the rooting zone. This concentration gradient will directly affect Zn uptake because of its effect on the average Zn concentration on the root surface. The Zn buffer power will affect this concentration gradient, because the rate of Zn depletion and/or replenishment is mirrored by it. In a sense the effective diffusion coefficient and the buffer power are analogous for nutrients which are principally absorbed by the plant root through diffusive processes (Nair, 1989). Hence, the crucial question to examine would be the role of Zn buffer power in influencing Zn availability for plant uptake.
BUFFERING POWER OF PLANT NUTRIENTS
273
B. QUANTIFYING ZN BUFFERPOWER Nair (1984) used a novel method to quantify Zn buffer power. Heavy metal pollution is a major environmental hazard in Europe, especially on the highways due to automobile exhausts; as a consequence, a number of soils are very highly contaminated. Such soils are unsuitable for a precise study on the impact of Zn buffer power on plant Zn uptake, and the choice of soil was narrowed to a heavy clay in the northern part of Belgium containing 1.1 ppm total Zn, 5.5 ppm DTPA extractable Zn, and 0.13 ppm electro-ultrafiltrable Zn. The soil was acidic (pH 6.6, 1 N KCl), low in organic carbon (1.8%), high in CaCO, (16.5%), containing beidellite as the predominant clay mineral. Zn buffer power was quantified by regressing DTPA extractable Zn (y) on electro-ultrafiltrable Zn (x) by a stepwise extraction of incubated soil, and the b value in the regression function represented the Zn buffer power. Electro-ultrafiltration has been employed for a rapid estimation of the intensity of a range of heavy metals, especially in central, western, and eastern European soils (Nemeth, 1979; Nemeth and Recke, 1982). However, earlier investigations of these authors did not give satisfactory results: analysis of only the alkaline cathode filtrate is insufficient to give precise information about availability indices, primarily due to the fact that the heavy metals are collected by the cathode filter as hydroxides and hydrated oxides, which are hardly soluble in the alkaline cathode filtrate. However, as these oxides comprise soluble and desorbable ions of the heavy metal in question, they are of importance to plant nutrition; Nair (1984) used a modified procedure to bring the oxides into an acid medium for subsequent analysis. The studies indicated that Zn intensity alone is insufficient to precisely predict Zn uptake by maize used as a test plant in the investigation. The relevant data are given in Table VII. Data in Table VII show that the precision of predicting Zn concentration in the
Table VII Correlation Coefficients ( r ) between Zn Intensity and Maize Top Zn Concentration and Total Zn Uptake by Plant Top at Harvest without (A) and with (B) Zn Buffer Power Integration (afterNair, 1%) Correlation coefficient Details
A
B
Zn intensity vs maize top Zn concentration Zn intensity vs total Zn uptake by maize top
0.290
0.812*** 0.618**
~
0.356
~~
****** Significant at
1 and 0.1% confidence levels, respectively.
274
K. P. PWHAKARAN NAIR
maize top increased remarkably, as substantiated by the increase in the coefficient of determination which works out to 65.9%. With regard to total Zn uptake by the maize plant, the increase is less marked, yet substantial. The data show that the major factor responsible for the variance in Zn concentration in the maize top and total Zn uptake by the maize top is the Zn buffer power, and not Zn intensity as quantified by electro-ultrafiltration. In a separate experimental set-up, Nair et al. (1984) monitored Zn transport by simulation experiments in a laboratory set-up using the same field soil and obtained proof to support the hypothesis that it is essentially the buffered Zn that contributes to the major uptake process by the test plant. Dissolved salts in a field soil must move with the liquid, the net movement from a given reference plane being dependent on the activity gradient of salt and water. This naturally brings us to the important question of solute movement within the soil matrix which is dependent on the soil conductivity. Buffered Zn would be in the bulk soil solution to be further transported to plant roots. This transport, though dependent on the soil conductivity, would also be linked to the quantum of Zn per se in the bulk soil solution, which, in turn, would be dependent on the Zn buffer power of the soil in question. Nair et al. (1984) quantified these relationships. In a field soil, it is the buffered Zn which maintains a concentration gradient between the bulk soil solution and the root surface sufficiently conducive for optimal plant uptake. The investigation also showed the ineffectiveness of DTPA extraction as a reliable soil test for Zn availability. Electro-ultrafiltration is being put to use increasingly for commercial soil testing in many countries in Europe, notably, for sugarbeet cultivation in Austria. Determination of Zn intensity singly may not provide reliable information regarding plant available Zn. Besides Zn intensity and Zn buffer power, other soil factors have only a minor influence on Zn absorption. Electro-ultrafiltration has the advantage of a rapid assessment of the Zn intensity factor. Combining the advantage of rapid assessment of the intensity factor of the nutrient with its corresponding buffer power might greatly enhance the reliability of fertilizer recommendations not only for Zn, but, perhaps, for other heavy metals as well, making electro-ultrafiltration a reliable tool in soil testing for heavy metals. Very recently Nair (1995, unpublished) employed an adsorption-desorption equilibrium and a soil incubation technique to quantify the Zn buffer power of some highly Zn deficient Turkish soils which have failed to show consistent response to Zn application based on DTPA extraction in wheat grown in these soils. Despite quite comparable DTPA extractable Zn, the Zn buffer power of these soils varied largely from a low value of 27 to a value as high as 200. It is being hypothesized that the Zn buffer power could, possibly, hold the key to explaining the differential response of wheat to Zn application in these problematic soils. The technique developed by Nair (1995, unpublished) dispenses with the need to employ highly cost intensive electro-ultrafiltration, which will cer-
BUFFERING POWER OF PLANT NUTRIENTS
275
tainly not be suitable to practical situations in many developing countries constrained by economic considerations, and in its place, employs a far simpler, and, simultaneously, far less costlier, technique of quantifying Zn intensity by an adsorption-desorption equilibrium technique using 0.01 M CaCl, solution or soil incubation technique.
C. OTHER HEAVY METALS One other important aspect to be considered, which is of relatively recent origin, is the heavy metal contamination of agricultural soils and their toxic effects on buffering. An element of particular interest is cadmium (Cd). The main sources of Cd contamination of agricultural soils is through the application of mine wastes, sewage sludges, and other residues. Kuo (1990) has reported on Cd buffering capacity and accumulation in Swiss chard (Beta vulgaris L.) in sludgeamended soils. He reported that while DTPA-extractable Cd did not show a high correlation with plant Cd, the Cd buffering capacity and plant Cd were very highly correlated.
D. MOLYBDENUM Dissolved soil molybdenum can buffer under a wide range (Barber, 1995). Values ranging from 9 for a sandy loam to 83 for a silty clay loam have been obtained (Lavy and Barber, 1964). If the concentrations of adsorbed molybdenum, which varied from 0.6 to 1350 p.g kg-I with solution molybdenum concentrations which varied from 1 to 13 kg L-1 are compared, a buffer power range from 1 to 100 is obtained as indicated by the data of Lavy and Barber (1964). By comparison, Australian soils indicated a buffer power range of 102000 (Barrow, 1970). Such very high buffer power values will greatly affect molybadenum uptake by plants growing in these soils. Since both mass flow and diffusion under differing molybdenum concentrations in the soil solution will affect its uptake, the precise quantification of the buffer power assumes greater relevance in predicting molybdenum availability. Since, for a given soil, molybdenum uptake by the plant growing in it will increase as its pH increases, ruminants which consume forage grown on such soils would develop molybdenosis, a disorder related to excessive intake of this element. A prior knowledge of the molybdenum buffer power will help manage better molybdenum fertilization of such soils. Additionally, the excess presence of both phosphate and sulfate will affect molybdenum availability in soils, but, their effect on molybdenum buffer power has to be clearly elucidated. There is, but scant, information on these aspects.
2 76
K. P. PRABHAKARAN NAIR
E. IRON The iron buffer power has to be related to both soil adsorbed iron and soil solution iron. Since redox reactions, precipitation, and presence of iron in organic complexes will affect the equilibrium between adsorbed and soil solution iron, it will be difficult to precisely quantify the iron buffer power of agricultural soils. O’Connor et al. (1971) obtained a buffer power value of 1000 for 59Fe*+ added to a thoroughfare loamy sand having a pH of 7.9. Since this value increased with time, it might have been affected by precipitation of iron oxides. Inasmuch as iron uptake from soils is concerned, one crucial factor would be root activity. The secretion of Fe-chelating phytosiderophores by Fe-deficient graminaceous species and further the production of siderophores which are low molecular weight iron chelating agents produced by almost all bacteria and fungi under iron limiting conditions (Buyer and Sikora, 1990) would make it difficult to quantify Fe flux by mass flow and diffusion, particularly the latter, on which Fe availabilitybased on its buffer power hinges. The effect of both phytosiderophores and siderophores will assume crucial significance in soils of high pH, wherein, these materials will solubilize iron and allow its movement both by mass flow and diffusion to the root surface. Since the ability of plant root to secrete phytosiderophores is both species and cultivar specific, there is an imminent need to understand better how this phenomenon might possibly affect Fe buffering at the soil-root interface.
F. MANGANESE The range in Mn buffer power is greater than those for most other nutrients. Investigations on Mn buffer power are scanty. Though Mn buffer power can be calculated from data on exchangeable Mn and soil solution Mn, the validity of such values is questionable because soil solution Mn has two components, namely, ionic and complexed, and their combined presence will render the Mn buffer power calculations invalid. Large variations have been found between exchangeable Mn and soil solution Mn. Barber et al. (1967) reported values ranging from 527 to 1.1 for these two fractions. Halstead et al. (1968) found narrower (65 to 0.4) ratios. Soils with low soluble Mn have been found to have high values while the reverse is true with regard to acid soils where Mn levels are high (Barber, 1995).
G. BORON The buffer power for B has been calculated from adsorption curves and values less than 3 have usually been found (Barber, 1995). For a loam soil, values of
BUFFERING POWER OF PLANT NUTRIENTS
277
I .35 and 1.7 have been found (Sulaiman and Kay, 1972). The absorption of B by plant roots would differ from that of other ionized species since B is present in solution as undissociated H,BO, (Barber, 1995). Since the buffer power values for B are generally small, it implies that the effective diffusion coefficient for B will be large and this in turn results in relatively easy long distance transport of B by mass flow and diffusion to reach plant root (Barber, 1995). Obviously, much of the B that is diffusible will reach the plant root during the period of plant growth.
VII. INFLUENCE OF HEAVY METAL CONTAMINATION ON BUFFERING OF MAJOR ELEMENTS With the increased use of sewage sludges and other residues in farming, there is a need to understand the effect of heavy metals carried in these materials on buffering of other major nutrients. Though interaction between heavy metals such as Zn and Fe with P has been studied in great detail (Nair and Babu, 1975) there is little information on the effect of heavy metals such as Cd or Cu on the buffering of other major nutrients. Yang and Skogkley, (1990) have studied the effect of Cu and Cd on K adsorption and K buffering capacity. When Cu or Cd are introduced into the soil as salts (Yang and Skogley 1989) or when metals accompany sludge applications (Page et al., 1987), soil-solution concentration and plant uptake of these metals is extremely low. Consequently, metal adsorption by soil components is important in regulating soil-solution metal concentration, bioavailability, and the toxicity that some of the heavy metals may cause in crop production. Since K is found to be highly vulnerable to adsorptivedesorptive (exchange) reactions, the large application of these metals might affect K dynamics. Yang and Skogley (1990) reported that K buffering was reduced by 20-32% by Cu addition and 7-20% by Cd addition. The rate of application was 400 mg Cu or Cd kg-1 soil. Cu and Cd changed the distribution of K, Ca, and Mg between the soil and solution. These results suggest that Cu or Cd addition would influence soil K availability by changing K adsorption kinetics, buffering capacity, and distribution.
VIII. POSSIBLE BUFFERING EFFECT ON PLANT ACQUISITION OF HEAVY METALS There is now growing evidence that root-induced changes are primarily responsible for the differential uptake of heavy metals such as Fe and Zn. In a
278
K. P. PWHAKARAN NAIR
recent review on Fe nutrition, Marschner and Romheld (1994) reported that two different types of root response strategies to Fe deficiency are identified in all plant species. In strategy I, occurring in all plant species except grasses, a plasma membrane-bound reductase is induced with enhanced net excretion of protons. In strategy 11, confined to grasses, there is an increase in the biosynthesis and secretion of phytosiderophores which form chelates with Fe3+. Compared to the progress made in identifying the cellular mechanisms of root responses in strategy I and strategy I1 plants, the understanding of processes taking place in the apoplasm of root rhizodermal cells and the role of low-moleclar-weight root exudates and siderophores in Fe acquisition of plants growing in soils of differing Fe availability is very limited (Marschner and Romheld, 1994); this suggests that there is a need to further investigate the buffering effect of Fe on root-induced differential uptake patterns. Such interactive investigations are conspicuously absent in the published literature. A similar situation exists with regard to the role of HCO, ion in affecting Fe and Zn nutrition of crop plants. Mengel (1994) hypothesized that it is the HCO, ion in association with the NO, ion, found in calcareous soils, that trigger Fe chlorosis. The chlorosis is induced by the impaired Fe uptake from root apoplast into the cytosol of root cells in the presence of excessive HCO, and NO, ions, and a similar situation exists in the case of leaves. Reduction of apoplast pH by spraying chlorotic leaves with an acid, such as H,SO,, led to a regreening of the leaves. The hypothesis of Marschner and Romheld (1994) stresses the physiologic effect, while that of Mengel (1994) stresses the chemical one. But in both cases, the possible Fe buffering effect visa-vis the root-induced changes merit much closer scrutiny than what has hitherto been attempted. Parallel to the above studies, the role of excess HCO, ion in the growth medium on Zn nutrition of wheat cultivars has recently been reported (Yang et af., 1994), indicating that excess HCO, ion in the growth medium leads to accumulation of organic acids and their insufficient compartmentation in Zninefficient cultivars as compared to Zn-efficient cultivars. These studies point to the lacuna in a proper understanding of the buffering effects vis-a-vis organic acid accumulation. It is not known what effects Zn buffering in the medium would have on excess accumulation of the organic acids in the Zn-inefficient cultivars as compared to the Zn-efficient cultivars.
IX. CONCLUDING COMMENTS AND mrrzTRE IMPERATIVES Historically, soil testing has been used to quantify availability of essential plant nutrients to field grown crops. However, contemporary soil tests are based
BUFFERING POWER OF PLANT NUTRIENTS
2 79
on philosophies and procedures developed several decades ago without significant changes in their general approach. For a soil test to be accurate, one needs to clearly understand physico-chemico-physiologic processes at the soil-root interface, and an understanding of soils and plant root systems as polycationic systems is essential. It is this knowledge that leads to sound prescriptive soil management practices inasmuch as nutrient availability vis-a-vis fertilizer application is concerned, because of all the factors which govern sustainability of crop production, the nutrient factor is the most important, as it is also the least resilient. This review focuses on the buffering power of plant nutrients, especially phosphorus, potassium, zinc, and some other heavy metals such as cadmium, on their availability to growing plants. Experimental evidence is presented from research carried on temperate and tropical soils which unequivocally establishes the fact that precise quantification of the nutrient buffer power holds the key to a clearer understanding of the plant availability of some of the more important essential plant nutrients, such as phosphorus, potassium, and zinc. The nutrient buffer power concept attempts to explain availability of plant nutrients on the basis of the diffusion model, as it is the principal process by which phosphorus, potassium, and zinc are taken up by plant roots. Possibly, other nutrients which are taken up by plant roots by the same process would also conform to the principles of this concept. To substantiate this, supportive experimental evidence is needed. There are similarities between the buffer power approach and quantityintensity ( Q l l ) approach in predicting plant availability of important nutrients such as phosphorus and potassium. While the Q/I approach has met with a fair measure of success in predicting P availability, with regard to K availability the Q l l relationship does not enjoy universal acceptance since a single relationship for all soils between K uptake by a given crop and ARK does not exist, perhaps due to the nature of the soil components regulating ARK. In substance, soil testing in a laboratory is meant to simulate what a plant root would accomplish in a field soil. All soil tests meet this requirement with different degrees of success. But, for a soil test to be universal in scope, it must attempt to quantify, as precisely as possible, factors which most crucially affect plant root uptake. Of all the factors involved in nutrient uptake, the most crucial is the average nutrient concentration on the root surface. In a dynamic state of plant growth it is virtually impossible to quantify precisely the average nutrient concentration on the root surface. The nutrient’s buffer power is an indirect measure of the average nutrient concentration on the root surface, and analytical methods which quantify the buffer power with precision would render the soil test in question dependable; hence, the fertilizer recommendation which follows would be reproducible. It is against this background that the nutrient buffer power concept merits wider application for field use. Fertilizer prices have dramatically escalated following the oil crisis of the mid- 1970s, and field experience
2 80
K. P. PRABHAKARAN NAIR
shows that recommendations emanating from routine soil tests have met with only limited success. A number of countries on the Asian and African continents spend enormous sums of money on importation of fertilizers, which is a great strain on their national exchequers. Most of these countries have no infrastructure for fertilizer manufacture and where there is any, the soil testing support service is far from satisfactory. Since fertilizer is a crucial input, second only to water, in modem farming, cost-effective fertilizer recommendations are an important prerequisite for successful agriculture. In countries such as India, fertilizer prices, especially of phosphate and potassium, have skyrocketed following governmental policies centered on a market economy. In such situations, farmer confidence can only be instilled and maintained by accurate soil testing and‘ reproducible fertilizer recommendations. The success of a new approach, to a large measure, rests with the ingenuity of those applying it to suit the demands of a new situation. This principle is no exception in making the “nutrient buffer power concept” a universal success. As compared to routine soil testing, this approach calls for the accurate determination of the buffer power of the nutrient in question, to begin with. Once this is done, the buffer power factor can be integrated with the routine soil test data and fertilizer recommendations can be .made on the basis of this new information. This implies that, in addition to obtaining routine soil test data, one also needs to know the buffer power. This review has shown how the buffer power is determined based on analytical methods varying from a simple adsorption-desorption equilibrium technique to elaborate electro-ultrafiltration. Only selected laboratories around the world have EUF facilities. Where they are absent, alternative analytical procedures to precisely quantify the buffer power need to be developed. Developing appropriate procedures to quantify the buffer power and modification of the protocol for routine soil testing to integrate buffer power data for precise fertilizer recommendations calls for both ingenuity and drive on the part of the researcher, but that will be the future imperative of accountable soil testing.
ACKNOWLEDGMENTS With gratitude I dedicate this chapter to: Professor Dr. Ir. A. H. Cottenie, former Rector of the State University of Gent, Belgium and member of the Royal Academy of Science, Letters, and Fine Arts, one of the very few finest men I have had the good fortune to interact with during the last three decades; Professor Dr. Konrad Mengel, my former colleague and former Director of the Institute of Plant Nutrition, Justus von Liebig University, Giessen, Federal Republic of Germany, who first motivated me, more than a decade and a half ago, to embark on a research theme which led to the compilation of this chapter and which still keeps my scientific curiosity alive; my wife Pankajam, and our children Kannan and Sreedevi and Cheri, our German shepherd, who are my only sustenance during the difficult journey that life is; and, most of all, to my late parents, my father, Kuniyeri
BUFFERING POWER OF PLANT NUTRIENTS
281
Pukkulon Kannan Nair, and my mother, Kodoth Padinhareveetil Narayani Amma, both of whom left me an orphan at a very young age, but whose blessings made me what I am today.
REFERENCES Adams, F. (1971). Ionic concentrations and activities in soil solutions. SoilSci. Soc. Am. Proc. 35, 420-426. Adams, F. (1974). Soil solution. In “The Plant Root and Its Environment” (E. W. Carson, Ed.), pp. 441 -481. University Press of Virginia, Charlottesville. Argersinger, W. J . , Jr., Davidson, A. W., and Bonner, 0. D. (1950). Thermodynamics and ion exchange phenomena. Trans. Kans. Acad. Sci. 53, 404-410. Barber, S. A. (1974). Influence of the plant root on ion movement in soils. In “The Plant Root and Its Environment” (E. W. Carson, Ed.), pp. 525-564. University Press of Virginia, Charlottesville. Barber, S. A. (1984). Soil nutrient bioavailability: A mechanistic approach. Wiley, New York. Barber, S. A. (1995). Soil nutrient bioavailability: A mechanistic approach. Wiley, New York. Barber, S. A , , Halstead, E. H., and Follett, R. E. (1967). Significant mechanisms controlling the movement of manganese and molybdenum to plant roots growing in soil. Trans. Joint Meeting Commissions I1 and IV. International Soil Science Society, Aberdeen, Scotland, pp. 299-304. Barekzai, A. (1984). Alterung von wasserloslichem Phosphat-untersucht in Giefaa und Modellversuchen. Ph.D. Thesis, FB Ig, Justus Liebig University, Giessen. Barrow, N. 1. (1970). Comparison of the adsorption of molybdate, sulfate and phosphate by soils. Soil Sci. 109, 282-288. Barrow, N. J. (1983). On the reversibility of phosphate sorption by soils. J. Soil Sci. 34, 751-758. Beckett, P. H. T. (1971). Potassium potentials-A review. In “Potash Review,” pp. 1-41. Int. Potash Inst., Bern, Switzerland. Beckett, P. H. T. (1972). Critical cation activity ratios. Adv. Agron. 24, 376-412. Beckwith, R. S. (1965). Sorbed phosphate at standard supernatant concentration as an estimate of the phosphate needs of soils. Aust. J. Exp. Agric. Anim. Husb. 5, 52-58. Bertsch, P. M . , and Thomas, G. W. (1985). Potassium status of temperate region soils. I n “Potassium in Agriculture” (R. D. Munson, Ed.), pp. 131-162. Am. Soc. Agron., Madison, WI. Boguslawski, E. V., and Lach, G . (1971). Die K-Nachlieferung des Bodens im Pflanzenexperiment im Vergleich mit dem austauschbaren Kalium. Z. Acker fjlunzenbau 134, 135-164. Boul, S. W. (1994). Soil, Society’s yoke to the Earth. In “Proc. XV of the ISSS Symposium, Acapulco, Mexico, July 10-16, 1994,” Vol. I , pp. 89-104. Borlaug, N. E. (1990). The challenge of feeding 8 billion people. Farm Chern. Inr. Summer, 1990. Borlaug, N. E., and Dowswell, C. R. (1993). Fertilizer: To nourish infertile soil that feeds a fertile population that crowds a fragile world. Better Crops Plant Food 77(3), 6-7. Bowman, R. A,, and Olsen, S . R. (1985a). Assessment of phosphate buffering capacity. 1 . Laboratory methods. Soil Sei. 140, 287-291. Bowman, R. A , , and Olsen, S. R. (1985b). Assessment of phosphate buffering capacity. 2. Greenhouse methods. Soil Sci. 140, 387-392. Bray, R. H.,and Kurtz, L. T. (1945). Determination of total organic and available form of phosphorus in soils. Soil Sci. 59, 39-45. Brouder, S. M., and Cassman, K. G. (1994). Evaluation of a mechanistic model of potassium uptake by cotton in vermiculitic soil. Soil Sci. Soc. Am. J . 58, 1174-1 183. Buyer, J. S., and Sikora, L. J. (1990). Rhizosphere interactions and siderophores. Plant. Soil 129, 10I - 107.
2 82
K. P. PRABHAKARAN NAIR
Cassman, K. G. (1990). The role of soil fertility research in developing sustainable food production systems. Better Crops Plant Food 74(4), 16-19. Claassen, N. (1990). “Die Aufnahme von Nahrstoffen aus dem Boden durch die hohere Pflanze als Ergebnis von Verfiigbarkeit und Aneignungsvermogen.” Severin-Verlag, Gottingen, Germany. Claassen, N., and Barber, S. A. (1976). Simulation model for nutrient uptake from soil by growing plant root system. Agron. J. 68, 961-964. Claassen, N., Hendriks, K., and Jungk, A. (1981). Rubidium-Verarmung des wurzelnahen Bodens durch Maispflanzen. 2. Pj7anzenern. Bodenk. 144, 533-545. Claassen, N., Syring, K. M., and Jungk, A. (1986). Verification of a mathematical model by simulating potassium uptake from soil. Planr Soil 95, 209-220. Cope, J. T., and Evans, C. E. (1985). Soil testing. Adv. Soil Sci. 1, 201-228. Cordero, A,, and Espinosa, J. (1994). Maximum economic rice yield in Costa Rica. Research for Maximum yield in harmony with nature. In “XV World Congress of Soil Science, Acapulco, Mexico, July 10-16,” pp. 29-39. Curtin, D., and Smillie, G . W. (1983). Soil solution composition as affected by timing and incubation. Soil Sci. SOC.Am. J . 47, 701-707. DeWit, C. T. (1992). Resource use efficiency in agriculture. Agric. Systems 40, 125-151. Drew, M. C., Nye, P. H., and Vaidyanathan, L. V. (1969). The supply of nutrient ions by diffusion to plant roots in soil. I. Absorption of potassium by cylindrical roots of onion and leek. Planr Soil 30, 252-210. During, C., and Duganzich, D. M. (1979). Simple empirical intensity and buffering capacity measurements to predict potassium uptake by white clover. Plant Soil 51, 167- 176. Elgawhary, S. M., Lindsay, W. L., and Kemper, W. D. (1970). Effect of EDTA on the self-diffusion of zinc in aqueous solution and in soil. Soil Sci. SOC. Am. Proc. 34, 66-70, Evangelou, V. P., Wang, J., and Phillipis, R. E. (1994). New developments and perspectives on soil potassium quantitylintensity relationships. Adv. Agron. 52, 173-227. Fohse, D., Claassen, N., and Jungk, A. (1991). Phosphorus efficiency of plants. 11. Significance of root radius, root hairs, and cation-anion balance for phosphorus influx in seven plant species. Planr Soil 132, 261-272. Fox, R. L., and Kamprath, E. J. (1970). Phosphate sorption isotherms for evaluating the phosphate requirement of soils. Soil Sci. SOC.Am. Proc. 34, 902-907. Fox, R. L. (1981). External phosphorus requirements of crops. In “Chemistry in the soil environment,” (R. H. Dowdy er al., Eds.) pp. 223-239. ASA Spec. Publ. 40, ASA and SSSA, Madison, WI. Fresco, L. O., and Kroonenberg, S. B. (1992). Time and spatial scales in ecological sustainability. Land Use Policy July 1992, 155-167. Frissel, M. J. (Ed.). (1978). “Cycling of Mineral Nutrients in Agricultural Ecosystems. Elsevier, Amsterdam. Gapon, E. N. (1933). Theory of exchangeable adsorption in soil. J. Gen. Chem (USSR)3, 144-163. [Abstract in Chem Abs. 28, 41-49, 1934.1 Grimme, H. (1974). Potassium release in relation to crop production. In International Potash Institute (ed) “Potassium Research and Agricultural Production,” pp. 131- 136. Int. Potash Inst., Berne, Switzerland. Grimme, H. (1979). Factors controlling potassium availability to plants. In “Proc. Int. Potassium Workshop, Cairo, Egypt” (A. Saurat and M. M. El-Fouly, Eds.), pp. 33-47. Halstead, E. H., Barber, S. A., Warncke, D. D., and Bole, I. (1968). Supply of Ca, Sr, Mn and Zn to plant roots. Soil Sci. SOC.Am. Proc. 32, 69-72. Hegde, D. M., and Dwivedi, S. S. (1992). Nutrient management in rice-wheat cropping system in India. Fertil. News 37, 27-42.
BUFFERING POWER OF PLANT NUTRIENTS
283
Hendriks, L., Claassen, N., and Jungk, A. (1981). Phosphatverarmung des wurzelnahen Bodens und Phosphataufnahme von Mais und Raps. 2. fjunzenern. Bodenk. 144, 486-499. Hinsinger, P., Jaillard, B., and Dufey, J. E. (1992). Rapid weathering of a trioctahedral mica by roots of ryegrass. Soil Sci. Soc. Am. J . 56, 977-982. Hoagland, R., and Martin, J. C. (1933). Absorption of potassium by plants in relation to replaceable, non-replaceable and soil solution potassium. Soil Sci. 36, 1-34. Holford, I. C. R. (1988). Buffering of phosphate in the soil solution during growth of ryegrass compared with buffering by sorption. Plant Soil Ill, 3-9. Jardine, P. M., and Sparks, D. L. (1984). Potassium-calcium exchange in a multi-reactive soil system 11. Thermodynamics. Soil Sci. SOC.Am. J. 48, 45-50. Keating, M. (1993). “The Earth Summit’s Agenda for Change. A Plain Language Version of Agenda 21 and the Other Rio Agreements. Centre for our Common Future, Geneva. Keerthisinghe, G . , and Mengel, K. (1979). Phosphatpufferung verschiedener B d e n und ihre Veranderung infolge Phosphatalterung. Mitt. Dtsch. Bodenk. Ges. 29, 217-230. Kottgen, P., and Diehl, R. (1929). Lieber die anwendung der dialyse und electroultrafiltration zur bestimmung des nahvstoff bedurfnisses des bodens. Z. fflanzenernaehr. Bodenkd. 12,65- 105. Kovar, I. L., and Barber, S. A. (1988). Phosphorus supply characteristics of 33 soils as influenced by seven rates of phosphorus addition. Soil Sci. Soc. Am. J . 52, 160-165. Kuchenbuch, R., and Jungk, A. (1984). Wirkung der Kaliumdungung auf die Kalium verfiigbarkeit in der Rhizosphare von Raps. Z. fflanzenernaehr. Bodenkd. 147,435-448. Kuhlmann, H., and Wehrmann, J. (1984). Kali-Dungeempfehlung auf der Grundlage von 81KDungeversuchen zu Gertreide und Zuckeniiben. Z. fflanzenernuhr. Bodenkd. 147,349-360. Kuo, S. (1990). Cadmium buffering capacity and accumulation in Swiss chard in some sludgeamended soils. Soil Sci. SOC. Am. J . 54, 86-91. Larsen, S. (1967). Soil phosphorus. Adv. Agron. 19, 151-210. Lavy, T. L., and Barber, S. A. (1964). Movement of molybdenum in the soil and its effect on availability to the plant. Soil Sci. SOC. Am. froc. 28, 93-97. Lewis, D. G., and Quirk, J. P. (1967). Phosphate diffusion in soil and uptake by plants 111. 31P-movement and uptake by plants as indicated by 32P-autoradiography.f fanr Soil 26, 445453. Lu, S . , and Miller, M. H. (1994). Prediction of phosphorus uptake by field-grown maize with the Barber-Cushman model. Soil Sci. SOC.Am. J . 58, 852-857. Lumbanraja, J . , and Evangelou, V. P. (1990). Binary and ternary exchange behavior of potassium and ammonium at low exchangeable fractional loads. Soil Sci. SOC. Am. J. 54, 698-705. Lumbanraja, J., and Evangelou, V. P. (1992). Potassium quantity-intensity relationships in the presence and absence of NH, for three Kentucky soils. Soil Sci. 154, 366-376. Maas, E. V. (1969). Calcium uptake by excised maize roots and interactions with alkali cations. flanr fhysiol. 44, 985-989. Marschner, H . (1994). Micronurr. News Informarion 14, 3-5. Marschner, H., and Romheld, V. (1994). Strategies of plants for acquisition of iron. Planr Soil 165, 261 -274. Martin, H . W., and Sparks, D. L. (1983). Kinetics of nonexchangeable potassium release from two coastal plain soils. Soil Sci. SOC.Am. J . 47, 883-887. Mattingly, C. E. G., Russell, R. D., and Jephcott, B. M. (1963). Experiments on cumulative dressings of fertilizers on calcareous soils in Southwest England. 11. Phosphorus uptake by ryegrass in the green house. J. Sci. Food Agric. 14,629-637. McLean, E. 0.. and Watson, M. E. (1985). Soil measurement of plant-available potassium. In “Potassium in Agriculture” (R. D. Munson. Ed.), pp. 277-308. ASA, CSSA, and SSA, Madison, WI.
284
K. P. PRABHAKARAN NAIR
Mehlich, A. (1984). Mehlich 3 soil test extractant. A modification of the Mehlich 2 extractant. Commun. Soil Sci. Plant Anal. 15, 1409-1416. Mengel. K. (1985). Dynamics and availability of major nutrients in soils. Adv. Soil Sci. 2, 65-131. Mengel, K. (1994). Iron availability in plant tissues-Iron chlorosis on calcavcous soils. Plant Soil 165, 275-283. Mengel, K., and Kirby, E. A. (1980). Potassium in crop production. Adv. Agron. 33, 59-110. Mengel, K., and Busch, R. (1982). The importance of the potassium buffer power on the critical potassium level in soils. Soil Sci. 133, 27-32. Mengel, K., and Steffens, D. (1985). Potassium uptake of ryegrass (Loliumperenne) and red clover (Trifolium prarense) as related to root parameters. Biol. Fen. Soils 1, 53-58. Mengel, K., and Uhlenbecker, K. (1993). Determination of available interlayer potassium and its uptake by ryegrass. Soil Sci. Soc. Am. J . 57, 761-766. Meyer, D., and Jungk, A. (1993). A new approach to quantify the utilization of nonexchangeable soil potassium by plants. Plant Soil. 149, 235-243. Mitsios, 1. K., and Rowell, D. L. (1987). Plant uptake from exchangeable and non-exchangeable potassium. I. Measuring and modelling for onion roots in a chalky boulder clay soil. J. Soil. Sci. 38, 53-63. Morgan, M. F. (1935). A universal soil testing system. Conn. Agr. Exp. Stn. Bull. 372. Nair, K. P. P. (1984). Towards a better approach to soil testing based on the buffer power concept. In “Proc. 6th Int. Colloquium for the Optimization of Plant Nutrition, Vol. 4, 2-8 September, Pierre-Martin Prevel, Montpellier, France,” pp. 1221- 1228. Nair, K. P. P. (1992). Zinc buffer power as an important criterion for a dependable assessment of plant uptake. Planf Soil 81, 209-215. Nair, K. P. P. (1989). Comments on “Phosphorus supply characteristics of 33 soils as influenced by seven rates of phosphorus addition.” Soil Sci. Soc. Am. J . 53, 984. Nair, K. P. P. (1990). Quantification of P and Zn replenishment capacity of soils as a dependable alternative to routine soil tests in assessing their bio-availability to growing plants. I n “Transactions of the 14th International Congress of Soil Science, Kyoto, Japan, August 12-18,” Comm. IV, pp. 493-494. Nair, K. P. P. (1992). Measuring P buffer power to improve routine soil testing for phosphate. Eur. J. Agnon. 1(2), 79-84. Nair, K. P. P. (1992). Soil testing procedure to economise on fertilizer use in land management. In “The Spirit of Enterprise-The 1993 Rolex Awards,” pp. 323-325. Nair, K. P. P., and Babu, G.R. (1975). Zinc-phosphorus-iron interaction studies in maize. Plant Soil 42, 517-536. Nair, K. P. P., and Prabhat, G. (1977). Differential response of tropical maize genotypes to zinc and manganese nutrition. Plant Soil 47, 149- 159. Nair, K . P. P., and Mengel, K. (1984). Importance of phosphate buffer power for phosphate uptake by rye. Soil Sci. Soc. Am. J . 48, 92-95. Nair, K. P. P., Sadanandan, A. K . , Hamza, S., and Abraham, J. (1994). Measuring K buffer power-The most important determinant of K availability in cardamom (Eletfariacardarnomurn M.)growing soils of India. Eur. J. Agron., in press. Narang, R. S., and Bhandari, A. L. (1990). Integrated nitrogen management in rice-wheat system. Paper presented at the rice-wheat workshop. Directorate for Cropping System in Punjab, India. Proceedings of Symposium on Maximum Yield Research, Nov 16-18, 1990, New Delhi, India. Narang, R. S., Tiwanria, U. S., and Dev, G. (1994). Maximum yield research studies in rice-wheat system and soil productivity-The Indian experience. Research for Maximum Yield in harmony with nature. Satellite symposium sponsored by PPI, pp. 46-55. XV World Congress of Soil Science, Acapulco, Mexico, July 10-16.
BUFFERING POWER OF PLANT NUTRIENTS
285
Nemeth, K. (197 I ). Die characteriesierung des K-haushaltes von boden mittels K-desorptionskurven. Geoderma 5 , 99- 101. Nemeth, K. (1972). The determination of desorption and solubility rates of nutrients in the soil by means of electroultrafiltration (EUF). froc. Collq. Inr. Potash Insr. 9, 171-180. Nemeth, K. (1979). The availability of nutrients in the soil as determined by electroultrafiltration (EUF). Adv. Agron. 31, 155-188. Nemeth, K., and Recke, H. (1982). Reproducibility of Ca, K, Na, and P contents in the different EUF fractions. Plant Soil 64, 25-33. Nye, P. H. (1972). Diffusion of ions and uncharged solutes in soils and clays. Adv. Agron. 31, 225272. O’Connor, G . A,, Lindsay, W. L., and Olsen, S. R. (1971). Diffusion of iron and iron chelates in soil. Soil Sci. Soc. Am. Proc. 35, 407-410. Okajima, H., Kubota, H.,and Sakuma, T. (1983). Hysteresis in the phosphorus sorption and desorption process of soils. Soil Sci. flanr Nurr. 29, 271-283. Olsen, S . R., Cole, C. V., Watanabe, F. S., and Dean, L. A. (1954). Estimation of available phosphorus in soils by extraction with sodium bicarbonate. USDA Cir. No. 939. Page, A. L., Logan, T. J., and Ryan, J. A. (1987). Land application of sludge-Food chain implications. Lewis Publ. Chelsea, MI. Pagel, H., and van Huay, H. (1976). Wichtige Parameter der Phosphat-Sorptionskurven einiger Boden der Tropen und Subtropen und ihre Zeitliche Veranderung durch P-Diingung. Arch. Acker. Pjlanzenbau. Bodenk. 20, 765-778. Pratt, P. F. (1965). Potassium. In “Methods of Soil Analysis” (C. A. Black er a[., Eds.), Part 2, pp. 1023-1031. Agron. Monogr. 9. ASA, Madison, W1. parameters Raven, K. P., and Hossner, L. R. (1994). Sorption and desorption-Quantity-intensity related to plant-available soil phosphorus. soil Sci. Soc. Am. J. 58, 405-410. Robison, L. R., Johnston, N. P., and Hill, J. M. (1981). Small-scale agriculture, an untapped giant. Benson Insrit. 4(2), 4-7. Roy, H. K., Kumar, A,, and Sinha, K. (1990). Response of wheat to K in red loam (Alfisols) soils of Ranchi. J . Potassium Res. 6 , 23-28. Sanchez, P. A. (1994). Tropical soil fertility research: Towards the second paradigm. In “Proc. XV ISSS Symposium, Acapulco, Mexico, July 10-16, 1994,” Vol. I , pp. 89-104. Schachtschabel, P. (1937). Aufnahme von nicht austauschbarem Kalium durch die Pflanze. Z. fjlanzenernahr. Bodenkd. 48, 107- 133. Schachtschabel, P. (1961). Fixierung und Nachlieferung von Kalium-und Ammonium-Ionen. Beurteilung und Bestimmung des Kaliumversorgungs-grades von Boden. Landwirfsch Forsch. 14, 29-47. Schiiller, H. (1969). Die CALMethode, eine neue Methode zur Bestimmung des pflanzen verfugbaren Phosphates in Boden. Z. fjlunzenernuhr. Bodenkd. 123, 48-63. Schofield, R. K. (1955). Can a precise meaning be given to “available” soil phosphorus? Soil Ferril. 18, 373-375. Selim, H. M. (1992). Modeling the transport and retention of inorganics in soils. Adv. Agron. 47, 331-384. Sharpley, A. N . I., Pierzynki, G., and Sims, J. T. (1993). Innovative soil phosphorus availability indices: Assessing inorganic phosphorus. In “Agronomy Abstracts.” American Society of Agronomy, Madison, WI. Singh, G. B., and Narnbiar, K. K. M. (1986). Crop productivity and fertility under intensive use of chemical fertilizers in long-term field experiments. Indian J. Agron. 31, 115-127. Simrad, R. R., Tran, T. S., and Zizka, J. (1991). Evaluation of the electro-ultrafiltration technique (EUF) as a measure of the K supplying power of Quebec soils. Plant Soil 132, 91-101.
286
K. P. PRABHAKARAN NAIR
Simrad, R. R., and Tran, T. S. (1993). Evaluating plant available phosphorus with the electroultrafiltration technique. Soil Sci. SOC. Am. J. 57, 404-409. Shuman, L. M. (1975). The effect of soil properties on zinc adsorption by soils. Soil Sci. SOC. Am. Proc. 39, 454-458. Shu-Yan, C., and Sposito, G. (1981). The thermodynamics of ternary cation exchange systems and the subregular model. Soil Sci. SOC. Am. J. 45, 1084-1089. Soil Science Society of America (Committee Report) (1965). Glossary of soil science terms. Soil Sci. SOC. Am. Proc. 29, 330-351. Sparks, D. L. (1987). Potassium dynamics in soils. Adv. Soil Sci. 6, 2-63. Sparks, D. L. (1989). “Kinetics of Soil Chemical Processes.” Academic Press, San Diego, CA. Sparks, D. L., and Huang, P. M. (1985). Physical chemistry of soil potassium. In “Potassium in Agriculture’’ (R. D. Munson, Ed.), pp. 201-276. American Society of Agronomy, Madison, WI. Sposito, G. (1981a). Cation exchange in soils. A historical and theoretical perspective. In “Chemistry in the Soil Environment” (R. H. Dowday, Ed.). Soil Sci. Soc. Am., Madison, W1. Sposito, G. (1981b). “The Thermodynamics of the Soil Solution.’’ Clarendon Press, Oxford. Sposito, G., Holtzelaw, K. M., Johnston, C. T., and Le Vesque, C. S . (1981). Thermodynamics of sodium-copper exchange on Wyoming bentonite at 198°K. SoilSci. SOC.Am. J . 45, 1079-1084. Sposito, G., Holtzelaw, K. M., Charlet, L., Jonany, C., and Page, A. L. (1983). Sodium-calcium and sodium magnesium exchange on Wyoming bentonite in perchlorate and chloride background ionic media. Soil Sci. SOC. Am. J . 47, 51-56. Stammer, L. B. (1992). Study finds serious harm to 10 percent of world’s best soil. Los Angeles Times, January (quotations from World Resources Institute of U.N. Environmental Program). Steffens, D. (1994). Phosphorus release kinetics and extractable phosphorus after long-term fertilization. Soil Sci. SOC. Am. J . 58, 1702-1708. Steffens, D., and Mengel, K. (1979). Das Aneignungsvermogen von Lolium perenne im Vergleich zu Trifolium prurense fur Zwischenschilht-Kalium der Touminerale. Lundw. Forsch. S H . 36, 120127.
Steffens, D., and Mengel, K. (1981). Vergleichende Untersuchungen zwischen Lolium perenne und Trifolium prurense uber das Aneignungsvermogen von Kalium. Mirr. Drsch. Bodenk. Ges. 32, 375-386.
Sulaiman, W., and Kay, B. D. (1972). Measurement of the diffusion coefficient of boron in soil using a single cell technique. Soil Sci. SOC. Am. Proc. 36, 746-752. Swarup, A., and Singh, K. N. (1987). Movement of potassium in a sodic soil profile as influenced by long-term use of inorganic fertilizers under rice-wheat rotation. J. Indian SOC.Soil Sci. 35,253256.
van der Paauw, F. (1971). An effective water extraction method for the determination of plantavailable soil phosphorus. Plant Soil 34, 467-48 I . Vanselow, A. P. (1932). Equilibria of the base-exchange reactions of bentonites, permutities, soil colloids, and zeolites. Soil Sci. 33, 95-1 13. von Reichenbach, H. (1972). “Factors of mica transformation. In “Potassium in Soil,” pp. 33-42. Int Potash Institute, Berne, Switzerland. Whitmore, A. P., and van Noordwijk, M. (1994). Bridging the gaps between environmentally acceptable and agronomically desirable nutrient supply. In “Proceedings of the Long Ashton Symposium: Agriculture in the 21st Century’’ (D. Glen, Ed.), in press. Wilkinson, H.F., Loneragan, J. F., and Quirk, J. P. (1968). The movement of zinc to plant roots. Soil Sci. SOC. Am. Proc. 32, 831-833. Wood, L. K., and De Turk, E. E. (1941). The adsorption of potassium in soils in replaceable forms. Soil Sci. SOC. Am. Proc. 5, 152-161.
BUFFERING POWER OF PLANT NUTRIENTS
287
Woodruff, C. M. (1955a). Ionic equilibria between clay and dilute salt solutions. Soil Sci. Soc. Am. Proc. 19, 36-40. Woodruff, C. M. (1955b). The energies of replacement of calcium by potassium in soils. Soil Sci. SOC. Am. Proc. 19, 167-171. Xu, M. L., and Liu, Z. Y. (1983). The nutrient status of soil-root interface. 11. Potassium accumulation and depletion in rhizosphere soils. Acta Pedol. Sinica 20, 295-302. Yates, R. A,, and Kiss, A. (1992). Using and sustaining Africa’s soils. In “Agricultural and Rural Development Series,’’ No. 6. World Bank, Washington. Yang, 1. E., and Skogley, E. 0. (1989). Influence of copper or cadmium on soil potassium availability properties. Soil Sci. Am. J. 53, 1019-1023. Yang, J. E., and Skogley, E. 0. (1990). Copper and cadmium effects on potassium adsorption and buffering. Soil Sci. Soc. Am. J. 54, 739-744. Yang, X., Romheld, V., and Marschner, H. (1994). Effect of bicarbonate on root growth and accumulation of organic acids in Zn-inefficient and Zn-efficient rice cultivars (Oryza sativa L.) Plant Soil 164, 1-7.
This Page Intentionally Left Blank
OVERVIEW OF VERTISOLS: CHARACTERISTICS AND IMPACTS ON SOCIETY Clement E. Coulombe, Larry P. Wilding, and Joe B. Dixon Department of Soil and Crop Sciences, Texas A&M University, College Station, Texas 77843
I. Introduction 11. Distribution A. World Distribution B. Distribution in the United States of America 111. Formation of Vertisols A. Parent Material B. Climate C. Topography D. Vegetation E. Time IV.Morphological Properties A. Texture B. Color C. Structure and Special Physical Features V. Pedogenic Processes in Vertisols VI. Classification:From Marbut to Soil Taxonomy VII. Mineralogical Properties A. Phyllosilicates B. Minerals Other Than Phyllosilicates VIII. Chemical Properties A. pH B. Cation Echange Capacity and Exchangeable Cations C. Cation and Anion Behavior M.Biological Properties A. Organic Constituents: Distribution and Biology B. Clay-Organic Complexes X. Physical Properties A. Bulk Density and Coefficient of Linear Extensibility (COLE) B. Consistence and Atterberg Constants C. Shrinkage Curve and Moisture Retention Characteristics D. Gas Diffusion E. Hydraulic Conductivity F. Soil Structure and Porosity
289 Aduanccs in Agnmomy, Vdunv F7
Copyright Q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
290
C. E. COULOMBE, L. P. WILDING, AND J. B. DMON XI. Management of Vertisols A. Management of Vertisols with Low-Input Cultural Systems: Case of Vertisols of West Africa with Special Reference to Northern Cameroon B. Management of Vertisols with High-Input Cultural Systems with Special Reference to Texas XII. Summary and Concluding Remarks References
I. INTRODUCTION Shrinking and swelling of soils with changes in moisture conditions impact their use for agricultural, engineering and environmental purposes. Shrinking and swelling processes occur in all soils; however, clayey soils such as Vertisols and vertic intergrades show a greater expression of these phenomena than most other soils. Due to this shrinking and swelling behavior, Vertisols do not express uniform horizonation as in other soils. However, this does not imply they are simple homogeneous soil systems. Vertisols exhibit extensive spatial and temporal variability in their properties and remain the most difficult land resource systems in the world to manage successfully. For many regions of the world, Vertisols are important for agriculture production. For instance, they are considered highly productive and sustainable resources in India and China where they have been cultivated for hundreds and even thousand of years and are still productive. Yet in other continents such as Africa, Australia, and America, Vertisols are undergoing degradation due to intensive and inadequate management practices after only decades of cultivation. In drier climates, they often serve as the lifeline to subsistence agriculture where crops on more droughty soils fail. Vertisols have not been perceived as soil resources subject to degradation. Their clay content and intrinsic properties make them resilient soil systems. Furthermore, their shnnk-swell behavior upon wet-dry cycles favors the development of a fine granular mulch layer at the soil surface that overlays larger peds, generally prismatic in kind. Degradation processes of these resources are inevitably linked to a modification of the soil structure and can be chemically, biologically, and/or physically induced. Soil structural degradation is commonly expressed by decreases in organic matter, decreases in biological activity and diversity, increases in salinity, increases in bulk density, and associated loss in macroporosity. The final effect of degradation is reflected by a decrease in soil fertility and crop productivity. In developing countries, farmers eventually are forced to use marginal lands for food, fuel, and fiber production. Interestingly,
OVERVIEW OF VERTISOLS
291
degradation processes occur worldwide in Vertisols regardless of parent material, climatic conditions, and cultural input invested in their management. Agriculture is not the only land use impacted by these shrink-swell soils. Vertisols are commonly involved in engineering applications. However, their shrink-swell activity, even under a low range of soil moisture conditions, is sufficient to induce soil movement and cause destabilization of structural foundations. Consequently, a considerable amount of money is spent every year for the repair and maintenance of damage to buildings as well as roads, sidewalks, pipelines, transmission lines, and other Vertisol-based installations damaged by soil instability. Likewise, considerable investments are made in the construction phases for soil stabilizers (e.g., activated lime) to overcome these constraints. Injury and loss of human life resulting from collapse of soil excavations deserve serious consideration when dealing with Vertisols. For example, from 1980 to 1985, 50 construction workers were killed when these soil materials were destabilized by construction activities (USDA-SCS, 1986). Thus shoring materials and other safety precautions must be taken. Shrink-swell activities may contribute to environmental hazards in other ways. Loss of fertilizers, pesticides, and other toxic inorganic and organic chemicals via overland transport and/or transmission along desiccation cracks and other macropores may contaminate surface water and groundwater and negatively impact soil, water, and air quality. Environmental impacts of Vertisols are not restricted solely to agricultural and engineering activities but extend to society in general. A simple example is the disposal of household wastes. When acetone, ethanol, or other highly polar organic compounds are disposed of in a landfill, leakage may occur. The clay liner with low permeability to water can become very permeable to these organic solvents (Brown and Thomas, 1987; Brown and Daniel, 1988). Once these organic chemicals reach the clay liner, desiccation of the clay occurs. This results in structural alteration of the liner and promotes desiccation cracks and macropores. The increased liner permeability fosters subsequent transmission of aqueous and nonaqueous liquid pollutants into vadose-zone aquifers. A voluminous amount of literature about Vertisols has been published with regard to their properties and management, e.g., Dudal (1965), Dixon (1982), Ahmad (1983), McGarity et al. (1984), Probert et al. (1987), IBSRAM (1987, 1989), Jutzi et al. (1988), Wilding and Puentes (1988), Hirekerur et al. (1989), ICRISAT (1989), Jones and Gerik (1990), Coulombe et al. (1996), and Ahmad and Mermut (1996). Even with the remarkable increase in the knowledge base, especially in the past 15 years, Vertisols remain a challenge in their utilization and management, Information on Vertisols has been considerably extended and updated; however, several concepts are still not fully understood. It remains imperative to understand structural changes due to management and soil utilization. Such foundational research should promote more effective stabilization
292
C. E. COULOMBE, L. P. WILDING, AND J. B. DMON
technologies and the development of improved management practices to maintain or enhance structural stability. Increased sustainability of Vertisol resources should promote enhanced soil, air, and water quality. These are vital components to our society and the quality of life on this planet. Much of this review is based on the literature published about Vertisols but also includes recent developments and concepts concerning Vertisols with regard to their distribution, formation, pedogenesis, and classification, their morphological, mineralogical, chemical, biological, and physical properties, and their management as a soil resource in the world.
11. DISTRIBUTION A. WORLDDISTRIBUTION A first estimate of Vertisol distribution was reported by Dudal (1965) at about 257 million ha. Vertisols are currently estimated to compose about 308 million ha of the Earth’s surface, i.e., 2.23% of the global land area (USDA-SCS, 1994). The increased number of soil surveys globally and new developments in concepts of Vertisols have contributed to the knowledge base and increased extent of Vertisols (Soil Survey Staff, 1960, 1975; 1990; 1992; 1994). It is noteworthy that a reliable estimate of global distribution of Vertisols is still not possible because many countries have yet to make detailed inventories. In other cases, the area of Vertisol distribution within a given soil survey area is too small to be shown at the scale of the map compilation. Nonetheless, Vertisols and vertic intergrades have been reported in more than 80 countries, 6 of which can account for over 75% of their extent: India (25%), Australia (22%), Sudan (16%), the United States of America (6%), Chad (5%), and China (4%) (Dudal and Eswaran, 1988; Wilding and Coulombe, 1996). Figure 1 shows the global distribution of Vertisols in the world. From a climatic standpoint, Vertisols are reported to occur under all temperature and moisture regimes. They are particularly more abundant in the tropic and subarid regions. For instance, 60% of the Vertisols are in the tropics, 30% in the subtropics, and 10% in cooler regions (Dudal and Eswaran, 1988; Wilding and Coulombe, 1996). Similarly, 13% occur in humid and subhumid regions, 65% subarid, 18% arid, and 4% in Mediterranean climate.
B. DISTRIBUTION IN THE UNITEDSTATESOF AMERICA Vertisols have been reported in 25 States, Puerto Rico, and the Virgin Islands (Fig. 2). Vertisols in the U.S.A. are estimated at about 12 million ha. They occur
Rgure 1 Global distribution of Vertisols and associated vertic soils (modified from Dudal and Eswaran, 1988).
2 94
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
Flgure 2 Distribution of Vertisols in the United States (Data kindly provided by Dr. R . J. Ahrens, USDA-NRCS, Lincoln, Nebraska).
in all temperature and moisture regimes. By far, Texas comprises the major extent of Vertisols in the U.S.A. with about 6.5 million ha. A large proportion of Vertisols also occur in the northcentral and western United States (Fig. 2). The area of Vertisols extends from Dallas to San Antonio in Central Texas and along the Texas Gulf Coast region (Fig. 3). Hence, they significantly impact urban as well as rural populas. South Dakota, California, and Montana comprise the next largest extent with 1.5, 1.0, and 0.6 million ha, respectively. The remaining states and territories have less than 0.25 million ha.
III. FORMATION OF WRTISOLS The classical soil forming factors, i.e., parent material, climate, topography, vegetation, and time, established by Dokuchaev, are still the best elements to comprehend the formation of Vertisols. These soil forming factors are interdependent and highly variable and, consequently, they influence the properties of Vertisols in multiple ways. The difference in intensity of processes as condi-
OVERVIEW OF VERTISOLS
295
Figure 3 Distribution of Vertisols in Texas.
tioned by each of these factors within and among regions is responsible for the global diversity of Vertisols. Each of these factors is briefly discussed below.
The type of parent material is the most important factor involved in the formation and distribution of Vertisols. Parent materials of Vertisols are highly variable in origin. The parent materials may originate from sedimentary, igneous, or metamorphic origins (Dudal, 1965; Ahmad, 1983; Probert etal., 1987; Coulombe et al., 1996). Examples of sedimentary origin are loessial, fluvial, colluvial, lacustrine, and marine deposits, and marl, chalk, limestone, coral and shale bedrock. Igneous and metamorphic origins refer to Vertisols developed from weathering products of volcanic ash and tuff, schist, granite, gneiss, basalt, gabbro, norite, trachyte, andesite, amphibolite, diabase, dolerite, and serpentinite. Parent materials of some Vertisols are presented in Table I. The parent materials must provide, from inheritance or weathering, a high content of clay with high surface area and generally a high base status (Dudal,
Table I Selected Parent Materials from Which Vertisols Are Derived ~~
Parent materials Sedimentary Aluvium/Colluvium Alluvium/Colluvium Alluvium Alluvial clay Alluvium (calcareous) Alluvium, lacustrine sediments Clay sediments (basalt or mudstone) Shale (Beaumont formation) Marl Marl Marl (weathered into red clays) Marl and calcareous silstone Marl, chalk, and limestone Limestone Limestone Limestone Coral limestone Dolomite Dune sand and aeolian dust Sandstone
Age
Regions
References
Mio. IPleisto. Holo./Pleisto. Holo. IPleisto. n/aa Pleistocene Quaternary Quaternary Pleistocene Miocene Quaternary Cretaceous Miocene Cretaceous Cretaceous Tertiary Oligocene Holo. /Pleisto. n/a nla Quaternary
California Sudan Oregon Southern Turkey Iraq Cameroon Uruguay Texas (coast) Spain Argentina Morocco Trinidad Texas Jordan Jamaica Antigua Barbados Israel Israel Sudan
Southard and Graham (1 992) Finck (1961); Blokhuis et al. (1968/1969) Parsons et al. (1973) Giizel and Wilson (1981) Hussain et al. (1984) Yerima (1986); Yerima et al. (1988) Lugo-Lopez et al. (1985) Kunze et al. (1963) Sollis and Torrent (1989b) Stephan et al. (1983) Badraoui and Bloom (1990) Ahmad and Jones (1969) Nelson et al. (1960); Templin et al. (1956) Shadfan (1983) Ahmad and Jones (1969) Ahmad and Jones (1969) Ahmad and Jones (1969) Feigin and Yaalon (1974) Feigin and Yaalon (1974) Yousif et al. (1988)
Igneous and metamorphic Volcanic ash, tuff, andesite Volcanic ashltuff (noncalcareous) Tuffaceous sandstone Granitelgneiss
Holocene Pleistocene Oligocene n/a
El Salvador Dominica Utah India (south)
Gneiss, granodiorite, andesite Gneiss, schist, eclogite, hornblende, garnet, pyroxene Gneiss: biotte/amphibolite Basalt
nla nla
Australia Ghana
n/a Pliocene
Togo Israel
Basalt, augite, dolerite
nla
India
Basalt, volcanic cinders, olivine, augite, diorite, andesite, granite and schist Basalt and phonolite Basalt: norite, plagioclase dolerite, olivine, augite Schist Chlorite schist
nla
Arizona
nla nla
Kenya South Africa
Precambrian nla
Cameroon India
a
n/a, not available.
Yerima (1983); Yerima et al. (1985) Ahmad and Jones (1969) Graham and Southard (1983) Roy and Barde (1962); Hirekurubar et al. (1991) Bottinger (1992) Stephen (1953) Kounetsron et al. (1977) Dan and Koyumdjisky (1963); Feigin and Yaalon (1974) Roy and Barde (1962); Agrawal and Ranamorthy (1970) Johnson et al. (1962)
Kantor and Schwertmann (1974) Menve and Heystek (1955); Fitzpatrick and Le Roux ( 1 977) Yerima et al. (1988) Hirekurubar et al. (1991)
2 98
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
1965; Probert ef al., 1987; Coulombe et al., 1996). The distinction between inherited and neoformed clay minerals is still speculative, especially in the case of sedimentary deposits. If primary minerals have persisted, it may be possible to suggest sources or origins of parent material that contributed to these soils. A high base status favors formation and stability of clay minerals abundant in Vertisols. However, the base status may be relatively low in certain types of parent materials (e.g., volcanic tuffs, granite, gneiss, and schists). With time, the base status may increase due to release of bases from mineral weathering and/or from external sources, e.g., leaching from uplands, flood waters, base-rich water tables, or eolian deposits.
B. CLLMATE Climate is the second most important factor in formation and distribution of Vertisols. The critical element is that Vertisols must undergo seasonal soil moisture stress which can be expressed as distinct wet-dry seasons or by water deficit during part or all of the growing season. Broadly speaking, climate involves temperature and moisture regimes. As mentioned previously, Vertisols are found under a wide range of climatic conditions. Mean annual temperatures vary from 0 to >25"C (cryic to hyperthermic). With respect to moisture regimes, Vertisols differ in the amount of water they receive. In terms of annual precipitation, the range is from 250 to 3000 mm but most commonly 500-1500 mm. Periodicity, duration, and intensity of rainfall events, or other important hydrological secondary effects such as surface runoff, surface run-in, fluctuation of groundwater table or flooding may influence Vertisol formation. Consideration of these secondary factors has been incorporated to some degree within the definition of moisture regimes for Vertisols (Soil Survey Staff, 1975, 1994). With the exception of Aquerts, the soil moisture regimes are based on size, depth, and duration of the cracks (Soil Survey Staff, 1994) rather than the moisture conditions of the soil. The definition of soil moisture regimes for Vertisols needs to be revised using better quantitative measurements for the evaluation of seasonal and temporal changes. Influence of microclimatic conditions also needs attention particularly in the formation of some features, e.g., gilgai microrelief, which in turn controls the microclimatic condition of the given region. The temperature and moisture regimes discussed above are contemporaneous ones. The paleoclimatic conditions which may have been favorable for the genesis of Vertisols in the past are not considered. Recognition and interpretation of Paleo-Vertisols was reported by some authors, e.g., Retallack (1986), Gray and Nickelsen (1989), Gustavson (1991), and Joeckel (1994, 1995). Information obtained from these paleosols may be used to predict current and future behavior of Vertisols associated with climatic conditions.
OVERVIEW OF VERTISOLS
299
C. TOFWGRAPHY A flat or gently sloping topography occurring at low elevation is typical of Vertisols. However, Vertisols may occur at higher elevations, like in the Ethiopian plateau, or at higher slopes like in the West Indies. In the latter cases, Vertisols are highly subject to erosion by water and also tend to develop shallower profiles than those found on more gentle gradients at lower elevations. Topography has little effect on Vertisol formation and distribution. However, the topographic position influences microtopography (gilgai pattern) and drainage conditions. For instance, Vertisols that occur at higher elevations and with better drainage conditions will have a tendency to produce a higher content of kaolinite and iron oxides which will be evident by the presence of a higher chroma of the soil matrix, lower cationic exchange capacity (CEC), and higher phosphorus sorption than in black Vertisols (see Coulombe er al., 1996). On the other hand, Vertisols that occur in depressional, nearly planar, or concave positions would favor accumulation and retention of bases and silica, thus promoting the formation and chemical equilibrium of smectite (see Coulombe et al., 1996). Such a soil will likely have a high CEC and a smectitic mineralogy, and its color will depend on the character of the parent material. On a large scale, undulating microtopography called “gilgai,” that consists of circular mounds or linear ridges and depressions, is a feature that may occur in Vertisols. The formation of this microtopography is not very well understood. Nonetheless, the depth distribution of organic carbon, carbonates, salts and other properties generally differ between the mounds and depressions of gilgai topography (Wilding and Tessier, 1988; Wilding ef al., 1991; Wilding and Coulombe, 1996). These features cause vertical and horizontal spatial variability in Vertisols within a few meters or less.
D. VEGETATION Little can be said about the influence of vegetation on pedogenesis or Vertisol distribution. Uncultivated Vertisols are associated with grasslands and savannas as native vegetation (Dudal, 1965; Probert et al., 1987). Evidence of mixed pine and deciduous forests has been reported in some regions such as in East Texas. At the present time, most Vertisols are being influenced by postcultural human activities, e.g., fire, agriculture practices, and engineering construction which makes the inference and identification of the native vegetation difficult. In northem Cameroon, a decrease of vegetative cover and diversity and/or change of vegetative species are indicators of different stages of Vertisol degradation (Donfack, 1992). Differences in vegetation are good indicators of surface variability. In the case of Vertisols that have a gilgai microrelief, generally xerophytic plants occur on
3 00
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
mounds while more mesophytic vegetation occupies depressions (Hubble, 1984; Wilding and Coulombe, 1996). This gives rise to distinct tonal patterns on aerial photographs. Even in the absence of gilgai microrelief, differences in color pattern of the vegetation may highlight zones of spatial variability (e.g., soil drainage, chlorosis due to carbonate-Fe interactions, nutrient deficiencies) induced by shrink-swell phenomena. Furthermore, vegetation can reflect soil degradation. Vegetation usually is not considered an active factor in Vertisol development. Its major impact is on patterns of soil water extraction through evapotranspiration and the development of soil cracking patterns upon soil desiccation. Vertisols under native grasslands and pastures would develop a polygonal cracking pattern while with row crops, such as maize and sorghum, a linear cracking pattern parallel to the row is likely to form or to be more dominant than polygonal cracking patterns. The parallel cracking pattern is attributed to differential desiccation on the ridge relative to the furrow. This is possibly due to the higher elevation of the ridge compared to the furrow and a consequent increase in water potential from the ridge toward the furrow center.
E. TIME Time is often inferred from the age of the underlying parent material where the soil has developed. Most Vertisols are derived from Cenozoic Era materials which include Tertiary (2 to 65 million years BP) and Quaternary (less than 2 million years BP) Periods in the geological time scale (Table I). Some Vertisols are developed from much older bedrock materials such as those derived from Cretaceous (65 to 144 million years BP) sedimentary deposits of the Mesozoic Era in the Blacklands regions of Texas, Mississippi, and Alabama (Templin et al., 1956; Hawkins and Kunze, 1965; Dixon and Nash, 1968), in Morocco (Badraoui and Bloom, 1990), and in Jordan (Shadfan, 1983). Vertisols derived from truncated basement complex geologic surfaces also exist. For instance, in Cameroon, some Vertisols are derived from Precambrian schist (570 to 3800 million years BP; Yerima, 1986). However, the age of the parent material gives only a maximum chronological point; age of the geomorphic surface and of the soil would be much less, perhaps several thousand to hundreds of thousands of years old. Even though the parent materials may have formed in older geological time periods, Vertisols have generally developed in recent times. Some unconsolidated parent materials may require only a few hundred years to develop because high clay contents are indigenous, while for consolidated parent materials sufficient time is needed for weathering, clay formation, and shrink-swell dynamics to develop. Nevertheless, the relative age of a soil may be estimated by recording the development of certain features or by using biogeochemical tools. An over-
OVERVIEW OF VERTISOLS
301
view of these different methods is discussed in Meyer (1987) and Sposito and Reginato ( 1992), and includes radioisotopic, microscopic, spectroscopic, chemical, mineralogical, geomorphic, and correlative methods. Furthermore, these methods also can be used for global climatic change studies.
IV. MORPHOLOGICAL PROPERTIES Vertisols can be recognized easily from other soil orders by their characteristic signatures reflecting dynamic morphological properties. A voluminous literature is available to verify a number of these properties which has been most recently summarized by Dudal and Eswaran (1988) and Coulombe er al. (1996). Likewise, numerous soil names have been used to identify these soils (see Section VI). Oakes and Thorp (1950), who originally proposed the Grumusol great group name which later became the Vertisol order in the 7th Approximation (Soil Survey Staff, 1960), recognized these soils as mainly fine in texture, dark in color, commonly lacking distinct horizonation, and possessing characteristically extreme shrink-swell attributes with changes in soil moisture content. The following discussion will focus on some morphological markers, e.g., texture, color, and structure and features as currently recognized in Soil Taxonomy (Soil Survey Staff, 1975, 1994).
A. TEXTURE According to Soil Survey Staff (1975, 1994), Vertisols have a clay (<2 pm) content equal to or greater than 30% to a depth of 50 cm or more. The total clay content may be as high as 90%. The total fine clay (<0.2 km) content may represent up to 80% of the total clay fraction. Smaller clay-size particles have high specific surface areas (from 100 to 400 m2 g-1, up to 800 m2 g- I ) and consequently increased shrink-swell potential (Wilding and Tessier, 1988). Surface area measurements complement clay-size distribution and reactivity for inferences of shrink-swell potential. Because in nature, clay minerals are organized in an “aggregate fashion” rather than occurring as single particles, it is difficult to measure accurately clay-size distribution. In fact, when we measure clay-size distribution, we measure the size of these aggregates called domains and quasicystals. Such microstructures will have a remarkable impact on properties and behavior of Vertisols and other clayey soils as well. This aspect will be discussed further in Section V. At this point let us simply note that soils with a high clay content and containing clays with a high surface area will provide a greater proportion of small interparticle pores and, consequently, higher shrink-swell potential.
3 02
C. E. COULOMBE, L. P. WILDING, AND J. B. DMON
Vertisols may have variable colors, e.g., black, gray, brown, or red, throughout the profile but commonly have poor horizon differentiation. Soil color was recognized at high categorical levels (great groups) in earlier versions of Soil Taxonomy (Soil Survey Staff, 1975) as pellic and chromic categories. More recently the dark-colored pellic soils have been considered the normal great group categories and the chromic Vertisols have become intergrades at the subgroup level (Soil Survey Staff, 1992, 1994). The reason that color was found less useful is that it does not consistently discriminate soil drainage conditions between Vertisols; pellic and chromic were to indicate the poorly and the betterdrained Vertisols, respectively, but in reality were poorly related to oxidation states (Comerma et al., 1988). Likewise, these great groups poorly differentiated differences in organic matter contents. They more commonly reflect differences in parent material or geomorphic soil stability. In the case of pellic Vertisols, the organic carbon content is generally low and ranges from 0.3 to 3% but may be as high as 6% in temperate and humid regions. The color is attributed to the strong clay-organic complexes rather than the organic carbon content (Singh, 1954, 1956). Other agents such as iron, titanium, and manganese oxides may also be responsible for the pigmentation of dark Vertisols. For instance, magnetite and ilmenite in the silt fraction were found responsible for the dark color instead of organic matter content in Vertisols of Hawaii (Raymundo, 1965). Fungi can also be a possible agent responsible for a low chroma in poorly drained Vertisols. A more brownish or reddish color in Vertisols is attributed to the presence of iron oxides or oxyhydroxides. Some of the factors that may contribute to the color are a higher topographic position that promotes leaching and oxidizing conditions, a higher iron content in the parent material, the dissolution of ironrich smectites in slightly acidic environments, and coatings of iron oxides on mineral and ped surfaces inherited from the parent material.
Structure and special physical features are the most striking morphological markers of Vertisols. These markers are indicative of pedogenic processes, particularly shrink-swell phenomena, that occur in Vertisols. A cross-sectional schematic of a Vertisol with structure and other physical features sketched illustrates the morphological markers of soil movement and horizon profile development (Fig. 4). In surface horizons, Vertisols generally develop a granular structure in the upper 10 cm. This granular structure is attributed to self-mulching. Self-
Figure 4 Morphological structure and special physical features of Vertisols.
3 04
C. E. COULOMBE, L. P. WILDING, AND J. B. DMON
mulching is the ability of a soil to form small aggregates at the surface due to shrink-swell and/or freeze-thaw phenomena (Soil Survey Staff, 1960). A more rigorous definition of self-mulching was proposed by Grant and Blackmore (1991): “The ability of a soil to reform a fine, water-stable granular structure (consisting of aggregates having diameters less than 5 mm), upon drying and rewetting after this structure has been destroyed by puddling.” The definition eliminates the impact of freeze-thaw phenomena on self-mulching as well as the influence of biotic factors on formation and maintenance of a granular structure. Nonetheless, the self-mulching behavior is a way for clayey soils to reduce loss of water by evaporation. Vertisols may also develop a massive structure at the surface. The contrast between a granular and massive structure was recognized in the 7th approximation (Soil Survey Staff, 1960). The formation of a massive structure is apparently a process of soil degradation. However, it is not clear at this point if it is the intrinsic soil properties, soil climatic conditions, or management that play a major role on the formation of a massive structure. Some soil regions appear more subject to develop a massive structure than others. Nonetheless, between granular and massive conditions, a range in structure types exists in a given region with similar parent material and climatic condition but under different land use systems (Coulombe, 1996). Little is comprehensively known about variation in structure of surface horizons of Vertisols with respect to parent material, climate, and various types of land use systems. The surface horizon with a granular structure often overlays a horizon that may be 10 to 50 cm thick with a prismatic structure (Fig. 4). These prisms that isolate large ped units are formed due to cracking upon soil desiccation. Even if the soil is tilled, this prismatic structure is commonly restored within the cropping season. From 25 to 125 cm, wedge-shaped structural aggregates, also called parallelepipeds, are present. These result from the intersection of slickensides. Slickensides and wedge-shaped aggregates were first described as lentils by Krishna and Perurnal (1948) and subsequently as cuneate and bicuneate structures by other authors, e.g., Worrall (1957), Brewer (1964), and De Vos and Virgo (1969). These features result from shrink-swell phenomena which will be discussed in the next section. Slickensides, periodic cracking, and a minimum clay content serve as the morphogenetic markers and major diagnostic characteristics common to all Vertisols in Soil Taxonomy (Soil Survey Staff, 1975, 1992, 1994). Topographic features that may occur, but are not common to all Vertisols, are gilgai and diapiric structures (diapir) (see Fig. 5). The diapir is a protrusion of subjacent soil material which penetrates into the overlying horizons and approaches or reaches the surface. This protrusion can be contrasted with the adjacent material by texture and color. From Paton (1974), gilgai is a microtopographic feature that consists of a mound, the higher part of the microrelief, a
OVERVIEW OF VERTISOLS
305
Figure 5 (a) Nuram (circular) a-type gilgai under ponded conditions on a Vertisol of Texas; (b) cross-sectional view of a Vertisol of Texas. Note the slickensides forming a synclinorium and the diapiric structure. The scale bar represents approximately I rn.
depression, the lower part of the microrelief, and a shelf, a planar or subplanar area intermediate in elevation between the two other elements. The term gilgai is an Australian aboriginal term meaning small water hole. It refers to the seasonal
3 06
C.E. COULOMBE, L. P. WILDING, AND J. B. DIXON
accumulation of water in the lower part of these topographic features (Paton, 1974). These features were first described in Australia by Jensen (191 1). They have stimulated considerable interest in an attempt to understand their morphology, genesis, and properties (e.g., Hallsworth et al., 1955; Stephen et al., 1956; Hallsworth and Beckmann, 1969; Paton, 1974; Knight, 1980; Hubble, 1984; Wilding and Tessier, 1988). Hallsworth and Beckmann (1969) described six types of gilgai: normal or round gilgai, melon-hole gilgai, lattice gilgai, linear or wavy gilgai, tank gilgai, and stony gilgai. Subsequently, Paton (1974) proposed a more rigorous classification of gilgai. The classification scheme was limited to two classes: linear and nuram (circular) gilgai. Nuram is another Australian aboriginal term meaning pock-marked, and refers to the general appearance of these features on aerial photographs (Paton, 1974). Example of a nuram (circular) gilgai is presented in Fig. 5a. Paton (1974) further subdivided each of the classes into four types (Fig. 6):
1
a-type
Figure 6 Gilgai types according to Paton (1974):a-type, mound and depression equally developed; P-type, mound of much greater extent than depression; y-type, depression of much greater extent than mound; and &type, mound, shelf, and depression all present.
OVERVIEW OF VERTISOLS a-type: P-type: y-type: 8-type:
3 07
Mound and depression equally developed, no shelf present; Mound of much greater extent than depression, no shelf present; Depression of much greater extent than mound, no shelf present; Mound, shelf, and depression all present.
Diapiric structures (diapirs) are subsurface features termed mukkara by Paton (1974) that may occur in a Vertisol independent of the presence of gilgai microrelief. If gilgai and diapir (mukkara) occur, the mound in gilgai is always developed over the diapir (Paton, 1974). An example of a diapiric structure is shown in Fig. 5b.
V. PEDOGENIC PROCESSES IN VERTISOLS Shrink-swell phenomena in Vertisols occur from macroscopic to molecular levels of resolution (Fig. 7A). Slickensides, wedge-shaped aggregates, cracks, gilgai, and special physical features discussed previously are observed at macroscopic levels (Fig. 7A). A striking example of a slickenside is shown in Fig. 8a. At a microscopic level, the presence of stress-oriented plasma (clay particles) in the form of plasma separations (sepic plasmic fabric of Brewer, 1964, 1976) or birefringence fabric (b-fabric of Bullock ef al., 1985) is an indicator of shrinkswell phenomena (Fig. 7b). These features are easily recognized under a light microscope (Fig. 8b) and are direct markers of preferential orientation of claysize particles or shear failure within the soil matrix. These plasma separations occur in the soil matrix, around skeleton grains, and/or along voids. Enhanced resolution by scanning and transmission electron microscopes provides useful tools to visualize submicroscopic structural features, and thus, helps elucidate mechanisms of shrink-swell phenomena (Figs. 7C and 7D). Enhanced knowledge about shrink-swell phenomena and clay microstructure is largely attributed to Tessier (1984). Clay-size crystallites may occur in soils as independent units (particles), in aggregates (domains), or forming network lattices (quasicrystals). Quartz, illite, and smectite are examples of single particles, domains, and quasicrystals, respectively. The stacking of these units will leave pores between the compound organizations (interparticle porosity; Fig. 7C) and/or within the quasicrystal structure (intraparticle porosity). In the case of smectite minerals, intraparticle porosity corresponds to pores left inside the quasicrystal and the interlayer space (Figs. 7D and 7E). Under specific conditions, interlayer expansion is partially responsible for shrink-swell behavior (Tessier, 1984; Wilding and Tessier, 1988). However, under most field soil conditions, the major proportion of water and volume changes is attributed to changes in microstructure and porosity at a submicroscopic level. The exceptions are sodium saturated, low-electrolyte
3 08
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
Figure 7 Shrink-swell features schematically illustrated at different scales of resolution.
~
~~
Figure 8 (a) Example of a slickenside in a Vertisol of Texas; (b) micrograph of oriented plasma (plasma separation) or birefringence fabric in a Vertisol of Texas. The scale bar represents approximately 5 cm for (a) and 100 p n for (b).
Figure 8
3 10
C. E. COULOMBE, L. P. WILDING, AND J. B. DMON
clay-water systems. Shrink-swell phenomena have been comprehensively reviewed by Quirk (1994), and Coulombe et al. (1996). The changes in microstructure and porosity upon changing soil moisture conditions are believed to induce soil movement which is directly related to the features observed from microscopic to macroscopic resolution scales. Several authors proposed models to explain genesis of these features. The first model, proposed by Hilgard (1906), concerned the formation of “hog-wallows.” This model is also known as “haplo’idizationby argillipedoturbation,” or simply called pedoturbation or self-mixing (Buol er al., 1980). The model states that during the dry season, cracks are formed. They extend to a depth of 1 m or more. While the cracks are open, surface soil material falls into them by biological activity, or as materials transported by wind, water or gravity. Upon rewetting, the cracks close due to the expansion of the clay in the soil and a space problem is generated. This results in the soil moving horizontally and vertically to accommodate the space of the infilled material. The final result is a development of weak horizonation, a homogenization of the whole profile, and a development of features at different morphological scales such as microshear, discontinuities, stress cutans, pressure faces on the peds, slickensides, and gilgai microtopography. This model is illustrated in Fig. 9 from two different perspectives: (a) soil displacement by heaving between cracks (Buol et al., 1980) and (b) soil displacement by heaving over a crack (Knight, 1980; Duchaufour, 1983). However, according to Wilding (1985) and Wilding and Tessier (1988), the self-mixing model does not explain observed attributes such as the systematic depth functions of organic carbon, carbonates, and salts, the increasing mean residence time of organic matter with depth, and the formation of albic and Bt horizons in Vertisols. These attributes should not exist in Vertisols formed primarily by the self-mixing model due to profile homogenization and turnover (inversion). Wilding and Tessier (1988) discussed three models, including the preceding one, for the genesis of Vertisols. The second model proposed is the “differential loading model” which states that gilgai in .Vertisols would result from a process whereby clays move from areas of higher confining pressures to areas of lower confining pressures under plastic-viscous flow (Paton, 1974). This model suggests that the overburden loads exceed the shear strength of the underlying material and enable flow to occur (Knight, 1980). This model would be rejected if slickensides were present because slickensides are evidence of shear failure planes. The differential loading model and the self-mixing model are only partially functional and do not explain the formation of other features such as slickensides (Wilding and Tessier, 1988). Finally, Wilding and Tessier (1988) used the Coulomb-Mohr theory of shear failure to explain formation of features in Vertisols. Knight (1980) also used the Coulomb-Mohr theory to explain the formation of gilgai at Boorook in Australia. This theory is based on an empirical equation, proposed by Coulomb about 1773-1776 and rekindled by Mohr in 1882:
OVERVIEW OF VERTISOLS
311
Figure 9 Haploidization or self-mixing models of Vertisols: (a) soil displacement by heaving between cracks (adapted from Buol er al.. 1980) and (b) soil displacement by heaving over a crack (adapted from Duchaufour, 1983; and Knight, 1980).
T
=c
+ u tan 4
(1)
where T = shear strength, c = cohesion, u = normal stress on the shear strength, and 4 = angle of internal friction. The shear strength is the resistance to deformation by continuous shear dis-
3 12
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
placement of soil along a surface of rupture (Cernica, 1982). According to Coulomb, shear strength is a linear function of the normal stress (Fig. 10a). Subsequently, Mohr proposed that the function is a curvilinear relationship (Fig. lob). Nonetheless, the Coulomb and the Mohr relationships are not significantly different. Cohesion (c) is the y-intercept of the function and is dependent on particle density, bulk density, clay content, clay mineralogy, temperature, and moisture content. The angle of internal friction ($) is related to the abundance, roughness, and interlocking of mineral grains (Wilding and Tessier, 1988). For instance, in a completely incohesive soil such as sandy soil, cohesion would be 0 and the shear strength function would be reduced to the product of normal stress (a)times the coefficient of internal friction (tan +). Similarly, the angle of internal friction could be 0 for the same soil when saturated for a prolonged period. Upon rewetting a dry soil, cohesion increases while the angle of internal friction decreases. Drainage conditions are taken into account in the relationship by using the pore water pressure (p,) as described below and illustrated in Fig. 11: T =
c
+ (a + p,) tan +
(2)
Under a relatively low normal stress value, e.g., a < ar (Fig. I I), the cohesion and the shear strength increase as the moisture content of the soil increases. However, when under a higher normal stress, e.g., (J’ < u < a” (Figure 1l), the shear strength of a dry soil is higher than under wet conditions. Under a higher normal stress value, e.g., (J > a”(Figure 1I), the shear strength of the soil is lower in the case of a saturated soil than under than either wet or dry
NORMAL STRESS
NORYAL STRESS
FEgure 10 Shear strength as a function of stress (a) according to the Coulomb theory and (b) according to the Mohr theory.
OVERVIEW OF VERTISOLS
I
313
Saturated
I-
PW a
rIn-
I,,
1,
U
NORMAL STRESS
Fcgure 11 Influence of pore water pressure (p) on shear strength (4) under dry, wet, and saturated field conditions, and at different normal stresses (u).
conditions. Consequently, soil moisture conditions and the normal stress of the soil need to be considered simultaneously in order to determine in which section of the graph shear failure is most likely to occur. This model works well for soils with low clay content. However, in the case of clayey soils, such as Vertisols, the model is not satisfactory because it would suggest that the shear strength is lower under dry conditions and low normal stress. Simple field observations reveal that cohesion of a clayey soil is much higher when dry than under moist or saturated conditions. Consequently, the shear strength is lower and shear failure is more likely to occur under moist or wet rather than dry conditions. Under saturation, the shear strength of a clayey soil is much lower than under dry and moist conditions, and plastic viscous flow is likely to occur even under low normal stress values. Therefore, formation of slickensides in Vertisols is likely to occur under conditions when the soil is moist or wet requiring a lower normal stress compared to the normal stress under dry conditions. In a soil profile, stresses act in a triaxial way: one in the vertical dimension (vertical stress) and two in the horizontal or lateral dimensions (lateral stresses). In order to keep the model simple, only two dimensions will be considered in Fig. 12. The failure plane occurs when vertical forces are confined and lateral
3 14
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
A
Lateral stress (oV)
B
Soil material
I
aN
OV
1 uv
t
OV
Figure 12 Slickenside formation according to the theory of shear failure when stresses exceed shear strength in a confined system: (A) Vertical and horizontal stresses acting on soil ped, and (B)orientation of shear plane at 45" to the principal stress (from Wilding and Tessier, 1988).
stresses exceed the shear strength of the soil. Theoretically the angle of the failure is at 45" to the horizontal minus half the angle of internal friction. The latter is considered negligible in Vertisols because of their high clay content but in practice shear planes are observed from about 20 to 60". This model would explain the formation of slickensides. Wilding and Tessier (1988) proposed a model which integrates the CoulombMohr theory of shear failure in the formation of features in Vertisols (Fig. 13). As previously shown in Fig. 4, slickenside patterns in Vertisols tend to follow the microtopography and form a synclinorium. Assuming the pattern is associated with major desiccation cracks, the formation of gilgai would proceed as follows: (i) formation of deep cracks during the dry period (Fig. 13A), (ii) upon rewetting a dry soil, the wetting front moves from the bottom of the cracks to the surface and the cracks close (Figs. 13B and 13C), (iii) slickensides are formed and induce the formation of thrust cones due to the polygonal network of cracks and possibly because the forces act in a three-dimensional way (Fig. 13D), (iv) formation of these thrust cones forces the soil to move up inducing a slightly
OVERVIEW OF VERTISOLS
315
Figure 13 Schematic illustration of possible stages (A-F) in the formation of slickensides, gilgai, and cyclic horizonation (from Wilding and Tessier, 1988).
undulating topography (Fig. 13E), and (v) the topography is amplified by drying-wetting cycles and/or differential leaching. The final result is a wavy horizonation geometry and a microtopography composed of mounds and depressions (Fig. 13F).This model accommodates the rapid formation of slickensides, systematic distribution of solutes, pedogenic structure and normal gilgai, and recognizes pedoturbation as a long-term process (Wilding and Tessier, 1988). It also explains the oblique translation of surface or underlying materials along the slickenside planes. However, the model proposed by these authors remains subject to controversy by advocates of the self-mixing model. Some mixing of A and Bss or BC materials can be found in soils within depths of cracking and adjacent to slickenside planes which support the pedoturbation model. However, slickensides are also found below the depths of normal field cracking, which tends to refute the need for infilled materials to initiate slickenside failure. Finally, pedogenic processes other than shrink-swell phenomena are functional in Vertisols. Several of these processes also are recognized in the recent Keys to Soil Taxonomy (Soil Survey Staff, 1994) and include: acidification (dystric great group); dissolution and accumulation of silica (duric),carbonates (culcic), gypsum (gypsic),sodium (nutric and sodic) and other soluble salt (sulk) great groups or subgroups; gleification (uquic suborder) with respect to saturation by water,
3 16
C. E. COULOMBE, L. P. WILDING, AND J. B. DMON
reduction, and formation of redoximorphic features at specific depths (epi and endo great groups); weathering transformation and neoformation of minerals; and degradation and distribution of organic matter. Several of these processes are discussed further in the next sections (e.g., mineralogical, chemical, and biological properties) but not all of the processes mentioned herein occur in every Vertisol. Their magnitudes and morphogenetic expressions are dependent on sets of soil forming factors and environmental conditions acting in concert to influence combinations of soil processes. Aspects that have received little attention are the soil degradation and regeneration processes. As with other soil forming processes, degradatiodregeneration dynamics affecting physico-chemical properties, pertinent to soil management, are deserving of greater attention than previously received. Identification and control of these processes could be helpful when proposing strategies to maintain and improve soil, water, and air quality.
VI. CLASSIFICATION: FROM MARBUT TO SOIL TAXONOMY Vertisols, as a soil order, have been recognized in many international classification systems, e.g., Soil Taxonomy (Soil Survey Staff, 1960, 1975, 1992, 1994), F.A.0.-Unesco (1974, 1982, 1988), C.P.C.S. (1967), and Referentiel Ptdologique (Baize and Girard, 1992). Prior to the formal adoption of the term Vertisol, different names were used around the world to designate these shrinkswell soils. For instance, Vertisols were called Rendzinas and Grumusols in United States (Baldwin et af., 1938; Oakes and Thorp, 1950), regur or black cotton soils in India (Krishna and Perumal, 1948; Simonson, 1954a, b), gray and brown soils, black earth and cracking clay soils in Australia (Prescott, 1931; Stace et af., 1968; Hubble, 1984), tirs in Morocco (Del Villar, 1944), and tropical black clays or dark clay soils in Africa (Stephen el al., 1956; Dudal, 1965). An exhaustive list of names used to designate Vertisols is reported in Dudal and Eswaran (1988). The variety of names demonstrates their wide occurrence in the world and their common peculiar properties that make these soils unique. This section discusses more specifically the recognition and evolution of Vertisols with respect to U.S. classification systems. The first recognized U.S. soil classification system was published by Marbut (1928). Marbut’s classification was heavily influenced by the Russian work translated by Glinka (1914). It emphasized the concepts of soil type, e.g., Pedalfers and Pedocals. Presumably, Vertisols would have been classified in the Pedocals of tropical zones at the higher categories and as soils with imperfectly developed profiles at lower categories. At that time, pedology was a young science and little was known about high shrink-swell phenomena of these soils.
OVERVIEW OF VERTISOLS
317
The 1938 Soil Classification System in the United States (Baldwin et al., 1938) emphasized the concept of soil zonality, e.g., zonal, intrazonal, and azonal soils. Vertisols would have been classified at the order of intrazonal soils, suborder of calcimorphic soils, and in the great group of Rendzinas. Rendzinas were defined as dark soils derived from calcareous parent materials and with profiles ranging from about 38 to 150 cm thick. It is noteworthy that the definition of Rendzinas in the 1938 Soil Classification system in the United States was different from the European sense of Rendzina soils which were considered as shallow to very shallow, gray to black soils derived from limestone or chalk (Oakes and Thorp, 1950). At the present time, the term Rendzinas, as defined by the Europeans and still utilized in some classification systems, e.g., F.A.0.-Unesco (1988), does not refer to Vertisols but is used for shallow calcareous soils. Major changes were proposed by Oakes and Thorp (1950) concerning a name and a tentative definition for these clayey, shrink-swell soils in order to eliminate the lack of uniformity for their designation and definition. The term Grurnusol, as a great group name, was introduced for these dark clayey soils derived under varied climatic conditions but generally with alternating wet and dry seasons. The name originates from the Latin roots grumus which means “little heap or hillock” and sol for “soils.” The term also refers to soils exhibiting a crumb structure at the surface. To meet the requirements of the Grumusol great group, Oakes and Thorp (1950) suggested 15 characteristics that these soils generally have in common. These attributes were as follows: 1. Clayey texture in the “typical” form; 2. No eluvial or illuvial horizons; 3. Moderate to strong granular structure in the upper 15 to 50 cm, becoming blocky or massive with depth; 4. Calcareous reaction normally but with acid to neutral intergrades to other groups; 5. High coefficient of expansion and contraction on wetting and drying; 6. Gilgai microrelief; 7. Extremely plastic consistence; 8. Exchange complex nearly saturated with calcium, or calcium and magnesium; 9. Clay minerals dominantly of the Montmorillonite (Smectite) group; 10. Parent material mostly calcareous, high in clay content, and nearly impervious; 11. Sola more than 25 cm thick and typically more than 75 cm thick; 12. Dark color of low chroma; 13. Medium to low content of organic matter (usually 1 to 3% in the surface) and the organic matter gradually decreasing with depth; 14. Stage of weathering relatively unadvanced or minimal; 15. Tall grass or savanna vegetation.
318
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
Even though these criteria were not rigid, they were used as guidelines and allowed differentiation of these soils from the Chernozems (Mollisols). During the 6th Congress of the International Society of Soil Science in 1956, the term Vertisols was proposed to qualify soils having high shrink-swell characteristics (Dudal and Eswaran, 1988). This term originates from Latin roots verzere which means “to chum or turn over” and sol for “soil.” This name was preferred over Terrasol, expressing the same meaning but originating from Greek roots (Dudal and Eswaran, 1988). The term Vertisols was subsequently adopted in the 7th Approximation (Soil Survey Staff, 1960) as a soil order to recognize shrink-swell clayey soils. This classification was the first to recognize Vertisols as a distinct group with unique properties. Table I1 shows an overview of the Vertisol categories from suborders to subgroups, according to the 7th Approximation (Soil Survey Staff, 1960). The Vertisol order had only two suborders, aquic and ustic, based on soil moisture regimes. At the great group level, the class differentia had a structural bias, i.e., grumic and mazic, which refers to a granular and massive structure, respectively. The grumic character originates from the proposed Grumusol great group proposed by Oakes and Thorp (1950). Finally, 17 subgroups represented the central, intergrade, and extragrade concepts toward other soil orders or suborders of the classification system. With the introduction of Soil Taxonomy (Soil Survey Staff, 1975), several major changes were made in the 7th Approximation (Soil Survey Staff, 1960) classification of Vertisols. According to Soil Taxonomy (Soil Survey Staff, 1975, 1990), Vertisols were defined as Mineral soils that have a frigid or warmer soil temperature regime; that do not have a lithic or paralithic contact or petrocalcic horizon or duripan within 50 cm of the soil surface; that after the upper soil to a depth of 18 cm has been mixed, have 30% or more clay in all horizons down to a depth of 50 cm or more; that at some period in most years have cracks that are open to the surface or to the base of a plow layer or surface crust and are at least 1 cm wide at a depth of 50 cm unless the soil is irrigated; and that have one or more of the
Table I1 Vertisol Categories in the 7th Approximation (Soil Survey Staff, 1960) Suborders
Great groups
Subgroups
Aquerts
Grumaquerts Mazaquerts Grumusterts Mazusterts
(Typic), orthic, entic, mollic (Typic), orthic, entic, natraquollic, natraqualfic, aquultic (Typic), orthic, entic (Typic), orthic, natrustalfic, natrargidic
Usterts
OVERVIEW OF VERTISOLS
3 19
following characteristics: a) gilgai; or b) at some depth between 25 cm and 1 m, slickensides close enough to intersect; or c) at some depth between 25 cm and 1 m, wedge-shaped (sphenoid), structural aggregates whose along axes are tilted 10" to 60" from the horizontal. At the suborder level, Vertisols had four taxa in Soil Taxonomy (Soil Survey Staff, 1975, 1990) instead of two in the 7th Approximation (Soil Survey Staff, 1960). The aquic suborder was dropped and the torric, udic and xeric suborders were added, based on soil moisture regimes. Reasons for the elimination of the aquic suborder were not documented. The determination of the moisture regime in Vertisols was based on the number of days that soil cracks remain open. For instance, xererts are Vertisols in which the cracks open and close once each year and remain open for 60 or more consecutive days during the 90 days following summer solstice. Torrerts are Vertisols that are usually dry in all parts of the solum and whose cracks are open to the surface more than 305 days in most years unless irrigated. Uderts are Vertisols that are usually moist but in which cracks open at sometime during the year but do not remain open for as many as 90 days cumulative each year. Usterts are Vertisols in which the cracks open and close more than once a year but remain open for a total of 90 or more days a year but not more than 305 days in most years. Cracks, that remained open for given times, were used to reflect soil moisture regimes in these soils because of the difficulty in determining a soil moisture control section in soils where water movement occurs as an irregular front along crack and slickenside zones. At the great group category, the bias was toward soil color instead of soil structure. With the exception of the Torrert great group, the class differentia were chromic, i.e., chroma value 2 1 . 5 in the upper 30 cm of the solum and pellic, i.e., a chroma value <1.5 in the upper 30 cm of the solum (Soil Survey Staff, 1975). The deletion of the structural bias at the great group level is not well documented. Probable reasons were problems related to quantification and variation of soil structure of surface horizons (Wilding and Williams, 1991). In the 4th edition of the Keys to Soil Taxonomy (Soil Survey Staff, 1990), the color limit for soil chroma became 2 rather than 1.5 to differentiate between pellic and chromic great groups. The rationale for this change was that the eye can more consistently determine soil color when a color chip is used rather than when the critical value is between color chips, so a matter of operator consistency was invoked. Finally, subgroup categories represented either the central concept or intergrade and extragrade concepts between other soil orders and suborders of the classification system. From 1975 to 1990, the number of subgroups increased from 24 to 27. An overview of the classification of Vertisols at the suborders, great groups, and subgroups in Soil Taxonomy (Soil Survey Staff, 1975, 1990) appears in Table 111. A vertic diagnostic horizon has been proposed for inclusion in Soil Taxonomy
320
C. E. COULOMBE, L. P. WILDING, AND J. B. DMON Table Ill Vertisol Categories in Soil Taxonomy (Soil Survey Staff, 1975, 1990) Suborders Torrerts Uderts Usterts Xerets
Great groups
Subgroups
Torrerts Chromuderts Pelluderts Chromusterts Pellusterts Chromoxererts Pelloxererts
Paleustollic, mollic, argidic,a typic Aquentic, aquic, entic, typic Entic, typic Udorthentic, udic, paleustollic, entic, typic Udorthentic, udic, entic, paleustollic,a typic Aquic,O entic, palexerollic, typic Chromic, entic, typic
Added in the 4th edition of “Keys to Soil Taxonomy” (Soil Survey Staff, 1990).
(Blokhuis et al., 1991; Mermut et al., 1991) but has not yet been accepted. However, a horizon with slickensides (Bss) as recognized in the field as a vertic condition has been accepted since the 4th revision of the Keys to Soil Taxonomy (Soil Survey Staff, 1990). Subsequent revisions in Soil Taxonomy published from 1975 to 1990 included few changes in the Vertisol order, but another major set of revisions for Vertisols occurred in Soil Taxonomy in 1992. The 5th edition of the Keys to Soil Taxonomy (Soil Survey Staff, 1992) again contained significant changes in the concept of Vertisols. These changes were proposed by ICOMERT (International Committee on Vertisols; Comerma et al., 1988). Many of the new categories recognized within the current classification of Vertisols were to accommodate their diversity in conditions, to aid in better interpretations for use and management, and to deemphasize soil color at the higher categorical levels. The definition of Vertisols remained similar but the presence of gilgai was no longer one of the auxiliary criteria that could be used to identify them. Gilgai is not a common feature to all Vertisols. However, when it is present, it is a good clue that Vertisols exist because no examples of gilgai have been found without slickensides present. The reverse, however, does occur. Thus on aerial photographs, the identification of gilgai is a useful indirect inference of Vertisols but does not currently serve as a criterion. Gilgai as a differention was deernphasized because there was too much confusion among different soil scientists as to what phenomena constituted gilgai (e.g., animal mounds, freeze-thaw microrelief, extreme expressions of selfswallowing microtopography). The Vertisol order in the 5th edition has 6 suborders, 23 great groups, and 153 subgroups (Table IV). The aquic suborder was reintroduced. Also, Vertisols of cold regions were recognized in the cryert suborder. The latter implies that the temperature regime, particularly short growing seasons and freeze-thaw phe-
OVERVIEW OF VERTISOLS
32 1
Table IV Vertisol Categories in the 5th and 6th Editions of the Keys to Soil Taxonomy (Soil Survey Staff, 1992, 1994) Suborders
Great groups
Subgroups
Aquerts
Salaquerts Duraquerts Natraquerts Calciaquerts Dystraquerts Epiaquerts Endoaquerts Humicryerts Haplocryerts Salitorrerts Gypsitorrerts Calcitorrerts Haplotorrerts Dystruderts Hapluderts Dystrusterts Salusterts Gypsiusterts Calciusterts
Aridic, ustic, leptic, entic, chromic, typic Aridic, xeric,“ ustic, aeric, chromic, typic Typic Aeric, typic Sulfaqueptic, alic, aridic, ustic, aeric, leptic, entic, chromic, typic Halic, sodic, aridic, xeric, ustic, aeric, leptic, entic, chromic, typic Halic, sodic, aridic, xeric, ustic, aeric, leptic, entic, chromic, typic Sodic, typic Sodic, chromic, typic Aquic, leptic, entic, chromic, typic Chromic, typic Petrocalcic, leptic, entic, chromic, typic Halic, sodic, leptic, entic, chromic, typic Alic, aquic, oxyaquic, leptic, entic, chromic, typic Lithic, aquic, oxyaquic, leptic, entic, chromic, typic Lithic, aquic, aridic, udic, leptic, entic, chromic, typic Lithic, sodic, aquic, aridic, leptic, entic, chromic, typic Lithic, halic, sodic, aridic, udic, leptic, entic, chromic, typic Lithic, halic, sodic, petrocalcic, aridic, udic, leptic, entic, chromic, typic Lithic, halic, sodic, petrocalcic, aridic, leptic udic, entic, udic, chromic udic, udic, leptic, entic, chromic, typic Halic, sodic, aquic, aridic, udic, haplic, chromic, typic Lithic, aridic, petrocalcic, leptic, entic, chromic, typic Lithic, halic, sodic, aridic, aquic, udic, leptic, entic, chromic, typic
Cryerts Torrerts
Uderts Usterts
Haplusterts Xererts
a
Durixererts Calcixererts Haploxererts
Added in the 6th edition of “Keys to Soil Taxonomy” (Soil Survey Staff, 1994).
nomena, is more important than the moisture regime in these Vertisols. The consideration of the temperature regime at the suborder category brings some redundancy because it is considered at the family level also. However, the family temperature regime is more specific because several kinds of cryic thermal classes occur under different land use and drainage conditions. The great group categories have a morphogenetic bias by considering processes such as salt accumulation (salic, natric, calcic, and gypsic great groups), silica accumulation (duric great group), and depth of saturation by water (epi and endo great groups). Finally, 153 categories were distinguished at the subgroup level in the 5th edition of the Keys to Soil Taxonomy (Soil Survey Staff, 1992). Some of the subgroups are intergrade or extragrade concepts toward other soil orders. How-
322
C. E. COULOMBE, L. P. WILDING, AND J. B. DMON
ever, the majority of the subgroups have a morphogenetic or a moisture regime bias. It is noteworthy to mention that the color bias, e.g., chromic subgroup which previously occurred at the great group level in the earlier versions of Soil Taxonomy, was relegated to a subgroup level in the recent keys. At the greatgroup levels, all Vertisols are considered to be dark-colored (pellic). The reason color had a decreased emphasis was that it was not a very good indicator of soil aeration, drainage or soil utilization. Commonly it reflected the nature of the parent material or landscape erosional processes rather than redoximorphic or organic matter accumulation processes. The 6th edition of the Keys to Soil Taxonomy (Soil Survey Staff, 1994) is similar to the 5th edition except the addition of the Xeric Duraquert subgroup (Table IV). This highly complex Vertisol classification scheme reflects the diversity in the world of Vertisols. But perhaps a negative aspect is the number of categories proposed at the subgroup level which makes the classification of Vertisols much harder to comprehend than the previous revisions. Nonetheless, this was done to accommodate the wide diversity in Vertisol properties globally and to be consistent with other soil orders in Soil Taxonomy (Wilding and Williams, 1991). Conceptual changes, perhaps with a complementary classification scheme, that would include degradation and regeneration processes and their impact on soil structure, porosity, and associated physical properties would add considerably to the Vertisol knowledge base (Coulombe, 1996).
VII. MINERALOGICAL PROPERTIES Many minerals in Vertisols originate from inheritance, transformation, and/or neoformation (see Coulombe et al., 1996). The variability in parent materials and environmental conditions leads to the complexity in mineralogy of Vertisols. This section briefly discusses recent concepts of mineralogy in Vertisols.
A. PHYLLOSILICATES Most mineralogical reports on Vertisols have emphasized clay minerals, i.e., phyllosilicates, due to their high clay content. Phyllosilicates are silicates with a layer structure. Classification and selected properties of common phyllosilicates are presented in Table V. Identification of these phyllosilicates may also give valuable information about the origin or conditions that promote Vertisol formation and evolution of other mineral phases. The amount and type of phyllosilicates are fundamental aspects involved in the microstructure of these soils and
OVERVIEW OF VERTISOLS
323
affect in turn soil behavior such as shrink-swell phenomena and many of the chemical and physical properties that are discussed later in this chapter. Smectite has been considered a major component of Vertisols in most reviews, with kaolinite of secondary importance (Dudal, 1965; Dixon, 1982; Hajek, 1985). However, kaolinite has been reported as a particularly abundant constituent of some Vertisols in El Salvador (Yerima et al., 1987), Sudan (Yousif et al., 1988), Hawaii (Ikawa, 1985), and Australia (Norrish and Pickering, 1977). Hubble (1984) mentioned that kaolinite was co-abundant with illite in the clay fraction of Northern Australian Vertisols. In the case of Hawaiian Vertisols, electron microscopy revealed that the mineral believed to be kaolinite was dehydrated halloysite (Malik, 1990; Malik and Jones, 1991). Halloysite, like kaolinite, is a 1 :1 clay mineral but differs by the presence of a monolayer of water in the interlayer space. Halloysite has been reported quite often as a minor constituent in Vertisols, yet Vertisols containing abundant halloysite have been reported only in Northern Cameroon (Simon et al., 1975). Many vertic intergrades developed from volcanic deposits are halloysitic in mineralogy, e.g., Central America and regions of volcanic islands. Mite is another clay mineral reported as abundant in Vertisols. By definition, illite is a clay-size mica. Micas in Vertisols may occur in the sand, silt, and clay fractions. Their relative abundance is quite variable from one Vertisol to the next. For instance, Dixon (1982) mentioned that mica generally occurred in small amounts (about 5-14% mica in Texas Vertisols). On the other hand, illite was reported abundant in South Australian Vertisols (Norrish and Pickering, 1977) and coabundant with kaolinite in many Vertisols in Northern Australia (Hubble, 1984). Another clay mineral, reported by Murthy (1988), as abundant in Vertisols of Karnataka, India, was hydroxy-interlayered smectite (HIS), rich in iron and with high tetrahedral charge. Apparently, these Vertisols contained practically no kaolinite and mica (Murthy, 1988). The process responsible for precipitation of aluminum hydroxy-polymers under these Vertisol conditions was not reported. Many other phyllosilicates such as vermiculite, chlorite, hydroxy-interlayered vermiculite (HIV), and palygorskite have been reported as minor or trace components in Vertisols (see Coulombe et al., 1996). Furthermore, several types of interstratified clay minerals have been reported: kaolinite-smectite in El Salvador (Yerima, 1983; Yerima et al., 1985), West Indies (Ahmad and Jones, 1967), and Australia (Norrish and Pickering, 1983; Bottinger, 1992); mica-vermiculite in Northern Utah (Graham and Southard, 1983) and Cameroon (Yerima et al., 1988); mica-srnectite in Jordan (Shadfan, 1983), France (Tessier et al., 1991), and Hawaii (Malik, 1990; Malik and Jones, 1991); vermiculite-smectite and vermiculite-chlorite in Texas (Kunze et al., 1963; Carson and Dixon, 1972). Nevertheless, smectite remains the most abundant and common phyllosilicate reported in Vertisols around the world. Montmorillonite, beidellite, and nontronite are smectites reported in Vertisols. Montmorillonite and beidellite are the
Table V Classification and Selected Properties of Common Pbyllosilicates(from Coulombe ef ul., 19%)
Mineral
Type
Arrangement of tetrahedral and octahedral sheets
Layer charge (half unit cell)
Cation exchange capacity (cmol kg-l)
Surface area (m2 g-')
11
0
?
e
s
8
P
I
V
I
s
11
e
2m
P
tt
0
ijia 1.1
324
3
.. ..
3
N N
I
5 P
d 0
2
-
11
3
N
a
3
al Y
U
2
'ti
X
i
N
325
326
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
two most common ones that have been verified thus far. A nontronite was reported in a Vertisol of Upper Volta (Trauth et al., 1967). Soil smectites frequently contain an appreciable amount of iron even though they seldom qualify as nontronite. Generally, the occurrence of iron-rich smectite is related to the higher iron content in the parent material containing abundant ferromagnesian minerals such as those derived from igneous origin. The layer charge is an important aspect of smectite minerals. Ideally, the layer charge ranges between 0.3 and 0.6 electron per half unit cell in smectites. This is located in the octahedral sheet for montmorillonite and in the tetrahedral sheet for beidellite and nontronite. However, Tessier and Pedro (1987) reported that smectites having a high layer charge deficit, i.e., between 0.45 and 0.60 electron per half unit cell, are common in soils. These smectites with a high layer charge also have the propensity to fix potassium and ammonium cations. Moreover, smectites with a layer charge in the range of vermiculite (0.6-0.9 electron per half unit cell) also were reported in Vertisols (Chen et al., 1989; Badraoui and Bloom, 1990). The tetrahedral charge contributes more than 50% of the total charge. A high layer charge and the site of charge deficiency, especially when coming from the tetrahedral sheet, are very important attributes contributing to potassium and ammonium fixation. Besides the layer charge and the iron content, soil smectites such as those reported in Vertisols exhibit other characteristics that make them different from geologic or reference smectites. One of these characteristics is that soil smectites are thermodynamically more stable under acid conditions than other reference smectites (Carson et al., 1976). To maintain their stability, smectite requires a high silicic acid concentration. After equilibrated, the concentration of silicic acid of a Houston Black soil smectite was above the level required for amorphous silica and evidently higher than the level of reference smectites Belle-Fourche and Aberdeen (Carson et al., 1976). Desilication o f trace mica, possibly as an interstratified component, was considered responsible for maintaining this high level of silicic acid. Another characteristic of soil smectites is that they are generally small in size and exhibit high specific surface areas. As discussed previously, the mechanism of shrink-swell is mainly attributed to the size of clay assemblages and their specific surface areas. Smectites that have higher specific surface areas exhibit higher shrink-swell potential. Kaolinite, illite, and other clay minerals may exhibit significant shrink-swell potential if they have fine clay particle sizes and high specific surface areas (Tessier, 1984).
B. MINERALS OTHER THAN PHYLLOSILICATES Silicates other than phyllosilicates, oxides-oxyhydroxides, carbonates, sulfides, sulfates, etc., are also common mineralogical constituents of Vertisols (see
OVERVIEW OF VERTISOLS
327
Coulombe et al., 1996). These components may occur in diverse particle-size fractions. Their occurrence depends heavily on the origin of the parent material and past and current environmental conditions. Several of these minerals are discussed briefly below. Quartz is the most common and dominant mineral of the silicate group reported in Vertisols. This mineral can be inherited from acidic igneous and metamorphic parent rocks, sedimentary deposits, and/or neoformed in soils. Feldspars also are very common in Vertisols. Alkali feldspars (K-feldspar) and plagioclase feldspars (Na-Ca feldspar series) weather to form clay minerals and supply potassium, sodium, and calcium cations as plant nutrients. Also reported, in minor and trace abundance, are zeolites in the silt fraction of Vertisols of West Indies (Ahmad and Jones, 1969) and Northern Utah (Graham and Southard, 1983), and garnet, epidote, and zircon in the sand and silt fractions of a Vertisol derived from gneiss in Ghana (Stephen, 1953). Hornblende, pyroxene, and sphene also occur in Vertisols derived from basic igneous parent materials. Poorly crystallized silica such as opal-A and opal-CT have been reported in Vertisols of El Salvador (Yerima, 1983; Yerima et al., 1985), Alabama and Mississippi (Dixon, 1978), and Australia (Bottinger, 1992). Amorphous glass (lechatelierite), allophane, and imogolite have not been reported in Vertisols; however, a significant proportion of vertic soils are derived from volcanic ash, tuff, and glass so their occurrence is probable in soils derived from pyroclastic deposits. It is likely that such minerals are transformed progressively to halloysitic and eventually clay-rich smectite in these vertic soils. Oxides and oxyhydroxides of iron, manganese, and titanium contribute to the pigmentation of Vertisols. Iron can contribute either a low chroma or high chroma depending on the mineral phase, while manganese and titanium oxides are generally black pigmenting agents. For instance, magnetite (Fe,O,) and ilmenite (FeTiO,) of the sand and silt fractions were found responsible for the dark color instead of organic matter in Vertisols of Hawaii (Raymundo, 1965). Iron and manganese are also two elements influenced by redox conditions in Vertisols subject to aquic moisture regimes. Many of these mineral phases are characterized as short-range order minerals. Wang ef al. (1993) identified ferrihydrite, lepidocrocite, and fine goethite in the clay fraction of vertic soils of East Texas. Iron and manganese also may be in the form of glaebules (concretions or nodules) in association with other mineral forms such as silicates and carbonates. Sullivan and Koppi (1992, 1993) also reported accumulation of a manganese oxide (bimessite) in structural pores of Vertisols of Australia. The fibrous manganese oxide todorokite is the major oxide in nodules from a Vertisol of northern Guatemala (Dixon et al., 1994). Carbonates are common minerals in neutral and alkaline Vertisols. They commonly occur either as calcite (CaCO,), Ca-Mg carbonates (Mg-calcite and dolomite), and undifferentiated Na-carbonates. Huntite [Mg,Ca(CO,),] and a
328
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
polymorph of calcite called aragonite were reported in Australian Vertisols (Veen and Amdt, 1973). Sodium carbonates are indicators of very alkaline conditions (pH > 9.5). Sodium carbonates also are considered as more soluble salts than gypsum and will be discussed below. Carbonates glaebules have been extensively studied in Vertisols (Singh and Lal, 1946; Blokhuis et al., 1968/1969; Rajan et al., 1972; Singh and Singh, 1972; Fitzpatrick and Le Roux, 1977; Mermut and Dasog, 1986). Commonly, nodules contain in addition to carbonates other minerals such as iron and manganese oxides and silicates of the soil matrix. The concentration of iron and manganese oxides inside carbonate glaebules is highly variable. Sulfides, sulfates, and salts more soluble than gypsum that occur in Vertisols may suggest extreme pH conditions. Sulfide minerals such as pyrite (FeS,) form in Vertisols under anaerobic conditions in coastal regions of the West Indies, Venezuela, and Texas Coast region (Ahmad, 1985; Comerma, 1985). Soil pyrite typically forms from bacterial activity under reducing conditions. Under drained conditions, pyrite oxidation proceeds and produces sulfuric acid (H,S04) and sulfate minerals such as jarosite [KFe,(OH)6(S04),], natrojarosite [NaFe,(OH)6(S04),], and gypsum (CaS04-2H,0). Sulfate minerals also may be inherited in Vertisol parent materials. Na minerals such as halite (NaCl), Nasulfates, and Na-carbonates are highly subject to dissolution and reprecipitation upon changing soil moisture conditions. Finally, phosphate minerals in Vertisols have been reported in India by Sarma and Krishna-Murty (1970). They identified a mineral of the plumbogummite group, a secondary hydrous phosphate, which is believed to result from the weathering of apatite. Apatite was reported in the parent material but not in the profile of Vertisols of Ghana (Stephen, 1953) and Morocco (Badraoui and Bloom, 1990). Several of these mineral phases discussed above are particularly sensitive to soil moisture and environmental conditions. Dissolution, translocation, and precipitation due to wet-dry cycles and the subsequent redistribution of these various mineral phases in the soil profile may impact soil structure and porosity negatively due to clogging of the pores as well as affecting the soil chemical and physical properties. The intensification of these wet-dry cycles may lead to the formation of (i) a duripan due to silica accumulation; (ii) argillic/natric horizons due to the translocation and immobilization of silicate clays; (iii) redoximorphic features such as iron and manganese oxides and sulfidic minerals; (iv) a sulfuric horizon by the presence of sulfide and sulfate minerals under low pH conditions; and (v) calcic, gypsic, and salic horizons formed by precipitation of calcium carbonates, gypsum, and very soluble salts, respectively. Horizons with concentrations of these chemical constituents will have diverse impacts on pH, cation exchange capacity, phosphorus sorption, structure and porosity, aeration, drainage, and hydraulic conductivity.
OVERVIEW OF VERTISOLS
329
VIII. CHEMICAL PROPERTIES Chemical properties of Vertisols were recently reviewed by Coulombe et al. (1996). Properties such as pH, cation exchange capacity, and exchangeable cations and behavior of some cation and anion species are discussed below.
A. PH Vertisols may be acid, neutral, or alkaline in reaction. The concept of acid Vertisols was reported earlier by Kunze et al. ( 1963) in Texas coastal regions and subsequently by other authors, e.g., Ahmad and Jones (1969), Parsons et al. (1973), Duchaufour (1983), Ahmad (1983, and Comerma (1985). Acid conditions may occur in the upper horizons or in the whole profile. Acid Vertisols may be due to acidic parent materials, dissolution of carbonates by leaching (decarbonatation) of the profile in a more humid climate, continuous leaching conditions, or alternating oxidation/reduction conditions (ferrolysis). Ferrolysis (Brinkman, 1978) involves destabilization of mineral structure by proton attack. Another process called “acid sulfate weathering,” which is the oxidation of sulfide minerals that produce sulfuric acid and sulfate minerals previously mentioned, is operational in Vertisols formed near coastal fringes (Ahmad, 1985; Comerma, 1985). Acid Vertisols are now recognized in Soil Taxonomy (Soil Survey Staff, 1992, 1994) with the dystric great group that requires pH values 5 4 . 5 CaCI, (55 in 1:l water) and an electrical conductivity of less than 4.0 dS m-* at 25°C of the saturation extract. The majority of Vertisols are neutral or alkaline because they are mostly derived from calcareous or base-rich parent materials. However, alkali conditions (pH > 8.3) may destabilize structural peds and lead to very slow permeabilities which are not desirable for agriculture purposes. The presence of sodium carbonates is an indication of alkali conditions. High pH conditions (8.5 to 9.5) promote an increase of silicon activity and dissolution of silicate minerals. Furthermore, organic matter is subject to dispersion. Both of these processes contribute to accelerated degradation of Vertisols.
B. CATIONEXCHANGE CAPACITY AND EXCHANGEABLE CATIONS Generally, Vertisols have a relatively high cation exchange capacity (CEC) which ranges from 20 to 45 cmol kg-I(soil) or even higher. The amount and type of clay, in particular the smectitic content, and the organic matter content are the determinant factors.
330
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
Cations that occupy the exchange sites vary with soil environmental conditions. In neutral Vertisols, the exchange sites are occupied mainly by calcium and magnesium and to a lesser extent by potassium and sodium. If sodium occupies more than 15% of the exchange complex (ESP), the soil is considered sodic (Soil Survey Staff, 1975, 1994); the high ESP negatively impacts soil properties and plant growth for salt-sensitive plants. However, Australians recognized that conditions are sodic between 6 and 15% ESP and extremely sodic above 15% ESP (Northcote and Skene, 1972). In fact, E S h <15% may have a more deleterious effect on soil dispersion than ESPs >15%, especially if the latter are associated with high electrolyte concentrations. In the latter case, the soil still remains flocculated while at ESPs <15%, soil particles may become dispersed due to low electrolyte concentration and saturating cations (Wilding and Tessier, 1988). However, flocculation is most likely to occur in the presence of flocculating agents, such as gypsum, even at low ESPs and sodium adsorption ratios (SAR). Even though ESP and SAR are important determinants in soil dispersion and shrink-swell potential, coupled values of electrical conductivity (EC) with ESP and/or SAR are necessary to accurately predict probable clay dispersion (Wilding and Tessier, 1988). Furthermore, if sodium ions occur in association with carbonate in soils, or react with atmospheric carbonic acid (H,CO,), sodium carbonate may form. This will commonly increase the soil pH above 8.5 and contribute to dispersion of mineral and organic components (see Coulombe er al., 1996). Aluminum, magnesium, and exchangeable acidity (H) increase inversely to replace calcium on the exchange sites under acid conditions. Acid Vertisols are very unique in this regard. For example, at equivalent pH values, only 25 to 40% of the exchange sites may be occupied by exchangeable aluminum compared to 80% by Spodosols, Ultisols, and Oxisols (Ahmad, 1985, 1988). As acid conditions persist, minerals such as kaolinite form, the CEC decreases, exchangeable aluminum, magnesium, and acidity increase on the exchange sites, and these attributes progressively lead to the chemical degradation of Vertisols.
C. CATIONAND ANIONBEHAVIOR Adsorption, precipitation, and diffusion are processes that influence cation and anion behavior in Vertisols. Potassium, ammonium, nitrate, phosphate, and sulfate are of particular interest due to their importance as nutrients for plant growth and groundwater contamination. Behavior of these ions is discussed briefly below. 1. Potassium
Micaceous clay minerals and alkali feldspars are the main sources of potassium in soils. Potassium is a major nutrient for plant growth and sometimes
OVERVIEW OF VERTISOLS
331
considered limiting in certain Vertisols. The adsorption/release of potassium in soils is influenced by many factors: (i) the amount and type of clay; (ii) site and amount of charge deficit of clay minerals; (iii) variation in moisture and temperature conditions; (iv) topographic position; (v) the presence of other cations; and (vi) electrolyte concentration (see Coulombe et al., 1996). The most significant factor is the layer charge when it is greater than 0.45 electron per half unit cell in the mineral structure, e.g., smectites, vermiculite, micas, or interstratified components that are subject to potassium retention (Tessier and Pedro, 1987). Potassium retention is even more effective if the layer charge is mainly located in the tetrahedral sheet instead of the octahedral sheet, due to the closer proximity of the charge deficiency to the interlayer space where potassium resides. Potassium was also reported to have a dispersive effect on kaolinitic, halloysitic, and smectitic minerals (Ahmed, 1969; Tessier, 1984; Delvaux et al., 1992). This aspect is relevant particularly for crops that require heavy potash fertilization. Potassium may become anhydrous under relatively dry conditions but can be very hydratable under wet conditions. The dispersing effect was observed under low electrolyte concentration with clay minerals that had a low layer charge. In the case of smectite, the layer charge was less than 0.45 per half unit cell. More research is needed to elucidate the potassium dispersive behavior on clay minerals.
2. Ammonium Ammonium is a form of nitrogen in soils. Ammonium retention has been reported in Vertisols by several authors (Rodriguez, 1954; Ahmad et al., 1972; Said, 1973; Feigin and Yaalon, 1974; Chen et al., 1989). Similar to potassium, ammonium is subject to strong retention by clay minerals having a high layer charge. Potassium and ammonium have similar behavior because they are similar in charge, size, and hydration energy. However, they differ by their polarizability; polarization of ammonium is higher than that of potassium (Sparks, 1980). Therefore, ammonium is generally preferred over potassium on the exchange sites. Ammonium retention is not the only way to reduce ammonium availability to plants. Other pathways include: (i) losses by volatilization of the ammonia species, (ii) leaching of nitrate after bacterial transformation,and (iii) losses by denitrification. Also, ammonium commonly is lost in calcium rich soils because it is not able to compete for calcium dominated exchange sites. The ability for plants to extract ammonium also is a factor to consider. Certain plants such as rice are more effective than others for the extraction of ammonium in soils.
3. Nitrate The nitrate anion may originate from nitrification of ammonium or from fertilizer sources (organic and mineral). The assimilable form of nitrogen is nitrate for
332
C . E. COULOMBE, L. P. WILDING, AND J. B. DMON
most plants. Nitrate is a mobile ion in soils and, consequently, an environmental hazard for surface water and groundwater pollution. In Vertisols, nitrate can transfer in solution phase downward from the plow layer along macropores of structural interfaces and bypass solution in the meso-micropores of the soil matrix. Kissel et al. (1974) reported that larger interconnected pores enhanced movement of nitrate and may result in a greater nitrification rate than in the rest of the soil mass. Denitrification due to flooded conditions is another possible pathway for nitrate loss and inefficient use in Vertisols. The type of land use systems may also contribute to decreased nitrate in the soil profile. For instance, Standley et al. (1990) reported that sorghum (Sorghum bicolor) and Rhodes grass (Chloris gayana), a perennial, progressively depleted nitrate in the subsoil of a Vertisol over a 7-year period that had accumulated in a previous cropping system over 17 years down to a depth of 1.8 m. The perennial grass was more effective than sorghum, which is possibly attributed to the fact that the roots of the perennial were active for the most of the year compared to sorghum roots that would be physiologically active only during the growing period. Regardless the fertilizer type, the factors that deserve major consideration in nitrate accumulation/depletion in Vertisols are: (i) the type of land use systems, i.e., fallow, continuous, or rotation; (ii) the type of crop, i.e., perennial, annual, legumes; (iii) the crop nitrogen demand; (iv) the root distribution patterns; and (v) extent of bypass flow through macropores.
4. Phosphate Phosphorus is a major element that is essential for plant growth but is limiting in many soils because it is present at very low concentrations in the Earth’s crust and commonly is not readily available due to its form and sorption phenomena. The term sorption includes adsorption, precipitation, and diffusion processes. Most soluble phosphorus studies use sorption isotherms with Langmuir and Freundlich equations. These equations describe adsorption and precipitation but give little or no information on the kind of chemical mechanism involved (Veith and Sposito, 1977; Sposito, 1982). Phosphorus content was reported low in Vertisols. For instance, the range for total phosphorus varies between 0.012 and 0.11% (116-1051 pg P g-1) in Vertisols of Texas (Kunze et al., 1963; Hawkins and Kunze, 1965; Raven, 1992; Coulombe, 1996), India (Simonson, 1954), Northern Cameroon (Yerima et ul., 1988) and Spain (Solis and Torrent, 1989b). Ahmad and Jones (1967) reported a higher range in values, for example from 0.014 to 0.048% (138-4750 pg P g-1) in Vertisols of West Indies. Total phosphorus content is highly dependent on the type and origin of the parent material and the amount of organic matter (see Coulombe et al., 1996). Phosphorus in soils is generally taken up by plants as phosphate anions
OVERVIEW OF VERTISOLS
333
(P04-3). Factors that influence the availability of phosphate in soils include: pH,
iron oxide content, aluminum oxide content, carbonates, organic matter, clay content, clay mineralogy, flooding, shrink-swell phenomena, and phosphorus concentration, to name a few. Sorption of available phosphorus in Vertisols was found to be high (Burnham and Lopez-Hernandez, 1982; Nychas and Kosmas, 1984; Yerima et al., 1988; Solis and Torrent, 1989a; Raven, 1992). Raven (1992) also reported that a Vertisol (Houston Black) had a very high buffering capacity compared to other Texas soils. This high buffering capacity is definitely a limiting factor for available phosphorus in Vertisols. High clay contents and pore tortuosity decrease phosphorus diffusion rates in Vertisols (Yerima et af., 1988). 5. Sulfate
Sulfate is an essential nutrient for plants but is required in much lower amounts than nitrogen, phosphorus, and potassium. Sources of sulfate anions are from (i) oxidation of sulfides in acid soils, (ii) dissolution of sulfate minerals, and (iii) acid rain. Sulfate behavior is not fully understood. For instance, Patil et al. (1989) observed no evidence of sulfate retention in Vertisols while Oxisols and Alfisols did demonstrate this phenomena. The content of hydrous sesquioxides and organic matter was considered responsible for this difference. Therefore, it is expected that Vertisols that have a relatively high organic matter content and/or derived from iron-rich parent materials would limit movement of sulfate ions in the soil profile.
IX. BIOLOGICAL PROPERTIES With the voluminous literature published about Vertisols, the topic of biological properties is certainly the one that has received the least emphasis. However, biological properties are considered not only essential but among the most important components of these soil systems. Management of biological properties is essential for the sustainability of Vertisols due to their impact on fertility, soil structure, and porosity.
A. ORGANIC CONSTITUENTS: DISTRIBUTION AND BIOLOGY Organic matter is an important constituent of Vertisols and affects morphological, chemical, and physical properties and management. Organic matter content reported in Vertisols ranges from 5 to 100 g kg-I (3 to 60 gC kg-l) but is commonly from 5 to 50 g kg-1. Generally, the distribution of organic matter
3 34
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
content is (i) higher in humid environments than in arid and semiarid environments; (ii) higher in virgin soil conditions than in cultivated areas (Kunze and Templin, 1956; Skjemstad and Dalal, 1987; Coulombe, 1996); (iii) higher in microlow topographic elements than in microhigh gilgai positions (Kunze and Templin, 1956; Wilding and Tessier, 1988); and (iv) decreases gradually with depth unless the soil has been overturned (Wilding and Tessier, 1988; Southard and Graham, 1992; Coulombe, 1996). A high organic matter content is often inferred from a black color in soils; however, this is not necessarily valid for Vertisols. It is well established that dark Vertisols may contain less than 10 g kg-1 organic matter. Dark colors generally extend throughout the whole profile and hinder differentiation of horizons. Several agents may be responsible for a dark color in Vertisols such as mineralogy and drainage conditions (see Coulombe ef al., 1996), but it is generally attributed to strong clay-organic complexes of smectite and humic materials (Singh, 1954, 1956; Skjemstad ef al., 1986; Skjemstad and Dalal, 1987). Clay-organic complexes are fundamental to soil structure and are discussed more specifically in the next subsection. The humus type associated with Vertisol topsoils is likely a mull with a carbon:nitrogen ratio ranging from about 10 to 25 under different land use systems (Coulombe, 1996). The turnover rate of the organic matter is generally rapid in surface horizons. However, the mean residence time of organic matter increases with depth from about 1000 to 4000 and up to 12,000 years B.P. (Yaalon and Kalmar, 1978; Stephan et al., 1983). This suggests that organic matter is more protected as depth increases. Yaalon and Kalmar (1978) mentioned that the mean residence time of organic matter with depth in a Vertisol exhibits a linearly increasing relationship as do other soils that do not crack. This result also suggests that a complete homogenization of the soil profile is not a dominant process accounting for pedogenesis in Vertisols. However, it was noted that the Vertisol curve had a steeper slope showing lower mean residence time values for the same depth compared to the other soil types (Yaalon and Kalmar, 1978). The formation of resistant smectite-humus complexes rather than the chemical recalcitrance of humic materials to microbiological degradation is reported to be responsible for the protection of the complex (Skjemstad et al., 1986; Skjemstad and Dalal, 1987). However, shrink-swell behavior may physically disrupt the strength of the clay-organic complexes and could explain the lower mean residence times in subsoils of Vertisols compared to other soils. The pedobiology of Vertisols is active and varied; earthworms, fungus, mycelium, bacteria, protozoa, thecamoebae, diatoms, enchytreides, and collemboles are examples (Stephan et al., 1983; Coulombe, 1996). However, one of the most drastic impacts on biological activity and diversity is the type of land utilization (Coulombe, 1996). Coulombe (1996) reported that diversity of biological activity is maximal under virgin conditions. For instance, earthworm activity is intense and evident under native conditions but gradually disappears, suc-
OVERVIEW OF VERTISOLS
335
ceeded by mites under cultivated conditions (Coulombe, 1996). However, after 20 years of fallow, earthworm activity returns but is not as optimal as under virgin conditions. Changes in biological conditions could be used in the assessment of soil structure with the identification and occurrence of the casts left by various organisms. With regard to the microbial biomass, Jocteur Monrozier et al. (1991) reported that it was higher in a Vertisol than in an Alfisol in Australia. It was estimated to be 5% of total organic carbon. Also, Chotte et al. (1994) reported that most of the biomass is associated with aggregates >200 pm for a Vertisol of Martinique (Lesser Antilles). However, only 14% of the total biomass was associated with the clay fraction. Supplementary research is needed regarding biology and loci of organic matter in Vertisols derived from different parent materials and under different environmental conditions and land use systems.
B. CLAY-ORGANIC COMPLEXES Clay-organic complexes are fundamental to the formation and stability of aggregates and soil structure from molecular to macroscopic structural levels. Humic materials, roots, fungal hyphae, polysaccharides, and lipids are common organic constituents that react with clays in order to form complexes and/or physically stabilize these compound units. Humic materials, e.g., humic acid, fulvic acid, and humin, constitute 60-80% of the organic matter in soils. The humic material is the result of random biochemical processes. Information concerning the chemical composition of humic materials from Vertisols is relatively limited. Holder and Griffith (1983) analyzed humic materials from acid and neutral-alkaline Vertisols. They reported that yields of fulvic acid were higher than yields of humic acids in all Vertisols; fulvic acid is the more highly reactive humic fraction. In general, humic materials, associated with polyvalent metal cations, are considered dominant and persistent binding agents in natural soil aggregates smaller than 250 pm (Tisdall and Oades, 1982; Oades, 1984). Their stability is a function of intrinsic properties of the soil rather than management practices (Tisdall and Oades, 1982). Numerous authors have reported an interlayer association where humic materials are adsorbed in the interlayer space of smectites. However, most of this literature refers to reference smectites and smectitic soils other than Vertisols. The possible formation of an interlayer clay-organic complex with smectite is still controversial. It is undisputed that smectite has a high affinity for humic substances resulting in the formation of microaggregates of generally less than 250 km. Jocteur Monrozier et al. (1991) reported an abundance of microaggregates, of 2 to 20 pm effective diameter, in surface horizons of an Australian Vertisol. These aggregates contained 60% of the total carbon, nitrogen, and biomass carbon of
336
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
the total soil. Clay, humic materials, polysaccharides, and calcium carbonates were the major components of these aggregates. Smectites can also promote the formation of large aggregates (250 to 1000 pm) even under conditions of low organic matter content (Krisna-Murti et al., 1977). Stability of these macroaggregates is dependent on climatic conditions, in particular wet-dry and/or freeze-thaw cycles. In general, formation of macroaggregates is the result of temporary and transient binding agents such as root exudates, fungal hyphae, and polysaccharides. These components constitute a small proportion (<20%) of the total organic matter but are the important agents involved in the formation and stability of macroaggregates. The influence of a fungal polysaccharide (scleroglucan) was reported for calcium-saturated kaolinite and montmorillonite (Chenu et al., 1987; Chenu, 1989; Chenu and Jaunet, 1990; Chenu and GuCrif, 1991). For both reference clay minerals, the loci of scleroglucan was at the interparticle levels and increased water aggregate stability and mechanical strength. Lipids, which constitute about 0.5 to 4% of the total organic matter, also may form complexes with clays. These substances are generally hydrophobic. Kaolinite was reported to adsorb more lipids (oleic acid) resulting in increased water retention at 0.1 MPa soil water potential compared to montmorillonite (Jambu et al., 1987). In the case of montmorillonite, the same authors mentioned that oleic acid decreased water retention and promoted dehydration of the system. The maintenance of a high organic matter content and stable aggregation is influenced by management practices and land use systems. A loss in organic matter content is detrimental to soil structure and results in degradation of other physical attributes. For instance, Skjemstad et al. (1986) reported a loss of 50, 6 1, and 66% organic carbon after respectively 20,35, and 45 years of continuous cultivation in Australian Vertisols. Similar results were observed with Vertisols cultivated for a long period compared to virgin soils (Skjemstad and Dalal, 1987; Coulombe, 1996). The amount and type of clay minerals apparently buffer these soils against degradation and increases resilience. However, inappropriate intensive management practices that promote biological oxidation, particularly continuous cultivation, progressively lead to degradation of clay-organic complexes and soil structure regardless of the parent material and climatic conditions.
X. PHYSICAL PROPERTIES Physical properties of Vertisols are major constraints to their optimal utilization. Several of the physical properties vary with moisture content and associated shrink-swell phenomena. In reality, shrinking and swelling are processes that occur in all soils but Vertisols and vertic intergrades show a greater expression of these phenomena.
337
OVERVIEW OF VERTISOLS
Shrink-swell phenomena in Vertisols require reappraisal of theoretical phase relationship. For instance, under a rheologically stable mineral condition, a moist soil would contain about half the volume as solid phase and the other half is shared in equal proportions by liquid and gaseous phases (Fig. 14). If the solids were inert and static, the bulk volume would not change when hydration and dehydration occur. The liquid and gaseous phases would interchange without significantly affecting the volume and other rheological soil properties. However, observed bulk volume changes occur upon hydration and dehydration due to changes in porosity and water content (Fig. 14). When the sample is rewetted, the volume occupied by air is replaced by water, but the soil will increase also in pore volume and, consequently, its water content (Tessier, 1988). As a result, shrinkswell phenomena in Vertisols impact soil physical properties. These dynamic attributes may be used to interpret soil structural conditions as discussed below.
A. BULKDENSITY AND COEFFICIENT OF LINEAR EXTENSIBILITY (COLE)
-
Bulk density is defined as the mass of soil per unit volume. This property is commonly used as an index of soil physical condition. However, bulk density of
PFHYDRATION
HYDRATION
L Observed Figure 14 Theoretical and observed aspects of dehydration/hydrationin shrink-swell soils.
338
C . E. COULOMBE, L. P. WILDING, AND J. B. DIXON
clayey soils varies with moisture content and consequently shrink-swell phenomena. Bulk density values reported for Vertisols from different regions are presented in Table VI. In addition to water content, factors which influence bulk density, particularly in surface horizons, can be attributed to: (i) particle density of mineral fraction; (ii) organic matter content; (iii) land use systems; and (iv) the method used to determine bulk density. Interestingly, Hirekurubar et al. (1991) reported little difference in bulk density between Vertisols derived from different parent materials. Variation observed was attributed to differences in clay content between sites. These results could also be explained by the fact that the Vertisols studied were under similar land use systems. The type of soil utilization is certainly a major factor that may
Table VI Ranges in Selected Physical Properties of Some Vertisols Properties
References
Bulk Density (Mg m-3) 0.33 MPa: 0.9-1.2 1.5 MPa: 1.6-2.0 COLE (cm cm-I): 0.07-0.20
Brown (1977), Yule and Ritchie (1980a), Coulombe (1996) Wilding and Coulombe (1996), Coulombe (1996) Ahmed er al. (1969), ElSwaify er al. (1970), Yule and Ritchie (1980a). Hirekurubar et al. (1991) El-Swaify er al. (1970), Yule and Ritchie (1980a). Ahmad (1982), Hirekurubar et al. (1991)
Atterberg constants (% weight) Liquid limit: 48-84 Plastic limit: 17-43 Plasticity index: 24-41 Moisture retention (% weight) 0.33 MPa: 35-75 1.5 MPa: 15-35 Saturated hydraulic conductivity (pm sec- 1 ) Column method (Bouma 1982): 1.44-49.1 1 Ponded infiltrometer (Prieksat et al., 1992): 6.05-30.60 Disc permeameter (Perroux and White, 1988): 21.96-193.22
Tension (mm water) 20 m m 30 m m 60 m m
Coulombe (1996) Kenneth N. Potter (personal communication) Hangshen Lin (personal communication)
Unsaturated hydraulic conductivity (pm sec-I) Tension Infiltrometer of Disc Permeameter of Ankeny er al. (1992) Perroux and White (1988) (K.N. Potter, (H.Lin, personal personal communication) communication) 2.54-7.40 1.50-4.30 0.80-2.W
7.61-18.93 3.76-7.73 0.95-2.66
OVERVIEW OF VERTISOLS
339
explain variation in bulk density from one site to another. Densification can be attributed to compaction or consolidation. The latter occurs naturally due to wetdry or freeze-thaw cycles. Sometimes bulk density may not be a very sensitive index to structural change when compared to other methods such as microfabric, shrinkage curve, oxygen diffusion, and infiltration measurements (McGarry, 1987; Hodgson and MacLeod, 1989b; Puentes, 1990; Coulombe, 1996). For bulk density, the measurement method, e.g., core vs. Saran-coated clods, the size and number of samples collected, and the moisture content at time of sampling are critical factors influencing bulk density results between management treatments. Coulombe ( I 996) reported differences in bulk density of surface horizons of Texas Vertisols between five different land use systems (Fig. 15): (i) virgin, (ii) restored meadow for 20 years, (iii) improved pasture, (iv) cereal (oat), and (v) row crop (soybean). Bulk density at field capacity (0.33 MPa) ranges from 0.86 to 1.18 mg m-3 for all treatments (Fig. 15a). A low bulk density at field capacity suggests high water holding capacity due to changes in microstructure geometry and dominance of micropores. Oven dry bulk density for these same land use systems ranged from 1.33 to 1.74 mg m-3 (Fig. 15b). Interestingly, the restored meadow exhibited high bulk density values at all depths. This site was conventionally cultivated for more than 50 years previously. The effect of cultivation on increased bulk density is also shown for the two cultivated sites, e.g., cereal and row crop, at 15-30 cm. As suggested by the restored treatment, intensive cultivation practices cause irreversible changes to structure and porosity as well as other properties, such as water holding capacity and shrinkswell behavior. Coefficient of linear extensibility (COLE) was introduced by Grossman et al. (1968) as an index of soil shrinkage. COLE is estimated from relative differences between the length of a given soil column when moist and after drying, or from bulk density values of natural clods at field capacity and under oven dry conditions (Grossman et al., 1968). COLE values can also be measured by the change in length of a ribbon of soil under different soil moisture contents (Schafer and Singer, 1976). The use of COLE was proposed by DeMent and Bartelli (1969) as a diagnostic property to identify vertic intergrades in Soil Taxonomy (Soil Survey Staff, 1975, 1990, 1992, 1994). COLE values reported for selected Vertisols are presented in Table VI. COLE values are correlated with total clay content, fine clay content, surface area, water retention at field capacity, and exchangeable sodium percentage (Anderson et al., 1973; Yerima, 1988; Yerima et al., 1989).
B. CONSISTENCE AND ATTERBERG CONSTANTS Consistence is a very critical property of Vertisol management. Vertisols have a very narrow range under favorable soil moisture conditions and consistence that
340
C . E. COULOMBE, L. P. WILDING, AND J. B. DIXON
figure 15 (a) Bulk density at field capacity (BDfc) of surface horizons of Texas Vertisols under different land use systems. (b) Oven dry bulk density (BDod) of surface horizons of Texas Vertisols under different land use systems (from Coulombe, 1996).
allow workability and trafficability. Consistence of Vertisols varies from plastic and sticky when wet, friable when moist, to hard and very hard when dry (Fig. 16). Cultivation of Vertisols even at relatively low moisture contents (e.g., 22%) can be detrimental to soil structure and porosity (McGarry, 1989; Hodgson and MacLeod, 1989a,b). Mechanical force used for cultivation may easily overcome the shear strength of the soil and cause compaction. On the other hand, developing countries commonly do not have access to mechanical power, so Vertisols are cultivated by hand or with animal traction. Thus, Vertisols of many areas are
341
OVERVIEW OF VERTISOLS Resldual Shrlnkage
Llnear
Structural
Shrlnkage
Shrlnkage
f
2 W
s
3
0 5
Hard
I
FrIable
I
Plastic
CONSISTENCE
I
Sticky
I
MOISTURE CONTENT (%)
Figure 16 Schematic shrinkage curve in relation to moisture content and consistence
underutilized because of high energy requirements. Consistence remains an empirical property but nonetheless, prediction of consistence as a function of water content as well as other properties such as shear and tensile strength would provide better information for timing of tillage practices, design of tillage equipment, optimal cultivation methods, and power requirements for Vertisols. The Atterberg constants, e.g., liquid limit, plastic limit, and plasticity index, also are empirical tests used by agriculturalists and engineers to characterize the mechanical behavior of a soil. The liquid limit is defined as the water content, at which the soil just begins to flow, expressed as percentage by weight of the oven dry soil (Allen, 1942; Brown, 1977). Values reported for liquid limit of Vertisols are presented in Table VI. Factors affecting liquid limit are the amount and type of clay; liquid limit increases with clay content. For a given clay content, the liquid limit decreases in the order: smectite > illite > kaolinite. El-Swaify et al. (1970) also reported liquid limits to be affected by saturating cations; sodium gave the lowest, potassium the highest, and calcium and magnesium gave intermediate or equal liquid limit values. The effect of these cations on the clay microstructure is the probable reason. Even though most Vertisols are calciumand magnesium-saturated, the impact of sodium originating from irrigation and drainage waters or potassium from potash-rich fertilizer sources might alter their aggregation and permeability. These monovalent cations have a dispersing effect on clays (Ahmed and Jones, 1969; Tessier, 1984; Delvaux et al., 1992).
3 42
C . E. COULOMBE, L. P. WILDING, AND J. B. DIXON
The plastic limit is defined as the water content at which the soil can be rolled into threads of approximately 3 mm in diameter without breaking apart, expressed as percentage weight of the oven dry soil (Allen, 1942; Brown, 1977). As for the liquid limit, the plastic limit increases with increasing clay content. Ranges in plastic limits reported for Vertisols are presented in Table VI. Variations are attributed to clay content and clay type. Saturating cations are reported to have little effect on plastic limit of Vertisols (Ahmed er af., 1969). The role of organic matter in Vertisols on plastic limit has not been comprehensively documented. Finally, the plasticity index is the difference between the liquid limit and the plastic limit (Allen, 1942; Brown, 1977). Plasticity indices reported for Vertisols are presented in Table VI. Properties that affect the liquid limit and the plastic limit also impact plasticity index.
C. SHRINKAGE CURVE AND MOISTURE RETENTION CHARACTERISTICS Shrinkage is a fundamental process of Vertisols and results from changes in water potential and moisture content. Three stages are distinguished in the shrinking phenomenon (Fig. 16). The first stage is called “structural shrinkage” and corresponds to a slight volume change when water is removed under low suction constraints (0-0.03 MPa) from large soil pores (Yule and Ritchie, 1980a). The second stage represents a volume reduction which is proportional to water loss in the system. This stage is called “linear shrinkage,” where for every unit of volume of water loss there is a unit change in shrinkage. This occurs at 0.03 to 1 MPa water potential, and at even lower values than 1.5 MPa in the case of smectitic soils (Yule and Ritchie, 1980a). Finally, under high suction constraints, the decrease in volume is less than the water loss and represents the region of “residual shrinkage.” Electrostatic repulsion, strongly bound water, and particle rigidity are responsible for preventing further collapse. According to Yule and Ritchie (1980a,b), the shrinkage is equidimensional, i.e., equal in the x, y, and z axes. Furthermore, the size of the cores had little effect on the shrinkage, i.e., small and large cores exhibited similar shrinkage behavior (Yule and Ritchie, 1980b). The shrinkage curve generally differs from the swelling curve due to hysteresis of the clay material (Chang and Warkentin, 1968; Tessier, 1984). The amount of water held at a specific water potential is generally lower upon rewetting cycles. The amount and type of clay associated with the saturating cation, electrolyte concentration, and the number, rate, and severity of dry-wet cycles are major factors involved in the hysteresis phenomenon (Tessier, 1984). In general, as the number and severity of wet-dry cycles increase, the size of the clay assemblages also increases, thus decreasing the porosity available for water storage.
OVERVIEW OF VERTISOLS
343
Indices derived from shrinkage curves were used to evaluate structural conditions of Vertisols (McGarry and Daniells, 1987; McGany and Malafant, 1987; McGany, 1988; McGany and Smith, 1988). Shrinkage curve indices are based on regression parameters related to bulk density, total porosity, and shrinkage potential. A low bulk density and high total porosity would cause the soil to have a greater shrinkage potential. Moisture retention characteristics are properties associated with shrink-swell phenomena. Water is retained by capillary forces for pores having an equivalent spherical diameter of less than 100 pm. Generally, Vertisols have a relatively high moisture retention at all water potentials. Several authors reported the capacity for Vertisols to adsorb considerable amount of water on wetting and retaining it under low suction and tension forces (e.g., El-Swaify ef al., 1970; Yule and Ritchie, 1980a; Ahmad, 1983; Hirekurubar et al., 1991). Ranges in moisture retention values are presented in Table VI for selected studies. The amount, type, and surface area of the clay fraction are major factors involved. Other factors are organic matter content, biological activity, shrink-swell potential, and land use systems (Coulombe, 1996). Freeze-thaw phenomena may also affect size and arrangement of structural aggregates and porosity of the system.
D. GASDIFFUSION Gas diffusion in Vertisols is relevant for gaseous exchange between the atmosphere and the pedosphere, especially for biological activity. Waterlogging and wet cultivation may affect the porosity and the aeration status in Vertisols. Hodgson and MacLeod (1989a,b) reported that the assessment of soil structural conditions by determination of oxygen flux density was more sensitive than the use of oxygen content, air-filled porosity, and bulk density parameters. The reason is that oxygen flux density relates to tortuosity and continuity of pores. The technique presented by these authors is simple, rapid, and adequate for routine sampling. Combination of this technique with other methods would provide better assessments of structural conditions.
E. HYDRAULIC CONDUCTMTY Hydraulic conductivity of Vertisols is an important parameter used to assess plant growth, soil aeration, soil water recharge, surface runoff, erosion, and evapotranspiration as well as the potential of surface water and groundwater pollution by organic and inorganic constituents. Hydraulic conductivity is separated into saturated hydraulic conductivity (I&) and unsaturated hydraulic conductivity (K,,,,,).Under saturated conditions, all pores are filled with water and hydraulic conductivity is maximum. When the soil desaturates, pores empty
344
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
progressively from macropores to mesopores to micropores, which results in decreasing hydraulic conductivity. Polymodal porosity is common in Vertisols and other structured soils (Bouma, 1991). Macroporosity is particularly important in water movement under saturated and unsaturated conditions. Previous studies reported the importance of macropores (biopores, cracks, slickensides) on movement of water and solutes in Vertisols (Ritchie et al., 1972; Kissel et al., 1973, 1974). Even though the role of macropores may be important in shrink-swell clay soils, meso-micropores in a given cross-section can conduct much more liquid when continuous than macropores that are large and discontinuous which do not contribute to flow (Bouma, 1982). Messing and Jarvis (1990) also mentioned that macropores may remain hydraulically passive in the field since they may be discontinuous or not open to the air-water interface. Dye tracing studies of two Texas Vertisols revealed that 5- 10% of the soil total porosity contributed to water flow at water tensions of less than 24 cm (Lin, 1995). Infiltration first proceeds into large macropores and between structural units before diffusing into the soil matrix (Lin, 1995). Water movement in Vertisols and other shrink-swell soils remains complex due to this macropore-micropore duality. Hydraulic conductivity ( K ) is a very dynamic and variable property. Factors that influence K values are numerous: texture, structure, soil moisture contents, biological activity, electrolyte concentration of the soil solution, soil temperature, depth, management practices, place and time of measurement, etc. The simultaneous control of all these factors is practically impossible and is not cost effective. Furthermore, repeated measurements are necessary in order to obtain confident results. For these reasons, determination of K in Vertisols is generally labor, time, and cost intensive. Values of K,,, were reported in Vertisols by Ritchie et al. (1972). These authors mentioned average values of 2.5 cm day-1 (0.289 pm SKI),0.3 cm day-1 (0.035 pm s-l), and 0.1 cm day-' (0.012 pm s - I ) for field basins (2.5 and 10 m2), undisturbed cores (21 and 73 cm in diameter), and disturbed cores (17 cm in diameter), respectively. The value for disturbed cores can be considered representative of hydraulic conductivities mostly in the absence of macropores, i.e., matrix flow. The K,,, value for field basins suggests the impact of major macropores on water flow in Vertisols. Hydraulic conductivity can be variable, especially in the surface horizons, due to the influence of biotic factors and management practices. For instance, Puentes (1990), using a double-ring infiltrometer, reported a range of 6.05 to 8.12 cm day-1 (0.70 to 0.94 pm s-1) for areas in pasture for 15 to 25 years and 1.21 cm day-1 (0.14 pm s-1) for sites under continuous cropping for 15 and 25 years. Recent work involving surface horizons of Texas Vertisols under various land use systems revealed that the methodology used is a major factor in the determination of hydraulic conductivity. Different values of K,,, using different methodologies are presented in
OVERVIEW OF VERTISOLS
345
Table VI. Variability was almost as high within as among treatments, which is relatively common in the case of hydraulic conductivity. With six replications for each treatment, Coulombe (1996) reported a coefficient of variation up to 95% with the column method (Bouma, 1982). The size of the cross-sectional area, the confinement of hydraulic conductivity measurement, and the type of land use systems are primarily responsible for the differences in saturated hydraulic conductivity values between treatments. Select values of unsaturated hydraulic conductivity (K,,,,,) also are reported in Table VI. Regardless of the technique used, K,,,,, decreased drastically at a tension 230 mm of water, suggesting a bimodal pore-size distribution. Differences between the values for the same land use are attributed to the size of the cross-sectional area, the methodology, and the spatial variability of the sites.
F. SOILSTRUCTURE AND POROSITY Structure and porosity of Vertisols were previously discussed as morphological markers. This subsection emphasizes methods of quantification of soil structure and porosity. Structure of surface horizons especially at the pedosphere-atmosphere interface are critical to biological, chemical, and physical processes. This was originally recognized in the 7th Approximation (Soil Survey Staff, 1960) as grumic and mazic great groups of the Vertisol order. Grumic (granular) and mazic (massive) were used to differentiate between Vertisols with the ability to self-mulch and those without that attribute. The grumic and mazic great groups were eliminated in the subsequent versions of Soil Taxonomy (Soil Survey Staff, 1975, 1990, 1992, 1994) probably due to problems of quantification of this phenomenon, the diversity of soil structure associated with climatic conditions, and soil utilization effects (Wilding and Williams, I99 1 ; Coulombe, 1996). As previously discussed in Section IV.C, the definitions proposed for selfmulching given by Soil Survey Staff (1960) and Grant and Blackmore (1991) emphasized physical processes, e.g., wet-dry and freeze-thaw cycles, without the influence of biological activity or land utilization on soil structure. The selfmulching index is similar to a slaking test for evaluating structural stability. It considers only intrinsic properties of soils. Another disadvantage of the selfmulching index is that it discriminates between only two structural conditions (granular vs. massive). Granular and massive structural conditions are two endmembers and other intermediate types of structure exist for Vertisols due to the influence of biological activity and land utilization. An integral component of soil structure is porosity. Evaluation of soil structure and porosity is often inferred from indirect properties and methodologies, e.g., bulk density, shrinkage curves, moisture retention characteristics, gas diffusion,
3 46
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
hydraulic conductivity, and mercury porosimetry. Several of these properties have been discussed previously. However, these measurements are only indirect qualitative and quantitative inferences of structural and porosity conditions. Direct qualitative and quantitative evaluation of soil structure and porosity can be done using meso-micromorphological and image analysis techniques. Assessment of structure and porosity with these techniques have been done successfully in the study of Vertisols, e.g., Bui and Mermut (1989), Puentes (1990), Puentes and Wilding (1990), Koppi and McBratney (1991), Koppi et al. (1992), Wild et al. (1992), and Coulombe (1996). Puentes et al. (1992) discussed concepts of sampling and variability of some parameters in porosity measurements of Vertisols. Variability of measured parameters are site and method specific. The number of samples necessary to obtain a confidence level is a function of the method used and the parameter of interest (Puentes et al., 1992). Mesomicromorphology and image analysis techniques still remain underutilized. Some of the reasons include the time required for processing and analyzing samples and the choice and complexity of terminology used in micromorphology. Nonetheless, these techniques, if used systematically and coupled with indirect methods such as those discussed previously, would provide a better understanding of the differences in soil structure and porosity. Figures 17 to 21 present striking examples of how structure and porosity differ between land uses and depths. From 0 to 5 cm of the surface horizons, Vertisols under native conditions exhibit extensive biological activity as indicated by the presence of numerous earthworm casts and compound packing voids (Fig. 17a). Tillage on Vertisols generates a complex structure made of fine and large aggregates (Fig. 17b). The fine aggregates, in particular, are more subject to slaking and erosion processes. Tillage practices disturb the steady state of the soil by exposing organic matter to oxidation which results in disintegration of the soil structure. A fine desiccation layer formed by deposition of the fine eroded particles generally occurs at the atmosphere/pedosphere interface (Fig. 18a). Upon continuous and intensive cultivation, a massive structure may form which results in the absence of pedality, and loss in density, continuity, and connectivity of macropores (Fig. 18b). At a depth of 20 to 30 cm, the pedality and porosity of Vertisols under native conditions are optimum (Fig. 19a). The effect of cultivation is again shown by the lack of pedality and porosity attributes (Fig. 19b). However, restoration of pedality, porosity, and biological activity can occur if Vertisols are subjected to improved pasture and restored meadow conditions (Figs. 20a and 20b). Restoration of pedality, and density, continuity, and connectivity of macropores is the first stage in regenerating the physical attributes of Vertisols. Similar phenomenon occur under controlled traffic, no-tillage management systems. The impact of chisel tillage, to prepare the soil, disturbs the pedality and breaks the density, continuity, and connectivity of the micropores which is not restored even after
OVERVIEW OF VERTISOLS
347
Fcgure 17 Structure and porosity of surface horizons (0-5 cm) of Vertisols under different land use systems (from Coulombe, 1996). (a) Granular (vermic) structure produced by intensive earthworm activity under native conditions of a Vertisol of Texas. The rounded ped units are earthworm casts and resistant to slaking under wet-dry cycles. (b) Complex structure made of small and large aggregates produced by tillage under continuous cultivation of a Vertisol from Mexico. The scale bar represents 1 cm.
Figure 18 Structure and porosity of surface horizons (0-5 cm) of Vertisols under different land use systems. (a) Desiccation crust that overlays a self-mulching structure under a row crop system on a Vertisol of Texas. The dessication crust is believed to be due to accumulation and deposition of eroded particles (from Coulombe, 1996). (b) Massive structure produced by intensive and continuous cultivation of dry season sorghum (muskwari) in Northern Cameroon. Note the lack of individual ped units and the decrease in density, continuity, and connectivity of macropores compared to the previous plates. (Courtesy of Dr. Mathieu Lamotte, ORSTOM, Niger.) The scale bar represents 1 cm.
OVERVIEW OF VERTISOLS
3 49
Rgure 19 Structure and porosity of surface horizons (20-30 cm) of Vertisols under different land use systems (From Coulombe, 1996). (a) Subangular blocky structure produced under native conditions on a Vertisol of Texas. (b) Coarse subangular blocky to massive structure produced under cultivated conditions of a cereal (oats) crop on a Vertisol of Texas. Note the contrast in biological activity, pedality, and porosity between the two conditions. The scale bar represents 1 cm.
Figure 20 Structure and porosity of surface horizons (20-30 cm) of Vertisols under different land use systems (From Coulombe, 1996). (a) Subangular blocky structure produced under improved pasture conditions on a Vertisol of Texas. (b) Subangular blocky structure of a meadow, after 20 years of restored conditions, that had been conventionally cultivated for more than 50 years on a Vertisol of Texas. Note in both plates the density, continuity, and connectivity of macropores with chambers containing infilled (casts) left by biological activity in the chambers. The restoration of these attributes is the first step involved in the regeneration process of Vertisols. The scale bar represents 1 cm. 350
Figure 21 Structure and porosity of surface horizons (20-30 cm) of Vertisols under different land use systems (From Coulombe, 1996). (a) Coarse subangular blocky to massive structure produced after 1 year of controlled-traffic, no-tillage management on a Vertisol of Texas. (b) Subangular blocky structure produced after 10 years of controlled traffic no-tillage management on a Vertisol of Texas. Note again the recovery of biological activity and regeneration of the pedality, and the density, continuity, and connectivity of macropores compared to the above plate. The scale bar represents 1 cm.
351
3 52
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
one cropping season (Fig. 21a). However, after 10 years under controlled traffic, no-tillage management, pedality and porosity attributes are restored despite an increase in bulk density due to consolidation (Fig. 21b). Again, these examples illustrate that bulk density may not be a sensitive index to structural change. Geographic information systems (GIS) are also potential tools to study structure and porosity of Vertisols at different scales of resolution. To our knowledge, no work has been reported on this aspect. However, the use of GIS coupled with global positioning systems (GPS) could be used to monitor the changes in structure and cracking patterns of Vertisols on a spatial, seasonal, and temporal basis. GIS and GPS systems are currently being developed for precision farming to optimize nutrient management based on spatial variability of soil properties as well as to reduce the impacts by agricultural activities on non-point source pollution. A new science that is gaining interest globally is the “Chaos Theory,” also referred to as the science of disorder. Several authors described texture, structure, and porosity of clayey materials using fractal geometry (Rieu and Sposito, 1991a,b; Young and Crawford, 1991; Van Damme and Ben Ohoud 1990; Ben Ohoud and Van Damme, 1990; Perrier, 1994; Van Damme, 1995). Fractals are geometrical shapes that are completely irregular; however, the same degree of irregularity is expressed on all scales of resolution (Mandelbrot, 1993). Fractal geometry brings another way of looking at the hierarchy of the organization of structure and porosity of soils. Current efforts are underway to characterize different types of aggregates and porosities in surface horizons of Vertisols under different land use systems. Correlation of fractal geometry coupled with micromorphology, image analyses, and other methods would contribute to the advancement of knowledge in Vertisol systems.
XI. MANAGEMENT OF VERTISOLS Vertisol management is a subject that has stimulated considerable interest globally, e.g., Dudal (1963, Ahmad (1983), McGarity et al. (1984), Probert et al. (1987), IBSRAM (1987, 1989), Jutzi et al. (1988), Wilding and Puentes (1988), Hirekerur et al. (1989), ICRISAT (1989), Jones and Gerik (1990), Ahmad and Mermut (1996). Vertisols have several positive attributes and are considered good productive resources. However, as mentioned previously, physical properties are constraints to optimum management of these resources. At the present time, there is no simple or single solution for the proper management of Vertisols. Degradation processes in Vertisols occur worldwide regardless of the parent material, climatic conditions, and level of cultural input. Strategies for the maintenance or regeneration of soil properties have been proposed and their success varies within and among regions. Vertisols require speci-
OVERVIEW OF VERTISOLS
353
ficity in terms of management practices and land utilization because of (i) their highly variable morphological, chemical, and physical properties within and among regions, (ii) the variability in climatic conditions from one region to another which impacts their behavior, and (iii) the difficulty in technology transfer from one region to another. The following discussion contrasts two levels of cultural input for Vertisol management and as it relates to soil degradation and technology used for the maintenance and regeneration of soil properties. The discussion is partly based on experiences by the authors in Cameroon, Texas, and Mexico.
A. MANAGEMENT OF VERTISOLS WITH LOW-INPUT CULTURAL SYSTEMS: CASEOF VERTISOLS OF WESTAFRICAWITH SPECIAL REFERENCETO NORTHERN CAMEROON Vertisols are significant land resources in the Sudano-Sahelian zone of West Africa, e.g., 0.1 million ha in Niger, 4 million ha in Nigeria, 1.2 million ha in Cameroon, and 16.5 million ha in Chad (Dudal and Eswaran, 1988). These soils are in complementary association with Alfisols, Inceptisols, and Entisols. While Vertisols are flooded during the rainy season which lasts from May through September, the other soil resources are utilized for production of food and fiber crops such as sorghum, maize, millet, cotton, peanut, cowpea, and pigeon pea. During the dry season, production of transplanted sorghum (vernacular name “muskwari”) occurs on Vertisols and vertic intergrades of West Africa. Transplanted sorghum utilizes residual soil moisture during the dry season. Thus, recharge of the soil with water during the rainy season is essential to production of transplanted sorghum. Cultivation of dry-season sorghum has been done successfully on Vertisols of West Africa despite harsh climatic conditions. Climate is characterized by a marked wet and dry season with annual precipitation that varies from 300 to 1200 mm within the region. This makes the management of soil and water resources difficult compared to temperate and/or humid regions where precipitation is more uniformly distributed throughout the year. Nonetheless, cultivation of transplanted sorghum allows people and animals to survive during the dry season. Due to its importance and success in West Africa, knowledge about cultivation of transplanted sorghum can be beneficial to regions with Vertisols under similar climatic conditions. The following discussion refers to Vertisols management of Northern Cameroon. Vertisols of Cameroon are mostly distributed in the northern part of the country (Fig. 22). Toward the last month of the rainy season, i.e., late August-early September, sorghum seeds are planted at a high density in a nursery on Alfisols or Inceptisols. During that period, most Vertisols are flooded because of their occurrence in flat or slightly depressional areas. Soil preparation begins in late September to early October or as soon as
3 54
C . E. COULOMBE, L. P. WILDING, AND J. B. DLXON
Figure 22 Distribution of Vertisols and vertic intergrades in Northern Cameroon.
Vertisols begin to dry. The grass is cut with a machete and commonly burned on site. It may also be removed for animal feed or left as a mulch. Burning of the grass has several advantages over the two other options: (i) it reduces and eliminates potential pathogens to the sorghum; (ii) it reduces competition be-
OVERVIEW OF VERTISOLS
355
tween sorghum and weeds for soil nutrients and water; and (iii) it provides a source of soil nutrients for sorghum. After the grass is cut and removed, sorghum plants from the nursery are carefully removed. The stage at which the sorghum is removed for transplanting is very crucial, e.g., at the development of the 4th leaf and/or before the first grain is formed. The leaves at the top of the plant are cut to limit moisture loss and to aid in subsequent root development. After extraction, plants may be kept in water and preserved for about 1 week. When it is time for transplanting, generally about October, farmers use a tree limb of approximately 15 cm in diameter and 180 cm in length cut from a hard wood, such as Belanites aegyptiaca. The limb is sharpened to a tip at one end and covered with metal (Fig. 23a). When constructed, the limb is used as a dibble to make a planting hole in the soil for transplanting the sorghum. The depth of the hole may range from 15 to 30 cm. Upon sorghum transplanting, the immediate area is hand watered. This procedure is common in Northern Cameroon. In Niger, farmers simply lift up a soil clod and transplant the sorghum sandwiched diagonally between the base of the clod and the subjacent surface (Moussa Gaoh GoubC, personal communication). Sorghum plants are transplanted about 1 m apart (Fig. 23b). The planting density may vary depending on the ability of the soil to store water for the growing season. Dry-season sorghum is generally harvested from January to February. The grain is harvested for human consumption while the stover residue is fed to cattle and goats. The flour produced from muskwari (dry-season sorghum) grain is considered very palatable by natives and preferred for human consumption over other sorghum varieties. The dry-season sorghum does not receive fertilizer or pesticide amendments and is not in rotation with other crops. Yields of dry season sorghum are reported to be around 1.5 Mg ha-'. For the remaining dry season, these Vertisols areas are used as unimproved pasture grazing lands for cattle and goats. In contrast, Vertisols with irrigation systems along river valleys or water reservoirs in West Africa also are utilized for rice production during the rainy season. In the region of Yagoua, Northeastern Cameroon, systems have been developed to produce rice continuously throughout the year at an industrial scale. However, the declining economy of the country has not been able to sustain this industry. Nonetheless, from this discussion it is clear that Vertisol resources and water conservation systems are utilized by African natives to produce rice either seasonally or continuously throughout the year. The rice-growing season may vary from 90 to 150 days, depending on the variety, and yields may reach 5 to 6 , even up to 7 Mg ha-1 without fertilizer nutrient inputs (Haman Oumar, personal communication). Rice yields are in fact similar to those of Texas Gulf Coast region with high-input cropping systems. This illustrates the production potential of Vertisols as sustainable agricultural resources. Irrigated Vertisols also have
Figure 23 (a) Limb of a hardwood sharpened to a tip, covered with metal, and used as a dibble to make a planting hole in the soil for transplanting sorghum. (b) Sorghum plants transplanted 1 m apart in Vertisols of northern Cameroon.
OVERVIEW OF VERTISOLS
357
been used for production of fruits and vegetables. Soya was recently introduced into rotation with rice to help feed the increasing population and to maximize the utilization of Vertisol resources year-round (Aboubakan Yacoubou, personal communication). Vertisols of West Africa are resources at risk. Erosion and degradation of these soils are the two major problems coupled with secondary constraints such as decreases in organic matter, nutrient deficiencies, and salinity. Sealing and densification of the soil surface, loss of macroporosity, decreased infiltration rates, decreased water recharge, enhanced surface water runoff, and soil erosion are consequences of this degradation (Fig. 24). Furthermore, with the population increasing 3% annually, Vertisols are under high productivity pressure. Specifically in Northern Cameroon, Seiny-Boukar (1992) reported that 85,000 ha of dryland Vertisols are degraded. Decreases in vegetative cover and diversity and changes in vegetative species reflect stages of degradation (Donfack, 1992). The advanced stages of soil degradation of these resources result in abandonment of crop lands which are no longer productive, and subsequent land use pressures are migrating toward the southwest area of north Cameroon. Degradation of these resources has forced redistribution of cropping to marginal lands. Included in these marginal lands are Vertisols and Alfisols at an intermediate stage of degradation. Efforts are being made to maintain or regenerate soil properties, and conservation of water of these marginal Vertisols with the installation of earthen structures such as water reservoirs, dikes, microcatchments, and earth bunds (casiers) to promote enhanced water infiltration and soil profile recharge (Fig. 25). In the case of Vertisols under advanced degradation, regionally called harde, regeneration of soil properties is unlikely without major physical, chemical, and biological technology input. On the other hand, irrigated Vertisols such as those used for rice production can be sustained for longer productivity with lower risks of degradation. Little is known about the specific dynamics and processes of Vertisol degradation. The type of land utilization is no doubt a factor. Another factor, which has received relatively little attention, is the number of dry-wet cycles on degradatiodregeneration of soil physical properties.
B. MANAGEMENT OF VERTISOLS WITH HIGH-INPUT CULTURAL SYSTEMS WITH SPECIAL REFERENCETO TEXAS Vertisols of developed countries are used extensively for production of cotton, grain sorghum, maize, soybeans, safflower, and small grains (Harris, 1989). Production of rice occurs in regions where Vertisols are subject to aquic moisture conditions such as the Gulf Coast Region of Texas (De Datta et al., 1988). A small proportion of Vertisols is either under fallow after several years of intensive cultivation or continuously cropped, or remains uncropped as prairies, range-
3 58
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
Figure 24 (a) Typical example of a Vertisol at an advanced stage of degradation (hard@ in Northern Cameroon. (b) The advanced stage of degradation of Vertisols of this region is exhibited by a sealed, dense, and “hard-setting” soil surface.
OVERVIEW OF VERTISOLS
3 59
Figure 25 Examples of soil and water conservation/regeneration interventions for marginal Vertisols in Northern Cameroon include: (a) Water reservoir built to serve as a water storage facility in developing sorghum for transplanting (muskwari), and (b) earth bunds (casiers) built to promote enhanced water infiltration and soil profile recharge.
3 60
C. E. COULOMBE, L. P. WILDING, AND J. B. DMON
lands, or forests (Puentes et al., 1988). In fact, land use is similar to that of developing countries except for the level of cultural input invested. In developed countries, Vertisols generally receive high cultural input, e.g., mechanized farming with high-energy machinery, extensive use of fertilizer and pesticide amendments, development of water conservation practices (grassed waterways, terraces, contours, no-tillage residue management), and development of irrigatioddrainage systems (Harris, 1989). Vertisols with these inputs may produce maximal crop yields despite evidence of soil-fertility decline and physical/biological degradation. Declines in crop yields and soil fertility are commonly overcome by modifying the level of cultural input, e.g., increasing energy needed for tillage practices, increased utilization of optimal levels of fertilizers and pesticides, conservation tillage, and increased irrigation frequency and efficiency. As a result, crop yields may be sustained in spite of apparent degradation in morphological, biological, chemical, and physical soil properties. Vertisols that receive high cultural input exhibit symptoms of degradation after less than a century of cultivation. Prior to that period, most Vertisols in Texas were maintained as pastures and rangelands, small areas that might have been cultivated by hand or with animal traction (Sharpless and Yielderman, 1993). Intensification of crop production on these soils is closely linked with the invention and utilization of mechanical equipment for cultivation. Conventional or traditional cultivation practices for Vertisols, as well as other soil types, are a combination of plowing, disking, and harrowing. These cultivation practices are necessary for preparation of an adequate seedbed, trafficability, weed control, and enhancement of root system development. Vertisols require proper timing and high energy for tillage practices. This is due to their narrow soil moisture content range for optimal workability and trafficability, a consequence of peculiar intrinsic properties. This aspect is critical and often difficult to achieve by farmers with large management holdings (several hundred hectares) without large equipment to provide timely operations under climatic constraints. This constraint becomes more acute in moving from hand traction to animal traction to mechanical traction. As a result, cultivation is done sometimes when the soil is too wet (plastic state) which compacts the soil surface, increases bulk density, and negatively affects the macropore density and continuity (Drescher et al., 1988; Kooistra and Boersma, 1994; Kooistra and Tovey; 1994; Soane and Ouwerkerk, 1994). The impact of wet cultivation is restricted not only to the time of tillage for planting, seedling establishment, and weed control, but to the time of harvesting as well. Operations of heavy equipment during harvest under inappropriate wet soil conditions will have a negative impact on soil properties and crop yields for the successive cropping season (McGany, 1989; Hodgson and MacLeod, 1989a,b). The impact of conventional and continuous cultivation on degradation of Vertisols has been documented (Skjemstad et al., 1986; Skjemstad and Dalal,
OVERVIEW OF VERTISOLS
361
1987; Puentes, 1990; Puentes and Wilding, 1990; Oleschko et al., 1993; Coulombe, 1996). Conventional and continuous cultivation lead to a decrease in organic carbon content, excessive salt accumulation, increase in bulk density and soil strength, decrease in infiltration, and degradation of macropore space and pedality. This degradation also increases susceptibility to surface runoff and erosion (Potter and Chichester, 1993). Conservation tillage systems, such as no-tillage, were developed as an altemative to conventional tillage practices to help alleviate soil degradation. The objectives of conservation tillage are to: (i) control soil water and wind erosion; (ii) reduce compaction by tillage and vehicular traffic; (iii) increase the amount of crop residue left on the surface; and (iv) maintaining or augmenting crop yields. Dalal (1989) reported an increase in organic carbon, total nitrogen, and aggregation index and a decrease in pH, electrical conductivity, and exchangeable sodium percentage (ESP) in the no-tillage system compared to conventional tillage systems. Gerik and Morrison (1984) reported similar grain sorghum yields and soil water contents between no-tillage and conventional tillage systems. Thus, no-tillage was considered as an attractive management alternative for Vertisols. However, negative effects of no-tillage management on physical properties were reported. For instance, it has been shown that no-tillage increases bulk density and decreases total porosity compared to conventional tillage (McGany, 1988; McGany and Smith, 1988; Mella, 1991). Likewise in wetter Vertisols, the soils are delayed in spring warm-up with no-tillage conservation systems because of the added residues which decrease evaporation and loss of water. Consequently, a conservation tillage farming system for clayey soils was developed by Morrison et al. (1990) that restricted traffic to assigned lanes (Fig. 26). Neither yields of sorghum, cotton, and wheat, nor soil properties such as bulk density, total porosity, and soil strength were significantly different in the nontraffic area compared to conventional tillage. The research was on a shrinkswell clayey soil (Austin Series) cultivated over a 3-year period (Gerik and Morrison, 1985; Gerik et al., 1987). After 6 years, Morrison and Chichester (1994) reported that yields of wheat, grain sorghum, and maize in rotation were not significantly different between a chisel tillage and a controlled traffic notillage system. However, Potter and Chichester (1993) reported that, over a 10year period of controlled traffic, no-tillage management, bulk density, and soil strength below 5 cm depth increased with time since primary tillage. Restoration of the density (number per area), continuity, and connectivity of macropores occurs despite densification by consolidation (Coulombe, 1996). Similar conclusions were reported by Norton and Schroeder (1987) on a silt loam. Restoration of these macropore attributes is the first stage of structural regeneration in Vertisols (Coulombe, 1996). Nonetheless, controlled traffic, no-tillage management is one of the strategies proposed for the maintenance or regeneration of Vertisol properties. Other strate-
3 62
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
Figure 26 Conservation farming systems for clayey soils developed by Morrison er al. (1990) that restrict traffic to assigned lanes: (a) controlled traffic for a row crop system, (b) combined controlled traffic and no-tillage for a cereal system.
OVERVIEW OF VERTISOLS
363
gies involve gypsum application (Wild et al., 1992), return to permanent meadow or fallow for a significant period (Puentes, 1990; Puentes and Wilding, 1990; Hulme et al., 1991;Oleschko et al., 1993), and change in the cropping system or the use of a crop to promote deep ripping (Hipp and Gerard, 1973; Hodgson and Chan, 1984; Hulme et al., 1991). in Texas, structural and biotic systems of these soils can be partially regenerated by returning cultivated areas to permanent meadows for a duration of 15 to 20 years (Puentes, 1990; Coulombe, 1996). Nonetheless, little is known about the dynamics and processes of degradation and regeneration of Vertisol properties derived from different parent materials and subject to different climatic conditions, land use systems, and level of cultural inputs. Proper management of Vertisols remains site specific and requires comprehensive knowledge of degradation and regeneration processes prior to proposed management strategies.
X I . SUMMARY AND CONCLUDING REMARKS Vertisols are clayey soils that shrink and swell extensively upon changing soil moisture conditions. They occur globally under various parent material and environmental conditions. Vertisols exhibit unique morphological properties such as the presence of slickensides, wedge-shaped aggregates, diapir (mukkara), and gilgai. Shrink-swell phenomena are the dominant pedogenic processes in Vertisols and are attributed to changes in interparticle and intraparticle porosity with changes in moisture content. This is in contrast to the commonly invoked process of clay interlayer hydration-dehydration to explain shrinkswell phenomena. However, models proposed to explain the genesis of Vertisol features have not received universal agreement. The Vertisol order was recognized early as a distinctive group of soils in the United States’ classification systems, e.g., the Grumusol amendment to the 1938 Soil Classification system (Baldwin et al., 1938; Oakes and Thorp, 1950) and the 7th Approximation (Soil Survey Staff, 1960). With the current keys of Soil Taxonomy (Soil Survey Staff, 1994), the classification scheme may be criticized for containing too many subgroups. Nonetheless, this was done to accommodate the wide diversity in Vertisol properties globally and to be consistent with other soil orders in Soil Taxonomy (Wilding and Williams, 1991). Due to their clay content, Vertisols are global resources that are resilient to degradation compared to other soils. Unfortunately, degradation of Vertisols has occurred and has been reported worldwide regardless of the parent material, environmental conditions, and level of cultural input. Mineralogy of Vertisols is not restricted to smectites. Other phyllosilicates are dominant or codominant and exhibit significant shrink-swell activities. Chemical properties of Vertisols are
3 64
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
dependent on environmental conditions. Physical properties are major constraints to their optimal management due to shrink-swell activities and short periods of favorable soil moisture conditions for optimal tillage and trafficability. Management of biological properties is necessary for successful maintenance of Vertisol properties. Fundamental understanding of the dynamic and temporal behavior of soil structure and porosity derived from various parent materials, under different climatic conditions, and land use systems could considerably improve the comprehension of Vertisol management. Management of Vertisols remains site specific and requires comprehension of degradation and regeneration processes prior to proposed management strategies. Even with the remarkable increase in the knowledge base, utilization and management of Vertisols remains a challenge. Due to complex and interrupted horizonation, they often require complex sampling schemes to verify their attributes. Nonetheless, Vertisols are significant global resources that serve as the lifeline in subsistence agriculture due to their high productivity. Efforts toward comprehension and successful utilization are imperative for continued productivity and long-term sustainability of these resources for current and future civilizations.
REFERENCES Agrawal, R. P., and Ranamoorthy, B. 1970. Morphological and chemical characteristics of alkali and normal soils from black and red soils of India. Geoderma 4, 403-415. Ahmad, N. 1983. Vertisols. In “Pedogenesis and Soil Taxonomy. 11. The Soil Orders” (L. P. Wilding, N. E. Smeck, and G . F. Hall, Eds.), pp. 91-123. Development in Soil Sci. 119, Elsevier, Amsterdam, Netherlands. Ahmad, N. 1985. Acid Vertisols of Trinidad. In “Proceedings of the 5th International Soil Classification Workshop, Sudan 1982,” pp. 141- 151. Soil Survey Administration, Khartoum, Sudan. Ahmad, N. 1988. Management of Vertisols in the humid tropics. In “Vertisols: Their Distribution, Properties, Classification and Management” (L. P. Wilding, and R. Puentes, Eds.), pp. 97-1 15. Tech. Mono. No. 18, Texas A&M Printing Center, College Station, TX. Ahmad, N., and Jones, R. L. 1967. Forms and occurrence of inorganic phosphorus and its chemical availability in the limestone soils of Barbados. Soil Sci. Soc. Am. Proc. 31, 184-188. Ahmad, N., and Jones, R. L. 1969. Genesis, chemical properties and mineralogy of Cambean Grumusols. Soil Sci. 107, 166-174. Ahmad, N., and Mermut, A. R. 1996. “Vertisols and Vertisol Technology.” Elsevier, Amsterdam, Netherlands. Ahmad, N., Davis, C. E., and Jones, R. L. 1972. Ammonium fixation and its effect on mineralogical properties of West Indian clay soils. Trop. Agric. 49, 347-354. Ahmed, S . , Swindale, L. D., and El-Swaify, S. A. 1969. Effects of adsorbed cations on physical properties of tropical red earths and tropical black earths. I. Plastic limit, percentage stable aggregates, and hydraulic conductivity. J. Soil Sci. 20, 255-268. Allen, H. 1942. Classification of soils and control procedures used in construction of embankments. Public Roads 22, 263-282.
OVERVIEW OF VERTISOLS
365
Anderson, J. U., Fadul, K. E., and O’Connor, G. A. 1973. Factors affecting the coefficient of linear extensibility in Vertisols. Soil Sci. SOC.Am. Proc. 37, 296-299. Ankeny, M. D., Kaspar, T. C., and Horton, R. 1988. Design for an automated tension infiltrometer. Soil Sci. Soc. Am. J . 52, 893-896. Badraoui, M., and Bloom, P. R. 1990. Iron-rich high-charge beidellite in Vertisols and Mollisols of the High Chaouia Region of Morocco. Soil Sci. SOC.Am. J . 54, 267-274. Baize, D., and Girard, M. C. 1992. “RCfkrentiel PCdologique; Principaux Sols d’Europe.” Collection Techniques et Pratiques, INRA Editions. Baldwin, M., Kellogg, C. E. and Thorp, J. 1938. Soil Classification. In “Soils and Man,” pp. 9791001. USDA Yearbook of Agriculture. Ben Ohoud, M., and Van Damme, H. 1990. La texture fractale des argiles gonflantes. C. R. Acad. Sci. Paris 311, 665-670. Blokhuis, W. A., Pape, T., and Slager, S. 1968/1969. Morphology and distribution of pedogenic carbonate in some Vertisols of the Sudan. Geoderma 2, 173-200. Blokhuis, W. A , , Wilding, L. P., and Kooistra, M. J. 1991. Classification of vertic intergrades: Macromorphological and micromorphological aspects. In “Characterization, Classification and Utilization of Cold Aridisols and Vertisols,” Proceedings of the Sixth Int. Soil Correlation Meet (VI ISCOM), ( J . M. Kimble, Ed.), pp. 1-7. USDA-SCS, National Soil Survey Center, Lincoln, NE. Bottinger, J. L. 1992. Genesis,mineralogy and geochemistry of a red-black (Alfisol-Vertisol) complex, northeastern Queensland, Australia. Ph.D. Dissertation, University of California, Davis, CA . Bouma, J. 1982. Measuring the hydraulic conductivity of soil horizons with continuous macropores. Soil Sci. SOC. Am. J . 46, 438-441. Bouma, J. 1991. Influence of soil macroporosity on environmental quality. Adv. Agron. 46, 1-37. Brewer, R. 1964. “Fabric and Mineral Analysis of Soils.” Wiley, New York. Brewer, R. 1976. “Fabric and Mineral Analysis of Soils.’’ Robert E. Krieger, Huntington, NY. Brinkman, R. H. 1978. “Ferrolysis.” Elsevier, New York. Brown, K. W. 1977. Shrinking and swelling of clay, clay strength, and other properties of clay soils and clays. In “Minerals in Soil Environments” (J. B. Dixon, and S. B. Weed, Eds.). pp. 689707. Soil Sci. SOC.Am. Pub., Madison, WI. Brown, K. W., and Daniel, D. E. 1988. Influence of organic liquids on the hydraulic conductivity of soils. In “Land Disposal of Hazardous Waste. Engineering and Environmental Issues” ( J . R. Gronow, A. N. Schofield, and R. K. Jain, Eds.), pp. 235-244. Ellis Horwood, Chichester, England. Brown, K. W., and Thomas, J. C. 1987. A mechanism by which organic liquids increase the hydraulic conductivity of compacted clay materials. Soil Sci. SOC. Am. J . 51, 1451-1459. Bui, E. N . , and Mermut, A. R. 1989. Orientation of planar voids in Vertisols and soils with vertic properties. Soil Sci. SOC. Am. J. 53, 171-178. Bullock, P., Fedoroff, N., Jongerius, A., Stoops, G., and Tursina, T. 1985. “Handbook for Soil Thin Section Description.” Waine Research Publications, Wolverhampton. Buol, S . W., Hole F. D., and McCracken, R. J. 1980. “Soil Genesis and Classification,” second ed. The Iowa State University Press, Ames, IA. Burnham, C. P., and Lopez-Hernandez, D. 1982. Phosphate retention in different soil taxonomic classes. Soil Sci. 134, 376-380. Carson, C. D., and Dixon, J. B. 1972. Potassium selectivity in certain montmorillonitic soil clays. Soil Sci. SOC.Am. Proc. 36, 838-843. Carson, C. D., Kittrick, J. A,, Dixon, J. B., and McKee, T. R. 1976. Stability of soil smectite from a Houston Black clay. Clays Clay Miner. 24, I5 I- 155.
3 66
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
Cernica, J. N. 1982. “Geotechnical Engineering.” Holt, Rinehart and WinstonKBS College, New York. Chang, R. K., and Warkentin, B. P. 1968. Volume change of compacted clay soil aggregates. Soil Sci. 102, 106- 11I . Chen, C. C., ’hrner, F. T., and Dixon, J. B. 1989. Ammonium fixation by high-charge smectite in selected Texas Gulf Coast soils. Soil Sci. SOC. Am. J . 53, 1035-1040. Chenu, C. 1989. Influence of a fungal polysaccharide, scleroglucan, on clay microstructures. Soil Biol. Biochem. 21, 299-305. Chenu, C., and Guerif, J. 1991. Mechanical strength of clay minerals as influenced by an adsorbed polysaccharide. Soil Sci. Soc. Am. J . 55, 1076-1080. Chenu, C., and Jaunet, A. M. 1990. Modifications de I’organisation texturale d’une montmorillonite calcique Mes a I’adsorption d’un polysaccharide. C.R. Acad. Sci. Paris 310, 975-980. Chenu, C., Pons, C. H., and Robert, M. 1987. Interaction of kaolinite and montmorillonite with neutral polysaccharide. In “Proceedings, Int. Clay Conf., Denver, 1985” (L. G. Schultz, H. Van Olphen and, F. A. Mumpton, Eds.), pp. 375-381. Clay Miner. SOC.,Bloomington, IN. Chotte, J. L., Villemin, G., Guillork, P., and Jocteur Monrozier, L. 1994. Morphological aspects of microorganism habitats in a Vertisol. In “Soil Micromorphology: Studies in Management and Genesis” (A. J. Ringrose-Voase, and G. S. Humphreys, Eds.), pp. 395-403. (Proc. IX Int. Work. Meet. Soil Micromorphology, Townsville, Australia (1992), Developments in Soil Science 22). Elsevier, Amsterdam, Netherlands. Comerma, J. A. 1985. Vertisols of Venezuela. In “Proceedings of the 5th International Soil Classification Workshop, Sudan, 1982,” pp. 153-161. Soil Survey Administration, Khartoum, Sudan. Comerma, J. A,, Williams, D., and Newman, A. 1988. Conceptual changes in the classification of Vertisols. In “Vertisols: Their Distribution, Properties, Classification and Management” (L. P. Wilding and R. Puentes, Eds.), pp. 41-54. Tech. Mono. No 18, Texas A&M Printing Center, College Station, TX. Coulombe, C. E. 1996. Surface properties of Vertisols of Texas and Mexico under different land use systems. Ph.D. Dissertation, Texas A&M University, College Station, TX. Coulombe, C. E., Dixon, J. B., and Wilding, L. P. 1996. Mineralogy and chemistry of Vertisols. In “Vertisols and Vertisol Technology” (N. Ahmad, and A. R. Mermut, Eds.). Elsevier, Amsterdam, Netherlands. C.P.C.S. (Commission de Pkdologie et de Cartographie des Sols) 1967. “Classification des sols.” E.N.S.A. Grignon, France. Dalal, R. C. 1989. Long-term effects of no-tillage, crop residue, and nitrogen application on properties of a Vertisol. Soil Sci. SOC. Am. J. 53, 1511-1515. Dan, J., and Koyumdjisky, H. 1963. The soils of Israel and their distribution. J. Soil Sci. 14, 12-20. De Datta, S. K., Raymundo, M. E., and Greenland, D. J. 1988. Managing wet Vertisols in ricebased cropping systems. In “Vertisols: Their Distribution, Properties, Classification and Management” (L. P. Wilding, and R. Puentes, Eds.), pp. 117-127. Tech. Mono. No 18, Texas A&M Printing Center, College Station, TX. Delvaux, B., Tessier, D., Herbillon, A. J., Burtin, G., Jaunet, A. M., and Vielvoye, L. 1992. Morphology, texture and microstructure of halloysitic soil clays as related to weathering and exchangeable cation. Clays Clay Miner. 40, 446-456. Del Villar, E. H. 1944. The Tirs of Morocco. Soil Sci. 57, 313-339. DeMent, J. A., and Bartelli, L. J. 1969. The role of vertic subgroups in the comprehensive soil classification system. Soil Sci. SOC.Am. Proc. 33, 129-131. De Vos, J. H., and Virgo, K. J. 1969. Soil structure in Vertisols of the Blue Nile clay plains, Sudan. J. Soil Sci. 20, 189-206. Dixon, J. B. 1978. X-ray diffraction analysis. In “Analyses of Clay, Silt and Sand Fractions of
OVERVIEW OF VERTISOLS
3 67
Selected Soils from Southeastern United States” (R. 1. Barnhisel, Ed.), pp. 11-22. Southern Cooperative Series Bulletin 219. Dixon, J. B. 1982. Mineralogy of Vertisols. I n “Vertisols and Rice Soils of the Tropics,” pp. 48-59. Symposia Papers 11. 12th ICSS, New Delhi, India. Dixon, J. B., and Nash, V. E. 1968. “Chemical, Mineralogical and Engineering Properties of Alabama and Mississippi Black Belt Soils.” 69 Southern Coop. Ser. No 130, Ala. Agr. Exp. Sta., Auburn, AL. Dixon, 1. B., Jacob, J. S., and White, G. N. 1994. Todorokite in manganese oxide nodules of a Guatemalan Vertisol. Agron. Absrr., 409-410. Donfack, P. 1992. Etude de la dynamique de la vCgCtation aprks abandon de la culture au NordCameroun. Thtse de doctorat, Universite de YaoundC, Yaounde, Cameroon. Drescher, J., Horn, R., and De Boodt. M. 1988. “Impact of water and external forces on soil structure.” Catena 11, Cremligen-Destedt, Germany. Duchaufour, P. 1983. “PCdologie 1. PedogenBse et classification,” second ed. Edition Masson, Paris. Dudal, R., 1965. “Dark Clay Soils of Tropical and Subtropical Regions.” FA0 Development Papers No 83. Dudal, R . , and Eswaran, H. 1988. Distribution, properties and classification of Vertisols. I n “Vertisols: Their Distribution, Properties, Classification and Management” (L. P. Wilding, and R. hentes, Eds.), pp. 1-22. Tech. Mono. No. 18, Texas A&M Printing Center, College Station, TX . El-Swaify, S. A,, Ahmed, S., and Swindale, L. D. 1970. Effects of adsorbed cations on physical properties of tropical red earths and tropical black earths. 11. Liquid limit, degree of dispersion, and moisture retention. J. Soil Sci. 21, 188-198. F.A.0.-Unesco. 1974. “Soil Map of the World.” World Soil Resources, Rome-Paris. F.A.0.-Unesco. 1982. “Document D: Draft Definitions of Soil Units at High Level.” World Soil Resources, Rome. F.A.0.-Unesco. 1988. “Soil Map of the World: Revised Legend,” pp. 1-1 19. World Soil Resources, Report No. 60, Rome, Italy. Feigin, A., and Yaalon, D. H. 1974. Non-exchangeable ammonium in soils of Israel and its relation to clay and parent materials. J. Soil Sci. 25, 384-397. Finck, A. 1961. Classification of Gezira clay soil. Soil Sci. 92, 263-267. Fitzpatrick, R. W., and Le Roux, J. 1977. Mineralogy and chemistry of a Transvaal Black clay toposequence. J . Soil Sci. 28, 165-179. Gerik, T. J., and Momson, J. E. Jr., 1984. No-tillage of grain sorghum on a shrinking clay soil. Agron. J. 76, 71-76. Gerik, T. J., and Morrison, 1. E., Jr. 1985. Wheat performance using no-tillage with controlled wheel traffic on a clay soil. Agron. J. 77, 115-1 18. Gerik, T. J., Morrison, J. E., Jr., and Chichester, F. W. 1987. Effects of controlled-traffic on soil physical properties and crop rooting. Agron. J. 79, 434-438. Glinka, K. D. 1914. “Die typen der bodenbildung, ihre Klassifikation und geographische verbreitung.” Borntraeger, Berlin. Graham, R. C., and Southard, A. R. 1983. Genesis of a Vertisol and an associated Mollisol in northern Utah. Soil Sci. SOC.Am. J . 47, 552-559. Grant, C. D., and Blackmore, A. V. 1991. Self-mulching behaviour in clay soils: Its definition and measurement. Aust. J. Soil Res. 29, 155-173. Gray, M. B., and Nickelsen, R. P. 1989. Pedogenic slickensides as indicators of strain and deformation processes in redbed sequences of Appalachian Foreland. Geology 17, 72-75. Grossman, R. B., Brasher, B. R., Franzmeier, D. P., and Walker, J. L. 1968. Linear extensibility as calculated from natural-clod bulk density measurements. Soil Sci. SOC. Am. Proc. 32, 570-573.
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON Gustavson, T. C. 1991. “Arid Basin Depositional Systems and Paleosols: Fort Hancock and Camp Rice Formations (Pliocene-Pleistocene), West Texas and Adjacent Mexico.” Bur. Econ. Geology, University of Texas, Austin, TX. Giizel, N., and Wilson, M. J. 1981. Clay-mineral studies of a soil chronosequence in Southern lbrkey. Geoderma 25, 113-129. Hajek, B. F. 1985. Mineralogy of Aridisols and Vertisols. In “Proceedings of the 5th International Soil Classification Workshop, Sudan 1982,” pp. 221-229. Soil Survey Administration, Khartoum, Sudan. Hallsworth, E. G., and Beckmann, G. G. 1969. Gilgai in the Quaternary. Soil Sci. 107,409-420. Hallsworth, E. G., Robertson, G. K., and Gibbons, F. R. 1955. Studies in pedogenesis in New South Wales. VII. Gilgai soils. J . Soil Sci. 6 , 1-31. Harris, B. L. 1989. Management of cropland Vertisols in Texas. In “Vertisols Management in Africa,” (Proceedings of a Workshop Held at Harare, Zimbabwe), pp. 253-267. IBSRAM Proceedings No. 9, printed in Thailand. Hawkins, R. H., and Kunze, G. W. 1965. Phosphate fractions in some Texas Grumusols and their relation to soil weathering and available phosphorus. Soil Sci. SOC. Am. Proc. 29, 650656. Hilgard, E. W. 1906. “Soils.” Macmillan, New York. Hipp, B. W., and Gerard, C. J. 1973. Influence of cropping system on salt distribution in an imgated Vertisol. Soil Sci. SOC. Am. Proc. 65, 97-99. Hirekerur, L. R., Pal, D. K., Sehgal, J. L., and Deshpande, S . B. 1989. Classification, management and use potential of swell-shrink soils. In “Transactions, International Workshop on SwellShrink Soils.” Oxford-IBH, Nagpur, India. Hirekurubar, B. M., Doddamani, V. S., and Satyanarayana, T. 1991. Some physical properties of Vertisols derived from different parent materials. J . Indian SOC. Soil Sci. 39, 242-245. Hodgson, A. S . , and Chan, K. Y. 1984. Deep moisture extraction and crack formation by wheat and safflower in a Vertisol following imgated cotton rotations. In “The Properties and Utilisation of Cracking Clay Soils” (J. W. McGarity, E. H. Hoult, and H. B. So, Eds.), pp. 299-304. Reviews in Rural Science 5, Armidale, New South Wales. Hodgson, A. S., and MacLeod, D. A. 1989a. Use of oxygen flux density to estimate critical air-filled porosity of a Vertisol. Soil Sci. SOC. Am. J . 53, 355-361. Hodgson, A. S . , and MacLeod, D. A. 1989b. Oxygen flux, air-filled porosity, and bulk density as indices of Vertisol structure. Soil Sci. SOC. Am. J. 53, 540-543. Holder, M. B., and Griffith, S . M. 1983. Some characteristics of humic materials in Caribbean Vertisols. Cnnad. J . Soil Sci. 63, 151-159. Hubble, G. D. 1984. The cracking clay soils: Definition, distribution, nature, genesis and use. In “The Properties and Utilisation of Cracking Clay Soils” (J. W. McGarity, E. H. Hoult, and H. B. So, Eds.), pp. 3-13. Reviews in Rural Science 5, Armidale, New South Wales. Hulme, P. J., McKenzie, D. C., Abbott, T. S., and MacLeod, D. A. 1991. Changes in the physical properties of a Vertisol following an irrigation of cotton as influenced by the previous crop. Ausf. J . Soil Res. 29, 425-442. Hussain, M.S . , Amadi, T. H., and Sulairnan, M. S. 1984. Characteristics of soils of a toposequence in northeastern Iraq. Geoderma 33, 63-82. IBSRAM (International Board for Soil Research and Management Inc.). 1987. Management of Vertisols under semi-arid conditions. In “Proceedings of the First Regional Seminar on Management of Vertisols under Semi-Arid Conditions, Held at Nairobi, Kenya.” IBSRAM Proceedings No. 6, printed in Thailand. IBSRAM. 1989. Vertisols management in Africa. In “Proceedings of a Workshop Held at Harare, Zimbabwe.” IBSRAM Proceedings No. 9, printed in Thailand. ICRISAT (International Crops Research Institute for the Semi-Arid Tropics). 1989. Management of
OVERVIEW OF VERTISOLS
3 69
Vertisols for improved agricultural production. In “Proceedings of an IBSRAM Inaugural Workshop.” ICRISAT Center, India. Ikawa, H. 1985. The Vertisols of Hawaii. In “Proceedings of the 5th International Soil Classification Workshop, Sudan 1982,” pp. 125-139. Soil Survey Administration, Khartoum, Sudan. Jambu, P., Coulibaly, G., and Bemer, J. 1987. Etude expbrimentale du r6Ie des lipides dans la microorganisation des argiles. I n “Micromorphologie des Sols-Soil Micromorphology” (N. Fedoroff, L. M. Bresson, and M. A. Courty, Eds.), pp. 475-480. Proceedings VII Int. Work. Meet. Soil Micromorphology, Paris, France. Jensen, H. 1. 191 I . The nature and origin of gilgai country. Proc. R. Soc. N.S.W. 45, 337-358. Jocteur Monrozier, L., Ladd, J. N., Fitzpatrick, R. W., Foster, R. C., and Raupach, M. 1991. Components and microbial biomass content of size fractions in soils of contrasting aggregation. Geoderma 50, 37-62. Joeckel, R. M. 1994. Virgilian (Upper Pensylvanian) Paleosols in the upper Lawrence Formation (Douglas group) and in the Snyderville shale member (Oread Formation, Shawnee group) of the northem midcontinent, USA: Pedologic contrasts in a cyclothem sequence. J. Sedimentary Res. 64, 853-866. Joeckel, R. M. 1995. Paleosols below the Ames marine unit (Upper Pensylvanian, Conemaugh group) in the Appalachian Basin, USA: Variability on an ancient depositional landscape. J . Sedimentary Res. 6 5 , 393-407. Johnson, W. M., Cady, 1. G., and James, M. S. 1962. Characteristics of some Brown Grumusols of Arizona. Soil Sci. SOC.Am. Proc. 26, 389-393. Jones, C. A,, and Gerik, T. J. 1990. Vertisol Management Workshop: International Collaboration in Research, Training and Extension-Final Report. Jutzi, S. C., Haque, I . , McIntire, J., and Stares, J. E. S. 1988. Management of Vertisols in subSaharan Africa. Proceedings of a conference held at Ilca, Addis Ababa, Ethiopia. Kantor, W., and Schwertmann, U. 1974. Mineralogy and genesis of clays in red-black soil toposequences on basic igneous rocks in Kenya. J. Soil Sri. 25, 67-78. Kissel, D. E., Ritchie, J. T., and Bumett, E. 1973. Chloride movement in undisturbed swelling clay soil. Soil Sci. SOC.Am. Proc. 37, 21-24. Kissel, D. E., Ritchie, J. T., and Burnett, E. 1974. Nitrate and chloride leaching in a swelling clay soil. J . Environ. Qualify 3, 401-404. Knight, M. J. 1980. Structural analysis and mechanical origins of gilgai at Boorook, Victoria, Australia. Geoderma 23, 245-283. Kooistra, M. J., and Boersma, 0. H. 1994. Subsoil compaction in Dutch marine sandy loams: Loosening practices and effects. Soil Tillage Res. 29, 237-247. Kooistra, M. J., and Tovey, N. K. 1994. Effects of compaction on soil microstructure. In “Soil Compaction in Crop Production” (B. D. Soane and C. van Ouwerkerk, Eds.), pp. 91-111. Elsevier, Amsterdam, Netherlands. Koppi, A. J., and McBratney, A. B. 1991. A basis for soil mesomorphological analysis. J. Soil Sci. 42, 139-146. Koppi, A. J., Douglas, J. T., and Moran, C. J. 1992. An image analysis evaluation of soil compaction in grassland. J. Soil Sci. 43, 15-25. Kounetsron, O., Robert, M., and Berrier, J. 1977. Nouvel aspect de la formation des smectites dans les vertisols. C.R. Acad. Sci. Paris 284, 733-736. Krishna, P. G . , and Perumal, S. 1948. Structure in Black Cotton soils of the Nizamsagar project area, Hyderabad State, India. Soil Sci. 66, 29-38. Krishna Murti, G. S. R., Singh, G., and Rengasamy, P. 1977. The nature of soil clays and the stability of microaggregates. Aust. J . Soil Res. 15, 115-1 19. Kunze, G. W., and Templin, E. H. 1956. Houston Black Clay, the type Grumusol: 11. Mineralogical and chemical characterization. Soil Sci. SOC.Am. Proc. 20. 91-96.
3 70
C. E. COULOMBE, L. P. WILDING, AND J. B. DMON
Kunze, G. W., Oakes, H., and Bloodworth, M. E. 1963. Grumusols of the coast prairie of Texas. Soil Sci. SOC.Am. Proc. 27, 412-421. Lin, H. 1995. Hydraulic properties and macropore flow of water in relation to soil morphology. Ph. D. dissertation, Texas A&M University, College Station, TX. Lugo-Lopez, M. A,, Camelli, J. P., and Acevedo, G. 1985. Morphological, physical and chemical properties of major soils from Calagua in northwestern Uruguay. Soil Sci. SOC.Am. J . 49, 108113.
Malik, H. U. 1990. Characterization and mineralogical interpretation of shrink-swell behavior of Hawaiian kaolinitic Vertisols. Ph. D. dissertation, University of Hawaii, Honolulu, HI. Malik, H. U . , and Jones, R. C. 1991, Two scenarios for the genesis of the kaolinitic Vertisols on the Hawaiian Islands of Lanai and Molokai. Abstr. Clay Minerals Soc.. 108. Mandelbrot, B. 1993. Fractals-A geometry of nature. In “Exploring Chaos. A Guide to the New Science of Disorder” (N. Hall, Ed.), pp. 122-135. Norton, New York. Marbut, C. F. 1928. A scheme for soil classification. Proc. First Inr. Cong. Soil Sci. IV, 1-31. McGarity, J. W., Hoult, E. H., and So, H. B. 1984. The properties and utilization of cracking clay soils. Reviews in Rural Science 5 , Armidale, New South Wales. McGany, D. 1987. The effect of soil water content during land preparation on aspects of soil physical condition and cotton growth. Soil Tillage Res. 9, 287-302. McGany, D. 1988. Quantification of the effects of zero and mechanical tillage on a Vertisol by using shrinkage curve indices. Aust. J. Soil Res. 26, 537-542. McGany, D. 1989. The effect of wet cultivation on the structure and fabric of a Vertisol. J. Soil Sci.
40, 199-207. McGany, D., and Daniells, I. G. 1987. Shrinkage curve indices to quantify cultivation effects on soil structure of a Vertisol. Soil Sci. SOC. Am. J . 51, 1575-1580. McGany, D., and Malafant, K. W. J. 1987. The analysis of volume change in unconfined units of soil. Soil Sci. SOC. Am. J . 51, 290-297. McGany, D., and Smith, K. J. 1988. Indices of residual shrinkage to quantify the comparative effects of zero and mechanical tillage on a Vertisol. Ausr. J . Soil Res. 26, 543-548. Mella, W. I. I. 1991. Effects of conservation tillage and conventional tillage on selected physical and chemical properties of a Blackland prairie soil in Texas. M.Sc. Thesis, Texas A&M University, College Station, TX. Mermut, A. R., and Dasog, G. S. 1986. Nature and micromorphology of carbonate glaebules in some Vertisols of India. Soil Sci. SOC. Am. J. 50, 382-391. Mermut, A. R., Acton, D. F., and Tarnocai, C. 1991. A review of recent research on swelling clay soils in Canada. In “Characterization, Classification and Utilization of Cold Aridisols and Vertisols,” Proceedings of the Sixth Int. Soil Correlation Meet. (VI ISCOM), (J. M. Kimble, Ed.), pp. 112-121. USDA-SCS, National Soil Survey Center, Lincoln, NE. Merwe, C. R. van der, and Heystek, H. 1955. Clay minerals of South African soil groups: 11. Subtropical black clays and related soils. Soil Sci. 79, 147-158. Messing, I., and Jarvis, N. J. 1990. Seasonal variation in field-saturated hydraulic conductivity in two swelling clay soils in Sweden. J. Soil Sci. 41: 229-237. Meyer, R. 1987. “Paleoaltkrites et Paleosols. L‘empreinte du continent dans les sCries s6dimentaires.” Manuels et Methodes No. 13, Editions BRGM, Orleans, France. Morrison. J. E., Jr., and Chichester, F. W. 1994. Tillage system effects on soil and plant nutrient distributions on Vertisols. J. Prod. Agric. 7 , 364-373. Morrison, J. E., Jr., Gerik, T. J., Chichester, F. W., Martin, J. R.,and Chandler, J. M. 1990. A notillage farming system for clay soils. J. Prod. Agric. 3, 219-227. Murthy, A. S. P. 1988. Distribution, properties and management of Vertisols of India. Adv. Soil Sci. 8, 151-214. Nelson, L. A., Kunze, G. W.,and Godfrey, C. L. 1960. Chemical and mineralogical properties of San Saba Clay, a Grumusol. Soil Sci. 89, 122-131.
OVERVIEW OF VERTISOLS
371
Norrish, K., and Pickering, J. G. 1977. Clay mineralogic properties. In “Soil Factors in Crop Production in a Semi-arid Environment” (J. S. Russell, and E. L. Greacen, Eds.), pp. 33-53. University of Queensland Press, Australia. Nomsh, K., and Pickering, J. G. 1983. Clay minerals. In “Soils, An Australian Viewpoint,” pp. 281-308. Commonwealth Scientific and Industrial Research Organization, Melbourne/ Academic Press, London. Northcote, K. H., and Skene, J. K. 1972. Australian soils with saline and sodic properties. CSIRO Soil Pub. No. 27, Australia. Norton, L. D., and Schroeder, S. L. 1987. The effect of various cultivation methods on soil loss: A micromorphological approach. In “Micromorphologie des Sols-Soil Micromorphology” (N. Fedoroff, L. M. Bresson, and M. A. Courty, Eds.), pp. 431-436. Proceedings VII Int. Work. Meet. Soil Micromorphology, Paris, France. Nychas, A. E., and Kosmas, C. S. 1984. Phosphate adsorption by dark alkaline Vertisols in Greece. Geoderma 32, 319-327. Oades, J. M. 1984. Soil organic matter and structural stability: Mechanisms and implications for management. Planr Soil 76, 319-337. Oakes, H., and Thorp, 1. 1950. Dark-clay soils of warm regions variously called Rendzina, Black Cotton Soils, Regur, and Tirs. Soil Sci. SOC. Am. Proc. 15, 348-354. Oleschko, K. J. D., Etchevers, B., and Osorio-I., L. 1993. Pedological features as indicators of the tillage effectiveness in Vertisols. Soil Tillage Res. 26, 11-31. Parsons, R. B., Moncharoan, L., and Knox, E. G. 1973. Geomorphic occurrence of Pelloxererts, Willamette Valley, Oregon. Soil Sci. SOC. Am. Proc. 37, 924-927. Patil, S. G., Sarma, V. A. K., and Van Loon, G. W. 1989. Acid rain, cation dissolution, and sulphate retention in three tropical soils. J . Soil Sci. 40, 85-93. Paton, T. R. 1974. Origin and terminology for gilgai in Australia. Geoderma 11, 221-242. Pemer, E. 1994. Structure gkomktrique et fonctionnement hydrique des sols: Simulations exploratoires. These de Doctorat (Ph. D.), Univ. Paris VII, Paris, France. Perroux, K. M., and White, I. 1988. Design for disc permeameters. Soil Sci. SOC.Am. J. 52, 12051215. Potter, K. N., and Chichester, F. W. 1993. Physical and chemical properties of a Vertisol with continuous controlled-traffic, no-till management. Trans. ASAE 36, 95-99. Prescott, J. A. 1931. The soils of Australia in relation to vegetation and climate. Aust. Commonw., Counc. Sci. Ind. Res. Bull. No. 52. Prieksat, M. A,, Ankeny, M. D., and Kaspar, T. C. 1992. Design for an automated, self-regulating, single ring infiltrometer. Soil Sci. SOC. Am. J . 56, 1409-141 I . Probert, M. E., Fergus, 1. F., Bridge, B. J., McGany, D., Thompson, C. H., and Russel, J. S. 1987. “The Properties and Management of Vertisols.” C.A.B. International. Puentes, R. 1990. Soil structure restoration in Vertisols under pastures in Texas. Ph. D. dissertation, Texas A&M University, College Station, TX. Puentes, R., and Wilding, L. P. 1990. Structural restoration in Vertisols under pastures in Texas. Proc. 14th ISSS, Kyoto, Japan VII, 244-249. Puentes, R., Harris, B. L., and Victora, C. 1988. Management of Vertisols of temperate regions. I n “Vertisols: Their Distribution, Properties, Classification and Management” (L. P. Wilding and R. Puentes, Eds.), pp. 129-145. Tech. Mono. No. 18, Texas A&M Printing Center, College Station, TX. Puentes, R., Wilding, L. P., and Drees, L. R. 1992. Microspatial variability and sampling concepts in soil porosity studies of Vertisols. In “Digitization, Processing and Quantitative Interpretation of Image Analysis in Soil Science and Related Areas” (A. R. Mermut and L. D. Norton, Eds.). Geoderma 53, 373-385. Quirk, J. P. 1994. Interparticle forces : a basis for the interpretation of soil physical behavior. Adv. Agron. 53, 121-183.
3 72
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
Rajan, S. V. G., Murthy, R. S., Kalbande, A. R., and Venugopal, K. R. 1972. Micromorphology and chemistry of carbonate concretion in black clay soils. Indian J. Agric. Sci. 42, 1020-1023. Raven, K. P. 1992. Characterization of the phosphorus supplying capacity of soils by desorption Q/I relationships and kinetics. Ph. D. Dissertation, Texas A&M University, College Station, TX. Raymundo, M. E. 1965. The properties of the black earth of Hawaii. Ph.D. dissertation, University of Hawaii, Honolulu, HI. Retallack, G. J. 1986. The fossil record of soils. In “Paleosols: Their Recognition and Interpretation” (Wright, V. P., Ed.), pp. 1-57. Blackwell Sci., Oxford. Rieu, M., and Sposito, G. 1991a. Fractal fragmentation, soil porosity, and water properties: I. Theory. Soil Sci. Soc.Am. Proc. 55, 1231-1238. Rieu, M., and Sposito, G . 1991b. Fractal fragmentation, soil porosity, and water properties: 11. Applications. Soil Sci. SOC. Am. Proc. 55, 1239-1244. Ritchie, J. T., Kissel, D. E., and Burnett, E. 1972. Water movement in undisturbed swelling clay soil. Soil Sci. SOC. Am. Proc. 36, 874-879. Rodriguez, G. 1954. Fixed ammonia in Tropical soils. J . Soil Sci. 5, 264-274. Roy, B. B.. and Barde, N. K. 1962. Some characteristics of the black soils of India. Soil Sci. 93, 142- 147. Said, M. B. 1973. Ammonium fixation in the Sudan Gezira soil. Plant Soil 38, 9-16. Sarma, V. A. K., and Krishna-Murti, G. S. R. 1970. Plumbogummite minerals in Indian soils. Geoderrnu 3, 321-327. Schafer, W. M.,and Singer, M. J. 1976. A new method of measuring shrink-swell potential using soil pastes. Soil Sci. Soc. Am. J. 40, 805-806. Seiny-Boukar, L., Floret, C., and Pontanier, R. 1992. Degradation of savanna soils and reduction of water available for the vegetation: The case of northern Cameroon Vertisols. Cunud. J. Soil Sci. 72, 481-488. Shadfan, H. 1983. Clay minerals and potassium status in some soils of Jordan. Geodermu 31,41-56. Sharpless, M. R., and Yelderman, J. C., Jr. 1993. “The Texas Blackland Prairie. Land, History, and Culture.” Baylor University, Waco, TX. Simon, A. B., Bidlo, G., and Liautaud, G. 1975. On the black cotton soils of North Cameroon. Eng. Geol. 9, 351-357. Simonson, R. W. 1954a. The Regur soils of India and their utilization. Soil Sci. SOC. Am. Proc. 18, 199-203. Simonson, R. W. 1954b. Morphology and classification of the Regur soils of India. J . Soil Sci. 5 , 275-288. Singh, S . 1954. A study of the black cotton soils with special reference to their coloration. J . Soil Sci. 5, 289-299. Singh, S . 1956. The formation of dark-coloured clay-organic complexes in Black Soils. J. Soil Sci. 7, 43-58. Singh, D., and Lal, G . 1946. Kankar composition as an index of the nature of soil profile. Indian J. Agric. Sci. 16, 328-342. Singh, L., and Singh, S. 1972. Chemical and morphological composition of Kankar nodules in soils of the Vindhyan region of Mirzarpur, India. Geoderrnu 7, 269-276. Skjemstad, J. O., and Dalal, R. C. 1987. Spectroscopic and chemical differences in organic matter of two Vertisols subjected to long periods of cultivation. Aust. J. Soil Res. 25, 323-335. Skjemstad, J. 0.. Dalal, R. C., and Barron, P. F. 1986. Spectroscopic investigations of cultivation effects on organic matter of Vertisols. Soil Sci. SOC. Am. J. 50, 354-359. Soane, B. D., and van Ouwerkerk, C. 1994. “Soil Compaction in Crop Production.” Elsevier, Amsterdam, Netherlands. Soil Survey Staff. 1960. “Soil Classification, A Comprehensive System, 7th Approximation.” USDA, Washington, DC.
OVERVIEW OF VERTISOLS
3 73
Soil Survey Staff. 1975. ‘‘Soil Taxonomy. A Basic System of Soil Classification for Making and Interpreting Soil Surveys.” USDA-SCS/Wiley, New York. Soil Survey Staff. 1990. “Keys to Soil Taxonomy,” fourth ed. SMSS Technical Monograph No. 6. Pocahontas Press, Blacksburg, VA. Soil Survey Staff. 1992. “Keys to Soil Taxonomy,” fifth ed. SMSS Technical Monograph No. 19, Pocahontas Press, Blacksburg, VA. Soil Survey Staff. 1994. “Keys to Soil Taxonomy,” sixth ed. USDA-SCS, U.S. Government Printing Office, Washington, DC. Solis, P., and Torrent, J. 1989a. Phosphate sorption by calcareous Vertisols and Inceptisols of Spain. Soil Sci. SOC. Am, J. 53, 456-459. Solis, P., and Torrent, J. 1989b. Phosphate fractions in calcareous Vertisols and Inceptisols of Spain. Soil Sri. Sor. Am. J . 53: 462-466. Southard, R. J., and Graham, R. C. 1992. Cesium-I37 distribution in a California Pelloxerert: Evidence of pedoturbation. Soil Sci. SOC. Am. J. 56, 202-207. Sparks, D. L. 1980. Chemistry of soil potassium in Atlantic coastal plain soils: A review. Commun. Soil Sci. Plant Anal. 11, 435-449. Sposito, G. 1982. On the use of the Langmuir equation in the interpretation of the “adsorption” phenomena: 11. The “two-surface” Langmuir equation. Soil Sci. Soc. Am. J. 46, 1147-1152. Sposito, G., and Reginato, R. J. 1992. “Opportunities in Basic Soil Science Research.” Soil Sci. SOC.Am., Madison, WI. Stace, H. C. T, Hubble, G. D., Brewer, R., Northcote, K. H., Sleeman, J. R., Mulcahy, M. J., and Hallsworth, E. G. 1968. “A Handbook of Australian Soils.’’ Rellim Tech., Glenside, South Australia. Standley, J., Thomas, G. A,, Hunter, H. M., Webb, A. A,, and Berthelsen, S. 1990. Decreases over seven years in subsoil nitrate in a Vertisol with grain sorghum and grass. Plant Soil 125, 1-6. Stephan, J., Berrier, J., De Petre, A. A , , Jeanson, C., Kooistra, M. J., Scharpenseel, H. W., and Schiffmann, H. 1983. Characterization of in situ organic matter constituents in Vertisols from Argentina, using submicroscopic and cytochemical methods-First report. Geodermu 30, 21-34. Stephen, 1. 1953. A petrographic study of a tropical black earth and grey earth from the Gold Coast. J . Soil Sci. 4, 21 1-219. Stephen, I., Bellis, E., and Muir, A. 1956. Gilgai phenomena in Tropical Black clays of Kenya. J. Soil Sri. 7 , 1-9. Sullivan, L. A , , and Koppi, A. J. 1992. Manganese oxide accumulations associated with some structural pores. 1. Morphology, composition and genesis. Ausr. J. Soil Res. 30,409-427. Sullivan, L. A,, and Koppi, A. J. 1993. Manganese oxide accumulations associated with some structural pores. 11. Composite coatings and translocation. Ausr. J. Soil Res. 31, 227-223. Templin, E. H., Mowery, 1. C., and Kunze, G . W. 1956. Houston Black clay, the type Grumusol: I. Field morphology and geography. Soil Sci. SOC. Am. Proc. 20, 88-90. Tessier, D. 1984. Etude exp6rimentale de I’organisation des matCriaux argileux. Hydratation, gonflement et structuration au cours de la dessication et de la rkhumectation. Thkse Doctor es Sciences, Univ. Paris VII, Paris, France. Tessier, D. and PCdro, G., 1987. Mineralogical characterization of 2:1 clays in soils: Importance of the clay texture. In “Proceedings Int. Clay Conf., Denver, 1985” (L. G. Schultz, H. Van Olphen, and F. A. Mumpton, Eds.), pp. 78-84. Clay Miner. Soc., Bloomington, IN. Tessier, D., Bruand, A., and Cabidoche, Y. M. 1991. Vertisols of France. In “Characterization, Classification and Utilization of Cold Aridisols and Vertisols” (J. M. Kimble, Ed.), pp. 227231. Proceedings of the Sixth Int. Soil Correlation Meet. (VI ISCOM), USDA-SCS, National Soil Survey Center, Lincoln, NE. Tisdall, J. M., and Oades, J. M. 1982. Organic matter and water-stable aggregates in soils. J. Soil Sci. 33, 141-163.
3 74
C. E. COULOMBE, L. P. WILDING, AND J. B. DIXON
Trauth, N., Paquet, H., Lucas, J., and Millot, G. 1967. Les montmorillonites des vertisols lithomorphes sont femfkres: consequences gtochimiques et stdimentologiques. C. R . A c d . Sci. Paris 264, 1577-1579. USDA-SCS. 1986. ‘‘Soil Cave-in: A Fatal Slip.” Fact Sheet. USDA-SCS. 1994. “Global Soil Regions.” World Soil Resources, Soil Survey Division of the Soil Conservation Service, U.S. Dept. of Agriculture. Van Damme, H. 1995. Scale invariance and hydric behaviour of soils and clays. C. R . Acad. Sci. Paris 320, 665-681. Van Damme, H., and Ben Ohoud, M. 1990. From flow to fracture and fragmentation in colloidal media: 2. Local order and fragmentation geometry. In “Disorder and Fracture” (J. C. Charmet et a / . , Eds.), pp. 105-1 16. Plenum, New York. Veen, A. W. L., and Arndt, W. 1973. Huntite and aragonite nodules in a Vertisol near Katherine, Northern territory, Australia. Nature 241, 37-40. Veith, J. A., and Sposito, G. 1977. On the use of the Langmuir equation in the interpretation of “adsorption” phenomena. Soil Sci. SOC. Am. J . 41, 697-702. Wang, H. D., White, G. N., Turner, F. T., and Dixon, J. B. 1993. Ferrihydrite, lepidocrocite, and goethite in coatings from East Texas Vertic soils. Soil Sci. SOC.Am. J . 57, 1381-1386. Wild, M. R., Koppi, A. J., McKenzie, D. C., and McBratney, A. B. 1992. The effect of tillage and gypsum application on the macropore structure of an Australian Vertisol used for irrigated cotton. Soil Tillage Res. 22, 55-71. Wilding, L. P. 1985. Genesis of Vertisols. In “Proceedings of the 5th International Soil Classification Workshop, Sudan, 1982,” pp. 47-62. Soil Survey Administration, Khartoum, Sudan. Wilding, L. P., and Coulombe, C. E. 1996. Expansive soils: Distribution, morphology and genesis. In “Proc. NATO-ARW on Clay Swelling and Expansive Soils’’ (P. Baveye, and M. B. McBride, Eds.). Kluwer Academic, Dordrecht, Netherlands. Wilding, L. P., and Puentes, R. 1988. Vertisols: Their distribution, properties, classification and management. Tech. Mono. No. 18, Texas A&M Printing Center, College Station, TX. Wilding, L. P., and Tessier, D. 1988. Genesis and Vertisols: Shrink-swell phenomena. In “Vertisols: Their Distribution, Properties, Classification and Management” (L. P. Wilding and R. Puentes, Eds.), pp. 55-81. Tech. Mono. No. 18, Texas A&M Printing Center, College Station, TX. Wilding, L. P., and Williams, D. 1991. Reply to “Self-mulching as a classification criterion at the Subgroup or Family level.” Soil Sci. SOC. Am. J. 55, 1805-1806. Wilding, L. P., Williams, D., Miller, W., Cook, T., and Eswaran, H. 1991. Close interval spatial variability of Vertisols: A case study in Texas. In “Characterization, classification and utilization of cold Aridisols and Vertisols” (J. M. Kimble, Ed.), pp. 232-247. Proceedings of the Sixth Int. Soil Correlation Meet. (VI ISCOM), USDA-SCS, National Soil Survey Center, Lincoln, NE. Worrall, G. A. 1957. Features of some semi-arid soils in the district of Khartoum, Sudan. I. The high level dark clays. J. Soil Sci. 8, 193-206. Yaalon, D. H., and Kalmar, D. 1978. Dynamics of cracking and swelling clay soils: Displacement of skeletal grains, optimum depth of slickensides, and rate of intra-pedonic turbation. Earth Surface Processes 3 , 3 1-42. Yerima, B. P. K. 1983. Weathering and genesis of Volcanic ash-influenced Vertisols and vertic-like soils of El Salvador. M.Sc. Thesis, Texas A&M University, College Station, TX. Yerima, B. P. K. 1986. Soil genesis, phosphorus and micronutrients of selected Vertisols and associated Alfisols of Northern Cameroon. Ph. D. dissertation, Texas A&M University, College Station, TX. Yerima, B. P. K., Calhoun, F. G., Senkayi, A. L., and Dixon, J. B. 1985. Occurrence of interstratified kaolinite-smectite in El Salvador Vertisols. Soil Sci. SOC.Am. J. 49, 462-466. Yenma, B. P. K., Hossner, L. R., Wilding, L. P., and Calhoun, F. G. 1988. Forms of phosphorus and phosphorus sorption in northern Cameroon Vertisols and associated Alfisols. In “Vertisols:
OVERVIEW OF VERTISOLS
375
Their Distribution, Properties, Classification and Management” (L. P. Wilding and R. Puentes, Eds.), pp. 147-164. Tech. Mono. No. 18, Texas A&M Printing Center, College Station, TX. Yerima, B. P. K., Wilding, L. P., Calhoun, F. G., and Hallmark, C. T. 1987. Volcanic ashinfluenced Vertisols and associated Mollisols of El Salvador Vertisols: Physical, chemical and morphological properties. Soil Sci. SOC.Am. J. 51, 699-708. Yerima, B. P. K., Wilding, L. P., Hallmark, C. T., and Calhoun, F. G. 1989. Stastistical relationships among selected properties of northern Cameroon Vertisols and associated Alfisols. Soil Sci. SOC.Am. J. 53, 1758-1763. Young, I. M., and Crawford, J. W. 1991. The fractal structure of soil aggregates: Its measurement and interpretation. J. Soil Sci. 42, 187-192. Yousif, A. A., Mohamed, H.H. A,, and Ericsson, T. 1988. Clay and iron minerals in soils of the clay plains of Central Sudan. J. Soil Sci. 39, 539-548. Yule, D. F., and Ritchie, J. T. 1980a. Soil shrinkage relationships of Texas Vertisols: 1. Small cores. Soil Sci. SOC.Am. J . 44, 1285-1291. Yule, D. F., and Ritchie, J. T. 1980b. Soil shrinkage relationships of Texas Vertisols: 11. Large cores. Soil Sci. SOC. Am. J . 44, 1291-1295.
This Page Intentionally Left Blank
HYBRID RICE S. S. Virmani International Rice Research Institute, Manila, Philippines
I. Introduction
II. Heterosis in Rice A. Genetic Basis for Heterosis B. Prediction of Heterosis 111. Genetic Tools for Developing Rice Hybrids A. Cytoplasmic-Genetic Male Sterility B. Fertility Restoration C. Environment-Sensitive Genic Male Sterility D. Wide Compatibility Genes E. Apomixis Iv. Breeding Procedures for Developing Rice Hybrids A. Development of Parental Lines B. Seed Production of Experimental Hybrids C. Evaluation of Hybrid Combinations D. Biotechnological Applications V. Accomplishments A. In China B. Outside China VI. Agronomic Management VII. Disease/Insect Resistance WI. Grain Quality IX.Adaptability to Stress Environments X. Hybrid Seed Production A. Plant Characteristics Influencing Outcrossing in Rice B. Flowering Behavior in Relation to Outcrossing in Rice C. Floral Traits Influencing Outcrossing in Rice D. Natural Outcrossing Mechanism in Rice E. Guidelines for Hybrid Seed Production F. Practices for Hybrid Seed Production G. Performance of Hybrid Seed Production XI. Economic Analysis XII. Technology Transfer and Policy Issues XIII. Future Outlook XW. Conclusions References
377 Advmca in A p n q , f i l m r f?
Copyright &1 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
378
s. s. VIRMANI I. INTRODUCTION
The successful development and utilization of maize hybrids beginning about 1930 was a landmark in crop breeding. It provided impetus to plant breeders to explore commercial exploitation of hybrid vigor or heterosis phenomenon in other crops. Sorghum, pearl millet, cotton, sun flower, tomato, eggplants, chillies, onion, sugar beet, and rice are some examples of crops of which hybrids are now extensively used. Rice is the premier food crop in the world. It is cultivated in about 146 million ha which produce about 520 million tons of grains. It is the staple food providing about two-thirds of the calories for more than two billion people in humid and subhumid Asia and one-third of the calorie intake of the nearly one billion people in Africa and Latin America. Rice is also a major source of protein for the masses of Asia. Since the mid-sixties, when the release of semidwarf varieties triggered the green revolution, rice production has increased at 2.4%per year, faster than the rate of population growth. Achieving self-sufficiency in rice production and maintaining price stability are important political objectives in low-income countries because of the importance of this crop in providing national food security and generating employment and income for low-income people (Hossain, 1995). Most Asian countries have done remarkably well in meeting the food needs of the growing population over the last quarter century. However, the future poses even more challenging and ambitious tasks. The yield ceilings of rice varieties of the green revolution era must be lifted yet again because by 2020 the world must produce 350 million more tons of rice than it produced in 1992, to meet the demand created by increasing populations and rising incomes. This production increase must be achieved on less land, with less labor, less water, and less pesticides, and it must be sustainable (Anonymous, 1993). Using hybrid rice is one of the strategies to meet this immense challenge (Virmani and Dedolph, 1993). Although heterosis in rice has been known since 1926 (Jones, 1926), its commercial exploitation was demonstrated only when Chinese rice scientists developed and used commercial rice hybrids which increased rice varietal yields by about 20% over semidwarf varieties (Yuan, 1977; Lin and Yuan, 1980; Yuan et al., 1989, 1994). This yield advantage of rice hybrids was harnessed without any significant difference in input use (He et al., 1984, 1987a). In 1991 hybrid rices were cultivated in about 17 million ha in China which constituted 55% of its total rice area, and contributed 66% of total rice production of the country and 20% of the total rice production of the world. Developments in China encouraged the International Rice Research Institute and several national rice research programs during the 1980s to initiate research on exploring potentials and problems of hybrid rice technology (Virmani er al.,
HYBRID RICE
3 79
1981, 1982; Luat et al., 1985; Jachuck etal., 1988; Suprihatno, 1986; Chitrakon et al., 1986; Heu et al., 1984; Yuan and Virmani, 1988). Virmani and Edwards (1983) reviewed the status and prospects for breeding hybrid rice and concluded that exploitation of heterosis through hybrid breeding offered an important option to increase varietal yields in this self-pollinated crop. During the past 12 years, considerable progress has been made in the development and use of this technology both in and outside China. This chapter reviews work on various aspects of hybrid rice research since 1983 and provides a future outlook on this subject.
11. HETEROSIS IN RICE Kim and Rutger (1988) and Virmani (1994a) reviewed literature on heterosis in rice until 1990 and reported the presence of significant standard heterosis for yield and yield components. They also reviewed heterosis studies for several other traits, viz. plant height, days to flower, dry matter production, harvest index, root characteristics, photosynthesis, respiration, net assimilation rate, leaf area index, tolerance to temperature, and other stresses and biochemical traits. Generally speaking, commercial rice hybrids were reported to show 20-30% standard heterosis for grain yield (Lin and Yuan, 1980; Yuan et al., 1989). A farm-level study (He et al., 1987a) showed a 15-16% yield advantage of hybrid rices over inbred rices conducted in the Jiangsu province of China. Comparison of average yield of hybrid rice and conventional rice varieties grown in the country during 1981-1990 (Yuan et al., 1994) showed a 29 to 45% yield advantage of hybrids. The above comparisons are sometimes reported to measure standard heterosis. It should not be accepted as such because these results are based on secondary data on production and yield of hybrids and commercial inbred varieties grown in a given county, perfecture, or the whole country which also include the differences caused due to (i) differential input use and (ii) differential hybrid vs inbred X environment interaction. It is, therefore, important to compare the hybrids and inbreds in carefully laid out trials at experiment stations in order to get realistic estimates of standard heterosis. Yield heterosis in rice hybrids evaluated along with inbreds in the experiment station trials conducted in Hunan Provincial Yield Trials, China, during 1979- 1981 are summarized in Table I. Standard heterosis values ranged from 11.1 to 20.4%. More recent results from China are not available because Chinese hybrid rice breeders now compare new experimental rice hybrids with old rice hybrids and not inbreds. Outside China, significant standard heterosis for yield in rice has been observed in trials conducted at IRRI (Table 11) and in several national programs, viz. Philippines, India, U.S.A., Korea, and Egypt (Table 111). At IRRI
s. s. VIRMANI
3 80
Table I Extent of Standard Heterosis for Yield Observed in Multilocation Yield Trials Conducted in Hunan Rovince, China, during 1979 and 1981 Second Crop Season (Ma0 Chang Xiang, Personal Communication) ~~
Hybrid
~
Yield (tlha)
1979 Trials (27 locations) V20AlIR26 6.5 FGAIGui 630 6.4 v201s 6.3 TYZA/Gui 154 6.1 6.0 V41lML XI 5.4 Dong Ting Wan Xian (ck) 1981 Trial (14 locations) V20AlIR26 6.2 FG A I IR26 6.1 Yuchi 231-8 (ck) 5.2
Standard heterosis (S) 20.4* 18.5*
16.7* 12.9a 11.1a 19.2* 17.3*
* Values statistically significant. a
Nonsignificant.
the best rice hybrids outyielded the best inbred varieties by about 17%. Significant heterobeltiosis and standard heterosis for yield was also observed outside the
IRRI. Heterotic rice hybrids were observed to possess varying growth durations ranging from 105 to 136 days (Virmani, 1987), indicating that growth duration did not correlate with expression of heterosis. Yield advantage of a hybrid over inbred check was higher at high-yielding environments than low yielding environments (Fig. l), although percentage heterosis may reduce due to the higher mean yield of the check varieties at higher yielding environments. Kim and Rutger (1988) also reported higher heterobeltiosis for grain yield at lower N levels than at higher N levels. Increased yield in rice hybrids has been attributed to their increased dry matter, resulting from higher leaf area index (LAI) and higher crop growth rate, and increased harvest index, resulting from their increased spikelet number and to some extent, increased grain weight (Ponnuthurai et al., 1984; Akita et al., 1986; Blanco et al., 1986; Agata, 1990). Heterosis in panicles per plant, spikelets per panicle, and spikelet fertility varied highly among crosses and cultivation conditions due to yield component compensation (Akita, 1988). Kabaki (1993) analyzed growth and yield of japonica/indica hybrid rice in Japan and concluded that the expansion of leaf area due to increased number of tillers was the main
Table I1 Comparison of Highest Yielding F, Rice Hybrids and Inbreds in Hybrid Rice Yield Trials Conducted at IRRI during 1986-1995
Season
1986 DS 1986 WS
Trial
Hybrid
Yield (tlha)
I I1 I I1
IR54754A/IRR6R IR54754A/ARC 11353R IR54752AIIR64 IR 19728A/IR25 167-9-2 IR46830AIIRSOR IR46830AIIR29723- 143-3-2-1R IR54755A/IR2797- 125-3-3-2R IR54752AIARC 1135313 IR46830A/IR9761-19-IR IR54752A/IR64R IR54752A/IR15324-13-3-3-2R IR54752A/IR 13146-45-2-3 IR46830A/IR9761-19-IR IR54752A/IR9761-19- IR IR54752A/IR28228-119-2-3-1-lR IR54752A/IR42686-C2- I 18-6-2R IR54752A/IR54742-22-19-3R IR62829AIIR9761- 19-1R IR62829A/IR3 1805-20-1-3 IR58025A/IR 30198-66-213 IR58025A/IR29723- 143-3-2-1R IR58025A/IR54752-22- 19-3R IR58025A/IR40750-82-2-2-3R IR62829AIIR35366-62- 1-2-2-3 IR58025A/IR54745-2-45-3-2-4R IR58025A/IR19058- 107- 1R IR62829A/IR473 10-94-4-3-1R IR62829AIIR54883-43- 1-3 IR62829AIIR46R IR58025A/IR46R IR64608A/IR42686-C2- 1 18-6-2 IR58025AIIR48725-B-B-141-2 IR58025A/IR377 12-90-3-3-3-2 IR58025AIIR54056-64-2-2-2 IR58025A/IR50404-57-2-2-3 IR58025AIIR34686- 179-1-2-IR IR58025AIBG915 IR58025A/IR51078-33 IR58025A/IR42221-14R PMS 8A/IR29723-143R
7.4 7.9 3.9 3.6 4.1 6.4 7.8 6.8 4.8 5.3 5.9 6.8 3.2 6.3 6.5 3.1 3.5 4.0 4.5 4.8 5.6 3.0 3.2 4.7 5.4 6.4 5.I 6.4 7.2 6.3 3.7 4.3 4.1 4.4 6.4 7.4 4.0 4.3 7.3 7.5
111
1987 DS
I
1988 DS
11 111 11 IV V
VI 1988 WS 1989 DS
1
1989 WS
I1 I 11
1990 DS
I I1 111
IV 1990 WS 1991 DS 1991 WS 1992 DS
I I1 I I1 I I1 I I1 111
1992 WS
I I1 111
1993 DS 1993 WS 1994 DS
IV I 11 I I1 I I1
Difference from the best check
1.2* 2.3* 1 .o*
0.6* 0.7* 1.8* 1.8* I .o* 0.0 0.9* 0.8 (ns) 1.1*
I .o* I .5* 1 .o*
0.4 (ns) 0.8* 0.6 (ns) -0.6 (ns) 1.5 (ns) I .o* 0.0 0.2 (ns) 0.7* I .2* I .2* 1.1*
0.5 (ns) 0.7 (ns) 0.8* -0.6 (ns) 0.2 (ns) 0.4 (ns) 0.7* 0.7 (ns) 0.8* 0.8* -0.7 (ns) 0.6 (ns) 0.5 (ns)
% of
check
119 I42 I34 120 I20 139 130 117 I00 120 116 119 145 131 119 114 131 117 88 112 121
Growth duration (days)
126 133 126 122 110 I12 130 128 108 120 122 116 I05 111
124 120 136 110
107 I18 128 123 128 I08
116 116 128 135 126 112 122 113 120 I12
111
115
114 86
117
105 111
116 119 I26
100
119 112 112 125 86 109 107
111
110
128 105
114 111 126 (continues)
s. s. VlRMANI
3 82
Table II (continued)
Season 1994 WS 1995 DS Mean DS
ws
Grand mean
Trial I I1 I I1
Hybrid IR58025A/IR59606-119-3 IR68275A/IR46R IR58025A/IR58103-62-3 IR58025A/RP633-76-1R
Yield (t/ha)
Difference from the best check
check
5.8 6.6 9.5 9.6
1 .O* 1.6* 1.6* 1.6*
121 132 120 120
6.5 4.2 5.6
0.95 0.58 0.80
117 117 117
% of
Growth duration (days) 105 122 114 121
Note. DS, dry season; WS, wet season; ns, nonsignificant. * Significant at least at 5% level using LSD test.
factor for achieving heterosis in crop growth rate of hybrid rice during the 30 day period after transplanting. Though the degree of heterosis decreased thereafter, it rose again after heading, contributing to higher grain yield. Kabaki also concluded that hybrid rice displayed an efficient sink formation in terms of unit dry matter production as well as high potential of ripening due to vigorous dry matter production after heading. Thus high yield of hybrid was attained by the summation of the increase in each of the yield components. Yamauchi (1994) also observed that heterosis for dry matter accumulation was high at the vegetative and reproductive stages, lower at heading and partially restored at maturity. Dry matter accumulation was increased by larger leaf area in the vegetative stage; however, at the reproductive and ripening stages, it was controlled by net assimilation rate instead of leaf area. The role of photosynthetic activity in heterosis of hybrid rice has been examined by some researchers and both positive (Maruyamaet al., 1982) and negative (Kabaki er af., 1976; Kabaki, 1993; Yamauchi and Yoshida, 1985) results have been reported. Deng (1988) reviewed studies on the biochemical bases of heterosis in rice and reported differences between hybrids and their parents with regard to amylase and alpha-amylase activity in germinating seeds, RNA content of young roots, glycolic acid oxidase (related to photorespiration), and catalase and peroxidase enzymes.
A. GENETIC BASISFOR HETEROSIS An understanding of the genetic basis of heterosis is helpful to determine whether this phenomenon is fixable and/or predictable. Heterosis can result from
Table IIl Heterosis for Yield Observed in Promising Rice Hybrids in Trials Conducted in Some Countries during 1991-1994
Country
No. of crosses studied
Philippines
15
India India
8 6
Pakistan
6
U.S.A.
32
India
21
Korea
10
HeterosiP Trait
SH
Dry matter Harvest index Grain yield Grain yield Productive tiller per hill Panicle length Grains per panicle 100 grain weight Grain yield Panicles per plant Panicle length Grains per panicle 100 grain weight Grain yield Panicles per plant Panicle weight Grain yield Panicles per plant Panicle length Grains per panicle 100 grain weight Panicles per plant Panicle length Spikeletslpanicle
-40** to 58** -36** to 15 -59** to 34**
HB
Peng et al. (1991)
-41** to 69** -42** to 27** -lo** to 11** -26** to 9* -7** to 3 -31 ** to 46** -17** to 6 -1.3 (ns) to 18** -20** to 1 I * 0 to 13** -16 (ns) to 35** -23 to 14 -31 to 50 - 16 to 68 -19** -21** -59** -8** -29 -4 -20
to 31**
Reference
Reddy et al. (1991) Vivekanandan et al. (1992)
Mirza er al. (1993)
Gravois et al. (1993)b
Sreedhar et al. (1993)
to 16** to ll** to 1 1 * * to 36
Cho et al. (1994)b
to 8 to 6 (continues)
Table III (continued)
Country
No. of crosses studied
Korea
7
India
12
India
6
Egypt Egypt Egypt
64 198 85
HeterosiP Trait
SH
loo0 grain weight Milled rice yield Panicles per plant Panicle length SpikeIets/panicle Milled rice yield Productive tiller per hill Panicle length Grains per panicle 100 grain weight Grain yield Panicles per plant Grain per panicle 100 grain weight Grain yield Grain yield Grain yield Grain yield
-22 to 13 -29 to 20 -8 to 31 -5 to 30 53 to 135 18 to 70 -4** to 32** -2** to 14** -15** to 13** -24** to 2 6** to 65** - 16** to 34** -17** to 28** -13** to 30** 22** to 8 (ns) -78 to 99 -180 to 213 -81 to 51
* Significant at 5% level. ** Significant at 1% level. a HB, heterobeltiosis; SH, standard heterosis.
Authors did not report statistical significance of difference in the publication.
HB
Reference
Cho er al. (1994)b
Dhanakdi et al. (1994)
Geetha et al. (1994)
Mohammed et al. (1994)b Mohammed er al. ( 1994)b Mohammed et al. ( 1994)b
HYBRID RICE
385
Yield (t/ha) of test cultivars
1'1
A 'Re4616 OIR72
I
+ 1.088 xi (r= 0.866**) Y= 0.815 + 0.878 XI (r= 0.843**) Y= 0.335
L
2
3
4
5
6
7
8
Mean yield (t/ha) of all varieties at each location
Figure I Regression of yield of hybrid: IR64616H and inbred check (IR72) on location mean yield in National Coordinated Trials of irrigated rice cultivars in the Philippines. (Source: The Rice Varietal Improvement Group of the Philippines Seedboard, PhilRice, 1990 WS, 1991 DS, 1991 WS, 1992 DS, and 1992 WS).
partial to complete dominance, overdominance, epistasis, and a combination of these factors (Comstock and Robinson, 1952). If partial to complete dominance predominates it is theoretically possible to develop homozygotes with performance equal to or superior to hybrids. If overdominance or an overdominant type of epistasis predominates, then the highest-yielding lines must be heterozygotes (Sprague and Eberhart, 1977). Several biometrical models are used to study the genetic basis of heterosis. These include combining the ability test (Sprague and Tatum, 1942), the three- and six-parameter models (Jinks and Jones, 1958), the Joint Scaling Test (Mather and Jinks, 1971), Diallel cross mating (Hayman, 1954; Griffing, 1956a,b), and triple test cross analysis (Kearsey and Jinks, 1968). Critical studies on gene action on yield and yield components in rice are very few and most suffer from small population sizes, wider spacings than normal, and limited evaluations at one location or for one year (Kim and Rutger, 1988). Studies on combining ability in relation to heterosis have been reviewed by Kim and Rutger (1988) and Virmani (1994a). Most of these studies used the diallel mating design of Griffing (1956a,b) with or without a reciprocal. With a few exceptions all the studies showed significant GCA and SCA effects for yield indicating that both additive and nonadditive gene action were important in the inheritance of this trait. Relative proportions of GCA and SCA variances were
386
s. s. VIRMANI
found to vary in different studies, Many researchers also studied the combining ability of yield components and several other agronomic traits. These traits are also controlled by genes showing additive as well as nonadditive gene effects. As anticipated yield components usually showed a preponderence of GCA variances compared to SCA variances. Combining ability studies conducted at IRRI (Young, 1987; Peng and Virmani, 1990) have indicated that GCA effects of parents and SCA effects of hybrids for yield were highly significant. SCA variances were higher than GCA for yield over six environments and the latter showed a strong influence on GCA, implying that these parameters should be estimated over different environments before a generalization is made. Jinks (1983) stated that evidence of genuine over- dominance for quantitative traits does not exist, although apparent over-dominance due to nonallelic interaction and linkage disequilibrium is a common contributor to heterosis. Pooni and Treharne (1994), while supporting this viewpoint, stated that although modest levels of heterosis could be observed in most genetic situations, its exploitable magnitude depended very much on the presence of dispersion and directional dominance. Epistasis and interactions between the background and segregating loci may supplement this heterosis on many occasions. But as secondary sources of genetic variation they are not likely to be the sole cause of hybrid vigor, at least in the majority of the cases where the level of expressed heterosis is relatively high (above 30%). Since heterosis is essentially the expression of genes determining the size or rate of physiological processes, the acceptable explanation for a genetic basis of heterosis is fully dependent upon satisfactory elucidation of quantitative inheritance. Although biometrical analysis gives some light on the relative importance of dominance or overdominance, such analyses are not sufficiently powerful to identify the role of overdominanceif it exists only for a reduced part of the set of loci involved in the quantitative characters of interest (Gallais, 1988). Hauler and Miranda (1981) also pointed out that definitive proof for either of the hypotheses proposed for the genetic basis of heterosis will probably be difficult to establish. Because of the complexity of inheritance of quantitative traits such as yield, all types of gene interaction, both inter- and intraallelic, are probably involved. Gallais ( 1988) observed that heterozygosity for regulatory systems might be more significant for the explanation of heterosis. Heterozygosity for regulatory genes must lead to a greater homoestasis in a variable environment (internal or external), so heterosis will be a result of genotype X environment interaction and such a mechanism will be a fundamental property of the diploid level. He further stated that from a breeding viewpoint, in the long term, the best types of varieties could always be a single cross hybrid, but in the short term, according to a particular situation, it could be better to develop lines or hybrids. The homeostasis associated with hybrids must also be considered. Gallais (1988) further stated that in both auto- and allogamous plants
387
HYBRID RICE
it is difficult to exclude a role of ‘marginal’ overdominance which could justify hybrid variety production. The author proposes the following conceptual model for the manifestation of a quantitative trait (such as yield) in inbreds and hybrids.
+ + +
YH
+
Y 1 = p i ia ( i x e ) e + (i + h) + ha + ( h X e ) + e ,
= p,
where: p = general mean, i = inbred effect, h = heterosis effect, ha = effect of adaptability of hybrid, ia = effect of adaptability of inbred, h x e = hybrid x environment interaction, i X e = inbred X environment interaction, e = environment effect. In order to calculate heterosis, hybrids are evaluated with inbreds and differences in performance of the two are designated as a heterosis effect, i.e., YH YI = h. This would be true if ha = ia and (h x e) = ( i x e ) . But in reality ha and h x e may be unequal to ia and i X e, respectively. In such circumstances heterosis effects will be biased upward or downward depending on the positive or negative value of (ha - ia) and ( h X e) - ( i X e). Since it is difficult to separate the effect of ha from (i + h) and ia from i , a reliable estimate of the heterosis effect should be based on multilocation trials, data from which are analyzed using G x E analysis procedures. Since heterosis is a phenomenon of superior growth, development, differentiation, and maturation caused by the interaction of genes, metabolism, and environment, a simple explanation of heterosis based solely on the nuclear genome heterozygosity appears untenable (Srivastava, 1983). Bergounioux-Bunesset et a f . (1982) provided evidence for the effect of cytoplasm on heterosis in Petunia hybrida. The effect of cytoplasm on agronomic traits was reported in corn (Hunter and Gamble, 1968; Kalsy and Sharma, 1972; Flemming, 1975) and sorghum (Lenz and Atkins, 1981). In rice, several studies (Maruyama et al., 1985; Sasahara et a f . , 1986; Chen et al., 1987; Young and Virmani, 1990c) demonstrated the positive and negative effects of cytoplasm on agronomic as well as physiological traits. Chen er al. (1 987) observed that male sterility-inducing cytoplasms (CMS-WA and CMS-Gam) varied in their effects depending on the nuclear background of the hybrids, and some of the effects were considered nucleocytolasmic interactions, especially for yield, number of spikelets, number of filled grains, plant height, and panicle exsertion. Young and Virmani (1990~) observed that cytoplasmic effects were higher in heterosis for days to flower than in heterosis for yield and plant height. Wang and Wen (1995) studied the effect of sterility inducing cytoplasm using seven isogenic-allo-plasmic male sterile lines and their maintainer, Lu-Hongzao lB, and found significant positive effects on spikelets/panicle (CMS-K and CMS-WA cytoplasm) and significant negative effects on grain yield/panicle (CMS-L, CMS-J, and CMS-WA cytoplasm), seed
388
s. s. VIRMANI
set percentage (CMS-L, CMS-J, CMS-Y, CMS-S, CMS-D, CMS-K and CMSWA cytoplasm), 1000-grain weight (CMS-L, CMS-J, CMS-Y, CMS-S, and CMS-WA cytoplasm), and spikelets per panicle (CMS-J cytoplasm). From the foregoing review it is apparent that the genetic basis of heterosis in rice has not been clearly understood as in several other crops. Genetic and combining ability studies for yield and yield components indicate the role of both additive and nonadditive gene effects, the latter including the role of dominance and epistasis. Significant inbreeding depression has been reported by Ponnuthurai and Virmani (1985) and Reddy ef al. (1991). Attempts to develop double haploid (DH) inbred lines through anther culture of heterotic rice hybrids, yielding as high as the F, hybrid, were not successful and only up to 90% of heterosis was recovered in the DH lines (Siddiq ef al., 1994). Effects of cytoplasm on heterosis for yield, yield components, and several other agronomic traits are also apparent along with cytoplasmic-nuclear interaction. Considering the facts that 15-20% standard heterosis for yield is widely observed, hybrid seed production technology is available, and the pedigree breeding approach practiced during the past three decades has not helped in significantly increasing rice yields, one can say that the investment in hybrid breeding in rice is justified both genetically and economically.
B. PREDICTION OF HETEROSIS Performance of a hybrid depends on the choice of its parents. The most important challenge to hybrid variety breeders is to choose the parental lines which would result in heterotic combinations without necessarily making all possible crosses among the potential parents. Several methods tried with variable success include per se performance, combining ability, and mitochondria1complementation (see Virmani, 1994). In a hybrid rice breeding program in China several internationally known commercial rice varieties (viz. IR24, IR26, and Milyang 46 from Korea) and elite breeding lines (viz. IR661, IR9761-19-1R, etc.) have resulted in successful commercial rice hybrids (Lin and Yuan, 1980; Coffman and Virmani, 1983; Yuan et al., 1985). In trials conducted at IRRI, several commercial rice varieties, viz. IR42, IR46, IR50, and IR54 (Virmani et al., 1981, 1982) resulted in heterotic rice hybrids. The hybrids UTL 1 and UTL 2, released in Vietnam (Luat et al., 1995), and Rc 26H or Magat, released in the Philippines, involve the male parent IR29723-143-3-2-1, an elite IRRI line which has been released as a commercial variety in Vietnam. Per se performance is therefore an important attribute for selecting parental lines for developing heterotic hybrids in rice. Genetically diverse parents, within limit, are more likely to give heterotic
HYBRID RICE
3 89
hybrids than those genetically related. Therefore, the presence or absence of heterosis can be predicted on the basis of genetic diversity among the prospective parents. Genetic diversity can be related to geographic origin of parental lines. The geographical variation can be related to ecological and environmental variations which, in turn, dictate survival fitness, created by spontaneous and induced genetic variation in natural and directed-selection situations. Consequently, the parental lines derived from different geographic origins are considered to have more genetic diversity than those derived from the same geographic origin. During the past three decades, however, internationalization of plant breeding efforts and massive exchange of unimproved and improved germplasm throughout the world have altered the situation, and therefore differences in geographic origin of the parental lines may not always reflect genetic diversity among them. On the other hand, extensive hybridization practiced in several international and national crop breeding programs around the world has created vast genetic diversity among the lines developed under given geographical conditions and one can expect genetic diversity among parents from the same geographical origin. The presence of significant heterosis in crosses from IRRI-bred rice cultivars (Virmani et al., 1982) and the presence of a large amount of diversity within IRRI elite lines (Julfiquar et al., 1985) provide testimony to the above hypothesis. Mahalanobis generalized distance or D2 statistics (Mahalanobis, 1936) based on multivariate analysis has also been used to estimate genetic diversity by classifying prospective parents of the hybrids into various genetically diverse groups or clusters. The parental lines belonging to different and distantly located clusters have a higher probability of giving heterotic hybrids than those parental lines belonging to the same cluster group. This technique has been deployed in rice by Julfiquar et al. (1985), Vaidyanath and Reddy (1985), and Li and Ang (1988) to classify rice cultivars. Li and Ang (1988) also observed that characteristics of panicle and grain and growth duration played a major role in estimating genetic distance among parents. A highly significant correlation ( r = 0.55 to 0.76) between the magnitude of genetic distance values and the strength of heterosis in japonica rice helped them to predict heterotic crosses with an 80% success rate. However, at IRRI Peng et al. (1991) did not find such a relationship to enable prediction of heterosis in a hybrid rice breeding program. It is possible that D*estimates determined at IRRI (Julfiquar et al., 1985) were biased because of the inclusion of yield, a complex characteristic, among divergence. In another study (IRRI, 1994) hybrids derived from indica/tropical japonica crosses showed higher yield than hybrids derived from indica/indica and tropical japonica/tropical japonica crosses (Table IV). Tropical japonicas or javanica rices are japonicas adapted to tropical conditions and are genetically diverse from indica rices (Glaszmann, 1987; Khush and Aquino, 1994). On the other hand, hybrids derived from indica/temperate japonica crosses have shown lower heterosis in the tropics than indica/indica crosses (Virmani et al., 1991, 1994). This
s. s. VIRMANI
3 90
Table IV Total Biomass, Grain Yield, Harvest Index (HI), and 1000 Grain Weight of IntervarietalGroup and Intravarietal-GroupHybrids and Their Inbreds Evaluated at IRRI, 1993 WS, and 1994 DS ( s h a d , Virmani, and Khush, unpublished) Total biomass Group TJlI I/I TJlTJ I TJ TJ/I I/I TJlTJ I TJ
Number
(ah2)
Grains Wm2)
1993 WS (spacing 20 X 10 cm) 1816a 890a 1540b 710b 1489b 643c 1418b 603c 11 16c 412d 1994 DS (spacing 15 X 10 cm) 2047a 1030a 1834b 894b 1724bc 822c 1651c 726d 1453d 566e
HI
lo00 grain wt. (g)
0.49a 0.46b 0.43~ 0.42~ 0.37d
3 I .Ob 28.0~ 32.6a 26.3d 28.8~
0.50a 0.48b 0.48b
28. la 27.0b 27.6b 24.4~ 25.0~
0.44C
0.39d
Nore. TJ, tropical japonica; I, Indica.
observation implies that genetic diversity between parents alone is not adequate, and that it should be combined with adaptability and per se performance of parental lines to result in heterotic hybrids. During the past two decades biotechnological tools, viz. isozyme and restriction fragment length polymorphism (RFLP) and random amplified polymorphic DNA (RAPD) markers, have also been developed and deployed to estimate genetic diversity among rice varieties (Chu, 1967; Shahi et al., 1969; Pai et al., 1975; Fu and Pai, 1979; Second, 1982; Glaszmann, 1987; Mackill, 1995). Their use in selecting parents for developing heterotic rice hybrids has been reported separately in this review in the section on biotechnological applications. Combining ability analysis, deployed usually to study the genetic basis of heterosis, is also a powerful tool to test the value of parental lines to produce superior hybrids. Estimation of general combining ability (GCA) effects helps in selecting lines with high, average, and low combining ability. Generally speaking, choice of parental lines with high GCA effects increases the probability of getting heterotic hybrids in crop plants. A successful hybrid breeding program routinely evaluates parental lines for their GCA effects. In rice, several reports have been published from time to time on the combining ability of rice varieties (Virmani and Edwards, 1983; Virmani, 1994a). More recent reports are summarized in Table V. Some CMS and restorer lines showing high GCA effects at IRRI
Table V Summary of Reports Published on Combining Ability Analysis for Yield in Rice during 1989-1994 Combining ability effects
No. of crosses
+ parents studied 72 + 9 21
72
+7 +
16
42
+7
28 15 75
+8 +8 + 18
Mating design
9 x 9D 7 x 7a 16 X 16a 7 x -ID 8 X 8a 5x3LT 7 X I 1 LT
Variance due to ~
~
GCA
SCA
** **
** ** ** **
**
** ** ** **
**
GCA
6813.39b 0.0307 12.905b
? 63251.8b
SCA
SCAIGCA ratio
1658.36b 0.0717 130.17 ? 43319b
**
?
?
**
0.0046
0.0045
Note. D, diallel mating; LT, line X tester; **significant at 1% level; ns, nonsignificant. Without reciprocal. Value was computed from the data given in the publication.
1.26
1.o
Reference Heu et al. (1991) Lokaprakash et al. (1991) Sarawgi er al. (1991) Murthy and Shrivashankar ( 1992) Gravois (1994) Ramalingam er al. (1993) Peng and Virmani (1990)
392
s. s. VIRMANI
are listed in Table VI. Experience at IRRI indicated that crosses showing high heterosis were more frequently obtained when at least one of the parents has high GCA effects (Peng and Virmani, 1990), although occasionally heterotic combinations were derived from parents both having low GCA effects (Kumar and Saini, 1981). Srivastava and Seshu (1983) reported that sometimes parents showing high GCA did not result in heterotic combinations. These results indicated that prediction of heterosis on the basis of general combining ability may not always be accurate. Mitochondria1 complementation was considered as a possible tool to predict combining ability of parental lines selected for hybrid breeding (McDaniel and Sarkissan, 1966; Sarkissan and Srivastava, 1967; Srivastava, 1983). The attractiveness of the technique encouraged several laboratories to study this phenomenon in relation to prediction of heterosis. Contradictory results were reported (Sarkissan and Srivastava, 1969; Ellis et al., 1973; Sage and Hobson, 1973; Flavell, 1977). Riley et al. (1979) concluded that results from the mitochondria1 complementation technique were not reproducible at all. This technique has not been tested in rice. From the above results it is apparent that none of the methods for predicting heterosis are foolproof. However, hybrid rice breeders in China, at IRRI, and elsewhere have been generally relying on per se performance, combining ability, and genetic diversity of parental lines to develop heterotic combinations.
Table VI List of Some IRRI-Bred CMS and Restorer Lines Showing High General Combining Ability in Combining Ability Trials Conducted at IRRI during 1990-1995
CMS line
Restorer line
IR58025A IR68275A lR6828 1A IR68273A IR68888A IR68891A IR68893A IR68895A IR68897A IR68898A IR68899A IR68900A IR68902A
IR46R IR29273-3-2-1R IR34686-179 IR21567- 18-3R IR32809-26-3-3R 1154742-22-19-3R IR20933-68-21- 1-2R
HYBRID RICE
393
111. GENETIC TOOLS FOR DEVELOPING RICE HYBRIDS Rice, being a self-pollinated crop, must involve use of an effective male sterility system to develop and produce F, hybrids. The most popular male sterility system used for hybrid rice breeding in China and elsewhere is cytoplasmic-genetic male sterility (CMS), although environment-sensitive genetic male sterility systems are now being tested for their usability also. In order to exploit an enhanced level of heterosis in indica/japonica rice crosses, use of a wide compatibility gene was proposed by Ikehashi and Araki (1984). Since then wide compatibility genes have also become important genetic tools for breeding rice hybrids. Apomixis, asexual seed production, is the ultimate genetic tool for developing true breeding hybrids with permanently fixed heterosis. Serious research has been initiated in China and at IRRI to develop this tool. This section reviews the progress made in the development of the above mentioned genetic tools for hybrid rice breeding.
A. CYTOPLASMIC-GENETIC MALESTERILITY A cytoplasmic-genetic male sterility system, controlled by the interaction of cytoplasmic and nuclear genes, was discovered by Jones and Emsweller (1937) in onion (Aliurn cepa). The role of cytoplasm in causing male sterility in rice was first reported by Sampath and Mohanty (1954). Subsequently, a similar phenomenon was also observed by Katsuo and Mizushima (1958) and Kitamura (1962). Cytoplasmic male sterile lines in rice were developed by Sasashara and Katsuo (1965), Shinjyo and Omura (1966a,b), Erickson (1969), Watanabe (1971), Carnahan er al. (1972), Athwal and Virmani (1972), and Yabuno (1977). None of these researchers, however, deployed these CMS lines for developing hybrid rices. The first cytoplasmic male sterile line used to develop commercial F, rice hybrids was developed in China in 1973 from a male sterile plant occumng naturally in a population of wild rice (Oryza sariva f. sponranea) growing on the Hainan island in 1970 (Yuan, 1977). This plant was designated wild rice with aborted pollen (WA). Since then a number of CMS lines have been developed in China, at IRRI, and elsewhere from various wild and cultivated accessions (Lin and Yuan, 1980; Virmani et al., 1986; Li and Zhu, 1988; Virmani and Wan, 1988). Male sterility inducing cytoplasms were also identified in various geographic forms of Oryza perennis (Rutger and Shinjyo, 1980). The frequency of male sterile cytoplasms in Asian and American strains was about 64 and 4%, respectively. In China, Li and Zhu (1988) found 62 out of 300 strains of Oryza rufipogon possessing male sterility inducing cytoplasm. They also catalogued
3 94
s. s. VIRMANI
CMS sources identified in China and observed that most of these were found in 0. sativa f. spontanea and indica rice cultivars. Virmani and Edwards (1983) and Virmani and Wan (1988) listed some of the CMS sources identified in and outside China. These CMS sources are designated, in principle, according to the name of the cultivar from which the cytoplasmic factor inducing male sterility is derived. In some cases, different symbols have been assigned by different researchers for the same materials. For example, Shinjyo designated the cytoplasm of Chinsurah Boro I1 as [CMS-boro] or [CMS-bo] but the Chinese workers using the same materials designated it as BT (B for Chinsurah Boro I1 and T for Taichung 65, the nuclear donor source). Virmani and Shinjyo (1988), therefore, proposed an interim designation of various cytoplasmic sources known to induce male sterility in rice and proposed a model for identification of genetic differences among cytoplasms and restoring genes. Chen et al. (1995) analyzed the inheritance of WA, BT and HL CMS lines and confirmed that these were genetically different systems. During the past decade some new CMS sources in rice have been identified at IRRI. These include CMS-ARC (Virmani et al., 1985; Virmani and Dalmacio, 1987), 0. perennis (Dalmacio et al., 1995), Oryza glumepetala (IRRI, 1995), and IR62829B (IRRI, 1995; Virmani et al., unpublished). The CMS source, IR62829B, was induced by gamma irradiation of the maintainer line (IR62829B) of a WA CMS line, IR62829B, also bred at IRRI. In the M, generation of IR62829A two highly pollen-sterile plants were picked up; backcrossing to the unirradiated mother parent, IR62829B, maintained its sterility. After recurrent backcrossing a stable CMS line has been developed which has shown differential fertility restoration and maintenance behavior with some elite lines in comparison to WA-CMS lines. This line has been designated as IR68885A (IRRI, 1995; Virmani et al., unpublished data). In India, Pradhan et al. (1990) identified two new CMS sources in indica cultivars, viz. V20B (a maintainer of a commercially used WA CMS line, V20A, developed in China) and Sattari, in crosses with japonica rices. Most recently, an Oryza nivara accession has also been identified in India to be a source of CMS (Hoan and S. Siddique, personal communication). The new CMS sources identified in China during the last decade are given in Table VII. Zhang (1985) made the following conclusions from his studies on cytoplasmic male sterility in rice. 1. The occurrence of male sterility depended on affinity between cytoplasm and nucleus. The more distantly related the cytoplasm and substituted nucleus, the more frequently obtained the male sterile lines and their maintainers. 2. Different degrees of pollen abortion occurred depending upon the combination of cytoplasm and nucleus. In rice pollen abortion has been observed from the uninucleate stage before the first pollen mitosis, to the binucleate stage just
395
HYBRID RICE Table VII New CMS Sources Identified in China (Lin and Ming, 1991) CMS source
Derived from
Yian Chien red awned wild rice Liu Zhou awned wild rice Liu Zhou white awned wild rice He Pu wild rice Tain Dong wild rice No. 2 Ghunnia wil rice No. 5 Guanxi wild rice No. 37 Indian wild rice Hung county Yungbio wild rice Chao Yian No. I Seng Qi Nan Guang Zhan Gui Hua Huang Zhao Tong Beizigu Dong Lan Tai Dong awned Tai Dong Guznxi No. 16 Long An Guanxi No. 36-4
Wild X cultivated rice Wild X cultivated rice Wild X cultivated rice Wild X cultivated rice Wild X cultivated rice Wild X cultivated rice Wild x cultivated rice Wild X cultivated rice Wild X cultivated rice Spontaneous ms mutant Indica/japonica cross Indica/japonica cross Indica/japonica cross Indica/japonica cross Wild rice Wild rice Wild rice Wild rice Wild rice Wild rice
'
before anthesis. The earlier the stage of abortion, the more frequent the morphologically discernible pollen sterility. From the foregoing information, it is apparent that male sterility-inducing cytoplasmic factors are widely distributed in wild and cultivated rices. It is, therefore, feasible to develop CMS lines possessing diverse cytoplasmic and nuclear backgrounds. Despite this situation Zhou (1994) observed that most of the commercial rice hybrids in China are based on CMS-WA cytoplasm (87.9%) followed by CMS-Di (7.8%), CMS-DA (2.6%), CMS-IP (0.5%), and others, viz. CMS-bo, CMS-HL, and CMS-GA (1.2%). Since CMS-WA cytoplasm gives stable CMS lines for which a high frequency of restorers is conveniently found, the hybrid rice breeders tend to deploy this CMS system in hybrid rice breeding more frequently.
B. FERTILITY RESTORATION In order to deploy CMS systems to develop commercial rice hybrids, it is essential to have effective restorer lines. Fortunately, several fertility restorer
396
s. s. VIRMANJ
lines have been identified in rice. Shinjyo (1969) found that fertility restoration in the CMS-bo male sterile line was controlled by a single dominant gene (“Rf”) which had a gametophytic effect. This gene was present in many tropical indica rice cultivars (Shinjyo, 1972a,b). Rice cultivars from Japan and other temperate countries were mostly nonrestorer for this CMS system. For the CMS-WA system hundreds of effective restorer lines have been identified among cultivated rice varieties and elite breeding lines bred in China (Lin and Yuan, 1980; Yuan and Virmani, 1988; Yuan et al., 1994), at IRRI (IRRI, 1983a,b, 1984, 1986, 1988; Govinda Raj and Virmani, 1989; Virmani, 1994), in Indonesia (Suprihatno, 1986; Sutaryo, 1989; Suprihatno et al., 1994), India (Rangaswamy et al., 1987; Singh and Sinha, 1987; Saran and Mandal, 1988; Bijral et al., 1989; Tomar and Virrnani, 1990; Siddiq et al., 1994), the Philippines (Lara et al., 1994), Vietnam (Voc et al., 1990; Luat et al., 1994, 1995), Korea (Heu et al., 1994; Moon, 1988; Moon et al., 1994), Malaysia (Mohamad et al., 1987; Osman et al., 1988), Thailand (Amornsilpa et al., 1994), and Colombia (Munoz and Lasso, 1991; Munoz, 1994). Restorers have also been identified for the CMS-ARC system at IRRI (Virrnani et al., 1994), and CMS-Di, CMS-DA, CMS-IP, CMS-HL, and CMS-GA systems in China. However, no restorers have yet been identified for CMS-TN, CMS-MS 577, and CMS-0. perennis and CMS-glumipetula since these CMS systems have not been used for developing commercial rice hybrids. Frequency of restorer lines was higher among rice varieties originating in lower latitudes compared to those originating in higher latitudes. Also, restorer frequency was higher among indica rices compared to japonicas. In China, late maturing indica rices showed higher frequency of restorers than the early maturing indica rices (Yuan, 1985). However, outside China this correlation does not hold true because the elite lines have been bred by extensive hybridization among late and early rice cultivars (Yuan and Virmani, 1988). Li and Zhu (1988) observed that among the three ecotypic rice cultivars, viz. aman, aus, and boro (indica rices cultivated in the eastern region of the Indian subcontinent), aman and boro cultivars had a higher frequency of restorers as compared to aus cultivars. Similarly, among the two ecotypic varieties (viz., bulu and Tjereh from Java island of Indonesia) Tejereh cultivars showed higher frequency of effective restorers; bulu rices showed weak restoration. They further observed that in Asia, effective restorers were found mainly in South and Southeast Asian countries and Southern China, while nonrestorers were concentrated in Northern China and far-eastern Asia. Japonica rice hybrids commercialized in China and developed in Japan and the Democratic People’s Republic of Korea have been bred by breeding japonica restorers by incorporating in them restorer gene(s) from indica rices. Genetics of fertility restoration in rice has been studied for three CMS systems, viz. CMS-bo (Shinjyo, 1969; Shinjyo et al., 1974; Shinjyo, 1975; Hu and Li, 1985; Teng and Shen, 1994), CMS-D (Hu and Li, 1985), and CMS-WA
HYBRID RICE
397
(Wang, 1980; Gao, 1981; Zhou et al., 1983; Young and Virmani, 1984; Hu and Li, 1985; Fu, 1985; Li and Yuan, 1986; Virmani et al., 1986; Govinda Raj and Virmani, 1988; Singh and Sinha, 1988; Bharaj et al., 1991; Anandakumar and Subramaniam, 1992; Teng and Shen, 1994). The effect of restorer gene(s) for CMS-bo and CMS-D cytoplasm was gametophytic which caused partial pollen fertility, but normal spikelet fertility, in F, hybrids. On the other hand, the effect of restorer genes for CMS-WA cytoplasm was sporophytic which gave normal pollen and spikelet fertility in F, hybrids. The inheritance of fertility restoration in CMS-bo and CMS-D cytoplasm was monogenic and the two genes were allelic (Hu and Li, 1985). Teng and Shen (1994) reported fertility restoration for CMS-bo cytoplasm controlled by a dominant gene Rf-1 carried in restorer line C57 or by an incompletely dominant gene of ZH 157 which was different from Rf-1. Govinda Raj and Virmani(1988) found that mode of action of the two restorer genes for CMS-WA cytoplasm varied with the cross; certain crosses showed dominant epistasis, while others showed recessive epistasis or epistasis with incomplete dominance. Studies also showed that one of the two fertility restorer genes for CMS-WA cytoplasm was stronger in action than the other. Teng and Shen (1994) also confirmed that there were two restoring genes for WA type CMS, a dominant gene and an incompletely dominant gene. These restoring genes were different from Rf-1 gene of CMS-bo cytoplasm. Li and Yuan (1986) studied the distribution of two restorer genes (a restorer parent of several commercial rice hybrids in China) among its pedigree parental lines. One of the genes appeared to have been inherited from a late indica variety from China while another was from SLO 17, an Indian variety. Govinda Raj and Virmani ( 1988) conducted allelic tests for restorer gene(s) present in six restorer varieties, viz. IR26, IR36, IR54, IR9761-19-1, IR2797-105-2-2-3, and IR42, and identified four groups of restorers possessing different pairs of restorer genes. The existence of a large number of R genes explains this high frequency of R lines among the elite indica breeding lines for the CMS-WA cytosterility system. Govinda Raj and Virmani (1987) also provided evidence for the role of intervarietal hybrid sterility and/or inhibitory genes present in a CMS line of rice, in causing incomplete fertility restoration by some established restorer lines possessing CMS-WA cytoplasm. Shinjyo (1975) located the Rf gene for CMS-bo cytoplasm on chromosome C (using Dr. N . Iwata’s nomenclature) now designated as chromosome 10 using trisomic analysis. Yoshimura et al. (1982) located the Rf gene for CMS-bo cytoplasm on chromosome 7 using the translocation method. Shinjyo and Sato (1994) also located restorer gene F (Rf-2) for CMS-L cytoplasm on chromosome 2 using primary trisomic and linkage tester lines. Bharaj et al. (1995) located the two restorer genes of CMS-WA cytoplasm using trisomic analysis. The stronger restorer gene (Rf-WA-I) was located on chromosome 7 and the weaker restorer gene (Rf-WA-2) was located on chromosome 10. From the foregoing information it can be concluded that fertility restoration is
398
s. s. VIRMANI
not a constraint in developing commercial rice hybrids, at least in indica rices. The problem does exist in japonica rices which are mostly nonrestorers and therefore require restorer gene(s) to be incorporated into them from indica rice through a long breeding process. Consequently, genetic diversity among japonica restorers is limited, which results in weaker heterosis in japonica compared to indica rice hybrids. This makes breeding of japonica rice hybrids more difficult. That is why hybrid rice areas under japonica rice hybrids are very limited and other japonica rice-growing countries (viz. Japan and Korea) have not succeeded in developing commercially usable rice hybrids.
C. ENVIRONMENT-SENSITIVE GENICMALESTERILITY Although the CMS system has been found to be the most effective system for developing commercial rice hybrids it is cumbersome. Also, its use is restricted to those germplasms in which maintainers and restorers are abundant. Continuous use of a CMS system risks potential genetic vulnerability of the hybrids to a biological stress. Shi (1981, 1985; Shi and Deng, 1986) reported a novel genic male sterility in rice which was found to revert back to fertility under certain photoperiods. It was called photoperiod-sensitive genic male sterility (PGMS). Yang et al. (1989) induced a PGMS line (5460 PS) in indica rice IR54. Subsequently, temperature-sensitive genic male sterility (TGMS) was also discovered (Zhou et al., 1988, 1991; Wu et al., 1991; Virmani and Voc, 1991) which reverted to partial or full fertility under certain temperature regimes. Based on these systems, Yuan (1987) put forward a new strategy of hybrid rice breeding which did not involve a maintainer line, and hence was called the two-line method. Any fertile line could be used as pollen parent to develop a rice hybrid. Several PGMS and TGMS lines have been developed in China, Japan and the U.S.A., and by IRRI (Table VIII). Satyanarayana et al. (1995) also reported a new source of TGMS line in India among spontaneously occurring sterile plants from indica rice cultivar IET 10726, and Oard and Hu (1995) induced a PGMS mutant in rice cultivar M201 using the chemical mutagen ethyl methane sulfonate. PGMS lines Nongken 58s and X-88 (from Japan) are completely pollen sterile under long-day (above 13.75 h) conditions and show fertility reversion under short-day (daylength below 13.5 h) conditions. Similarly, TGMS lines show complete pollen sterility when the maximum day-temperature is above 2729°C and partial fertility when the maximum day temperature is lower. The critical temperature has been found to vary depending on the source of TGMS gene (Lu et al., 1994). One TGMS line, IVA, developed in the Yunan province in China, expresses sterility at low temperature (24°C) and fertility at higher temperature (27°C). Zhang et al. (1992b) observed that fertility alteration in some PGMS lines was regulated by both day-length and temperature regimes.
HYBRID RICE
3 99
Table VIII PGMS and TGMS Lines Developed in China, Japan, and the U.S.A. and at IRRI' (Lu el al., 1994) Line
Varietal group
Nongken 58 S
Japonica
Annong S
Indica
Hennong S 5460 s R59T S
Developed by
Developed at
Indica Indica Indica
Spontaneous mutation Spontaneous mutation Cross breeding Irradiation Irradiation
Hubei, China
Hunan, China Fujian, China Fujian, China
IR32364-20- 1-3-28
Indica
Irradiation
IRRI, Philippines
Norin PL 12
Japonica
Irradiation
Tsukuba, Japan
IVA
Indica
Cross breeding
Yunnan, China
Dianxin IA
Japonica
CMS
Yunnan, China
EGMSb
X88
Japonica Japonica
Cross breeding
U.S.A Japan
M201
Japonica
EMS treatment
Louisiana
Hunan, China
Fertility induction conditions Daylength lower than 13.75 h Temperature 27°C Temperature below 29°C Temperature below 29°C Temperature lower than 29°C Temperature lower than 29°C Temperature lower than 28°C Temperature lower than 24°C Temperature lower than 22°C Daylength lower than 13 h Daylength lower than 13.75 h Daylength 12 h
Nongken 58 S, EGMS, and X88 are E M S lines; the rest are TGMS lines. Environment-sensitive male sterile mutant.
When the temperature is above a certain point pollen becomes sterile, and when it is below another point, pollen becomes fertile, regardless of the photoperiod. These two temperature points are referred to as the critical temperature for fertility induction (upper limit) and the critical temperature for sterility induction (lower limit). The temperature span between the two critical temperature points is the temperature range in which pollen fertility alteration takes place: fertility decreases as the temperature rises, and it becomes higher as the temperature falls (Zhang et al., 1992b). The critical thermosensitive stage for fertility alteration in TGMS lines was 22 and 26 days before heading in Japan (Maruyama et al., 1991a) and 6-15 days after panicle initiation at IRRI (Borkakati, 1994). Even 1 h of high-temperature treatment during the critical stage could influence pollen fertility and thus affect seed fertility. Both PGMS (Wang et al., 1991a,b) and TGMS (Maruyama et al.,
s. s. VIRMANI
400
1991a; Borkakati and Virmani, 1993) are recessive traits controlled by a single gene. TGMS genes reported by Sun et al. (1989) and Maruyama ef al. (1991) have been designated as tms, and tms,, respectively (Kinoshita, 1992). Borkakati and Virmani (1995) found that the TGMS genes in mutant IR32364 TGMS developed at IRRI (Virmani and Voc, 1991) were nonallelic to tms,. The allelic relationship between IR32364 TGMS and tms, has not been studied yet due to the nonavailability of TGMS line 54608 possessing the tms, gene. Breeders in China have set a standard for the PGMS and TGMS lines to be used in two-line hybrid rice breeding. Besides high combining ability, high outcrossing potential and high disease and insect pest resistance, there is a system of fertility standards listed below based on photoperiod and temperature conditions in China (Lu et al., 1994). At least 1000 identical plants should be tested and the proportion of sterile plants should be 100%. Pollen sterility on sterile plants should be not less than 99.5%. The lines should have clearly defined fertility alteration conditions, and the continuous period for inducing complete sterility should last for at least 30 days in a year. Seed setting percentage should be more than 30% during fertility induction. The critical temperature for sterility induction should be 23°C or below for PGMS and 23.5"C for TGMS. Several PGMS and TGMS lines are currently being bred in China and Japan and by IRRI to develop two-line rice hybrids. This system gives a higher frequency of heterotic hybrids compared to the CMS system (Table IX) because there are less restrictions in the choice of parents. Unlike the CMS system, which requires restorers to restore fertility in hybrids, the TGMS system can be deployed using any rice cultivar as a pollen parent. The PGMS system is useful in temperate regions where striking day-length
Table IX Relative Frequency of Heterotic Hybrids Derived from TGMS and CMS Systems in Testcrosses Made at IRRI, 1993-1994 No. of crosses studied
Number and (% frequency) crosses
MS system
1993
1994
1993
1994
CMS TGMS
103 131
106 I15
17(16) 47(36)
64(31) 77(67)
HYBRID RICE
40 1
differences exist during rice growing season(s) whereas the TGMS system is more useful under tropical conditions where day-length differences are marginal and temperature differences between low and high altitude can be usefully exploited. Some two-line rice hybrids of japonica rice (such as N 5047S/R9-1, 5088/R9-1, 313013/1514, and 7001S/Lun hui 422) and indica rice (e.g., W 6154 S/Teqing, W 6154 S/312, W 6154 STesanai, K 9 9 0 3 or Jingguang 1, Pelai 64 S/Xiang Zaosiau 1, 5460 SIMingui 63 or Guangyou 6063, and 8902/Mingui 63) have been developed in China. They cover about 30,000 ha in the Yangtze River Valley and South China yielding 8.3-9.8 t/ha, 5-10% higher than the CMS-derived hybrid Shan You 63 and Shan You Gui 33 (Lu et al., 1994). At IRRI two-line rice hybrids derived from the TGMS system are still in the experimental stage. Further information on the current status of the two-line method of hybrid rice breeding can be found in Lu et al. (1994).
D. WIDECOMPATIBILITY GENES As discussed earlier, the level of heterosis for yield and/or any other agronomic trait in hybrids depends to some extent on genetic distance between parents. Most of the hybrids grown in China and elsewhere are based on indica rice germplasm which has adequate genetic diversity. Heterosis in japonica rice hybrids, developed in China, Japan, and DPR Korea, is relatively lower (usually less than 10%). In order to enhance the level of yield, heterosis in rice use of indicaljaponica crosses was proposed by Chinese and Japanese rice scientists (Anonymous, 1988; Maruyama, 1988; Ikehashi, 1991). Experiments conducted at IRRI (IRRI, 1989) also indicated enhanced level of heterosis in crosses involving semi-improved bulu rices (now designated as tropical japonicas) and indica rices. However, hybrids between indica and japonica rices show a variable degree of hybrid sterility. Ikehashi and Araki (1 984) discovered a genetic tool, designated as wide compatibility gene(s), to overcome this hybrid sterility problem. They showed that gamete abortion by an allelic interaction at a locus (designated as S,) caused hybrid sterility in S5i-S4 but not in S,n-S5i or S,n-S4 (S5irepresenting indica, S4 japonica, and S5n a neutral allele). Thus incorporation of the S," allele into one of the parents overcame sterility in the hybrids derived from it. Donors of a wide-compatibility allele (S,") were termed a widecompatibility variety (WCV). Ikehashi and Araki (1984) screened 74 rice varieties and identified some WCVs which originated from Indonesia or Bengal. Subsequently, the WC (S,n) locus was closely linked with marker genes C (chromogen for pigmentation) and wx (waxy endosperm) which are located in chromosome 6 (Ikehashi and Araki, 1986, 1987). Several WCVs have been identified in China (Luo et al., 1990) and at IRRI (Vijaya Kumar and Virmani,
s. s. VIRMANI
402
1992). Table X lists some WCVs identified in Japan and China and at IRRI. The S,n allele has been incorporated into japonica types and successfully used for obtaining indica-japonica hybrids in Japan (Araki et al., 1988; Ikehashi, 1991) and China (Wang et al., 1991). Yanagihara et al. (1992) identified a new locus between Est-9 and Rc (red pericarp) in chromosome 7, which was responsible for hybrid sterility in some
Table X List of Some WCVs Identified in Japan and China and at IRRI
WCV cultivar
Varietal group
Ketan Nangka Calotoc CP SLO 17 NK 4a Norin PL 9a Banten N 22 Dular
TJ TI TJ J
Padi Bujang Pendek Aus 373 DV 149 Kaladumani DV 52 AS 35 Lepudumai BPI 76 Lambayeque 1 Moroberekan Palawan Fossa HV
TJ Aus Aus Aus Aus Aua Aus I Aus TJ TJ TJ J TJ TJ TJ TJ
02428
Gogo Serah Nggonemal Tanggalasi Senatus Madumi Nava 76 Newbonnet Bluebelle Changnot
J TJ Aus Aus
Ib Ib Ib Ib
wc gene
Reference Ikehashi and Araki (1986) Ikehashi and Araki (1986) Ikehashi and Araki (1986) Araki et al. (1988) Araki et al. (1990) Yanagihara et al. (1992) Yanagihara et al. (1992) Ikehashi and Araki (1987); Vijaya Kumar and Virmani (1993) Ikehashi and Araki (1987) Ikehashi and Araki (1987) Ikehashi and Araki (1987) Ikehashi and Araki (1987) Ikehashi and Araki (1987) Ikehashi and Araki (1987) Ikehashi and Araki (1987) Vijaya Kumar and Virmani (1993) Vijaya Kumar and Virmani (1993) Vijaya Kumar and Virmani (1993) Vijaya Kumar and Virmani (1993) Vijaya Kumar and Virmani (1993) Wang et al. (1991) Luo et al. (1990) Luo et al. (1990) Luo et al. (1990) Luo et al. (1990) Luo er al. (1990) Luo el al. (1990) Luo et al. (1 990) Luo et al. (1990)
a This is a japonica rice cultivar bred in Japan which possesses S," gene from Ketan Nangka.
The rice cultivars bred in U.S.A. are known to have some tropical japonica germplasm in their pedigree.
HYBRID RICE
403
crosses between javanicas and a variety from the Indian subcontinent. With the finding of the second “S” locus, it was also stated by Ikehashi et al. (1994) that hybrid sterility in most crosses was caused at one of the loci and the alleles at the other loci remained neutral. Several new S loci were identified by Yanagihara et al. (1992) and Wan eta!. (1993). Currently, a total of six loci (viz. S-5, S-7, S-8, S-9, S-15, and S-16 located on chromosomes 6 , 4 , 6, 7, 12, and 1, respectively) are known which can cause hybrid sterility in intervarietal hybrids independently of each other and for which neutral alleles WC genes have been identified in different rice cultivars (Ikehashi, personal communication). These widecompatibility genes have opened up additional opportunities for exploiting heterosis in intervarietal group hybrids to increase yield potential in rice in temperate as well as tropical rice growing countries.
E. APOMIXIS Apomixis is a method of reproduction in which the embryo (seed) develops without the union of egg and sperm. It is a reproductive process that bypasses female meiosis and syngamy to produce embryos genetically identical to the maternal parent. Thus, it refers to asexual reproduction through seed. The limiting factor for wide-scale adoption of hybrid rice in the tropics and subtropics is the complexity of the seed production system based on cytoplasmic male sterility. The thermosensitive genic male sterility system, although relatively less complex, still has to be developed and tested. In either case, rice hybrids that are produced do not breed true, and they lose the yield advantage in subsequent generations. Therefore farmers have to buy hybrid seeds for each crop. The cost of hybrid seed, being 10-15 times higher than that of ordinary seeds of rice, discourages poor farmers from taking advantage of hybrid technology. The development of apomixis in rice hybrids will enable rice farmers to grow their crops from seeds produced from their commercial hybrid crop year after year, and even poor farmers will benefit from growing hybrid rice. Apomixis will also increase the efficiency of hybrid rice breeders in producing many new true breeding hybrids compared with those produced by using three-line or two-line hybrid breeding systems. The availability of large numbers of hybrids will help increase genetic diversity and reduce genetic vulnerability (Khush et al., 1994). Apomixis has been reported in Gramineae along with several other families of higher plants (Bashaw, 1980; Hanna and Bashaw, 1987). Among the major cereals, maize, wheat, and pearl millet have apomictic relatives, most of which are polyploids. The phenomenon is broadly classified into two types: (1) gametophytic apomixis and (2) adventitious embryony. In gametophytic apomixis embryo sacs are produced from unreduced initial cells and the egg cell develops
404
s. s. VIRMANI
parthernogenetically, giving rise to embryos and plants resembling the mother plant. Gametophytic apomixis may be obligatory (predominantly apomictic) or to some extent continued with sexuality (facultative), and is further divided into displospory and apospory. Indicators of apomixis include occun-ence of (1) identical maternal progeny from plants of cross-pollinated species, or progeny of F, crosses, (2) limited or no genetic variation in the F, population of a cross between two distinct parents, (3) recessive genotypes from a cross of parents with recessive genes pollinated with a parent possessing a dominant marker gene, (4) unusually high seed fertility in aneuploids, triploids, and wide crosses normally expected to be sterile, ( 5 ) aneuploid chromosome number or structural heterozygosity remaining constant from parent to progeny, and (6) multiple seedlings per seed, multiple stigma, multiple ovules per floret, and doubled or fused ovaries. In rice there are reports of mutants possessing twin seedlings per seed (Yuan et al., 1990; Sharma and Virmani, 1990) and multiple pistillate ovaries (Suh, 1985, 1988). Chen et al. (1988) have claimed to have found identified apomixis in rice in China from an interspecific cross; however, their claim has not been confirmed through appropriate genetic tests conducted independently. IRRI scientists have initiated a search for apomixis in rice using three strategies (Khush et al., 1994). These include: 1. Analysis of the tetraploid wild germplasm of Oryza. 2. Induction of mutations for apomixis. 3. Use of molecular approaches to engineer apomixis. Once apomictic germplasm is identified or developed, gene(s) for apomixis will be incorporated in heterotic rice hybrids. This will enable even resourcepoor rice farmers to benefit from hybrid rice technology.
IV. BREEDING PROCEDURES FOR DEVELOPING RICE HYBRIDS Breeding procedures used for developing hybrid rices are strikingly different from those used for breeding inbred rice varieties. Inbred line breeding accumulates productivity genes that perform well under homozygous conditioning while hybrid breeding assembles genes under heterozygous conditions in hybrids from the two parents. These genes show additive, dominant, and epistatic effects; their adverse effects, if any, due to repulsion-phase linkages occurring in parental lines, are overcome by their coexistence in hybrids. The procedures involve three major steps: development of parental lines, seed production of experimental hybrid combination, and evaluation of hybrid combinations. In recent years
HYBRID RICE
405
some biotechnological tools have also been identified, application of which in any of the above steps can improve hybrid rice breeding efficiency.
A. DEVELOPMENT OF PARENTAL Lms In order to develop parental lines for breeding rice hybrids it is extremely useful to establish a close linkage with inbred rice breeding programs to have continuous access to genetically diverse elite lines possessing good per se performance, multiple disease and insect resistance, and acceptable grain quality. In a three-line hybrid rice breeding system this step involves developing male sterile (A), maintainer (B), and restorer (R) lines. However, in the case of a two-line breeding system, the system involves the development of TGMS or PGMS lines and selection of appropriate pollen parents. The essential features of the parental lines are (1) stable male sterility of A or TGMS/PGMS lines and effective fertility restoration by the pollen parents, (2) adaptability to local conditions, (3) good outcrossing potential of male sterile lines, (4) good general combining ability, and ( 5 ) acceptable grain quality. Elite lines or varieties possessing good agronomic characteristics such as high yielding ability, pest and disease resistance and good grain quality are selected from the on-going inbred breeding program. These are then testcrossed as single plants with the available CMS lines to identify effective maintainers and restorers. The F,s showing 99-100% pollen sterility are backcrossed to the corresponding male parent and the completely male sterile plants in the backcross progenies are repeatedly backcrossed for five to seven generations on a single-plant basis to develop a new CMS line. The recurrent male parent becomes the maintainer of the new CMS line. If a testcross F, shows above 80% pollen and spikelet fertility, its male parent can be considered a prospective restorer for that CMS line. Such restorers must be reconfirmed and purified genetically by retestcrossing on a single plant basis. The fertile testcross F,s showing superiority over the male parent and check varieties are identified and later evaluated critically for yield heterosis. Testcross F,s showing partial pollen and spikelet fertility are discarded because their male parents cannot be converted into CMS lines and are difficult to develop into restorer lines. Any rice cultivar (irrespective of whether it is a maintainer, restorer, or partial restorer) can be converted into a TGMS/PGMS line by using conventional pedigree breeding procedures and selecting completely male sterile plants in segregating progenies (F3 onward), and by testing these for their fertility reversion under specific temperature or daylength regimes. The stable CMS and TGMS/PGMS and restorer lines are evaluated for their combining ability using the line/tester mating scheme (Kempthorne, 1957) and lines possessing good general combining ability are selected for making experimental rice hybrids.
406
s. s. VIRMANI B. SEEDPRODUCTION OF EXPERIMENTAL HYBIUDS
After some elite male sterile and pollen parents have been identified, seed of numerous experimental rice hybrids is required to be produced for critical evaluation in observation, preliminary, and advanced yield nurseries. In China, hybrid rice seeds for observation yield nurseries are produced by hand-crossing or by enclosing synchronously flowering A and R lines in chimney-like isolation plots while hybrid seeds for replicated yield nurseries are produced in small- to medium-scale isolated seed production plots created by constructing barriers (about 2 m high) using polythene sheet or other materials (Yuan and Virmani, 1988). These techniques, when used at IRRI, were not successful because the chimney-like isolation chambers or polythene barriers were extremely difficult to retain during the typhoon-prone rainy season and chimneys became too hot during the dry season to allow good seed set; the high temperature also encouraged development of several fungal and bacterial diseases on the enclosed plants. In order to tackle this problem an isolation free system for producing seeds of experimental rice hybrids (Virmani and Casal, 1993) has been developed at IRRI. In this system, various (R) lines are grown side by side in 5 X 3-m plots (Fig. 2). On the four sides of each plot four rows of R plants are planted at 20 X 20-cm spacing to form a border to provide some isolation from adjoining plots. In the center of each plot are four 40-cm-wide vacant spaces, interpersed by single rows of R plants. The spaces can accommodate up to 68 male sterile plants at flowering. Male sterile lines of experimental hybrids are planted separately 5 times at an 8-to 10-day interval. The staggered planting ensures a continuous supply of flowering male sterile plants to synchronize with flowering of R lines in different seed production plots. Male sterile lines are planted upwind from R pollen parents and near seed production plots to facilitate the transport of these plants. When primary tillers of male sterile and pollen parents start booting, their flag leaves are clipped. The two outermost border rows in the pollen parent plots are not clipped to demarcate the plots and to act as a barrier to pollen from adjoining plots. Three to 5 days after clipping of flagleaves, male sterile plants are uprooted and relocated to the vacant spaces. Uprooting is done between 0630 and 0800 h to reduce shock. Male sterile plants are not used if more than 20% of their spikelets have already bloomed. Pollen dispersal is increased at peak anthesis (1030 to 1100 h) by shaking pollen parent panicles with a bamboo stick. Supplementarypollination is done for 5-7 days or until pollen of a pollen parent is exhausted. Pollen parent plants from seed production plots are harvested first, followed by A plants bearing the (A X R) F, seeds. Two CMS lines (viz. IR58025A and IR62829A) yielded 4-5 g hybrid seed per plant during the wet season and 7- 10 g seed per plant during the dry season at IRRI using this system. Five to 10 plants of a male sterile line can produce 40 g of experimental hybrid seed for evaluation in an observation yield trial for two seasons. Twenty to 40
407
HYBRID RICE
~o~~~~ooooooooe~oo+oooooeeo ooooooooo~o~ooooooeoe+oooooe~
‘30000.0.
e00000000000.0.
000000
~ooooaeooooooooooooooooeoooo~
A0A00*~*~OOA060A0A00O~O@OOAOA A O A O 0 O O O O 0 0 A O A O A 0 A 0 0 Oe0000A 0A A 0 A 0 0 0 0 9 0 0 0 A 0 A 0 A 0 A 000 0 0 0A 0 A AOA00OO~O00A0A0AOA00O~OOOOAOA A O A O O O O O O O O ~ O A O A O A O O O O O @ O ~ A ~ ~ ~ O A 0 0 O ~ O O 0 O A O A O A 0 A 0 0 ~ O AO O O ~ ~ ~ O A 0 0 ~ ~ 0 ~ 0 Q A Q A 0 A 0 A 0 0 0 0 ~ @ O O h O A 0 0 ~ + O O 0 0 A 0 ~ 0 A 0 A 0 0 ~ O ~ O O O A O A A O h 0 0 ~ O O + 0 0 A 0 A O A O A ~ 0 O O O O O O ~ ~ A A 0A 0 0 0 0 000 A 0 0 0AOA00. I) 0 000 A 0 A
A~A~0~ooo00A~A0AoA00ooooOOAOA
A O ~ ~ 0 O O O O O O A ~ A O A O A O O ~ O O O O O A O A A 0 A 0 0 0 0 0 I, 0 0 A 0 A 0 A 0 A 0 0 0 0 0 0 0 A 0 A ~ ~ A 0 0 0 0 0 0 0 0 A 0 A 0 0 0 ~ 0 0 0 0 0 0 ~ ~ A O A O O O 0 I) * O O A C ) Q O A O A O O O 0 000 A 0 D A ~ l A 0 0 O ~ O O 0 0 A O A O A 0 A 0 O O O ~ @ O O A O 4 A 0A 00 0 0 0 0 00 A 0A 0 A 0A 00 0 0 0 0 00 A 0 A
....*.....
0 0 0 0 0 ~ ~ + * 0 0 0 0 0 0 0 ~ O ~ 0 8 0 0 0 0 ~
00000.e.~00000000000ooeo~~o~~ ~ ~ ~ 0 ~ ~ e 0 e 0 ~ ~ 0 0 e 0 ~
* ~ o . e. ~. ~ *
Restorer line 14-
Restorer line 2
6
0 Restorer plants (R lines) A
Uprooted relocated CMS plants (A lines)
Restorer line 3
4
Area of flag clipping
0
Unclipped R lines
Figure 2 Layout for Isolation-free system for producing expenmental nce hybnd seed (Virmanl and Casal, 1993)
plants can produce 150 g of experimental hybrid seed for evaluation in a replicated preliminary yield trial for two seasons. About 500 experimental rice hybrids are produced at IRRI every season using this system; seed contamination due to outcrossing from adjoining plots ranges from 0 to 8% (mean 1.7%) which is considered tolerable at the initial stage of evaluation of experimental hybrids. With the development and deployment of the isolation free system, the capacity of hybrid rice breeders to make new hybrid combinations can be increased at least threefold over the hand-crossing system. Hybrid seeds for multilocation advanced yield trials of experimental hybrids are produced in small to mediumsize isolated seed production plots practicing routine hybrid rice seed production procedures.
m
408
s. s. VIRMANI C. EVALUATION OF HYBRID COMBINATIONS
The experimental rice hybrids are evaluated against the best available inbred rice varieties in sequential yield trials (viz. observation yield nursery, preliminary yield trials, advanced yield trials, and multilocation regional yield trials) conducted routinely by plant breeders, and hybrids that are significantly higher yielding than the check varieties are selected and advanced to the higher order yield trials. The rice hybrids found consistently superior to check varieties are tested in on-farm trials and those found superior to check varieties are released for commercial cultivation. Figure 3 illustrates an organizational plan of a hybrid rice breeding program which includes several nurseries (viz. source nursery, testcross nursery, backcross nursery, retestcross nursery, combining ability nursery, seed increase nursery) and yield trials and indicates the seed production mechanism for various trials.
D. BIOTECHNOLOGICAL APPLICATIONS Recent advances in biotechnology, particularly in tissue culture and molecular biology, have opened new avenues in hybrid rice breeding (Brar et al., 1994). Developments include: the regeneration of indica and japonica rice plants from protoplasts, production of somatic hybrid and cytoplasmic hybrid (cybrid) plants and transfer of CMS into elite breeding lines through.protoplast fusion, DNA transformation and production of transgenic rices with new genetic properties, establishment of a comprehensive restriction fragment length polymorphism (RFLP) map consisting of more than 2000 DNA markers, tagging of more than 20 genes of agronomic importance and transfer of genes for diseaselinsect resistance from wild species across crossability barriers. Table XI summarizes application of biotechnology to improve efficiency of hybrid rice breeding procedures. Some salient examples of the application of biotechnology in hybrid rice breeding are listed below. Expeditious purification of A, B, and R lines through anther culture (Xu et al., 1988). Expeditious breeding of japonica rice restorers through anther culture by transferring restorer gene(s) from indica rices (H. P. Moon, personal communication). Extraction of high-yielding inbred lines from heterotic F, hybrids (Siddiq et al., 1994). Diversification of a cytoplasmic male sterility system through wide hybridization (Dalmacio et al., 1995) and induction of somatic mutagenesis through tissue culture (Ling et al., 1987).
Elite
SOURCE NURSERY To evaluate and testcross
lines
Elite lines from
7 different sources Lines from restorer and maintainer breeding nursery
To identify B and R lines
Restorer evaluation
Combining ability
New A and B lines
Seed production for observation vield trial
CMS lines maintenance and evaluation
-4
Observation yield trials Evaluation of A/R hybrids
Seed increase of elite A and 6 lines
Seed production of elite hybrids for preliminary
+
Nucleus seed of A and 6 lines
i
I
Breeder seed of A and 6 lines
Seed production for
L
+
I 1
Seed increase qf elite R lines
Advanced yield trials Replicated multilocationl
omfarm trials
Seed production agencies On-farm trials
-
agencies
Release m
F1 seeds for commercial cultivation
I
I
Figure 3 Operational Row-chart of hybrid rice breeding and seed production program used at IRRI.
s. s. VIRMANI
410
Table XI Application of Biotechnology in Hybrid Rice Improvement (Brar et al., 1994) Technique Anther culture Embryo rescue
Protoplast fusion Somatic embryogenesis Molecular markers
Genetic transformation
Application Extraction of high-yielding inbred lines from superior F, hybrids. Purification of male sterile, maintainer, and restorer lines. Overcoming incompatibility to produce hybrids between wild and cultivated species. Deriving back-cross progenies to develop alloplasmic lines for diversification of CMS sources. Transfer of genes for apomixis from wild species into elite breeding lines of rice. Expeditious transfer of CMS into elite breeding lines. Development of cybrids between otherwise sexually incompatible species. Production of artificial seeds for mass propagation of true-breeding hybrid varieties. Tagging genes for wide compatibility, fertility restoration, thermosensitive male sterility, apomixis, and identifying QTLs for heterosis to facilitate marker-based selection. Choosing parents based on RFLP diversity to obtain highly heterotic combinations. Transfer of cloned genes governing apomixis for producing truebreeding commercial hybrids. Exploitation of genetically engineered nuclear male sterility and fertility restoration systems to produce hybrid varieties.
Production of cybrids and transfer of CMS into elite breeding lines through protoplast fusion using the mitochondria1 genome of the CMS line and the nuclear genome of the fertile maintainer varieties (Akagi et al., 1989; Kyozuka et al., 1989; Akagi and Fujimura, 1994). Tagging of the WC gene (S,n) with RFLP marker RG 213 (IRRI, 1993) and RFLP markers RG 138, RG 64,and RG 456 (Zheng et al., 1992). Tagging of restorer genes (from IR24, a restorer of WA cytoplasm) with RFLP and RAPD markers: (Ning Huang, personal communication). Characterization of the genetic diversity of parental lines based on RFLP and RAPD polymorphism (Ishii and Tsunewaki, 1991; Wang and Tanksley, 1989; Wang et al., 1992; Mackill, 1995). Studying the relationship between heterosis for quantitative traits in hybrids and isozyme polymorphism in parents with a view to predicting heterosis and finding both positive relationships (Gupta and Singh, 1977; Xiao, 1981; Yi ef al., 1984; Deng and Wang, 1984) and negative relationships (Peng et al., 1988; Kato et al., 1994). Studying the relationship between heterosis for biomass of hybrids and the polymorphism of RFLP markers of parental lines and finding only a weak correlation between the two (Kato et al., 1994b) and a strong correlation
HYBRID RICE
41 1
between mid-parent heterosis for yield, seeds per panicle, and kernel weight and heterozygosity measured through RFLP and microsatellite polymorphism among parental lines (Zhang et al., 1994). Studying the molecular basis of CMS by analyzing mitochondria1 DNA modification (Kadowaki et al., 1986, 1988; Mignouna et al., 1987; Yang et af., 1988a). The above examples indicate that several biotechnological tools can also be deployed to increase the efficiency of hybrid rice breeding procedures.
V. ACCOMPLISHMENTS
A. IN CHINA Since 1976, when hybrid rice technology was commercialized in China, Chinese scientists have made tremendous progress in breeding commercial rice hybrids by using the CMS system (see Lin and Yuan, 1980; Coffman and Virmani, 1983; Virmani and Edwards, 1983; Yuan and Virmani, 1988; Yuan et al., 1989, 1994). Hundreds of A, B, and R lines have been developed over the years in about 50 research centers around the country for use in breeding numerous indica and japonica rice hybrids, which occupy about 55% of the total rice area in the country. Four phases of hybrid rice development in China have been recognized by Yuan et al. (1994). 1975- 1979-Experirnent, demonstration, and extension phase: first set of rice hybrids (such as Erjui Nan 1A/IR24, Zhen Shan 97A/IR24, and V41A/IR24) were developed and released to cover about 5 million ha with an average yield of 4.7 t/ha; 20% higher yields than with inbred varieties. 1980- 1981-Adjustment phase: the area remained stagnant but yield increased slowly to 5.3 t/ha with the development of some rice hybrids possessing resistance to diseases and insects. 1982- 1985-Continuous development phase: the area of hybrid rice increased to nearly 10 million ha and the yield increased between 5.9 and 6.5 t/ha. Short-duration hybrids, viz. V20AlCe 64,developed collaboratively with IRRI (Yuan et al., 1985) and Zhen Shen 9IAlMin Hui 63 played a significant role in this phase. 1986- 1991-Rapid and steady development phase: the area under hybrid rice increased to 17.6 million ha yielding on average 6.6 t/ha with the development of a series of new hybrids which showed advantages not only in maturity but also in yield and pest resistance.
412
s. s. VIRMAM
Indica rice hybrids, viz. Nan You 2 (Er-Jiu Nan 1A/IR24), Wei You 6 (V20A/IR26), Shan You 63 (Zhen Shan 97AIMin Hui 63), and Wei You 49, and japonica rice hybrid Li You 57 have made a significant impact on rice production in China. Yuan et al. (1994) observed that during 1976-1991 hybrid rice helped China to increase its production by nearly 200 million t. According to an FA0 estimate (Anonymous, 1989), higher productivity of hybrid rices enabled China to reduce its area from 35.2 m ha in 1978 to 32.5 m ha in 1988, while increasing its production from 140 to 173 million t during the same period. More recently, breeding of hybrids possessing fine grain quality has also been emphasized; for example, Zhang et al. (1995) bred a new Honglian type CMS line, Zhu Shan 97A, which possessed good grain quality and resulted in hybrids with improved grain quality. Zhou and Liao (1995) have also developed a quasiaromatic hybrid rice, Xiang You 63, possessing good grain quality and high yield. Several two-line hybrids have also been developed, some of which (e.g., Pei Ai 64slTeqing) are indicaljaponica hybrids.
The successful development and use of hybrid rice in China encouraged IRRI to revive hybrid rice research in 1979 to explore its prospects and problems in increasing rice varietal yields (IRRI, 1980). As work progressed at IRRI and some positive results were reported (Virmani et al., 1982) several countries established collaboration with IRRI to carry out research on hybrid rice. Currently hybrid rice research is in progress in 16 countries (Table XU).
1. Development of Parental Lines Forty-five CMS lines were developed at IRRI during 1980 to 1992 (IRRI, 1984, 1986, 1987, 1990, 1991, 1992), most of these by transferring CMS-WA cytoplasm into elite inbred lines bred at the institute and in some national programs. In 1989, however, it was observed that frequency of maintainer lines had declined considerably (from 5 to 0.3%) among elite lines bred in IRRI’s irrigated rice program (IRRI, 1990). Therefore a maintainer breeding program was specifically initiated to breed large numbers of maintainer lines from B X B crosses. By adopting this strategy IRRI’s capability to breed CMS lines has increased several fold during the past 3 years (Fig. 4). Some IRRI-bred CMS lines are developed using cytoplasm other than CMS-WA, viz. CMS-ARC and CMS-0. perennis; in the latter case, however, no restorers have been identified yet (Dalmacio et al., 1995). National programs, viz. Japan (Kato et a!., 1994a), India (Siddiq et al., 1994), Malaysia (Guok, 1994), and Korea (Moon et al., 1994), have also bred some CMS lines possessing CMS-WA, CMS-boro, and other
Table XI1 Countries Involved in Hybrid Rice Research Outside China (Adapted from Virmani, 1994c)
Country
Institution
DPR Korea Philippines
Rice Research Institute, Pyongyang IRRI PhilRice Cargill Seeds Ring Around Rice Tec University of California, Davis Several national and state institutes PHI-Biogene MAHYCO Seed Company PROAGRO Food Crops Research Institute, Sukamandi Cargill Seeds Crop Experiment Stations at Suweon, Honam, Milyang Cuu Long Delta Rice Research Institute, Omon National Institute of Agricultural Research, Hanoi Malaysian Agricultural Research and Development Institute, Bumbong Lima Rice Research Institute, Bangkhan Rice Research Institute, Sakha National Agricultural Research Center, Tsukuba; Zennoh Agricultural Technical Center, Hiratsuka Mitsui Chemical Company Rice Research Station, Batalagoda National Rice Growers Federal, Bogota National Research Center for Rice and Bean, EMBRAPA-CNPAF Granjas 4 Irmaos, SA Bangladesh Rice Research Institute National Agric. Research Center, Islamabad, Rice Research Station, Kala Shah Kaku, Punjab; Rice Research Station, Dokri, Sind
U.S.A.
India
Indonesia
Republic of Korea Vietnam
Malaysia
Thailand Egypt Japan
Sri Lanka Colombia Brazil
Bangladesh Pakistan
Year of initiation of research 1976 1979 I988 1981 1980 1988 1980 1981 1988 1990 1992 1992
Collaborator
Several NRS IRRI China until 1992 China China -
IRRI -
IRRI
1986 1982
China until 1992 IRRI
1985
IRRI
1992 1985
China (through FAO) and IRRI IRRI
1983 1987 1983
IRRI IRRI IRRP
1989 1991 1983
-
1984
IRRI, IRAT
1994 1993 1993
CIRAD IRRI IRRI
IRRI IRRI
a Collaboration limited to thermosensitive genic male sterility and wide compatibility in rice.
s. s. vIRh4ANI
414 CMS lines (no.) 30
26 25 21 20 14
15 9
10
6
5
5 0
I
,
1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 Year
Figure 4 Frequency of CMS lines bred at IRRI during 1983 to 1994.
cytoplasm. Over the years IRRI has developed about 100 CMS lines but only a few (Table XIII) possess traits suitable for developing commercial hybrids. As mentioned earlier there is no dearth of restorer among elite indica rice cultivars developed at IRRI and in several national programs. Therefore, the pace of breeding rice hybrids increases tremendously once commercially usable CMS lines are available. Two IRRI-bred CMS lines, viz. IR58025A and IR62829A, have been used widely in developing commercial rice hybrids in India, Philippines, and Vietnam (Virmani, 1994b). During the past 2 years IRRI has also developed some TGMS lines (IRRI, 1994, 1995) which should be helpful in developing heterotic rice hybrids more efficiently. Hybrids developed by the TGMS system are likely to show stronger heterosis because there are less restrictions in the choice of parents in comparison to the CMS system.
2. Development of Heterotic Rice Hybrids Table XIV lists some hybrid combinations yielding significantly more than the inbred rice varieties of corresponding growth duration in replicated yield trials conducted at IRRI during 1989-1995. The best hybrids (mean 6.3 t/ha) outyielded the best inbred varieties by about 20%. The yield advantage of some elite
415
HYBRID RICE Table XIII Some IRRI-Bred CMS Lines Possessing Traits Suited for Developing Commercial Rice Hybrids Score (1-9 scale)O for traits for commercial usability
CMS line
Maintainer line
A
B
C
D
E
1 3 1 1 1 2 1 1 1 1 1 1 1 1 1 1
3 1 5 5 3 1 5 5 4 4 2 5 4 3 5 3
3 3 3 3
1 5 3 3
1 3 3 3
~~
IR58025A IR62829A IR67683A IR67684A IR68280A IR68885A IR68886A IR68887A IR68888A IR68889A IR68893A IR68897A IR68887A IR68899A IR68900A IR68902A
Pusa 167-1203-2 IR29744-94-3-2-2-2-3-3 CR 1009 IR51824-31-2 IR54718-C1-257-2-2-1-2 IR62829B IR62832-58-7-8-1-8-12-4 IR62843- 14-2-6-6-2-2-3 IR62844- 15-6-1-10-3-4-1 1162849-28-4-4-1 I - 1-5-7 IR62852- 13-3-6-2-6-1 1-3 IR62856-15-3-1-1-7-5-10-3 IR62856-112-2-5-3-1-7-7 IR62856- 134-3-2-12-4-7-5-2 IR62856-147-2-8- 12-4-7-2-1 IR62856- 162-3-4-7-6-1-3
3 3 3 3 5 3 5 4 5 5 3 2
-
5 5
-
3 5 -
3 -
5 -
3 3 3 1 3 1
5 3 5 3 3 I
Note. A, Stability for complete pollen sterility; B, phenotypic acceptability under tropicaUsubtropica1conditions; C, outcrossing potential, D, general combining ability; E, restoration ability. a 1, Most suitable; 9, Unsuitable.
IRRI rice hybrids evaluated in India, Vietnam, the Philippines, Malaysia, and Pakistan is presented in Tables XV and XVI. Some heterotic rice hybrids have been identified and released for commercial cultivation in India, Vietnam, and Philippines (Table XVII). Some of these hybrids were bred at IRRI, while in the case of others IRRI-bred CMS lines were used as the female parent and locally bred cultivars were used as the pollen parent.
3. Yield Stability of Heterotic Rice Hybrids Young and Virmani (1990a) studied the stability of 140 rice hybrids and their parents (10 female and 7 male) using the Eberhart and Russel model, and observed that hybrids showed less stability than the parents and that their performance over environments was less predictable than the parents. Nine out of 10 top-yielding hybrids showed significant deviation from the regression line; one rice hybrid showed above-average response to environments and was stable.
s. s. VIRMANI
416
Table XIV List of Some Promising Rice Hybrids Identified at IRRI during 1989-1995 Season
Hybrid
Yield (t/ha)
% of check
Growth duration
1989 DS 1989 WS 1991 DS
IR58025A/IR29723-143R IR58025A/IR 19058-107R IR62829A/IR35366-62R IR62829A/IR3463 1-123R IR58025A/IR54742-1 I-IR IR58025A/IR35366-62R IR58025AIIR54745-2R IR58025A/IR39323- 182-21 IR58025AIIR19058- I 17- 1R IR62829A/IR473 10-94-4R IR58025A/IR46R IR58025A/IR54056-64-2 IR58025AjIR34686- I79 IR58025A/BG 915 IR58025AIIR59606- 119-3R IR58025A/IR60821-76-2-3R IR62829AIIR46R IR58025A/IR52774-B-B-6R IR58025A/Taichung Sen Yu 85 IR58025A/IR58029- 169-3-2-3R IR58025AIIR58773-35-3-1-2R IR58025AIIR58 103-62-3 IR58025A/IR58100-97-2-1 IR58025AIRP633-76- 1R IR58025AIIR58773-35 IR58025A/Sanghuangan No. 2 IR58025AIIR5.5838-82-2-3-2-3
6.2 2.9 4.7 4.9 4.7 4.7 5.4 4.9 6.4 5.I 6.3 4.4 7.4 4.0 5.8 5.7 6.6 6.6 6.3 5.9 5.9 9.5 8.9 9.6 9.4 9.2 9.2
113* 122* 118** 122** I18** 118** 129** 117* 123** 127** 114* 119* 112* 125* 121* 119* 132** 132** 126** 118* 118* 120* 111* 120* 118* 115* 115*
127 110 112 1I4 112 112 122 126 113 120 117 126 128
Mean DS
7.0
ws
5.5
I17 124 120
1991 WS 1992 DS 1992 WS 1993 DS 1993 WS 1994 WS
1995 DS
Grand mean
6.3
105 105
111 122 122 117 118 119 114 117 121 123 122 !?n
* Value statistically significant at 5% level. * * Value statistically significant at I % level.
Both parents of this hybrid were also stable. However, two other stable parents gave an unstable hybrid. It was therefore concluded that stable rice hybrids could be developed from stable parents but such parents do not necessarily generate stable hybrids. Young and Virmani (1990b) also observed higher heterosis in wet seasons (stress environment) than in dry seasons (favorable environment) at IRRI. The yield advantage of some elite rice hybrids evaluated in the Philip-
Table XV Yield Advantage of Some Elite IRRI-Bred Rice Hybrids Evaluated in Some Collaborating Countries during 1990-1993
Country India
Season 1990 WS 1991 DS 1991 WS
1992 WS
Vietnam
1990 WS 1991 DS 1991 WS 1992 WS
1993 DS
Philippines
1992 DS 1993 DS
Malaysia
1990 DS 1991 WS 1992 WS
Location Mandya Mandya Hyderbad Hyderabad Maruteru New Delhi Hyderabad Mandya Faizabad Hyderabad Mandya Coimbatore Faizabad Cuttack Omon Omon Omon Omon Omon Omon Omon Omon Omon Hanoi
Omon Tanhiep Binhauc Maligaya San Mateo Maligaya Group I Group II San Mateo Group I Group I1 Bumbong Lima Bumbong Lima Bumbong Lima Overall Mean
Hybrid
Yield (t/ha)
Percentage of the best check
IR58025A/IR9761-19- I R IR58025A/IR29723- 143-3-2-1R IR62829A/IR10 198-66-2R IR62829AIIR 10198-66-2R IR62829AIIR35366-40-3R IR62829AIIR35366-40-3R IR62829AjIR28238- 109-2R IR58025A/IR40750-82-3R IR58025A/IR54742-22- 19-3R IR58025A/IR40750-82-3R IR58025A/IR34686- 179-1-2-11 IR58025A/IR32419-28-3- 1-3 IR58025A/IR39323- 182-2-3-3R IR58025A/IR46R IR62829A/IR46R IR62829A/IR29723- 143-3-2-1R IR58025A/IR29723-143-3-2-1R IR62829A/IR29723-143-3-2-1R IR58025A/IR29723- 143-3-2-1R IR62829A/IR29723- 343-3-2-11 IR58025A/IR29723- 143-3-2-1R IR62829A/IR29723- 143-3-2-1R IR58025A/IR21567-18-3R IR58025A/IR52287-15-2-3-2R IR58025A/IR29723- 143-3-2-1R IR58025AIIR66R IR62829A/IR10198-66-2R IR62829A/IR47310-94-4-3-1R IR62829A/IR47310-94-4-3-1R IR58025A/IR54742-22-19-3R IR58025AIIR32419-28-3-1-3 IR62829A/IR20933-68-21- 1-2R
9.3 7.0 6.8 7.0 6.3 11.6 5.2 6.4 6.2 4.6 7.3 7.5 6.3 6.6 5.3 6.7 7.6 6.1 6.0 4.9 4.6 5.5 6.2 6.7
IR64608A/IR29723- 143-3-2-1R IR58025AIIR54742-22-79-3R
7.8 7.7
I22 131
IR64608AIIR29723- 143-3-2-1R IR58025A/IR34686-179-1-2-1R IR62829A/IR29723- 143-3-2-1R IR58025A/IR29723- 143-3-2-lR IR58025A/IR29723- 143-3-2-I R IR62829A/IR29723- 143-3-2-IR
6.3 8.1
102 112 141 124 140 I17 I25
5.1
5. I 5.7 5.8 7.3 7.5 7.9 8.0
5.8 5.1
5.6 5.6 6.5
112 125 128 117 112 123 121 121 117 126 109 114 131 135 143 129 146 122 120 132 124 112 112 144 150 150 167 I10 112 101
113 139
Table XVI Yield Advantage of Some Elite IRRI-Bred Rice Hybrids Evaluated in Some Collaborating Countries during 1993-1994
Country India
Location
Season
Hybrid
Hyderabad
1993 WS
IR58025A/IR54742 IR58025A/IR32809 IR58025A/IR 13419 IR58025A/IR72R IR62829A/IR40750 IR62829A/IR40750 IR58025A/IR2 1567 IR58025A/RP633-76- 1 IR62829A/IR29723-143R IR58025AIIR20933 IR58025A/IR 10I98 IR58025A/IR34686- 179 IR62829A/IR29723 PMS lOA/BR 827-35-R IR58025A/IR 10198 IR62829A/IR54883 IR58025A/IR32809 IR58025A/IR54791-19R IR58025A/IR13419 IR58025A/IR54742 IR58025A/RP 1057 IR62829A/IR40750 PMS 8A/IR46R IR58025AIIR54969-41 IR58025A/RP633-76- 1 IR58025A/IR 13419 IR58025A/IR25912-8 1 IR58025A/BR 827-351
6.4 5.8 5.6 5.6 5.2 7.0 7.5 7.7 7.7 9.6 9.5 5.2 9.6 6.6 5.6 6.2 5.6 7.9 6.6 7.1 10.0 8.7 8.7 9.2 8.0 7.6 7.3 3.2
1.1* 1.5* 1.3* 1.3* 0.8* 1.5* 1.2* 1.2* 1.2* 0.9* 0.8* 1.o* 2.4* 2.6* 1.o* 1.6* 1.3* 1.3* 1.3* 2.4* 2.5*
IR58025AIIR21567 IR58025A/IR21567 IR58025A/IR29723 IR68275A/IR46R PMS 8A/IR29723 IR58025A/IR59606-119 IR68275A/IR46R IR58025AIIR52774-B-B IR58025A/IR34686 IR58025A/IR259 12
6.9 5.1 6.4 6.5 8.7 5.8 6.6 6.6 7.8 7.3
2.4* 1.2* 0.6* 1.1* 1.1* 1.o* 1.3* 1.3* 1.9* 1.3*
1994 DS 1994 WS Karnal
1993 WS
Delhi Kapurthala
1994 WS 1993 WS 1994 WS 1993 WS 1993 ws 1993 WS 1994 WS 1993 ws
Chinsurah Coimbatore Wyra Mandya
1994 DS 1994 ws Faizabad
Pakistan
Philippines
Vietnam
Difference from check @/ha)
Yield (t/ha)
Karjat Kalashah Kaku Sialkot Farooqabad Maligaya San Mateo IRRI, Los Banos Omon
1993 ws 1994 WS 1994 ws 1994 WS 1994 WS 1994 WS 1994 WS 1994 DS 1994 WS
1994 DS
* Difference statistically significant.
1.5*
1.5*
2.8* I .6* 1.5*
1.7* 2.5*
419
HYBRID RICE Table XVII Heterotic Rice Hybrids Released for Commercial Cultivation during 1993-1994 Country Vietnam India
Philippines a
Hybrid IR58025AlIR29723 IR62829A/IR29723 IR58025AIVajram IR62829AIMTU 9992 IR62829AIIR10198 IR58025AIIR976 1- 19-1R IR62829AIIR29723- 143-3-2-1R
Released as
Year
UTL- 1 a UTL-2a APHR-1 APHR-2b MGR-1 KRH- 1 Rc26He/(or Magat)
1993 1993 1994 1994 I994 1994 1994
For regional adaptability trials in Mekong Delta. For Telangana and Rayalseema districts of Andhra F’radesh state. For May/June and Sep/Oct planting of Tamil Nadu State. For irrigated areas of Kamataka State. For Cagayan valley region.
pines, India, Vietnam, and Philippines (see Tables XV and XVI) was restricted to a single location, although some rice hybrids did show yield superiority over inbred check varieties at more than one location. Genotypes possessing specific adaptability are considered useful in overcoming the problem of potential genetic vulnerability associated with general adaptability.
VI. AGRONOMIC MANAGEMENT Yan (1988) observed that agronomic management of hybrid rice differed considerably from that of inbred rice varieties. Major differences were pointed out in (i) management of seedlings in the seedbed, (ii) management of the field during vegetative growth, and (iii) management of fertilizer. In China, the recommended amount of hybrid rice seed is 15-25 kg/ha, compared with 110-180 kg/ha for inbred rices. However, in the seedbed hybrid rices seedlings receive four to five times more space than the conventional rice. This favors tillering in the nursery and makes it possible to transplant a single hybrid rice plant/hill with three to four tillers compared with three or four inbred rice plantdhill without tillers. During vegetative growth, hybrid rice accumulates more dry matter which results in more spikelets per panicle, whereas inbred rice depends basically on the accumulation of assimilates after heading. In early maturing japonica rice varieties grown in Northern China, about 90% of grain carbohydrates comes from photosynthetic assimilation after heading; in hybrid rice, 30-40% of the grain carbohydrates comes from assimilation before heading and only 60-70% comes from check assimilation after heading.
s. s. vIRh4ANI
42 0
Large amounts of N applied on hybrid rice at vegetative growth sometimes result in overtillering. To control tillering, Yan (1988) recommended manipulation of the fertilizer dose or drying of the soil according to the rates of tillering at 20-25 days after transplanting. Drying of the soil decreases nonproductive tillers, regulates metabolism between C and N, and increases dry matter accumulation. Dry soil also helps establish a good canopy for late growth. Hybrid rices are more responsive to fertilizers than inbred rices are. For example, in the Huai Bei region of Jiangsu Province a short-statured inbred rice variety, Nong Ken 57, responded up to 140 kg N/ha, whereas the hybrid Shan You 3 responded up to 180 kg N/ha and the hybrid Gang Hua 2 showed even greater response to N and the highest yield potential (Fig. 5). Dry matter production at different growth stages showed different patterns for hybrid rice and inbred rice. While hybrid rice has more dry matter accumulation in the early and middle growth stages, inbred rice has more in the late growth stages. Nitrogen uptake of hybrid rice varies with growth stages. It was found to be 29.1% more of the total from recovery to tillering, and 34.3% more from tillering to panicle initiation compared to inbred rice (Yan, 1988). Nitrogen uptake of hybrid Nan You 3 at 40 days after transplanting is 71.2%; hence, for hybrid rice more fertilizer should be applied at
Yield (t/ha)
/ -
ShanYou 3
'
Nong-Ken 57
I 60
I
120
I 180
I 240
I
I
300
360
4 '0
Nitrogen (kg/ha)
Figure 5 N response of rice hybrids (Shan-You 3 and Gang-Hua 2) and rice inbreds (Nong-Ken 57 and Da-Che-Gen) in China (Yang,1988).
HYBRID RICE
42 1
early stages. Yan (1988) also stated that if 160 kg N/ha is used during a cropping season, it is recommended that 80 kg/ha be applied basally, 50 kg N/ha during tillering, and the remainder after panicle initiation. For inbred rice, more N should be applied during the middle and late stages than during the early stages. For high yield of hybrid rice, sink is not the limiting factor as it is in inbred rice (Yan, 1988). To increase hybrid rice yields, nutrient status should be improved at late growth, a better canopy structure should be set up, and the relation between source and sink should be regulated. To increase inbred rice yields, enough panicles and spikelets should be produced. Fertilizer management in inbred rice should focus on spikelet initiation, and for hybrid rice it should focus on spikelet filling (Yan, 1988). Hybrid rice relies mainly on tillers to obtain desirable populations while inbred rices rely mainly on number of seedlings planted. About 85-90% of productive panicles of hybrid rice comes from tillers; in inbred rice, their proportion is 30-40%. During the middle growth stages hybrid rice tiller populations that are not high enough to meet the optimum LA1 can be increased by applying N 30-35 days before heading. For inbred rices, N must be applied 5-10 days earlier to make nonproductive tillers productive. In the Cuu Long delta in Vietnam, hybrid rice UTL 1 and inbred check variety OM 90-9 showed similar responses to NPK fertilizer rates, although UTL 1 outyielded the inbred variety at all levels (Luat et al., 1995). Guong et al. (1995) studied N and P use efficiency in direct-seeded rice in the Mekong river delta and concluded that the hybrid rice UTL2 had a higher physiological efficiency (kilogram yield per kilogram N uptake) than inbred rice MTU 19 during the wet season. Results from India (Anonymous, 1995) also revealed that application of nitrogen at 120 kg/ha in three splits (50% basal 25% at tillering + 25% at booting) was the best, with higher yields irrespective of the season. At IRRI, studies on nitrogen management of hybrids IR64616H and IR72 showed a significantly higher response of hybrid to late season nitrogen application than IR72 (IRRI, 1993). The above studies indicate that hybrid rices require different strategies for nitrogen management than inbreds to maximize expression of their yield advantage.
+
VII. DISEASE/INSECT RESISTANCE Mew er al. (1988) reported that in China the incidence of stem borer, white back plant hopper, leaf roller, bacterial blight, sheath blight, and virus diseases is more frequent on hybrid rice than on inbred rices. Local outbreaks of diseases
s. s. VIRMANI
42 2
such as downey mildew, false smut, and kernel smut occur frequently on hybrid rice. Brown plant hopper (BPH) infestation, however, was negligible when IR26, possessing a dominant gene Bph 1 for resistance, was used as the R line. Although stem borer (SB) occurs often on hybrids, their growth, vigor, and higher tillering capacity appeared to compensate for the damage. Resistance of either the A or the R line appeared to be adequate for conferring resistance to the hybrid. IRRI entomologists and pathologists evaluated the resistance of a number of F, hybrids and parental lines under artificial inoculation or infestation of brown BPH, green leafhopper (GLH), and bacterial blight (BB). Results (Table XVIII) showed that hybrids were resistant if the parents were resistant or if one parent was resistant. The hybrids were resistant or susceptible, depending on whether the gene imparting resistance was dominant or recessive. If both parents were susceptible, then the hybrids were also susceptible. Elite rice hybrids have been found to possess resistance to diseases and insects (Table XIX), implying that rice hybrids with the required level of disease/insect resistance can be developed by appropriate choice of parental lines. Studies in China (Mew er al., 1988) and at IRRI (Virmani, 1994a) did not find any evi-
Table XVIII Parents and Hybrids Exhibiting SpeciSc Reaction to Brown Planthopper Biotype 2, Green Leamopper, and Bacterial Blight (IRRI, 1990) Hybrids (No.) showing specific reaction Parent reaction
Susceptible
RIR RIS SIR
0 0 7
SIS
I
R/R RIS SIR
0 0 1 2 3 6 0 Bacterial blight rice PXO 61 0 0 0 0 0 0 3 0
SIS R/R RIS SIR SIS
Moderately resistant
Brown planthopper 0 1
5
0 Green leafhopper 4
Note. R, Resistant; S, Susceptible.
Resistant
Table XIX Reaction of Some Elite IRRI-Bred Rice Hybrids and Check Varieties to Some Biotic and Abiotic Stresses (IRRI) Reaction (1-9 scale) to Hybridlvariety IR58025A/IR29723- 143-3-2-1Ra IR62829A/lR29723-143-3-2-IRa*b IR62829A/IR10198-66-2-R'*c IR58025A/IR54742-22- 19-3R IR58025AIIR34686-179-1-2-1R IR62829AIIR46R IR58025A/IR2 1567-18-3R IR58025AIIR54791-19R IR58025AIIR54969-41R IR58025A/IR25912-81R IR58025AIBR827-35R IR58025AIIR59606-119R IR72 IR74 PSB4 PSBlO
Tung
GLH
BPH
BBRcl
BBRc2
BL
YSB
CT
SAL
MS MS MS MR R R MR MS R MS MS R R MR MR R
3 3 9 3 3 5 3 5 3 3 9 3 3 5 9 5
3 3 9 9 9 5 9 3 9 9 9 9
7
7 7 1 7 7 7 7 7
1 1
5 5 5 3 3 3 3 3 3 5 3 5 5 3 5
5 5 3 5 3 5 5 5 7 7 5 5 9 9 9 9
5 5
1
5 I 1 1 1 1
1 1 7
1
1 1
3 3 3
5 7 1
Hybrid released for commercial cultivation in Vietnam. Hybrid released for commercial cultivation in the Philippines. Hybrid released in Tamil Nadu, India.
1
3 9 1
1 9
7
1
7 7 7 5 5 7 5
5 1 1
6 1 1 1
1
9 9 5 9 9 5 9 9 9 9 7 9 7
424
s. s. VIRMANI
dence to associate any disease or insect susceptibility in rice with the CMS-WA cytoplasm which has been deployed extensively to develop commercial rice hybrids in and outside China. Zhu et al. (1994) proposed a new strategy to improve disease/insect resistance of a currently grown rice hybrid which possesses high yield potential but is susceptible to certain biotic stresses. The strategy involves introducing an additional maintainer to the three lines (e.g., A , , B , , and R,) used currently to produce the commercial hybrid. The additional maintainer (B,) line is selected such that it possesses the desired resistance to the biotic stress(es) (controlled by dominant or partially dominant genes) and is identical to B , with regard to growth duration, grain type, and plant height. The procedure involves multiplication of A-line seeds by using the B, maintainers so that the CMS seed so multiplied will give rise to A,/B, plants which will possess the desired resistance. The A,/B, CMS seed, when used with R, restorer, will give rise to hybrid seeds which will be a mixture of A I R , and A, (derived from A,B,) R,; the latter component of the hybrid will possess resistance to the required stress(es). Thus the hybrid A,/B,/R, (equivalent to a three-way cross) shows better resistance to the desired biotic stress(es) than the original A I R , hbyrid. Zhu er al. (1994) reported several AB combinations using different B lines possessing resistance to different biotic stresses. As expected, rice hybrids, so produced, were not uniform; they showed segregation for plant height and heading dates. Some of these yielded somewhat better than the check hybrid, perhaps due to a higher level of resistance. This strategy may not be acceptable to those whose aim is to produce uniform rice hybrids possessing high grain quality standards. However, the strategy looks interesting and may be evaluated critically before a final judgement is made.
VIII. GRAINQUALITY Grains harvested from commercial F, rice hybrids represent F2 seed generation and, hence, segregate for some grain characteristics. Since rice is primarily consumed as whole grains, it is important to consider the effect of this segregation on grain quality. Major determinants of grain quality in hybrid rice are: (1) milling and head rice recovery, (2) size, shape, and appearance, and (3) cooking and eating characteristics. Khush et al. (1988) critically studied grain quality of several rice hybrids in comparison to their divergent parents and made the following conclusions. Hybrids have superior head rice recovery if both parents have high head rice recovery. Hybrids of medium grain size and shape can be produced by intercrossing
HYBRID RICE
42 5
long-, medium-, or short-grained parents. However, in order to produce long-grained rice hybrids both parents should have long grains. For developing hybrids with better appearance and market acceptability, parents having widely different endosperm appearance should not be crossed. Segregation for different starch characteristics in bulk F, samples does not pose any problem in eating and cooking qualities. The cooking characteristics of hybrid bulk grains are intermediate between those of parents. Therefore, to produce hybrids with desirable tenderness, cohesiveness, gloss, and aroma, parents should be selected so that the weighted averages of the grains in bulk sample match consumer preferences. Islam and Virmani (1990) also observed that genetic heterozygosity of hybrids did not impair grain quality in terms of their physical and chemical characteristics. Several rice hybrids introduced from China to the U.S.A. during the 1980s did not meet the grain quality standards of local commercial inbred rice varieties due to their low milling yield caused by larger grain size and excessive chalkiness (Bollich et al., 1988). The latter traits of the hybrids were inherited from the CMS lines, viz. V20A and Zhen Shan 97A, used to breed these hybrids. The Chinese rice hybrids introduced in Vietnam also possessed poor grain quality which, despite their higher yields, did not increase the farmers’ profitability (Virmani et al., 1994; Luat et al., 1995). However, IRRI-bred rice hybrids released in the Philippines and India possessed acceptable grain quality (Virmani, 1994a; Anonymous, 1995) because their parents possessed acceptable grain quality. The above reports suggest that it is possible to develop rice hybrids of acceptable grain quality by using parental lines of the desired grain quality. Hybrid rice breeders in India and at IRRI are also engaged in developing basmati rice hybrids from CMS and restorer lines possessing basmati grain quality. In recent years, Chinese scientists have also been developing rice hybrids with improved grain quality. Even in the U.S.A., where grain quality standards are quite high, heterotic rice hybrids with acceptable grain quality are in the pipeline in a private seed company.
IX.ADAPTABILITY TO STRESS ENVIRONMENTS Rice hybrids have shown better seedling tolerance to low temperature (Kaw and Khush, 1989, although they were more sensitive than the parents to extreme temperature at flowering. Hybrids have also shown superiority over parental lines for salt tolerance (Akbar and Yabuno, 1975; Senadhira and Virmani, 1987),
426
s. s. VIRMANI
flood tolerance (Singh, 1983), and ratooning ability (Chauhan et al., 1983; Siddiq et al., 1994). Preliminary results at IRRI (Bhuiyan er al., personal communication) also indicate higher water use efficiency of a rice hybrid compared to an inbred rice. Thus hybrid rice may also offer an opportunity to increase factor productivity of water, which is becoming a scarce resource.
X. HYBRID SEED PRODUCTION Rice, being a self-pollinated crop, shows limited outcrossing ranging from 0 to 6.8% (Butany, 1957), although several wild rices have shown significantly high outcrossing rates (Oka, 1988). The Asian forms of the 0. perennis complex or 0. rujpogon showed a range from 7 to 56%, tending to be higher in perennial than in annual types. The African annual species 0. breviligulara appeared to have a lower rate, ranging from 3 to 20%. Some of the 0. longistuminata accessions are self-incompatibleand show up to 100% outcrossing. Male sterile cultivated rices show outcrossing rates ranging from 0 to 44% (Athwal and Virmani 1972; Carnahan er al., 1972; Azzini and Rutger, 1982). In hybrid rice seed production experimental plots in China, outcrossing rates on male sterile lines have been reported to range from 14.6 to 53.1 (Xuand Li, 1988). At IRRI, up to a 37.0% outcrossing rate has been observed on male sterile lines used in hybrid seed production plots. In India natural outcrossing up to 42.8% has been reported on male sterile lines (Anonymous, 1995). Variability in the extent of natural outcrossing on male sterile lines of rice can be attributed to variations in flowering behavior, floral characteristics of male sterile and pollen parents and variations in environmental factors. Some plant characteristics, viz. plant height, flag leaf length, and angle, and panicle exsertion, also affect natural outcrossing.
A. PLANT CHARACTERISTICS INFLUENCING
OUTCROSSING IN RICE When the pollen parent is slightly taller than the male sterile parent, this is conducive to higher outcrossing. Also small and horizontal flag leaves are useful to enhance the outcrossing rate compared to long and erect flag leaves. Since the number of panicles per square meter and the number of spikelets per panicle are the main seed yield components, high tillering capacity and larger panicles with many spikelets are useful in the seed parent. A good panicle exsertion in the parent would expose a higher number of spikelets for outcrossing compared to a male sterile line showing incomplete panicle exsertion. Similarly, incomplete panicle exsertion in a pollen parent would result in lower pollen release into the
HYBRID RICE
42 7
air. Synchronization of days to flowering of seed and pollen parent is the key to attaining good outcrossing on the seed parent. Yuan (1983, Xu and Li (1988), and Virmani and Sharma (1993) have outlined strategies for synchronization of flowering of parental lines in seed production plots. These include staggered seeding and transplanting of male and female parents, sowing the male parent more than once to extend the duration for which pollen is available, and predicting and adjusting flowering dates. Rutger and Carnahan (1981) discovered a recessive gene in japonica rice for the elongated uppermost internode (eui) which effectively resulted in a recessive tall plant type. This gene has been incorporated in indica rice IR50 through backrossing (Virmani et al., 1988).
B. FLOWERING BEHAVIORIN RELATION TO OUTCROSSING IN RICE The flowering process in rice has been described by Rodrigo (1925) and Parmar et al. (1979). Peak blooming in rice occurs between loo0 and 1030 h; blooming duration has been reported to be 46-93 min (Virmani and Athwal, 1973) and 28-78 min (Parmar et al., 1979). Saran et al. (1971) reported that duration of floret opening was positively correlated with the percentage of sterility. At IRRI, several cytoplasmic male sterile lines showing complete pollen sterility were found to bloom longer than their corresponding isogenic maintainer lines possessing normal pollen fertility (Virmani, 1994). Observation made at IRRI on blooming behavior of several restorer lines indicated that 80% of the spikelets of the R lines bloomed during the first 3 days while 15% bloomed during the next 2 days. It was therefore concluded that the R line in no case should flower earlier than the A line; otherwise, there will be a drastic reduction in the outcrossing rate. The angle of opening of lemma and palea of a floret also influences outcrossing, and varietal differences have been observed for this trait (Bassi et al., 1992). Virmani (1994a) has reviewed the studies conducted to describe the mechanism and angle of floret opening in rice and concluded that wide variability existed in rice for components of flowering behavior; he also concluded that in order to attain a high percentage of outcrossing it was important to attain an ideal degree of synchronization in the period of flowering. An ideal female parent in the commercial hybrid rice seed production program should have relatively early flowering initiation and long duration of blooming.
C. FLORAL TRAITS INFLUENCING OUTCROSSINGIN RICE Virmani (1994a) reviewed the literature on floral traits (viz. stigma size, style length, stigma exsertion, anther length, filament length, and pollen num-
s. s. VIRMANI
42 8
ber/anther) influencing outcrossing and noted significant varietal differences for these traits. Oka (1988) stated that outcrossing in rice depended on the capacity of stigmas to receive alien pollen before self-pollination and the capacity of anthers to emit much pollen to pollinate other plants in the proximity. The former capacity would be a function of the time interval from flowering to pollen emission, stigma size, and extension of stigma from the flower as conditioned by the style plus stigma length. The latter capacity would depend on the number and disseminating ability of the pollen grains. Kato and Namai (1987a) observed that residual pollen grains per anther (pollen grains remaining inside the anther at the time of floret opening; normally most of the pollen grains shed out of the anther at this time) and blooming spikelets per unit area directly affected pollen shedding. Therefore, pollen parents for hybrid rice seed production needed numerous residual pollen grains per exserted anther and a large number of spikelets per plant that constantly bloom. Kato and Namai (1987a) also developed pollen sampling techniques using pollen samplers to determine pollen load (i.e., number of pollen grains per liter of air in a flowering paddy field). They further showed that anther length did not correlate with percentage or number of residual pollen grains per anther exserted from the spikelets. Tables XX and XXI give floral traits of some widely used A and R lines bred at IRRI. Inheritance studies on floral traits influencing outcrossing (Virmani and Athwal, 1974; Hassan and Siddiq, 1984) are rather limited . Attempts to transfer the long (5 mm) stigma from a wild rice genetic stock 6209 (IRRI, 1983) into the genetic background of some elite maintainer lines have been only partly successful. The undesirable linkage between long exserted stigma of wild rices and improved agronomic characteristics of cultivated rices is quite strong and needs to be broken to incorporate these traits into selected genotypes (Virmani, 1994a). Taillebois and Guimaraes ( 1988) used backcrossing combined with pedigree
Table XX Floral Traits of CMS Lines: IRS8025A and IR62829A, Used for Commercial Rice Hybrids (IRRI, 1991 WS) CMS line Trait Duration of floret opening (min) Angle of floret opening (") Pistil length (mm) Stigma size (mm) Stigma exsertion (8) Maximum stigma receptivity (%) (3rd/4th day)
IR58025A
* 16.9
103.8 28.1 2 3.36 -C 1.31 2 70.6 51.9
3.9 0.14 0.09
IR62829A 79.6 32.7 f 2.73 -C 1.21 f 68.1
37.2
13.6 3.5 0.12 0.09
42 9
HYBRID RICE Table XXI Floral Traits of Some Promising Restorer Lines of Rice (IRRI, 1991 DS) Trait Anther length Restorer line IR46R IR976 I R IR10198R IR28238R IR29723R IR54742R
(mm)
*
2.84 0.07 2.99 -+ 0.07 2.63 f 0.04 2.61 -+ 0.03 2.46 f 0.04 3.0 t 0.07
Pollen grain/anther 1628 f 1729 -+ 1071 f 1604 ? 1407 f 1667 t
104 45 88 160 80
64
Amount of residual pollen (%) 34.1 70.0 12.5 73.2 20.5 48.2
selection and recurrent selection procedures to transfer the long stigma of the wild species 0. longistaminata to 0. sativa with some success. Sat0 et al. (1994) reported that anther length was affected by temperature and governed by a single gene. The short anther trait was incompletely dominant over the long anther trait. Their results also suggested that the earliness gene Ef-I appeared to shorten the anther length pleiotropically because no recombinants showing early heading and long anther and late heading and short anther segregated in 487 F, plants and 973 B,F, individuals of a cross involving parents showing polymorphism for anther length and heading. Despite these observations they did not rule out the possibility that anther length was governed by a different gene tightly linked with Ef-I.
D. NATURAL OUTCROSSING MECHANISM IN RICE The success of hybrid rice seed production depends on the deposition of a sufficient number of pollen grains in the stigma lobes of each spikelet of the seed parent. This is achieved through sufficient pollen flow from the pollen parents to the seed parents. Kato and Namai (1987b) studied the effect of wind velocity during flowering time on air-borne pollen (pollen load) and on seed set percentage of a CMS line and concluded that wind velocity of 2-3 m/sec and more than 15 airborne pollen grains per liter per day were sufficient for economical hybrid rice seed production. Studies conducted at IRRI (Fig. 6) showed that the number of air-borne pollen/liter was negatively correlated (r = 0.756**) with distance from the pollen source. The environmental factors influencing outcrossing in rice include temperature, relative humidity, light intensity, and wind velocity. In China, conditions favorable for normal flowering are daily temperature of 24-
s. s. VIRMANI
43 0 Number of alr-bome pollen/liter
'1
Y = 13.331 - 0.049X
2 U
15
30
45
60
75
90
105
120
135
150
165
180
Distance from the pollen source (cm)
Figure 6 Effect of distance from the pollinator to male sterile lines on pollen load.
28"C, relative humidity of 70-80%, difference between day and night temperature of 8-10°C, and sunny days with a breeze (Xu and Li, 1988). Generally, with high temperature and low humidity or with low temperature and high humidity some glumes will not open, which will reduce outcrossing. Studies conducted at IRRI (Virmani, 1994a) indicated that seed set percentage and seed yield of a CMS line were negatively correlated to relative humidity. A high seed yield was obtained at IRRI when the seed and pollen parents flowered during the end of February to early March. At this time the relative humidity was 50-60% and wind velocity was above 2.5 m/sec.
E. GUIDELINES FOR HYBRIDSEEDPRODUCTION Seed yield obtained on a male sterile line used in a hybrid rice seed production plot is a function of (i) yielding ability of the male sterile line as determined by the yielding ability of its maintainer, the fertile counter part; (ii) proportion of the male sterile line in relation to the pollen parent, and (iii) outcrossing rate of the male sterile line. Improvement in any of the above components can help to increase hybrid rice seed yields. Extensive research in China and at IRRI have led to the identification of the
HYBRID RICE
43 1
following guidelines for successful hybrid rice seed production (Yuan, 1985; Mao, 1988; Virmani, 1994a). 1. Selection of seed and pollen parents with synchronized time of anthesis. 2. Selection of seed parents with long, exserted stigma, longer duration, and wider angle of floret opening. 3. Selection of a pollen parent with a high percentage of residual pollen per anther after anther exsertion. High pollen shedding potential is attained by getting 2000-3000 spikeletdm2 to bloom per hour during peak flowering period. 4. Synchronization of flowering time of the two parents by seeding them at different dates depending on their growth duration or estimated accumulated temperature requirements for initiation of flowering. 5. Use of optimum seed parent:pollen parent row ratio such that the ratio of spikelet number per unit area of seed parent and pollen parent is about 3.5:l. 6. Use of seed and pollen parents with small and horizontal flag leaves, or cutting long and erect flag leaves. 7. Use of gibberellic acid (GA,) to improve panicle exsertion and prolong duration of floret opening. 8. Planting of seed parent pollen parent rows across the prevailing wind direction and use of supplementary pollination with a rope or stick when wind velocity is below 2.5 m/sec. 9. Selection of optimum time of flowering of parental lines in seed production plots.
F. PRACTICES FOR HYBRID SEEDPRODUCTION Hybrid rice seed production using the CMS system involves two steps, viz. multiplication of A line and production of hybrid seeds. However, production of hybrid seeds with the EGMS system involves one step, i.e., production of hybrid seed; the ms line multiplication is done by selfing, which is achieved by growing EGMS lines under suitable daylength and/or temperature conditions. Practices involved in producing hybrid rice seeds using a male sterility system include the following (Yuan, 1985; Virmani and Sharma, 1993; Virmani, 1994a). 1. Selection of an isolated field which has fertile soil and has access to adequate irrigation water and drainage. The seed production plot can be isolated from other rice varieties by distance (50-100 m), time (3 weeks), or natural or artificial barrier isolation. Isolation distance can be decreased by increasing the number of pollen parent rows around a seed production plot. 2. Differential seeding and transplanting of parental lines to attain synchronization of their flowering during optimum weather conditions.
432
s. s. VIRMANI
3. Adjusting flowering date of parental lines by 2-5 days by differential application of nitrogeneous, phosphatic fertilizers or some other chemicals, viz. 2,4-D, IAA, or Ethel (Huang et al., 1994), if panicle initiation stage of parental lines is found nonsynchronous. 4. Planting of the parental lines across the anticipated direction of wind during flowering of a seed production plot. 5 . Using the optimum seed parent:pollen parent ratio depending on the spikelet number per panicle of parental lines and the magnitude of pollen load of the pollen parent. 6. Providing good cultural management of the seed production plot by controlling weeds, diseases and insects and applying an optimum fertilizer rate. 7. Clipping of flag leaves by one-half to two-thirds if the parental lines have long and/or erect flag leaves obstructing the pollen movement. 8. Spraying GA, at 45 g/ha using a knapsack sprayer or 15 g/ha using an ultralow volume (ULV) sprayer in two splits (the first at 1 5 2 0 % heading and the second 2 days later). Spraying is done on a calm, sunny day, usually in the afternoon, and is avoided if rain is expected within 24 h. 9. Shaking the panicles of the pollen parent by rope-pulling by two persons or stirring the canopy layer of the pollen parent rows with a bamboo stick by one person, taking care not to break off the panicles at the neck. 10. Roguing at maximum tillering, flowering, and maturity stages to prevent off-types from cross-pollination with true-to-the-type seed parent plants. 11. Harvesting, threshing, drying, and cleaning the pollen and seed parent separately.
By using the above practices hybrid seed yields of 1-2 t/ha can be obtained. These practices are constantly under review by seed growers and hybrid seed production experts. In China, these have been modified from time to time to increase outcrossing and seed yields (Table XXII). By introducing these modifications average seed yields in hybrid rice seed production plots have improved from 0.67 t/ha during the late 1970s to 2.6 t/ha during recent years. In countries where the hybrid seed industry has not been established hybrid rice technology is considered difficult to adopt. For such situations, Virmani et al. (1993a) proposed a self-sustaining system for hybrid rice seed production by which hybrid rice cultivators can produce their own hybrid seeds using 8- 10 days time isolation and no distance isolation. About 5% of the area of a rice farm is used for producing the next season’s hybrid rice seed; the rest is used for growing the commercial hybrid. The seed production plot is planted so that it flowers 810 days after the main crop flowers to reduce contamination from the pollen of commercial hybrids or inbred varieties. Locating the plot in the comer of the farm lessens the chance of contamination from the neighboring farm. The pollen parent is planted in a 3-m-wide strip (15 rows 20 cm apart) around the seed production plot; it covers about 50% of the plot. The actual F, seed production
43 3
HYBRID RICE Table XXII Progressive Modification of Techniques of Hybrid Rice Seed Production in China during the 1970s to 1990s to Increase Seed Yields (R. C. Yang, Personal Communication) Period
Technique
Row ratio Width of a female Parent strip (m) Plants/hill A line R line Transplanting density (cm) A line R line Hillslha ( X 1000) A line R line Plantslha ( x 1OOO) A line R line Plants/ha ( X IOOO) Dosage of GA, (g/ha) Concentration (ppm) Seed set percentage Clipping flagleaf Yield (t/ha)
Late 1970s
1-2:6-8 1.5
Early 1980s
Late 1980s and early 1990s
2 : 10-12 2.0
2 : 14-16 2.5
2
2 2
1
16.5 X 16.5 20 x 20 320 67 320 67 2100 0-45 20-40 10-15 Yes 0.67
13.3 X 13.3 20 x 20 450 50 900 50 2700 75- I20 40-60 25-30 Yes or no 2.10
12 17
X X
13.3 30
530 27
I060 54 3000 120-180 100 45-50 no 2.60
plot is with A (CMS):R lines in the row ratio found optimum (e.g., 10:3 at IRRI) in the area (Fig. 7). The R line strip provides an abundant pollen load to minimize contamination from other pollen sources (viz. the commercial hybrid rice crop or a neighbor’s inbred variety). Recommended seed production practices, as described earlier, are followed to achieve high outcrossing on the A line resulting in a high seed yield. At the rate of 1 t/ha seed yield, hybrid seeds produced in the self-sustaining seed production net plot (area of 250 m2) can amount to 25 kg which is enough to cover a 1-ha farm area in the following season. The R line can usually yield comparably to an inbred check variety. Using this procedure farmers need to buy only 400 g of the A line and 1 kg of the R line seed every season. The latter is not essential, but is advisable to ensure seed purity. The public seed farms in such countries can produce the required quantities of foundation seeds of A and R lines for selling to the farmers, who can adopt this technology with some training organized by the national program.
s. s. VIRMANI
434
Hybrid seed production plot
Rice farm ( 1 ha)
-
1 5 rows
-
... ... ... ... B.D .D. .B.
mmm
0 0 0 0 0
0 0 0 0 0
00000000 00000000 00000000 00000000 00000000
oooooooooo
0000000000 0000000000 0000000000 0000000000 ..D 0000000000 0000000000 ..D 0000000000 m m ~o o o o o o o o o o 0000000000 m ~ mo o o o o o o o o o mmm oooooooooo B..
...
oooooooooo 0-3 A to R ratio
mmm
3mt
L -
3m
3m
15 rows
15 rows
... ... ... ... ..B
mmm
15 rows
B.D
.BB
..a mmm m a. mmm
3m
I
&
20 m
..a
25 m
i
R lines
0 A lines
.5 X 1 5 c m spacing
Hgure 7 Proposed field layout of a self-sustaining system for hybrid rice seed production (Virmani et a/.. 1993a).
G. PERFORMANCE OF HYBRID SEEDPRODUCTION Since commercial production of hybrid rice began in China in 1976, tremendous efforts have been made to improve seed production technology. In 1992, 187,000 ha was used for hybrid rice seed production in China with an average yield of 2.25 t/ha; Hunan province had 22,000 ha with an average yield of 2.7 t/ha (Huang et al., 1994). In Lingling Prefecture of Hunan province an area of 2400 ha in seed production gave an average yield of 3.2 t/ha. The highest seed yield reported from a seed production plot in Hunan province was 5.8 t/ha with 66.7% seed set percentage on the seed parent. The high seed yields have been associated with skillful spraying and increased dosage of GA,, discontinuation of flag leaf clipping, increasing seed parent:pollen parent ratios, closer transplanting with multiple seedlings, and supplementary pollination at the peak time of anther dehiscence (Huang et al., 1994). Hybrid seed production plots outside China have given seed yields ranging from 0.2 to 2.5 t/ha (Table XXIII). Results from IRRI (Fig. 8) indicated that seed yields of the CMS line IR58025A improved from 153 to 2050 kg/ha between
43 5
HYBRID RICE Table XXIII
Hybrid Rice Seed Yields Obtained on IRRI-Bred CMS Lines in Some Countries, 1991-1994
Country
Location
Year
Yield (t/ha)
Reference
Philippines
PhilRice Cagayan Hyderabad Mandya Delhi Kamal Hyderabad
1991 1992-93 1990-9 1 1990-9 1 1990-91 1990-91 1993 1994 1993 1993 1994 1993 1994 1993 1994 1994 1992-94 1990-91
1.5-2.2 0.4-2. I 0.7-1.7 I .O-2.5 I .4-2.0 1.9-2.1 0.8-1.3 0.6-0.9 0.4-0.6 0.5-0.9 0.5- 1.1 1.O- 1.5 1.5-2.4 0.8- 1.4 0.6-0.8 0.2-0.4 0.7-1.6 1.2-1.5
Lara et al. (1994) Lara (personal communication) Siddiq er al. (1994) Siddiq et al. (1994) Siddiq el a / . (1994) Siddiq er al. (1994) Anonymous (1995) Anonymous (1995) Anonymous (1995) Anonymous (1995) Anonymous (1995) Anonymous (1995) Anonymous (1995) Anonymous (1995) Anonymous (1995) Anonymous (1995) Vidyachandra er al. (1995) Luat eral. (1994)
India
Bangalore Coimbatore Pantnagar
Vietnam
Karnal Maruteru Faizabad Mandya Mekong
1989 (when the line was developed) and 1994, with increased familiarity and experience with the CMS line. Besides, seed yields were higher in dry seasons than in wet (monsoon) seasons. Virmani et al. (1991) studied the effect of flagleaf clipping, GA, application, and supplementary pollination individually and in all possible combinations on hybrid seed yield; GA, application was found most effective. They also observed the effect of locations on the quality of hybrid seeds. Seeds produced at IRRI (with high rainfall, high relative humidity, and high wind velocity during flowering) had a higher incidence of discoloration (caused by Sarocladinum oryzae, curvularia spp., Trichoconiella padwickii, andlor Gerlachia oryzae) than those produced at Cagayan Valley Experiment Station, San Mateo, Isabela (relatively lower temperatures and relative humidities). The extent of synchronization of flowering of seed and pollen parents affected seed yields considerably (Virmani er a l . , 1993). Studies on the effect of ma1e:female row ratio and distance from the pollinator on seed set and seed yield of CMS lines at IRRI (Sharma and Virmani, unpublished) showed that a 10:3 A:R ratio was optimal considering the yield of A and R lines (Figs. 9 and 10). Seed yields of the seed parent were found to be directly proportional to increases in their area (ranging from 16 to 78%) after which these declined (Fig. 11). The above information provides clear evidence that bulk quantities (1 to 2
Figure 8 Yield of hybrid rice seed production plots involving IR58025A as seed parent (1989-1994, IRRI).
43 7
HYBRID RICE Seed yield (t/ha)
2.5 2 1.5
1 0.5 0
CMS/IR 1 row of R
I
I
CMS/3R 3 rows of R
A
7
LSD .06 (restorer) = 0.260
CMS/4R
A 4 rows of R I
1
I
6
5
0
I
I
8
10
9
I
I
11
12
Number of female rows
Effect of row ratio on seed yield of CMS IR58025A and restorer IR29723R (IRRI,
Fzgure 9
1991 DS).
t/ha) of hybrid rice seed can be produced in tropical, subtropical, and temperate countries by using suitable seed and pollen parents and by adopting appropriate seed production procedures. Seed yields can even be increased beyond 2 tlha by developing seed and pollen parents with higher outcrossing potential and higher
Seed yield of restorer (t/ha)
7 LSD (0.05)= 0.18 t/ha
0
n -
1
1 row of restorer
o 2 rows of res
I
1
I
I
2
3
4
6 7 8 Number of female rows 5
I
I
I
I
I
9
1 0 1 1 1 2
I
Figure 10 Effect of number of female rows on seed yield of restorer in hybrid seed production plots (IRRI,1991 DS).
s. s. VIRMANI
43 8 Seed yield (t/ha)
2'5
I
1.0 -
16.6
25.0
37.0
50.0 60.0
69.2
73.3
83.0
90.9
92.3
Female area (percent)
Figure 11 Effect of female area on seed yield (IRRI, 1991 DS).
pollen load, respectively, and by further improving seed production technology. The self-sustaining system for hybrid rice seed production (Virmani et al., 1993) has been tried at IRRI during the past 3 years; seed yields of up to 1.2 t kg/ha were obtained. Seeds so produced gave uniform hybrid crops comparable with ones raised from seeds produced in isolated seed production plots. Lu et al. (1982), Shi et al. (1983), and Glaszmann et al. (1987) have indicated the utility of isozyme markers to determine the purity of hybrid rice seeds.
XI. ECONOMIC ANALYSIS Major concerns generally expressed about hybrid rice technology pertain to its economic viability. He et al. (1984, 1987a) conducted household level studies in China and found that the yield advantage of hybrid rice over the semidwarf inbred rice was about 1 t/ha (i.e., 15%), without major differences in material costs and labor requirements. These results were substantiated further by Lin (1990) who conducted a survey of 500 farm households in Hunan for the 1988 crop season. While the mean yields of hybrids were significantly higher than conventional varieties for middle and late crop season rice, the difference was not statistically significant for early season rice (Table XXIV). The yield advantage of hybrid rice was found to be partly offset by added requirements for chemical inputs and more expenditure for seeds. Hybrid rice did not require more
Table XXIV Means (and Standard Deviation) of Inputs and Outputs for Hybrid and Conventional Rice (Lin, 1990) ~
~~~~~~~
~~
~~
Early season rice Conventional ( n = 392) Seed (kglha) Fertilizer cost (Yuanlha) Pesticide cost (Yuanlha) Labor (dayslha) Draft animals (days/ha)
173.13 (3.7) 313.43 (9.4) 76.12 (4.7) 228.36 (6.4) 22.39 (1.1)
Machine (dayslha) Rice output (tlha)
7.46 (0.8) 5261.18 (97.3)
Hybrid 13)
(n =
46.27 (2.2) 410.45 (7.9) 108.96 (2.7) 253.73 (3.8) 28.36 (1.0)
19.40 (1.4) 5752.24 ( 172.1)
Midseason rice Conventional ( n = 34)
102.99 (3.1) 222.39 (7.9) 61.19 (3.5) 332.84 (13.6) 44.78 (3.0) 20.90 (6.9) 4029.85 (117.8)
Hybrid (n = 116)
29.85 (0.9) 282.09 (10.4) 92.54 (4.1) 302.99 (7.4) 55.22 (2.1) 0.90 (0.31) 6455.22 ( 124.2)
~
Late-season rice Conventional ( n = 213)
92.54 (3.3) 338.81 (13.2) 85.07 (3.9) 207.46 (5.8) 16.42 (1.1)
8.96 (0.8) 4829.85 (90.2)
Hybrid ( n = 308)
29.85 (1.1)
368.66 (10.7) 108.96 (5.4) 214.93 (5.4) 16.42 (0.7) 8.9 (0.9) 5770.15 (85.3)
440
s. s. VIRMANI
labor inputs than inbred rices. In the final analysis increased yield did overcompensate for the extra investment on seeds and chemical inputs. Thus increased yield of rice hybrids interested both the individual farmer and the government in China. Widespread use of hybrid rice in the country is, therefore, understandable. However, the adoption rate of hybrid rice technology did show dramatic geographic differences which were attributed to differences in agricultural research infrastructure, adaptive research investments, and the share of rice in provincial agricultural output (Lin, 1990). It was also observed that agricultural research infrastructure in China was concentrated mostly in those provinces which have high population densities. These provinces tended to concentrate on research to increase yield per unit of land. Provinces with high labor to land ratios also have high levels of irrigation infrastructure and have high levels of modem variety adoption. Thus the supply of yield-increasing technologies tend to be positively associated with 1abor:land ratios. The proportion of rice area to total cultivated area in a province was also an important determinant of profitable hybrid seed production. The higher the portion of rice area, the lower the marketing costs for seed companies and, therefore, the higher the availability of hybrids (Lin and Pingali, 1994). An economic evaluation of hybrid rice seed production has been reported for China (He et al., 1987b, 1988), India (Govinda Raj, 1993), and the Philippines (Lara et al., 1994). All studies showed that hybrid rice seed production is profitable, if hybrid seed yields range between 1.5 and 2 t/ha. Similarly, economic evaluation of hybrid rice cultivation reported by He et al. (1984, 1987a, 1988) in China and Govinda Raj (1993) and Anonymous (1995) in India also shows a favorable cost:benefit ratio from the use of the technology. Surveys made in China showed that hybrid rice production fitted well in the farming system: it gave more grain as well as straw. The straw is used as cattle feed as well as for manure to improve soil texture and fertility. Because hybrid rice increases grain output per hectare, farmers can use more of their land to grow cash crops, such as oil seeds, cotton, and watermelon, and they may also engage in poultry and livestock farming (He et al., 1988). In southern Jiangsu and riverine areas, which were double rice-triple cropping areas before, the dominant cropping pattern has changed to hybrid rice-wheat (He et al., 1988). Under hybrid rice production, the harvesting date of rice moved from late October to early October, so that the wheat crop could be sown by the third week of October instead of mid-November. This increased wheat yields from 1.5 to 3.0 t/ha in the Zhen Jui region of Jiangsu. Thus hybrid rice has made an important contribution to grain production and enhancement of economic efficiency in China. During recent years, the area under hybrid rice in China has decreased by about 2 million ha; this has been attributed to increased consumer preference for japonica rice which commands a higher price in the market compared to indica
HYBRID RICE
441
rice hybrids. Besides, the government has discontinued subsidies on hybrid rice seed production which increased the cost of hybrid rice seeds, and for a few years seed companies produced smaller quantities of hybrid rice seeds (C. X. Mao, personal communication). This situation is, however, temporary because the higher seed yields (greater than 2 t/ha) would reduce unit seed production cost, which should reduce the impact of subsidy withdrawal. He et al. (1988) also pointed out that hybrid rice cultivation and seed production are labor intensive; the cost of labor accounted for 46% of the total cost. With the modernization of China, the country needs to free labor from cropping and transfer it to forestry, animal husbandry, fisheries and other rural enterprises. Therefore, He et al. (1988) recommended that hybrid rice research in China should aim toward mechanization to save labor inputs. From the economic analyses made on hybrid rice in China (Lin and Pingali, 1994) the following conclusions were made with regard to prospects of hybrid rice in countries outside China. 1. The technology would be relevant for those countries where the rice supply is projected to fall short of the rice demand with the increasing population and 1and:labor ratio. 2. It would be acceptable for irrigated rice ecosystems where yield gains associated with high-yielding inbred varieties have been exhausted. The proportion of irrigated rice land in a country is therefore an important determinant of the potential of hybrid rice. 3. Since the relative profitability of hybrid rice over conventional rice is determined by the ratio of rice price to hybrid seed price and given the high labor requirements in seed production, it is anticipated that in the first instance the technology will receive serious attention in those countries where wages are lower due to high 1abor:land ratios. 4. In tropical Asia, countries with a high 1abor:land ratio and a high proportion of irrigated area (e.g., India, Indonesia, Philippines, Sri Lanka, and Vietnam) are likely to have the highest potential demand for hybrid rice technology. On the other hand, countries with a high proportion of irrigated area but a low 1abor:land ratio (e.g., Malaysia and Pakistan) would not find hybrid rice production profitable, since agricultural wages would be relatively higher. Countries with a low proportion of irrigated rice area (e.g., Bangladesh, Nepal, Myanmar, and Thailand) would have a low potential demand for hybrid rice technology, irrespective of wage rates. Although in recent years the irrigated rice area in Bangladesh appears to have increased especially in boro season, this makes Bangladesh also suitable for development and use of hybrid rice technology. 5. With regard to potential supply of hybrid rice technology, India (with 31% of the total irrigated area in tropical Asia) and Indonesia (with 13% of the total area in tropical Asia) would be the most important suppliers of hybrid rice
442
s. s. VIRMANI
technology. Given the size of the market, seed producers in these countries would benefit from substantial economies of scale. Scale economies would accrue in the provision of hybrid seed adapted to specific ecological regions and input supplies, especially GA, and technical skills. 6. Region-specific research and seed production infrastructure for hybrid rice can be economically feasible to set up in large countries, viz. India and Indonesia; however, region-specific institutions would not be cost effective in countries with smaller irrigated areas. Such countries could possibly benefit from their neighbors’ efforts. Sri Lanka, for instance, could benefit from technological spillovers from hybrid rice research conducted in southern India. Similarly, Bangladesh could benefit from the efforts in eastern India and Malaysia, and from the efforts made in West Java and Sumatra in Indonesia. 7. With regard to the input supplies, unit cost of GA,, an important component of hybrid rice seed production technology, would decrease with an increase in the scale of production. 8. Before making a large scale commitment to hybrid rice-related infrastructure, a detailed assessment of the land and labor requirements ought to be made. In countries like India, given the low wages, high rural unemployment, and high price of rice, it may make sense to pursue the technology. However, in the case of Thailand, with high and increasing wage rates and the high opportunity cost of irrigated land for growing high-value crops for export, hybrid rice may not be the best option. Swaminathan (1993) stated that hybrid varieties in any crop have three distinct advantages. First, they increase average yield; second, they provide more skilled jobs in seed production in rural areas; and third, they also provide a tool to improve the crop management practices at the farm level.
XII. TECHNOLOGY TRANSFER AND POLICY ISSUES Transfer of a hybrid technology to the farmers is closely associated with the seed industry. Historically, the seed industry in industrialized countries has expanded by developing and selling hybrids. Nearly 40% of the global commercial seed business of about U.S. $15 billion is accounted for by hybrid sales of various crops (Sehgal, 1992). In tropical Asian countries where the prospects of hybrid rice are being explored, the seed industry is either in its embryonic stage of development (e.g., Indonesia, Philippines, Vietnam) or just beginning to enter the growth phase (e.g., India, Thailand). In recent years, several new national and multinational seed companies have also entered the seed business in developing countries due to two key factors (Sehgal, 1992).
HYBRID RICE
443
1. Greater availability of hybrids for several crop species. 2. Governments’ easing of policies pertaining to private sector participation in the seed industry.
Among the tropical rice growing countries which have released rice hybrids (e.g., India, the Philippines, and Vietnam), India has a reasonably adequate seed industry infrastructure in the public as well as the private sector to undertake production, processing, certification and distribution of hybrid rice seeds. Vietnam does not have an adequate seed industry set up, but it is mobilizing resources in the public and collaborative public-private sector to handle hybrid rice seeds. The Philippines has a private company modestly involved in hybrid rice research and development and the government has also started working with seed cooperatives to produce hybrid rice seeds. Indonesia also has a reasonably good seed industry infrastructure in both the public and private sectors which can be mobilized after suitable rice hybrids are identified by plant breeders. Sri Lanka and Bangladesh plan to develop their seed industries during the next 5 years. It will take at least this much time to develop suitable rice hybrids in these countries because hybrid rice research in these countries has just been initiated. Both countries can also benefit from the hybrid breeding research and seed industry operating in India. Experience in other crops has indicated that the seed industry in Asia is capable of providing quality seeds to farmers as they need it. The same will be true of hybrid rice (Sehgal, 1992). Given the prominence of rice in Asia, it is also believed that rice hybrids can have the same catalytic effect on the development of the region’s seed industry that hybrid corn had on the seed industry development in North America and hybrid sugar beet had on the seed industry development in Europe (Sehgal, 1992). Outside tropical Asia, prospects of hybrid rice are also being explored in Japan, Korea, the U.S.A., Brazil, Colombia, and Egypt. In Japan and Korea, major constraints in technology transfer are not the absence of seed industry infrastructure but the nonavailability of heterotic rice hybrids, possessing excellent japonica grain quality, and the high cost of labor to produce their seeds. In the U.S.A., where both japonica and indica rices are grown, heterotic rice hybrids with acceptable grain quality and mechanized seed production are desired before this technology can be transferred at the farm level. In Brazil, Colombia, and Egypt, the technology is still in the early stage of development; it may take at least 5 years to develop the suitable technology before it can be transferred to the farmers for commercial scale adoption. The pace with which hybrid rice technology transfer will take place will depend on the extent of Research investment in NARS, through participation of public and privatesector institutions, to breed heterotic rice hybrids and develop seed production technology which is economically viable and commercially acceptable.
444
s. s. VIRMANI Training facilities in hybrid seed production. Administrative coordination between research and development organizations dealing with the technology to balance the demand and supply of hybrid seeds. Government support to the seed industry to enable them to carry, if necessary, large stocks of hybrid seeds from one seasodyear to another. Establishment of a clear government policy on Intellectual Property Rights.
Policy makers need to provide incentives for development and expansion of the seed industry as is currently happening in several developing countries. The private sector should be taken as a partner, rather than an adversary, in an effort to develop the seed industry in developing countries. After all, a good seed industry is useful not only for hybrid varieties but also for inbred varieties.
XIII. FUTUREOUTLOOK Hybrid rice is already a great success in China and during the next few years some other countries, viz. India, Vietnam, and the Philippines also expect to benefit from it. The technology is useful not only due to its higher yield potential, but also because of its capability to increase rural employment opportunities through a hybrid seed industry. The potential of this technology is higher in countries with a high proportion of irrigated rice area and a high 1and:laborratio (Lin and Pingali, 1994). In countries with large irrigated areas, low 1and:labor ratios, and high wages, this technology would be commercialized after seed yields of 2.5-3 t/ha are obtained, seed production is mechanized, and seed cost is reduced. The current level of heterosis (15-20% or 0.75-1 t/ha) in indica rice hybrids is economically viable, but a higher level, observed in indica japonica hybrids, will be more attractive. This would require tackling certain associated problems, viz. intervarietal hybrid sterility, taller stature, late maturity, and grain quality. Some indica/japonica rice hybrids possessing 10- 15% higher yield than indica/indica and japonica/japonica rice hybrids are in on-farm demonstration trials in China; in tropical countries such hybrids can be developed during the next 5 years from indica/tropical japonica crosses. In countries such as Japan, Korea, and Egypt, which require japonica rice hybrids with high grain quality, hybrids derived from IRRI-bred tropical japonica lines (Khush and Aquino, 1994) and locally bred temperate japonica lines may show better performance than temperate japonica rice hybrids. Significant heterosis, commonly observed for vegetative vigor and root characteristics, suggests that hybrid rice should also be explored for certain stress
HYBRID RICE
445
environments (viz. rainfed, lowland, drought-prone) where transplanting is practiced or some seeding equipment is available for direct seeding with a reduced seed rate. Seed cost can also be reduced by taking a ratoon crop of hybrid rice because hybrids tend to ratoon better than inbreds (Chauhan et al., 1983; Siddiq et al., 1994). Therefore, prospects of ratooning should also be explored in rice hybrids, especially in those areas where incidence of virus diseases and insects is low. Hybrid rices have also shown remarkably good performance in boro season in Eastern India and Bangladesh where the crop is affected by low temperature at the vegetative stage. It is, therefore, important to develop hybrid rice technology for this ecosystem. The prospects of rice hybrids for increasing water-use efficiency in irrigated rice should also be studied further. Per se performance, genetic diversity, and the combining ability of parental lines are generally used to predict heterosis. However, with the availability of molecular marker technologies in rice, it is easier to determine genetic diversity among prospective parental lines and tag heterotic gene blocks with molecular markers and subsequently transfer these in selected parental lines to improve their combining ability. Physiologically, rice hybrids have been found to be more efficient than inbred parental lines and their growth pattern is also different in several ways. However, their yield potential is usually compared with inbred varieties grown side by side with the agronomic management practices recommended for inbred lines. This may not be suitable for the full expression of yield potential in hybrids. It is, therefore, necessary to study critically if hybrid rices require different or similar agronomic management to maximize their yield expression in comparison to inbred rices. The CMS system has been found to be the most effective and practical for developing rice hybrids; several CMS sources have been identified, but only a few of them have been deployed to develop commercial rice hybrids. It is important to critically analyze the available sources for their utility in developing commercial rice hybrids and more and more genetically diverse CMS sources should be deployed for breeding commercial rice hybrids. Although there is no dearth of restorers among indica rice elite cultivars bred by various international and national rice improvement programs around the world, restorers among japonica rices are scarce, at least for CMS-WA and CMS-Bo cytoplasm. It would be extremely useful for japonica rice hybrid breeding if a new CMS source could be identified for which sufficient restorers are available among japonica rice cultivars. Meanwhile, breeding and purification of japonica restorers can be done expeditiously by deploying anther culture techniques to indica/japonica crosses made for this purpose. Identification of a new CMS system and transfer of an available CMS system into the genetic background of maintainer lines involves five or six backcrossings to develop stable CMS lines. This is a long and cumbersome process. The
446
s. s. VIRMANI
success in protoplast culture in rice makes it possible to produce cybrids which enable immediate transfer of cytoplasmic male sterility into elite rice cultivars (Akagi and Fujimura, 1994; Brar er al., 1994). This approach should be explored widely to develop genetically diverse CMS lines more expeditiously. Efficiency of breeding of maintainer and restorer lines in indica rice can also be increased by developing genetically diverse maintainer and restorer populations using male sterility-facilitated recurrent selection (Virmani er al., 1994). Such populations are being developed at IRRI for sharing with national programs on a regular basis to enable them to select locally adapted improved maintainer and restorer lines, respectively and developing suitable CMS lines and pollen parents from them. Deployment of the two-line method of hybrid breeding using the TGMS system should also increase the pace of breeding heterotic hybrids. For indica/japonica and japonica/japonica hybrid breeding, the two-line hybrid breeding system will have a decided advantage over three-line hybrid breeding since the former does not require restorers as pollen parents of hybrids. Deployment of wide compatibility in breeding indica/japonica hybrids is essential; however, selection for this trait in a segregating breeding population is cumbersome. Tagging of WC genes with molecular markers to exercise marker-aided selection for this trait will be extremely useful. Similarly, tagging of restorer genes and TGMS genes with molecular markers and exercising marker-aided selection for these traits will also increase efficiency of hybrid breeding. Research is in progress on genetically engineered nuclear male sterility in rice at the Japan Tobacco Company in collaboration with Plant Genetic Systems (PGS) of Gent, Belgium, using the chimerase ribonulease gene TA 29 isolated and cloned from the tobacco genomic library (Mariani et al., 1990) and in several other research laboratories funded by the Rockefeller Foundation’s Rice Biotechnology Program. If successful, this would provide an additional genetic tool for hybrid rice breeding. Rice hybrids with desired disease and insect resistance and acceptable grain quality can be developed by an appropriate choice of parental lines. Therefore, while selecting parental lines, their level of disease/insect resistance and grain quality should be monitored critically. In order to develop rice hybrids possessing premier grain quality such as basmati, both parents of the hybrids must possess basmati quality (viz. long slender grains, high grain elongation ratio, intermediate amylose, and strong aroma). The low frequency of restorer lines among basmati lines is an important handicap which can be overcome by using the TGMS system to develop basmati rice hybrids. Although rice is a self-pollinated crop, significant cross-pollination occurs on male sterile plants, depending on their flowering behavior, floral characteristics, amount of pollen availability from the pollen parent, and prevailing weather conditions. Hybrid seed yields up to 5 . 8 t/ha have been reported in China and up to 2.5 t/ha have been reported in other countries. Experience from China has
HYBRID RICE
447
shown that both genetic improvement for flowering behavior and floral traits of seed and pollen parents and modification in seed production practices are useful in increasing seed yields beyond 2 t/ha. These strategies should be adopted to increase hybrid seed yields in other countries. GA, application is an important component of hybrid rice seed production technology. Its cost is prohibitive in many countries. Therefore, incorporation of the elongated uppermost internode gene in male sterile lines should be useful in improving their panicle exsertion without GA, application. Seed discoloration of hybrid rice seeds caused by several fungi occurring in the tropics is an important problem which needs to be tackled for packaging the seed production technology. In China and northwestern India, CMS lines in hybrid rice seed production plots have been found to have a higher incidence of seed-borne diseases (such as paddy bunt, caused by Neovassia horinda, and Tak and false smut caused by Ustilogonoids virens) compared to pollen parents. This can cause a serious outbreak of these diseases on commercial crops of hybrid rice and therefore needs closer attention. For countries (such as Japan, Korea, Malaysia, U.S.A) having a low 1abor:land ratio and high wages, hybrid rice seed production needs to be mechanized to make it cost effective and economically feasible. Maruyama and Oono (1983) proposed the use of a facultative female sterile line as pollinator for mecharl.zed hybrid rice seed production involving mixed planting of seed and pollen parents. Seed of a facultative female sterile can be produced in the environment in which there is a partial seed setting. A recessive mutant sensitive at complete flowering to the herbicide bentozone (3-isoprophyl-2,1,3-benzothiadiadin-4-one-2-2-dioxide) was identified by Mori (1984) and proposed for incorporation into the pollen parent which would be killed by application of the herbicide after flowering of the seed production plot (Maruyama et al., 1991b). This would enable mixed planting and mechanized harvesting of hybrid seed. Use of highly heritable grain width differences (at least 0.7 mm) among seed and pollen parents for mixed planting and mechanized harvesting has also been suggested by Maruyama et al. (1991b). The F, seed borne on the seed parent can be separated from the pollen parent by circular pore sieves. Some rice varieties possess a phenol reaction (Ph) gene by which paddy grains, treated with the solution of phenolic compounds, viz. phenol, catechol, hydroquinone, pyrogallol, and tyrosine, become uniformly black. If the Ph gene is incorporated into the pollen parent and the seed parent is devoid of this gene, both pollen and seed parents can be harvested mechanically from a hybrid seed production plot, the bulk harvest can be treated with a phenolic compound, and after the staining treatment, the mixture of brownish-whitegrains and black grains can be separated by a color sorter machine (Virmani and Maruyama, 1991). For countries lacking suitable seed industry infrastructure, the self-sustaining seed production system (Virmani et al., 1993a) can be tried.
448
s. s. VIRMANI
Apomixis is the ultimate genetic tool to develop true breeding hybrids and facilitate commercial exploitation of heterosis by even resource poor farmers. So far, there is no confirmed report of the occurrence of apomixis in rice, but research is in progress in China, at R R I , and elsewhere to discover, induce, and/or genetically engineer such rices. Now that hybrid rice has proven to be useful in increasing rice varietal yields it is important to accelerate development and use of this technology to the fullest extent possible. Strategies useful for the purpose (Virmani er al., 1994) are: Identification of target areas and environments. Development of suitable infrastructure for research. Establishment and/or strengthening of the seed industry in public and/or private sectors and its linkage with hybrid rice breeding centers. Establishment of international collaboration mechanisms, such as an International Task Force on Hybrid Rice.
XIV. CONCLUSIONS The use of hybrid rice is a strategy to lift the yield ceiling of rice to help the world meet the future projected demand, which will increase due to increasing populations and rising incomes. The technology has already made an impact on rice production in China during the past 20 years, where rice hybrids possessing a 15-20% yield advantage over inbred rices are cultivated in about 50% of the total rice area. In recent years, similar hybrids have also been developed for India, Vietnam, and the Philippines, and several other countries are involved in developing the technology for irrigated rice ecosystems. Rice hybrids with even higher yield potential can also be developed by exploiting heterosis in intervarietal group crosses (viz. indicakemperate japonica, indica/tropical japonica, temperate japonica/tropical japonica). In addition to an irrigated ecosystem, the technology may also be adaptable to certain unfavorable irrigated ecosystems affected by low temperatures at the vegetative stage (boro season in Eastern India, Bangladesh) and a rainfed lowland ecosystem affected by drought. In order to maximize the yield of rice hybrids, their agronomic management must also be considered. Although CMS is the most effective genetic tool to breed the currently grown heterotic rice hybrids, PGMS and TGMS systems appear to hold promise for the future. Wide-compatibility genes are also available to develop intervarietalgroup rice hybrids with an enhanced level of heterosis. A search for apomixis in rice has been initiated with the objective of developing true breeding rice hybrids which would be used even by resource-poor farmers. Biotechnological tools
HYBRID RICE
449
(viz. anther culture, protoplast fusion and culture, molecular-marker-aided selection, genetic transformation) hold promise to help increase the efficiency of hybrid rice breeding procedures. Although rice is a self-pollinated crop, significant cross-pollination occurs on male sterile lines, which has enabled the development of economically viable hybrid rice seed production systems yielding up to 1-3 t/ha in large-scale seed production plots. Prospects of mechanization are also being explored in Japan, the U.S.A., and Brazil to improve the efficiency of the currently available hybrid seed production technology. Encouraged by these developments, several private and public sector seed companies around the world have started investing in hybrid rice breeding and/or seed production to help in technology transfer. An economic analysis indicates that countries with a high 1abor:land ratio and a high proportion of irrigated area (e.g., India, Indonesia, Philippines, Sri Lanka, and Vietnam) are likely to have the highest potential demand for hybrid rice technology. This is confirmed by the fact that India, Vietnam, and the Philippines have already released some rice hybrids for commercial cultivation and Indonesia and Sri Lanka are actively involved in developing this technology. Several other countries around the world are also exploring prospects and problems of this technology. The success of their efforts will be determined by the extent of their commitment and support to the development of locally adaptable and acceptable rice hybrids and seed production technology. An international task force on hybrid rice involving interested national and international research and seed production programs (in both public and private sectors) will expedite development and use of this technology. Thus hybrid rice holds great promise to increase rice production and contribute significantly in meeting the projected global rice demand in the 21st century.
REFERENCES Agata, W. 1990. Mechanism of high yield achievement in Chinese F, rice compared with cultivated rice varieties. Jpn. J . Crop Sci. 59(extra 1):270-273. [Japanese] Akagi, H., and Fujimura, T. 1994. CMS transfer into japonica varieties with cybrid method. In “Hybrid Rice Technology: New Developments and Future Prospects” (S. S. Virmani, Ed.), p. 289. Int. Rice Res. Inst., Manila, Philippines. Akagi, H., Sakamoto, M., Negishi, T., and Fujimura, T. 1989. Construction of rice cybrid plants. Mol. Genet. 215: 501-506. Akbar, M., and Yabuno, T. T. 1975. Breeding for saline resistant varieties of rice. 11. Response of F, hybrids to salinity in reciprocal crosses between Jhona 349 and Magnolia. Jpn. J. Breed. 25(4):215-220. Akita, S. 1988. Physiological bases of heterosis in rice. In “Hybrid Rice,”pp. 67-77. Int. Rice Res. Inst., Manila, Philippines. Akita, S . , Blanco, L., and Virmani, S. S. 1986. Physiological analysis of heterosis in rice plant. Jpn. J . Crop Sci. 65(extra 1):14-15.
450
s. s. V
I M 1
Amornsilpa, S., Potipibool, S., and Noojoy, S. 1994. Hybrid rice research in Thailand. In “Hybrid Rice Technology: New Developments and Future Prospects,” (S. S. Virmani, Ed.), pp. 213216. Int. Rice Res. Inst., Manila, Philippines. Anandakumar, C. R., and Subramaniam, S. 1992. Genetics of fertility restoration in hybrid rice. Theor. Appl. Genet. 83994-996. Anonymous. 1988. Japanese scientists report prospects of indicaljaponica hybrids using wide compatibility genes. Hybrid Rice Newslett. 2(2): 1. Anonymous. 1989. Selected indicators of food and agricultural development in Asia Pacific Region, 1978-88. RAPA publication 1989/7, Food and Agriculture Organization, Rome. Anonymous. 1993. Rice research in a time of change: IRRI’s medium term plan for 1994-1998. Int. Rice Res. Inst., Manila, Philippines. Anonymous. 1995. Development and use of hybrid rice technology. Annual Report 1993-94, Work plan 1994-95, Directorate of Rice Research, Hyderabad, India. Araki, H., Ikehachi, H., Toya, K., and Matsumoto, S. 1990. Development of wide compatibility rice line Norin PL 9, Jpn. Agri. Res. Q . 24:78-81. Araki, H., Toya, K., and Ikehashi, H. 1988. Role of wide compatibility genes in hybrid rice breeding. In “Hybrid Rice,” pp. 79-83. Int. Rice Res. Inst., Manila, Philippines. Athwal, D. S., and Virmani, S. S. 1972. Cytoplasmic male sterility and hybrid breeding in rice. In “Rice Breeding,” pp. 615-620. Int. Rice Res. Inst., Manila, Philippines. Azzini, L. E., and Rutger, J. N. 1982. Amount of outcrossing on different male steriles of rice. Crop Sci. 22:905-907. Bashaw, E. C. 1980. Apomixis and its application in crop improvement. In “Hybridization of crop plants” (W. R. Fehr and H. H. Hadley, Eds.), pp. 45-63. American Society of Agronomy, Wisconsin, USA. Bassi, Geeta, Joshi, D. P., and Gill, S. S. 1992. Studies on some important floral traits of CMS and restc er lines of hybrid rice. Seed Res. 20(2):74-77. Bergounioux-Bunesset, C., Perennes, C., and Milouse, B. 1982. Analysis of nucleocytoplasmic interactions in the phenomenon of heterosis. In “Variability in Plants Regenerated from Tissue Culture,” (E. D. Earle and Y. Demarly, Eds.), pp. 331-339. Praeger, New York. Bharaj, T. S . , Bains, S. S., Sidhu, G. S., and Gagneja, M. R. 1991. Genetics of fertility restoration of ‘Wild Abortive’ cytoplasmic male sterility in rice, Oryza sativa L. Euphytica 56:199203. Bharaj, T. S., Virmani, S . S., and Khush, G. S. 1995. Chromosomal location of fertility restoring genes for ‘Wild Abortive” cytoplasmic male sterility using primary trisomics in rice. Euphytica 83: 169- 173. Bijral, J. S., Sharma, T. R., Singh, B., Gupta, B. B., and Kanwal, K. S. 1989. Isolation of maintainers and restorers for three cytoplasmic male sterile lines. IRRN 14(3):6. Blanco, L., Akita, S., and Virmani, S. S. 1986. Growth and yield of Fl rice hybrids in different levels of nitrogen. Jpn. J. Crop Sci. 55 (extra 1):12-13. Bollich, C. N., Webb, B. D., Marchetti, M. A., Scott, J. E., and Del Var Petersen, H. 1988. Performance of hybrid rices in Texas, USA. In “Hybrid Rice,” pp. 289-290. Int. Rice Res. Inst., Manila, Philippines. Borkakati, R. 1994. Genetics of thermo-sensitive male sterility in rice (Oryza sativa, L.).PhD thesis submitted to the Assam Agricultural University, Jorhat, India. (mimeo). Borkakati, R., and Virmani, S. S. 1993. Inheritance of a thermo-sensitive genic male sterile mutant of indica rice. Rice Genetics Newsl. 10:92-94. Brar, D. S., Fujimura, T., McCouch, S., and Zapata, F. J. 1994. Application of biotechnology in hybrid rice. In “Hybrid Rice Technology: New Developments and Future Prospects” (S. S. Virmani, Ed.), pp. 51-62. Int. Rice Res. Inst., Manila, Philippines. Butany, W. T. 1957. Natural crossing in rice. Rice News Teller 5(3):18-21.
HYBRID RICE
45 1
Carnahan, H. L., Erickson, J. R., Tseng, S. T., and Rutger, J. N. 1972. Outlook for hybrid rice in USA. In “Rice Breeding,” pp. 603-607. Int. Rice Res. Inst., Manila, Philippines. Chauhan, J. S., Virmani, S. S., Aquino, R. C., and Vergara, 8 . S. 1983. Evaluation of hybrid rice for ratooning ability. Inr. Rice Res. Newslett. 8(6):6. Chen, J. S., Lin, Y.,and Jun, C. W. 1988. A study of apomixis in rice. Genet. Manipulations Crops Newslert. 4(2):3-10. Chen, S . , Min, S., Xiong, Z., Wu, J., and Zhu, X. 1995. Analysis on inheritance of cytoplasmic male sterility in rice (Oryza sativa L.). Chinese J . Rice Sci. 9(2):115-118. Chen, Yinhui, Cai, Junmai, and Lu Haoran. 1987. Effects of male sterile cytoplasmic interaction on the genetic performance of rice. J . Fujian Agric. Coll. 16(3):190-197. Chitrakon, S., Khambanonda. P., and Senawong, P. 1986. Hybrid rice: Status and future in Thailand. Thai. Agric. Res. J . 4(2):149-157. Cho, S. Y., Moon, P., 11, Jun, B. T., Lee, S . K., and Virmani, S. S. 1994. Hybrid rice. In “KoreaIRRI Collaborative Project on Rice Research 1992-1993,” pp. 29-42. RDA, Suweon, Korea. Chu, Y. E. 1967. Variation in peroxidase isozymes of Oryza perennis and 0.cativa. Jpn. J . Genet. 42~233-240. Coffman, W. R., and Virmani, S. S. 1983. Advances in rice technology in the People’s Republic of China. In “Agricultural and Rural Development of China Today” (R. Barker and B. Rose, Eds.), pp. 1-26. Cornell University, Ithaca, NY. Comstock, R. E., and Robinson, H. F. 1952. Estimation of average dominance of genes. In “Heterosis” (J. W. Gowen, Ed.), pp. 494-516. Iowa State College Press, Ames, IA. Dalmacio, R. D., Brar, D. S., Ishii, T., Sitch, L. A,, Virmani, S. S . , and Khush, G. S. 1995. Identification and transfer of a new cytoplasmic male sterility source from Oryza perennis into indica rice (Oryza sariva L.). Euphyrica 82:221-225. Deng, H. 1988. Biochemical basis of heterosis in rice. In “Hybrid Rice,” pp. 55-66. Int. Rice Res. Inst., Manila, Philippines. Deng, H., and Wang, G. 1984. A study on prediction of heterosis in crops. 11. Analysis of heterosis of rice and its esterase zymogram patterns, complementary patterns and artificial hybrid zymogram patterns. Hunan Agric. Sci. 3:l-5. [Chinese] Dhanakodi, C. V., and Subramaniam, M. 1994. Heterosis for yield and its components in short duration rice (Oryza sariva L.) hybrids. Madras Agric. J . 81:313-315. Ellis, J. R. S., Brunion, C. J., and Palmer, J. M. 1973. Can mitochondria1 complementation be used as a tool for breeding hybrid cereals? Narure(Lond) 241:45-47. Erickson, J. R., 1969. Cytoplasmic male sterility in rice (Oryza sativa L.). Agron. Abst. 6. Flavell, R. B . , and Barratt, D. H. P. 1977. Mitochondria1 efficiency and grain yield in wheat. Ann. Eor(Lond) 41: 1323- I33 I . Flemming, A. A. 1975. Effects of cytoplasm on inheritance of hybrid maize. Crop Sci. 15570-573. Fu, P. Y.,and Pai, C. 1979. Genetic studies on isozymes in rice plant. 2. Classification and geographical distribution of cultivated rice through isozyme studies. J . Agric. Assoc. China 107:l- 16. [Chinese with English summary] Fu, Q. 1985. Studies on the inheritance of restoring genes for the wild-abortive cytoplasmic sterility in rice, (Oryza sativa, subsp. he sien tung). J . Fujian Agric. Coll. 14194-202. Gallais, A. 1988. Heterosis: Its genetic basis and its utilization in plant breeding. Euphyrica 39(3):95- 104. Gao, M. 1981. Preliminary analysis of genotypes of hybrid indica rice of ‘WA type. Acta Genet. Sin. 8(1):66-74. Geetha, S . , Kirubakaran Soundararaj, A . P. M., and Palaniswamy, S. 1994. Grain characteristics of rice hybrids. Crop Res. 7:303-305. Glaszmann, J. C. 1987. Isozymes and classification of Asian cultivated rice varieties. Theor. Appl. Genet. 7421-30.
s. s. VlRMANI
452
Glaszmann, J. C., Edralin, O., and Virmani, S. S. 1987. Isozyme markers to monitor seed purity in indica hybrid rice. Inr. Rice Res. Newslerr. 12(5):17. Govinda Raj, K. 1993. Economics of hybrid rice in India: A pre-commercial experience. In “Hybrid Rice: Food Security in India” (B. R. Barwale, Ed.), pp. 94-98. MacMillan India. Govinda Raj, K., and Virmani, S. S. 1988. Genetics of fertility restoration of WA type cytoplasmic male sterility in rice. Crop Sci. 28:787-792. Gravois, K. A. 1994. Diallel analysis of head rice percentage, total milled rice percentage, and rough rice yield. Crop Sci. 34(1):42-45. Gravois, K. A,, and McNew, R. W. 1993. Combining ability and heterosis in U. S. Southern longgrain rice. Crop Sci. 3383-86. Griffing, B. 1956a. Concept of general and specific combining ability in relation to diallel crossing systems. Ausr. J. Biol. Sci. 9463-493. Griffing, B. 1956b. A generalized treatment of the use of diallel crosses in quantitative inheritance. Herediry 10:31-50. Guok, H. P. 1994. Hybrid rice research in Malaysia. In “Hybrid Rice Technology: New Developments and Future Prospects” (S.S. Virmani, Ed.), pp. 207-212. Int. Rice Res. Inst., Manila, Philippines. Guong, V. T., Lap, T. T., Hon, N. M., Castillo, E. G., Padilla, J. L. and Singh, U. 1995. N use efficiency in direct-seeded rice in the Mekong River Delta: Varietal and Phosphorous Response. In “Vietnam and IRRI: A Partnership in Rice Research” ( G . L. Denning and V. T. Xuan, Eds.), pp. 151-159. Int. Rice Res. Inst., Manila, Philippines. Gupta, S. K., and Singh, S. S. 1977. Biochemical basis of hybrid vigor for grain weight of Oryza sariva L. Theor. Appl. Genet. 49:259-264. Hanna, W. W., and Bashaw, E. C. 1987. Apomixis: Its identification and use in plant breeding. Crop Sci. 221136-1139. Hassan, M. A,, and Siddiq, E. A. 1984. Inheritance of anther and stigma exsertion in rice (Oryza sariva L.). Indian J. Genet. 44544-547. Hauler, A. R., and Miranda, J. B. 1981. Quantitative genetics in maize breeding. Iowa State University Press, Ames, IA. Hayman, B. I. 1954. The analysis of variance of diallel tables. Biometrics 10(2):235-244. He, G.T., Te, A., Zhu, X., Travers, S. L., Lai, X., and Herdt, R. W. 1984. The economics of hybrid rice production in China. IRRI Res. Paper Ser. 101:14. He, G . T., Zhu, X., and Flinn, J. C. 1987a. A comparative study of economic efficiency of hybrid and conventional rice production in Jiangsu province, China. Oryza 24:285-296. He, G.T., Zhu, X.,and Flinn, J. C. 1987b. Hybrid seed production in Jiangsu province, China. Oryza 24.297-312.
He, G. T., Zhu, X., Gu, H. Z., and Flinn, J. C. 1988. The use of hybrid rice technology: An economic evaluation. In “Hybrid Rice,” pp. 229-241. Int. Rice Res. Inst., Manila, Philippines. Heu, M. H., Kim, H. Y., and Cho, Y. H. 1984. Development of hybrid rice facilitated by cytoplasmic genetic male sterility. 11. Responses of Korean cultivars to the Chinese cytoplasmic genetic male sterile lines. Korean J . Crop Sci. 29(3):227-231. Heu, M. H., Kim, H. Y., and Koh, H. J. 1989. Combining ability for yield and its components of several japonica rice cultivars. Korean J . Breeding 21( 1): 1-8. Hossain, M. 1995. Sustaining food security for fragile environments in Asia: Achievements, challenges and implications for rice research. In: “Fragile Lives in Fragile Ecosystems,” Proc. Int. Rice Res. Conf., Feb. 13-17, 1995, Manila, Philippines, pp. 3-23. Hu, J., and Li, Z. 1985. A preliminary study on the inheritance of male sterility of rice male sterile lines with four different kinds of cytoplasms. J. Huazhoung Agric. Coll. 4(2):15-22.
HYBRID RICE
45 3
Huang, P. J., Maruyama, K., Sharma, H. L . , and Virmani, S. S. 1994. Advances in hybrid rice seed production technology. In “Hybrid Rice Technology: New Developments and Future Prospects” ( S . S. Virmani, Ed.), pp. 63-70. lnt. Rice Res. Inst., Manila, Philippines. Hunter, R. B., and Gamble, E. E. 1968. Effect of cytoplasmic source on the performance of double cross hybrids in maize (Zea mays L.). Crop Sci. 8278-280. Ikehashi, H. 1991. Genetics of hybrid sterility in wide hybridization in rice (Oryza saliva L.). In “Biotechnology in Agriculture and Forestry. 14. Rice” (Y. P. S. Bajaj, Ed.), pp. 113-127. Springer Verlag, Berlin. Ikehashi, H., and Araki, H. 1984. Varietal screening of compatibility types revealed in F, fertility of distant crosses in rice. Jpn. J. Breed. 34:304-313. Ikehashi, H., and Araki, H. 1986. Genetics of F, sterility in remote crosses of rice. In “Rice Genetics,” pp. 119-130. Int. Rice Res. Inst., Manila, Philippines. Ikehashi, H., and Araki, H. 1987. Screening and genetic analysis of wide compatibility in F, hybrids of distant crosses in rice, Oryza sativa. Tech. Bull. 22. Tropical Agric. Res. Center, Japan. Ikehashi, H., Zou, J. S., Moon, H. P., and Maruyama, K. 1994. Wide compatibility gene(s) and indica-japonica heterosis in rice for temperate countries. In “Hybrid Rice Technology: New Developments and Future Prospects,” (S. S. Virmani, Ed.). Int. Rice Res. Inst., Manila, Philippines. IRRI (International Rice Research Institute). 1980. Annual Report for 1979. Int. Rice Res. Inst., P. 0. Box 933, Manila, Philippines. IRRI. 1983. Annual Report for 1981. Int. Rice Res. Inst., Manila, Philippines. IRRI. 1984. Annual Report for 1983. Int. Rice Res. Inst., Manila, Philippines. IRRI. 1986. Annual Report for 1985. Int. Rice Res. Inst., Manila, Philippines. IRRI. 1987. Annual Report for 1986. Int. Rice Res. Inst., Manila, Philippines. IRRI. 1990. Program Report for 1989. Int. Rice Res. Inst., Manila, Philippines. IRRI. 1991. Program Report for 1990. Int. Rice Res. Inst., Manila, Philippines. IRRI. 1992. Program Report for 1991. Int. Rice Res. Inst., Manila, Philippines. IRRI. 1993. Program Report for 1992. Int. Rice Res. Inst., Manila, Philippines. IRRI. 1994. Program Report for 1993. Int. Rice Res. Inst., Manila, Philippines. IRRI. 1995. Program Report for 1994. Int. Rice Res. Inst., Manila, Philippines. Ishii, T., and Tsunewaki, K. 1991. Chloroplast genome differentiation in Asian cultivated rice. Genome 34:818-826. Islam, M. A., and Virmani, S. S. 1990. Effect of heterozygosity on grain quality of rice (Oryza sativa L.). Bangladesh J. Crop Sci. 1(2):111-119. Jachuck, P. J., Mahadevappa, M., and Sidhu, G.S. 1988. The hybrid rice program in India. In “Hybrid Rice,” pp. 287-288. Int. Rice Res. Inst., Manila, Philippines. Jinks, J. L. 1983. Biometrical genetics of heterosis. In “Heterosis-Reappraisal of Theory and Practice” (R. Frankel, Ed.), pp. 1-46. Springer-Verlag. Berlin, Heidelberg, New York. Jinks, J. L., and Jones, R. M. 1958. Estimation of the components of heterosis. Genetics 43223234.
Jones, J. W. 1926. Hybrid vigour in rice. J. Am. SOC.Agron. 18:424-428. Jones, H. A., and Emsweller, S. L. 1937. A male sterile onion. Proc. Ann. Soc. Horr. Sci. 34582585. Julfiquar, A. W., Virmani, S. S., and Carpena, A. L. 1985. Genetic divergence among some maintainer and restorer lines in relation to hybrid breeding in rice (Oryza sariva L.). Theor. Appl. Genet. 70:671-678. Kabaki, N. 1993. Growth and yield of japonica-indica hybrid rice. JARQ 27:88-94. Kabaki, N., Akita, S.,Tanaka, I., and Amaniya, A. 1976. Photosynthesis and photorespiration of cultivated rice and F1 hybrid. Jpn. J . Crop Sci. 45(extra issue 2):177-178 (in Japanese).
454
s. s. VIRMANI
Kadowaki, K., Ishige, T., Suzuki, S., Harada, K., and Shinjyo, C. 1986. Differences in characteristics of mitochondria1 DNA between normal and male sterile cytoplasms of japonica rice. Jpn. J. Breed. 36:333-339. Kadowaki, K., Osumi, T., Nemoto, H., Hereda, K., and Shinjyo, C. 1988. Mitochondria1 DNA polymorphism in male sterile cytoplasm of rice. Theor. Appl. Genet. 75234-236. Kalsy, H. S., and Sharma, D. 1972. Study of cytoplasmic effects in reciprocal crosses of divergent varieties of maize (Zeu mays L.). Euphyticu 21527-533. Kato, H., Maruyama, K., and Uchiyamada, H. 1994a. Hybrid rice research in Japan. In “Hybrid Rice: New Developments and Future Prospects” (S. S. V h a n i , Ed.), pp. 149-156. Int. Rice Res. Inst., Manila, Philippines. Kato, H., Tanaka, K., Nakazumi, H., Araki, H., Yoshida, T., Yashuki, O., Yanagihara, S., Kishimoto, N., and Maruyama, K. 1994b. Heterosis of biomass among rice ecospecies and isozyme polymorphism and RFLP. Jpn. J. Breed. 44:271-277. Kato, H., and Namai, H. 1987a. Intervarietal variations of floral characteristics with special reference to F, seed production in japonica rice (Oryzu surivu L.). Jpn. J. Breed. 37: 75-87. Kato, H.,and Namai, H. 1987b. Floral characteristics and environmental factors increasing natural outcrossing rate for F, hybrid seed production in rice (Oryzu surivu L.). Jpn. J. Breed. 37:318330. Katsuo, K., and Mizushima, U. 1958. Studies on the cytoplasmic difference among rice varieties, Oryzu sutivu L. I. On the fertility of hybrids obtained reciprocally between cultivated and wild varieties. Jpn. J. Breed. 8:l-5. [Japanese] Kaw, R. N., and Khush, G. S. 1985. Heterosis in traits related to low temperature tolerance in rice. Philipp. J. Crop Sci. 10:93-105. Kearsey, M. J., and Jinks, J. L. 1968. A general method of detecting additive, dominance, and epistatic variation for metrical traits. J. Theor. Heredity 23:403-409. Kempthorne, 0. 1957. An introduction of genetical statistics, 1st ed., pp. 458-571. New York. Khush, G. S., and Aquino, R. C. 1994. Breeding tropical japonicas for hybrid rice production. In “Hybrid Rice: New Developments and Future Prospects” (S. S. Virmani, Ed.), pp. 33-36. Int. Rice Res. Inst., Manila, Philippines. Khush, G. S., Brar, D. S., Bennett, J., and Virmani, S. S. 1994. Apomixis for rice improvement. In “Apomixis: Exploiting Hybrid Vigor in Rice” (G. S. Khush, Ed.), pp. 1-21. Int. Rice Res. Inst., Manila, Philippines. Khush, G. S., Kumar, I., and Virmani, S. S. 1988. Grain quality of hybrid rice. In “Hybrid Rice,” pp. 201-215. Int. Rice Res. Inst., Manila, Philippines. Kim, C. H., and Rutger, J. N. 1988. Heterosis in rice. In “Hybrid Rice,” pp. 39-54. Int. Rice Res. Inst., Manila, Philippines. Kinoshita, T. 1992. Report of the committee on gene symbolization nomenclature and linkage groups. Rice Genet. Newslerf. 9:2-4. Kitamura, E. 1962. Studies on cytoplasmic sterility of hybrids in distantly related varieties of rice, Oryzu sutivu L. 1. Fertility of F, hybrids between strains derived from a certain Philippine X Japanese variety crosses and Japanese varieties. Jpn. J. Breed. 1281-84. Kumar, I., and Saini, S. S. 1981. Diallel analysis in rice combining ability for various quantitative characters. Genet. Agric. 35:243-252. Kyozuka, J., Kaneda, T., and Shimamoto, K. 1989. Production of cytoplasmic male sterile rice (Oryzu surivu L.) by cell fusion. Biorechnology R1171-1174. Lara, R. J., Dela Cruz, I. M., Ablaza, M. S., Dela CNZ, H. C., and Obien, S. R. 1994. Hybrid rice research in the Philippines. In “Hybrid Rice Technology: New Developments and Future Prospects” (S. S. Virmani, Ed.), pp. 173-186. Int. Rice Res. Inst., Manila, Philippines. Lenz, M. C., and Atkins, R. E. 1981. Comparison of agronomic and morphologic characters in sorghums having different cytoplasms. Crop Sci. 21:946-950.
HYBRID RICE
45 5
Li, Y. C., and Yuan, L. P. 1986. Genetic analysis of fertility restoration in male sterile of rice. In “Rice Genetics,” pp. 617-632. IRRI, Manila, Philippines. Li, C., and Ang, S . 1988. Genetic distance and heterosis in japonica rice. In “Hybrid Rice,” p. 257. Int. Rice Res. Inst., Manila, Philippines. Li, Z., and Zhu, Y. 1988. Rice male sterile cytoplasm and fertility restoration. In “Hybrid Rice,” pp. 85-102. Int. Rice Res. Inst., Manila, Philippines. Lin, J. Y. 1990. Hybrid rice innovation in China: A study of market demand-induced technological innovation in a centrally planned economy. Peking University, China. Lin, I. Y. 1994. The nature and impact of hybrid rice in China. In “Modern Rice Technology and Income Distribution in Asia.” (C. C. David and K. Otsuka, Eds.). pp. 375-408. Int. Rice Res. Inst., Manila, Philippines. Lin, J. Y., and Pingali, P. L. 1994. Economic assessment of the potential for hybrid rice in tropical Asia: Lessons from the Chinese experience. In “Hybrid Rice: New Developments and Future Prospects” (S. S. Virmani, Ed.), pp. 131-141. Int. Rice Res. Inst., Manila, Philippines. Lin, S. C., and Ming, S. K. 1991. Rice varieties and their geneology in China. Shanghai Scientific and Technical Publishers, Shanghai, China. Lin, S. C., and Yuan, L. P. 1980. Hybrid rice breeding in China. In “Innovative Approaches to Rice Breeding,” pp. 35-51. Int. Rice Res. Inst., Manila, Philippines. Ling, D. H., Ma, Z. R., Chen, W. Y., and Chen, M. F. 1987. Male sterile mutant from somatic cell culture of rice. Theor. Appl. Genet. 75127-131. Lokaprakash, R., Shivashankar, G., Mahadevappa, M., Shankare Gowda, B. T., and Kulkami, R. S . 1991. Combining ability for yield and its components in rice. Oryzu 28(3):319-322. Lu, S. W., Wang, B. C., Kuang, 2. B., and Dy, G. 1982. Study on the application of esterase isozyme for the measurement of hybrid purity of the hybrid rice. Sci. Agric. Sin. 510-16. Lu, X. G., Zhang, G., Maruyama, K., and Virmani, S. S. 1994. Current status of two-line method of hybrid rice breeding. In “Hybrid Rice Technology: New Developments and Future Prospects,” (S. S. Virmani, Ed.), pp. 37-50. Int. Rice Res. Inst., Manila, Philippines. Luat, N. V., Bong, B. B., and Chandra Mohan, J. 1985. Evaluation of F, hybrids in the Cuu Long Delta, Vietnam. IRRN lO(3):19. Luat, N. V., Minh, H. T., and Suan, N. V. 1994. Hybrid rice research in Vietnam. In “Hybrid Rice Technology: New Developments and Future Prospects,” (S. S. Virmani, Ed.), pp. 187-194. Int. Rice Res. Inst., Manila, Philippines. Luat, N. V., Suan, N. V., and Virmani, S. S. 1995. Current status and future outlook on hybrid rice in Vietnam. In “Vietnam and IRRI: A Partnership in Rice Research” (G. L. Denning and V. T. Xuan, Eds.), pp. 73-80. Int. Rice Res. Inst., Manila, Philippines, and Ministry of Agriculture and Food Industry, Hanoi, Vietnam. Luo, L. J., Ying, C. S., and Wang, Y. P. 1990. The use of photoperiod-sensitive genic male sterile line in screening for wide compatibility rice varieties. Chinese J. Rice Sci. 4(3):143-144. [Chinese with English summary] Mackill, D. J. 1995. Classifying japonica rice cultivars with RAPD. Crop Sci. 35889-894. Mahalanobis, P. C. 1936. On the generalized distance in statistics. Proc. Nutl. Inst. Sci. India B . 2~49-55. Mao. C. X. 1988. Hybrid rice seed production in China. In “Rice Seed Health,” pp. 277-282. Int. Rice Res. Inst., Manila, Philippines. Mariani, C., Benckebe, M. D., Truettver, J., Leemans, J., and Goldberg, R. B. 1990. Induction of male sterility in plants by a chimeric endonuclease gene. Nature (London) 347:737-741. Maruyama, K. 1988. Strategy and status for developing hybrid rice. Iden (Heredity) 42(5):28-31. [Japanese] Maruyama S., Norihama, Y., Miyazatok, K., and Nose, A. 1982. Studies on productivity of hybrid rice. I. Heterosis for leaf photosynthesis rate. Jpn. J. Crop Sci. 51 (extra issue 2):85-86.
456
s. s. VIRMANI
Maruyama, K., Araki, H., and Kato, H. 1991a. Thermosensitive genetic male sterility induced by irradiation. In “Rice Genetics 11,” pp. 227-232. Int. Rice Res. Inst., Manila Philippines. Maruyama, K., Kato, H., and Araki, H. 1991b. Mechanized production of F, seeds in rice by mixed planting. JARQ 24:343-252. Maruyama, S., Mijazato, K., and Nose, A. 1985. Basic studies on utilization of hybrid vigor in rice. V. Reciprocal difference and heterosis in hybrid seeds and F, plants. Jpn. J . Crop Sci. 54(4):3 18-323. Maruyama, K., and Oonoo, K. 1983. Induction of mutation in tissue culture and their use for plant breeding. VII. On the female and male sterile rice plants regenerated from seed calluses. Jpn. J. Breed. 33(Suppl. 1):24-25. [Japanese] Mather, K., and Jinks, J. L. 1971. “Biometrical genetics,” 2nd ed. Cornell Univ. Press, Ithaca, NY. McDaniel, R. G., and Sarkisson, I. V. 1966. Heterosis complementation by mitochondria. Science 152: 1640- 1642. Mew, T. W., Wang, F. M., Wu, J. T., Lin, K. R., and Khush, G. S. 1988. Disease and insect resistance in hybrid rice. In “Hybrid Rice,” pp. 189-200. IRRI, Manila, Philippines. Mignouna, H., Virmani, S. S . , and Briquet, M. 1987. Modichondrial DNA modifications associated with cytoplasmic male sterility in rice. Theor. Appl. Genet. 74666-669. Mirza, M. J., Faiz, F. A., and Majeed, A. 1993. Heterosis studies of some promising rice crosses. fak. J. Sci. Ind. Res. 36:99-100. Mohamed El-Keredy, S. 1994. Use of hybrid vigour to increase rice productivity: Final Report 198993. National Agriculture Research Project, Egypt. Guok, H. P., and Othman, 0. 1987. Hybrid rice breeding in Malaysia. Malaya Appl. Mohamad, 0.. B i d . 16:129-138. Moon, H. P. 1988. Hybrid rice research in South Korea. In “Hybrid Rice”. Proc. Int. Symp. on Hybrid Rice, Changsha, Hunan, China. Oct. 6-10, 1986. IRRI, Manila, Philippines. p. 290. Moon, H. P., Heu, M. H., and Kim, C. H. 1994. Hybrid rice research in the Republic of Korea. In “Hybrid Rice: New Developments and Future Prospects” (S. S. Virmani, Ed.), pp. 217-226. Int. Rice Res. Inst., Manila, Philippines. Mori, K. 1984. Inheritance of susceptible mutant in rice plant to herbicide buntzone. Jpn. J . Breed. 34(Suppl. 1):421-422. [Japanese] Munoz, D. B. 1994. Hybrid rice research in Colombia. In “Hybrid Rice Technology: New Developments and Future Prospects,” (S. S. Virmani, Ed.), Int. Rice Res. Inst., Manila, Philippines. Munoz, D., and Lasso, L. 1991. Identification of maintainer and restorers for two cytoplasmicgenetic male sterile rice lines. IRRN 16(1):7. Murthy, N., and Shivashankar, G. 1992. Combining ability analysis for yield and some physiological traits in rice (Oryza sariva L.).Indian J. Genet. Plant Breed. 52(3):321-324. Oard, J. H., and Hu, J. G. 1995. Inheritance and characterization of pollen fertility in photoperiodically sensitive rice mutants. Euphyrica 82: 17-23. Oka, H. I. 1988. “Origin of Cultivated Rice,” p. 254. Jpn. Sci. Soc. Press, Elsevier, Tokyo. Osman, M., Guok, H. P., and Bin Omar, 0. 1988. Hybrid rice breeding in Malaysia. In “Hybrid Rice”. Proc. Int. Symp. on Hybrid Rice., Changsha, Hunan, China, Oct. 6-10, 1996. IRRI, Manila, Philippines, p. 289. Pai, C., Endo, T., and Oka, H. I. 1975. Genic analysis for acid phosphatase isozymes in Oryzu perennis and 0. sutiva. Can. J. Genet. Cytol. 17:637-650. Parmar, K. S., Siddique, E. A., and Swaminathan, M. S. 1979. Variation in components of flowering behavior of rice. Indian J. Genet. 39542-550. Peng, J. Y., Glaszmann, J. C., and Virmani, S. S. 1988. Heterosis and isozyme divergence in indica rice. Crop Sci. 28561-563. Peng, J. Y., and Virmani, S. S. 1990. Combining ability for yield and four related traits in relation to hybrid breeding in rice. Oryzu 221-10.
HYBRID RICE
457
Peng, J. Y., and Virmani, S. S . , 1991. Heterosis in some intervarietal crosses of rice. Oryzu 28:31-36. Peng, J. Y., Virmani, S. S. and Julfiquar, A. W. 1991. Relationship between heterosis and genetic divergence in rice. Oryzu 28: 129- 133. Ponnuthurai, S., and Virmani, S. S. 1985. Yield depression in F, hybrids of rice, Oryzu sutivu L. Int. Rice Res. Newslen. 10(3):21. Ponnuthurai, S., Virmani, S. S., and Vergara, B. S. 1984. Comparative studies on the growth and grain yield of some F, rice (Oryzusufivu L.) hybrids. Philipp. J. Crop Sci. 9(3):183-193. Pooni, H. S . , and Treharne, A. J. 1994. The role of epistasis and background genotype in the expression of heterosis. Heredity 72:628-635. Pradhan, S. B., Ratho, S. N., and Jachuck, P. J. 1990. Development of new cytoplasmic genetic male sterile lines through indica X japonica hybridization in rice. Euphyficu 51:127-130. Ramalingam, J., Vivekanandan, P., and Vinniarajan, C. 1993. Combining ability analysis in lowland early rice. Crop Res. 6(2):228-233. Rangaswamy, M., Natarajamoorthy, K., Palanisamy, G. S., and Sreerangaswamy, S. R. 1987. Isolation of restorers and maintainers for two Chinese male sterile lines having wild abortive (WA) cytoplasm. IRRN 12(1): 13. Reddy, C. D. R., Nerkar, Y. S.,and Dhana Lakshrni, G. 1991. Heterosis and inbreeding depression for grain size and yield in rice (Oryzusurivu) under direct seeding. Indian J. Agric. Sci. 61:416419. Riley, R . , Miedema, P., van Gelder, W. M. J. 1979. Mitochondria1 complementation and combining ability. In “Plant breeding perspective” (J. Sneep and A. J. T. Hendriksen, Eds.), pp. 348-351. Cent. Agric. Publ. Doc., Wageningen. Rutger, J. N., and Carnahan, H. L. 1981. A fourth genetic element to facilitate hybrid cereal production-A recessive tall in rice. Crop Sci. 21:373-376. Rutger, J. N., and Shinjyo, C. 1980. Male sterility in rice and its potential use in breeding. In “Innovative Approaches to Rice Breeding,” pp. 53-66. Int. Rice Res. Inst., Manila, Philippines. Sage, G. C. M., and Hobson, G . E. 1973. The possible use of mitochondria1complementation as an indicator of yield heterosis in breeding hybrid wheat. Euphyricu 22:61-69. Sampath, S., and Mohanty, H. K. 1954. Cytology of semisterile rice hybrid. Curr. Sci. 23:182-183. Saran, S . , and Mandal, R. K. 1988. Some restorers and maintainers of WA cyto-sterile lines. IRRN 13( 1 ) 5 .
Saran, H. S., Sahu, S. P., and Azam, M. Q. 1971. Sterility due to abnormal functioning of lemma and palea in IR5/189-1-3 rice (Oryzu surivu L.). Indian J. Agric. Sci. 41515-519. Sarawgi, A. K., Shrivastava, M. N., and Chowdhary, B. P. 1991. Partial diallel cross analysis of yield and its related characters in rice (Oryzu sufivu L.) under irrigated and rainfed situations. Indian J . Genet. Plant Breeding 51( 1):30-36. Sasahara, T., Cui, C. H., and Kambayashi, M. 1986. Cold resistance in rice with some reference to cytoplasmic effects. SABRAO 18(1):69-71. Sasahara, T., and Katsuo, K. 1965. Studies on the cytoplasmic difference among rice varieties, 0. surivu L. 111. On the abortive pollen of Fujisaka No. 5-type plants with cytoplasm of Chinese wild variety 0. surivu f. spontuneu. Jpn. J. Breed. 15(3):291-296. Sato, S., Tamaki, A., and Shinjyo, C. 1994. In “Towards Enhanced and Sustainable Agricultural Production in the 2000s. Breeding Research and Biotechnology Proceedings (Huang er ul., Eds.), pp. 109-1 19. SABRAO. Satyanarayana, P. V., Kumar, I., and Reddy, M. S. S. 1995. A new source of thermosensitive genic male sterility for 2-line hybrid rice breeding. Inf. Rice Res. Newslefr. 20(1):10. Second, G. 1982. Origin of the genetic diversity of cultivated rice (Oryza spp.): Study of the polymorphism scored at 40 isozyme loci. Jpn. J . Genr. 57:25-57. Sehgal, S. M. 1992. Opportunities in hybrid rice development. Seed World Dec. issue: 20-25.
s. s. VIRMANI
458
Senadhira, D., and Virmani, S. S. 1987. Survival of some F, rice hybrids and their parents in saline soil. Int. Rice Res. Newslett. 12(1):14-15. Shahi, B. B., Morishima, H., and Oka, H. 1. 1969. A survey of variation in peroxidase, acid phosphatase, and esterase isozyme of wild and cultivated Uryza species. Jpn. J . Genet. 44:303319. Sharma, H. L., and Virmani, S. S. 1990. Some varietal sources of twin seedlings trait in rice. Rice Genet. Newslet?. 7: 134- 135. Shi, M. S. 1981. Preliminary report of later japonica natural 2-lines and applications. Hubei Agric.
Sci.7. Shi, M. S. 1985. The discovery and study of the photosensitive recessive male sterile rice (Oryza sativa L. subsp. japonica). Sci. Agric. Sin. 2:44-48. Shi, M. S., and Deng, J. Y. 1986. The discovery, determination and utilization of the Hubei photosensitive genic male-sterile rice (Oryza sativa subsp. japonica). Acta Genet. Sin. 13(2): 107112. Shi, S. Y., Jiang, J. R., and Yi, Q. H. 1983. esterase isozyme analysis for purity of hybrid rice seeds. Jiangsu Nongye Kexue 9: 18-20. Shinjyo, C. 1969. Cytoplasmic-genetic male sterility in cultivated rice, Oryza sativa L. 11. The inheritance of male sterility. Jpn. J . Genet. M149- 156. Shinjyo, C. 1972a. Distributions of male sterility inducing cytoplasms of fertility restoring genes in rice. 1. Commercial lowland rice cultivated in Japan. Jpn. J. Gene?. 47:237-243. Shinjyo, C. 1972b. Distributions of male sterility inducing cytoplasms and fertility restoring genes in rice. 11. Varieties introduced from sixteen countries. Jpn. Breed. 22329-333. Shinjyo, C. 1975. Genetical studies of cytoplasmic male sterility and fertility restoration in rice, Oryza sativa L. Bull. Coll. Agric., Univ. Ryukus, 221-57. Shinjyo, C., and Sato, S . 1994. Chromosomal location of fertility-restoring gene Rf-2. Rice Genet. Newslett. 11:93-95. Shinjyo, C., and Omura, T. 1966a. Cytoplasmic male sterility in cultivated rice, Oryza sativu L. I. Fertilities of F,, F, and offsprings obtained from their mutual reciprocal backcrosses and segregation of completely male sterile plants. Jpn. J. Breed. 16(Suppl. 1):179-180. Shinjyo, C., and Omura, T. 1966b. Cytoplasmic male sterility and fertility restoration in rice, Uryza sativa L. Sci. Bull. Coll. Agric. Univ. Ryukyus 22:l-57. Shinjyo, C., Nishime, R., and Watanabe, Y. 1974. Inheritance of fertility restoring gene, Rfx and Rf in male sterile cytoplasm derived from variety Lead rice (preliminary report). Jpn. J. Breed. 24(Suppl. I): 130-131. [Japanese] Srivastava, M. N., and Seshu, D. V. 1983. Combining ability for yield and associated characters in rice. Crop Sci. 23:741-744. Siddiq, E. A,, Jachuck, P. J., Mahadevappa, M., Zaman, F. U., Vijaya Kumar, R., Vidyachandra, B., Sidhu, G. S., Kumar, Ish, Prasad, M. N., Rangaswamy, M., Pandey, M. P., Panwar, D. V. S., and Ahmed, I. 1994. Hybrid rice research in India. In Hybrid Rice: New Developments and Future Prospects (S. S. Virmani, Ed.), pp. 157-171. Int. Rice Res. Inst., Manila, Philippines. Singh, U. S. 1983. Hybrids excel standard cultivar. Deep Water Rice 1:l. Singh, M. R. K., and Sinha, P. K. 1987. Identification and classification of fertility restorers and maintainers for cytoplasmic male sterile lines V20A. IRRN 12(5): 16-17. Singh, M. R. K., and Sinha, P. K. 1988. Genetics of fertility restoration for cytosterile line V20A. OTZU 25184-185. Sprague, G. F., and Eberhart, S. A. 1977. Corn breeding. In Corn and Corn Improvement (G. F. Sprague, Ed.), pp. 309-362. Am. Soc. Agron., Madison, WI. Sprague, G. S., and Tatum, L. A. 1942. General and specific combining ability in single crosses of corn. Am. Soc. Agron. 34:923-932. Sreedhar, M., and Kulkarni, N. 1993. Heterosis in rice. J. Res. APAU 21:138-141.
HYBRID RICE
459
Srivastava, H. K. 1983. Heterosis and inter-genomic complementation mitochondria, chloroplast, and nucleus. In “Heterosis, reppraisal of theory and practice” (R. Frankel, Ed.), pp. 260-286. Springer Berlin. Heidelberg, New York. Suh, H. S. 1985. Studies on multiple pistillate male sterile rice. Korean J. Breed. 17:439-443. Suh, H. S. 1988. Multiple pistillate male sterile rices for hybrid seed production. I n “Hybrid Rice,” pp. 181-187. Int. Rice Res. Inst., Manila, Philippines. Sun, Z. X., Min, S. K., and Xiong, Z. M. 1989. A temperature-sensitive male-sterile line found in rice. Rice Genet. Newslett. 6:116-117. Suprihatno, B. 1986. Hybrid rice: Its prospects and problems in Indonesia. IARD J. 851-58. Suprihatno, B. Sutaryo, B., and Silitonga, T. S. 1994. Hybrid rice research in Indonesia. I n “Hybrid Rice Technology: New Developments and Future Prospects” (S. S. Virmani, Ed.), pp. 195-206. lnt. Rice Res. Inst., Manila, Philippines. Sutaryo, B. 1989. Some Indonesian restorers and maintainers of ‘WA cyto-sterile lines. IRRN 14(3):9. Swaminathan, M. S. 1993. The role of rice in the national food security system. In “Hybrid Rice: Food Security in India” (B. R. Barwale, Ed.), pp. 147-170. MacMillan, India. Taillebois, J., and Guimaraes, E. P. 1988. Improving outcrossing rate in rice (Oryza sativa L.). In “Hybrid Rice,” pp. 175-180. Int. Rice Res. Inst., Manila, Philippines. Teng, L. S., and Shen, Z. T. 1994. Inheritance of fertility restoration for cytoplasmic male sterility in rice. Rice Genet. Newsleft. 11:95-97. Tomar, I. B., and Virmani, S. S. 1990. Identifying maintainers and restorers of CMS lines for hybrid rice breeding. IRRN 15(6):5-6. Vaidyanath, K., and Reddy, G. M. 1985. Studies on genetic divergence in the genus Oryza. J. Cytol. Genet. 20:59-68. Vidya Chandra, B., Radhakrishna, R. M., and Lingaraju, S. 1995. Development and use of hybrid rice technology. Achievements of Mandya Center 1989-94. Univ. of Agri. Sci., Bangalore Regional Station, V. C. Farm, Mandya. Vijaya Kumar, R., and Virmani, S. S. 1992. Wide compatibility in rice (Oryzu sativu L.). Euphyticu 64:71-80. Virmani, S . S., 1987. Hybrid rice breeding. In “Hybrid Seed Production of Selected Cereals and Vegetable Crops” (W. P. Flestritzer and A. F. Kelly, Eds.),pp. 35-53. FA0 Plant Production and Protection Paper No. 82. Virmani, S. S. 1994a. “Heterosis and Hybrid Rice Breeding.” Springer Verlag, Berlin. Virmani, S. S. 1994b. Prospects of hybrid rice in the tropics and sub-tropics. In “Hybrid Rice Technology: New Developments and Future Prospects” (S. S. Virmani, Ed.), pp. 7-19. Int. Rice Res. Inst., Manila, Philippines. Virmani, S. S. 1994c. IRRI’s program on international collaboration on hybrid rice. In “Hybrid Rice Technology: New Developments and Future Prospects” (S. S. Virmani, Ed.), pp. 275-281. Int. Rice Res. Inst., Manila, Philippines. Virmani, S. S., and Athwal, D. S. 1974. Inheritance of floral characteristics influencing outcrossing in rice. Crop Sci. 14350-353. Virmani, S . S., and Voc, P. C. 1991. Induction of photo- and thermo-sensitive male sterility in indica rice. Agron. Abst. 119. Virmani, S. S . , Aquino, R. C., and Khush, G. S. 1982. Heterosis breeding in rice, Oryza sativa L. Theor. Appl. Genet. 63:373-380. Virmani, S. S., and Casal, C. 1993. Isolation-free system for producing experimental hybrid seed for preliminary evaluation. Int. Rice Res. Newsleft. 18(3):6-7. Virmani, S. S., Chaudhary, R. C., and Khush, G. S. 1981. Current outlook on hybrid rice. Oryza 18:67-84. Virmani, S. S., and Dalmacio, R. D. 1987. Cytogenic relationship between two cytoplasmic malesterile lines. Int. Rice Res. Newslett. 12(1):14.
460
s. s. VIRMANI
Virmani, S. S., Dalmacio, R. D., and Lopez, M. T. 1988. ‘Eui’ gene for elongated upper most internode transferred to indica rice. Inr. Rice Res. Newsletr. 13(6):6. Virmani, S . S., and Dedolph, C. 1993. Reaping the benefits of hybrid rice in the tropics. World Agriculrure 2: 17-20. Virmani, S. S., and Edwards, I. B. 1983. Current status and future prospects for breeding hybrid rice and wheat. Adv. Agron. 36:145-214. Virmani, S. S., Govinda Raj, K., Casal, C., Dalmacio, R. D., and Aurin, P. A. 1986. Current knowledge of and outlook on cytoplasmic genetic male sterility and fertility restoration in rice. In “Rice Genetics,” pp. 633-647. Int. Rice Res. Inst., Manila, Philippines. Virmani, S. S., Khush, G. S., and Pingali, P. L. 1994. Hybrid rice for tropics: Potentials, research priorities and policy issues. I n “Hybrid Research and Development of Major Cereals in Asia Pacific Region” (R. S. Paroda and M. Rai, Eds.), pp. 61-86. FAO, Bangkok. Virmani, S. S., Manalo, J., and Toledo, R. 1993a. A self-sustaining system for hybrid rice seed production. Int. Rice Res. Newslert. 18:4-5. Virmani, S. S., Prasad, M. N., and Kumar, I. 1993b. Breaking yield barrier in rice through exploitation of heterosis. I n “New Frontiers in Rice Research” (K. Muralidharan and E. A. Siddiq, Eds.), pp. 76-85. Directorate of Rice Research, 500030, India. Virmani, S. S., and Sharma, H. L. 1993. “Manual for Hybrid Rice Seed Production.” Int. Rice Res. Inst., Manila, Philippines. Virmani, S. S., and Shinjyo, C. 1988. Current status of analysis and symbols for male sterile cytoplasms and fertility restoring genes. Rice Genet. Newsletr. 59-15, Virmani, S. S., and Wan, B. H. 1988. Development of CMS lines in hybrid rice breeding. I n “Hybrid Rice,” pp. 103-1 14. Int. Rice Res. Inst., Manila, Philippines. Virmani, S. S., Young, J. B., Moon, H. P., Kumar, I., and Flinn, J. C. 1991. Increasing rice yields through exploitation of heterosis. IRRI Research Paper Series Number 156. Int. Rice Res. Inst., Manila, Philippines. Vivekarandan, P., Rangathan, T. B., and Kadambavana-sundaran, M. 1992. Heterosis for yield and its components in rice (Oryza sariva L.) hybrids. Madras Agric. J . 79:5-8. Voc, P. C., Mui, P. T., and Luat, N. V. 1990. Identification of restorers and maintainers of male sterile lines at CLRRI. IRRN 15(2):5-6. Wan, J. M., Yanagihara, S., Kato, H., and Ikehashi, H. 1993. Multiple alleles at a new locus causing hybrid sterility between Korean indica variety and a javanica variety in rice (Oryza saliva L.), Jpn. J . Breed. 43:507-516. Wang, S . 1980. Inheritance of R genes in rice and methods for selection of new R lines. Agric. Sci. Technol. Hunan 4:1-4. Wang, C. L., Zou, J. S., Wang, 2. M., Li, C. G., and Li, H. B. 1991a. Identification of wide compatibility and heterosis in rice, strain 02428 Xuan. Chinese J. Rice Sci. 5(1):19-24. [Chinese with English summary] Wang, W. M., and Wen, H. C. 1995. New cytoplasmic male sterile line with lower negative effects of cytoplasm on some quantitative traits in rice. Inf. Rice Res. Newslen. 20(2):in press. Wang, X. M., Wang, M. Q., Mei, G. Z., Wu, H. Y., Duan, W. J., and Wang, W. J. 1991b. Photoperiod-conditioned male sterility and its inheritance in rice. In “Rice Genetics 11,” pp. 217-226. Int. Rice Res. Inst., Manila, Philippines. Wang, 2. W., Second, G., and Tanksley, S. D. 1992. Polymorphism and phylogenetic relationships among species in the genus Oryza as determined by analysis of nuclear RFLR. Theor. Appl. Genet. 83:565-581. Wang, 2. W., and Tanksley, S. D. 1989. Restriction fragment length polymorphism in Oryza sativa L. Genome 32:1113-1118. Watanabe, Y. 1971. Establishment of cytoplasmic and genetic male sterile lines by means of indicajaponica cross. Oryza 8(Suppl. 2):9-16.
HYBRID RICE
46 1
Wu, X. J., Yin, H. Q., and Yin, H. 1991. Preliminary study of the temperature effect of Annong S-1 and W6154S. Crop Res. (China) 5(2):4-6. Xiao, Y. H. 1981. Study on esterase isozyme of hybrid rice and “three lines.” Hubei Agric. Sci. 11. [ChineseJ Xu, S., and Li, B. 1988. Managing hybrid rice seed production, In“Hybrid Rice,”pp. 157-163. Int. Rice Res. Inst., Manila, Philippines. XU, D. X., Wang, M. L., and Song, R. D. 1988. Using another culture to develop parental lines for hybrid rice breeding. In “Hybrid Rice.” Int. Rice Res. Inst., Manila, Philippines. Yabuno, T. 1977. Genetic studies on the interspecific cytoplasm substitution lines of japonica varieties of Oryzu suriva L. and 0. glaberrimu Steud. Euphyticu 26:451-463. Yamauchi, M. 1994. Physiological bases of higher yield potential in F, hybrids. I n “Hybrid Rice Technology: New Developments and Future Prospects” (S. S. Virmani, Ed.), pp. 71-80. Int. Rice Res. Inst., Manila, Philippines. Yamauchi, M., and Yoshida, S. 1985. Heterosis in net photosynthesis rate, leaf area tillering and some physiological characters of 35 rice hybrids. J. Exp. Bor. 36( 136):274-280. Yan, Z. D. 1988. Agronomic management of rice hybrids compared with conventional varieties. In “Hybrid Rice,” pp. 217-223. Int. Rice Res. Inst., Manila, Philippines. Yanagihara, S . , Kato, H., and Ikehashi, H. 1992. A new locus for multiple alleles causing hybrid sterility between an aus variety and javanica varieties in rice (Oryza sutiva L.). Jpn. J. Breed. 42:793-801. Yang, R. C., Li, W. M., Wang, N. Y., Liang, K. J., and Chen, Q. H. 1989. Discovery and preliminary study on indica photosensitive genic male sterile germplasm 5460 ps. Chines J. Rice Sci. 3(1):47-48. Yang, Z. 0.. Shikanai, T., and Yamada, Y. 1988. Asymmetric hybridization between cytoplasmic male sterile (CMS) and fertile rice (Oryza surivu) protoplasts. Theor. Appl. Gener. 76801-808. Yi, 0. H., Shi, S. Y., and Jiang, J. R. 1984. Analysis of the esterase isozymes in three lines and F, in Oryzu surivu and prediction of heterosis. Actu Bor. Sin. (Eng. trans].) 26506-512. Young, J. B. 1987. Heterosis and combining ability over environments in relation to hybrid rice breeding. PhD thesis, University of the Philippines at Los Banos, Laguna, Philippines. Young, J. B., and Virmani, S. S . 1984. Inheritance of fertility restoration in a rice cross. Rice Genet. Newsl. 1:102-103. Young, I. B . , and Virmani, S. S. 1990a. Stability analysis for agronomic traits in rice hybrids and their parents. Oryza 27:109-121. Young, J. B., and Virmani, S. S. 1990b. Heterosis in rice over environments. Euphytica 51:87-93. Young, J. B . , and Virmani, S. S. 1 9 9 0 ~ Effects . of cytoplasm on heterosis and combining ability for agronomic traits in rice (Oryzu sutiva L.). Euphyricu 48:177-188. Yuan, L. P. 1977. The execution and theory of developing hybrid rice. Zhonggue Nongye Kexue (Chinese Agric. Sci.) 1:27-31. [Chinese] Yuan, L. P. 1985. A Concise Course in Hybrid Rice.” Hunan Technol. Press, China. Yuan, L. P. 1987. Strategy conception of hybrid rice breeding. Hybrid Rice 1:l-4. Yuan, L. P., Li, Y. C., and Deng, H. D. 1990. Progress of studies in rice twin seedlings. Paper presented at the Fourth Annual Meeting of the Rockefeller Foundation’s International Program on Rice Biotechnology, May 9-12, 1990. Int. Rice Res. Inst., Manila, Philippines. Yuan, L. P., and Virmani, S . S. 1988. Organization of a hybrid rice breeding program. In “Hybrid Rice,” pp. 33-37. Int. Rice Res. Inst., Manila, Philippines. Yuan, L. P., Virmani, S. S., and Khush, G. S. 1985. Wei You 64-An early duration hybrid for China. Int. Rice Res. Newslett. 10(5):11-12. Yuan, L. P., Virmani, S. S., and Mao, C. X. 1989. Hybrid Rice-Achievements and future outlook. In “Progress in Irrigated Rice Research,” pp. 219-235. Int. Rice Res. Inst., Manila, Philippines.
462
s. s. VIRMANI
Yuan, L. P., Yang, Z. Y., and Yang, J. B. 1994. Hybrid rice in China. I n “Hybrid Rice Technology New Developments and Future Prospects (S. S. Virmani, Ed.), pp. 143-147. Int. Rice Res. Inst., Manila, Philippines. Zhang, Z. G., Zeng, H. L., Yuan, S. C., Wang, B. X., Li, Y. Z., andZhang, D. P. 1992b. Studies on the model of photothermo reaction of fertility alteration in photo-sensitive genic male sterile rice. J . Hunzong Agri. Univ. 11(1):1-6. Zhang, J. Z., Lin, K. P., and Peng, J. H. 1995. Breeding of a new Honglian-type CMS line Zhu Shen A possessing good grain quality. Inr. Rice Res. Newsleft. 20(4): 10- 1 I . Zhang, T. B. 1985. Male sterility and cytoplasmic regulation of gene action in rice. A brief account of studies. SABRA0 J. 17(1):87-90. Zhang, Q., Gao, Y. J., Yang, S. H., Ragab, R. A., Saghai Maroof, M. A., and Li, Z. B. 1994. A diallel analysis of heterosis in elite hybrid rice based on RFLPs and microsatellites. Theor. Appl. Gener. 89:185-192. Zheng, K.I., Shen, B., Guian, H., and Wang, J. 1992. Tagging genes for wide compatibility in rice via linkage to RFLP markers. Chinese J. Rice Sci. 6(4):145-150. Zhou, K. 1994. Breeding of CMS lines in indica hybrid rice. Hybrid Rice J. 3(4):22-26. Zhou, K., and Liao, F. 1995. Xiang You 63-A quasi-aromatic hybrid rice with good quality and high yield. Int. Rice Res. Newsleft. 20:in press. Zhou, T. L., Shen, J. H., and Ye, F. C. 1983. A genetic analysis on the fertility of Shan type hybrid rice with wild rice cytoplasm. Acra Agron. Sin. 9(4):241-247. [Chinese with English summary] Zhou, T. B., Xiao, H. C., Lei, D. Y., and Duan, Q. X. 1988. The breeding of indica photosensitive male sterile line. J. Hunan Agric. Sci. 6:16-18. Zhou, Y. B., Yu, T. Q . , and Xiao, G. Y. 1991. Relationship between the ecological background of photoperiod-sensitive genic male sterile rice and induction effects of fertility transformation. J . Hunan Agric. CON. 17(2):99-105.
Index A Abri Pataud rock shelter at Les Eyzies, 9 A.B.’s Midden archaeological site case study methods, 60 overview, 57-60 results and discussion, 60-65 summary, 65-67 Acid detergent fiber in forage sorghums genetic variability of, 162 versus tannin content, 165 Acidity, soil at A.B.’s Midden archaeological site, 67 at El Mirador Bajo archaeological site, 32 phosphate rock dissolution rate versus soil Ca and P, 99-100 versus tritable, 94-98 Acids in chemical extraction tests of PR-treated soil, 135-136 p-coumaric in forage sorghums, 168 humic and fulvic in Vertisols, 335 organic PR dissolution, enhancement of, 104-105 zinc availability and, 272 phenolic, in forage sorghums, 165 Acidulation of phosphate rocks, partial, see Partially acidulated phosphate rocks Acid Vertisols description, 329 mineralogical properties, 328 replacement of calcium on cation exchange sites, 330 ADF, see Acid detergentjber Aggregates, Vertisol clay-organic complexes in, 335-336 clay-size crystallite structures, 307 clay-size distribution measurements, 301 formed by self-mulching, 302-304 Agriculture anthrosols produced by, 6-7 at Mayan El Mirador Bajo site, 30, 32
463
Agronomic effectiveness of phosphate rocks determining measurements, 120- I2 I overview, I 18- I 19 techniques, 119-120 quantifying performance of, 121-125 reactivity and, 79 residual effectiveness, 125-130 in Watkinson model testing, 117- I18 Agronomic traits, rice, see also Traits, genetic; specific traits cytoplasm effects on, 387-388 in parental line development, 405 wide compatibility genes and, 401-403 Alkaline Vertisols description, 329 mineralogical properties of, 327-328 Alluvial soils, archaeology and case studies of, 24-30 correspondence to site locations, 14 particle-size analyses laboratory methods, 18 southeastern U.S. case studies, 25 Aluminum in archaeological soils, 53, 55 Amazon basin, terra prera soils in, 8-9 Amendments to phosphate rocks composting with organic manures, 143 overview, 142-143 partially acidulated phosphate rocks, 145- 146 phosphate rock-sulfur assemblages, 143- 145 Ammonia loss of soil nitrogen in volatization of, 188 in nitrogen mineralization process, immobilization and, 192 Ammonium cation behavior in Vertisols, 331 in nitrogen mineralization process from animal manures, 199-200 in C:N values, 203 immobilization and, 192, 193 incubation in determining increase of, 20921 1
464
INDEX
Ammonium (continued) in measurements of gross, 216-217 potassium availability and, 260-261 Amylase activity in rice hybrids, 382 Animals excreta as mineralization source, 198, 200 nutrition of, see Forage sorghums; Ruminants production of, in measuring agronomic effectiveness of PR, 121 Anion behavior in Vertisols, 330-333 Anion exchange resin test description, 136- 137 proposed improvements to, 140- 141 Anthropic epipedon, human activities indicated by, 6; see also Anthrosols Anthrosols Amazonian ferra prefa soils, 8-9 archaeopedology in characterization of, 6-8 in formalization of archaeopedology subdiscipline, 12 soil classification techniques in analyses of, 9-10 Apatites, see also specific apatites commercial calcium phosphates, 92 DCPD model extension to dissolution rate of, 113-115 Dissolution Rate Function, 8 1-82 Olsen bicarbonate test and, 133-134 reactivity carbonate, 85-86 fluorapatite, 84-85 soil pH and plant effects versus P release, 103-104 titratable acidity versus dissolution rate, 94-98 in Vertisols, 328 Apomixis in rice current research on, 448 description, 403-404 Aquic Vertisols, 318, 319 Archaeobotany, soil science and, I 1 Archaeological sites case studies of soil-archaeological investigations A.B.'s Midden in Florida, 57-67 alluvial sequences, southeastem US.,2430 at El Mirador Bajo, 30-36 at Hadrian's Villa, chemical properties of soils at, 36-37
paleosols near Mt. Vesuvius, 37-42 Pineland, 48-57 Seminole Rest in Florida, 42-44 chronology of soil horizons at, 3, 41 predicting location of ground-penetrating radar, 17 landscape analyses in, 20-21 statistical predictive models, I3 prehistoric shell-bearing island in Florida, 57-67 radiometric '4C dating of anthrosols, 7 paleosols, 23 soil chemistry analyses, 18-20 soil morphology as stratigraphic marker in, 16-17 Archaeology applications of soil science to, 8-10 biological soil components and, 10-1 I changing emphasis in, 4-5 chemical properties of soils in, 36-37 formalization of archaeopedology subdiscipline, 11-12 geophysical tools, applicable, 17 interdisciplinary approach with soil science summary, 67-68 value of, 2-4 paleosols, 2 1-24 soils data in, see also Archaeopedology chemical properties, 36-37 landscape analyses, 20-21 micromorphology, 2 I soil laboratory analyses, 17-20 soil surveys and maps, 12-14 subfields of, modem, 5-8 underwater. 21 Archaeometry, 5-6 Archaeopedology anthrosol and paleosol characterization in, 6-8 biological soil components and, 10-1 1 formalization of, 11-12 human-modifed landscapes and settlement patterns, 8-9 Pineland site case study, 48-57 soil laboratory analyses chemical, 18-20 physical, 18 types of, 17-18 soil morphology, 14- 17 traces of daily life, interpretation of, 9-10
465
INDEX Architecture, nitrogen mineralization and soil, 207-208 Argillic horizons in archaeological studies, 16 Paleo-Indian Thunderhird site in Virginia, 25-28 Atmosphere, nitrogen cycle and mineralization field measurements and, 215 overview, 188 potential environmental impact, 189 Atterberg constants, 341-342 Auger samples from Snyder’s Mound at Seminole Rest site, 44,45 Avellino paleosol, 39, 40f. 42
B Backslope landscape position in archaeological reconstruction of landscapes, 20 Bacterial blight infestations in rice, 422 Bacteria in nitrogen mineralization aerobic versus anaerobic conditions, 204 nitrification performed by, 194 Bajos, see El Mirador Bajo archaeological site Barium in archaeological soils, 39, 40f Barley in chemical extraction studies of N mineralization, 212 in measurements of changes to mineral nitrogen, 214 nitrogen uptake by, 222r rhizosphere soil pH and PR dissolution, 104 Basmati rice hybrids, 425 Beidellite, 323; see also Smectite; Vertisols Bicarbonate in iron and zinc nutriiton of crop plants, 278 Bioarchaeology, emergence of, 4-5 Biologic properties, Vertisol clay-organic complexes, 335-336 organic constituents, 333-335 Biomass, see Soil biomass Biotechnology genetic engineering of forage sorghums, 177 rice and in estimating genetic distance among varieties, 390 hybrid breeding, 408-41 I Bloom on forage sorghum plants, 169-170, 173-175
Blooming behavior, hybrid rice, 427 Emr mutations, see Brown midrib mutants
Boron, buffer power of, 276-277 Boscoreale archaeological site, 37-42, 43f Botany, see Archaeobotany Boundaries, soil at Pineland site Citrus Grove feature, 53-54 tracing paleosol, 23 Erassica napus
L
EUF study of P kinetics, 269 phosphate uptake and acid secreted by, 104I05 Breeding forage sorghums FFV criteria, 162 germplasm sources, 173- 175 hybrid development, 171-173 selection techniques, 173 rice hybrids biotechnological applications, 408-41 1 evaluation of hybrid combinations, 408 maintainer and restorer lines, 445 overview, 404-405 parental line development, 405 seed production of experimental, 406-407 two-line, 400-401 Brown Gem Clam, 45, 47 Brown midrib mutants, forage sorghum description, 167- 169 germplasm sources, availability of, 173 isolating genes for, 177 Brown plant hopper infestions in rice, 422 Buffer capacity phosphate availability to plants from phosphate rock, 107-108 in Vertisols, 333 phosphate rock dissolution rate and calcium, 98, 100 Kirk and Nye model sensitivity to pH, 115 phosphate, 98-99 soil pH, 96-97 time of application and, 110 Buffer power curves electro-ultrafiltration in measuring P and K intensity for, 267-270 phosphate availability, for determining, 249250 Buffer power of soils concluding remarks and future imperatives, 278-280
466 Buffer power of soils (conrinued) heavy metals and contamination by, 277 plant acquisition of, buffering effect on, 277-278 precise availability prediction, 270-277 measuring P and K intensity for power curve construction, 267-270 nutrient availability and effect on, 242-246 quantifying and testing, 247-266 precise availability prediction, heavy metals and other heavy metals, 275 quantifying zinc buffer power, 273-275 zinc, 270-272 Bulk density changes in archaeological soils, 18 of Vertisols cultivation and, 361 description, 337-339 Burial sites, sand mounds at Pineland site as, 48 Buried paleosols, 22-23; see also Paleosols Buried soils in archaeological investigations, 16 as paleosols, 22
C Cadmium buffering power and uptake of, 275 effect on potassium buffering capacity and adsorption, 277 Calcareous soils, phosphate rock measurements in, 91 Calcite in dissolution of carbonate apatites, I14 as phosphate rock impurity, 85-86 Calcium, see also Calcareous soils in archaeological soils, 52-53, 55 exchange reactions with potassium, 260-261, 263-264 in measuring PR residues in soil dissolution from exchangeable, 90 extraction in inorganic P fractionation, 8788 organic matter and buffer capacity of, 100102 phosphate rock dissolution rate plants with high uptake of, 104
and soil pH, 95 in soil solution and exchangeable, 98 versus soil acidity and P, 99-100 in Vertisols cation exchange sites, 330 dominance over ammonium on exchange sites, 331 Calcium carbonate at A.B.’s Midden archaeological site, 67 reducing PR dissolution rate in soils, 92-93 Calcium phosphates in analyses of archaeological soils, 10 diffusion in soil versus pure solution, I13 dissolution of PAPRs and, 145-146 Calusa Indians, see Pineland archaeological site case study Cambic horizons, formation of Argillic horizons versus, 16 Cameroon halloysite abundance in Vertisols, 323 management of, case study of, 353-357 Vertisol degradation advanced stage of, 358f loss of vegetation, 299 water conservation and regeneration, 359f Canals at Pineland archaeological site, 48, 49 Carbohydrates in forage sorghums, 163 source of rice grain, 419 Carbon:nitrogen ratio in composted manure, 200 in determining mineralization/immobilization, 202-203 humus associated with Vertisol topsoils, 334 plants and, 206 Carbon at archaeological sites, A.B.’s Midden site, 62, 651, 67 in archaeological soil chemical analyses, 19 N transformations from animal sources, 199200 provided by dhurrin in forage sorghum leaves, I66 in Vertisols cultivation and loss of, 361 microbial biomass contribution, 35 pellic, 302 Carbonate apatites dissolution rate of, 113-1 14 substituting for phosphate, 85-86, 92
467 Carbonates, see also Calcite; Calcium carbonate in neutral and alkaline Vertisols, 327-328 in phosphate rocks in dissolution rate of carbonate apatites, 113-114 in measurement of reactivity, 8 I , 82 Carbon dating of archaeological site materials anthrosols, 7 burial mound at Pineland site, 56 paleosols, 23 sites near Mount Vesuvius, 39 in soil chemical analyses, 19 Cardamom, potassium availability to, 255-256 Cation exchange capacity of Vertisols, 329330 Cations in Vertisols behavior, 330-333 exchangeable, 329-330 liquid limits, effect on, 341 shrinkage curve and, 342 I4C dating, see Carbon dating CEC, see Cation exchange capacity Cedar Keys National Wildlife Refuge, see A.B.’s Midden archaeological site case study Cellulose in forage sorghum bmr mutants, 168I69 Cenozoic Era, Vertisol parent materials formed in, 300 Chaos Theory in Vertisol studies, 352 Chemical extraction defining chemical environment of roots, 243 EUF as alternative for extracting labile fractions of nutrients, 267 in nitrogen mineralization studies extractants used in, 21 1-212 indices of potentially available nitrogen, 208 of phosphate, discrepancies in, 249 in soil P tests, 133-136 for zinc availability to plants, 271 Chemical properties, soil, see also spect5cproperties of archaeological soils A.B.’s Midden site, 60, 62-65 Hadrian’s Villa site, 36-37 laboratory analysis methods, 18-20 Pineland site, 51-53, 55, 56-57 sites near Mount Vesuvius, 39 phosphate rock dissolution versus, 9 1-94
Vertisol cation and anion behavior, 330-333 cation exchange capacity and exchangeable cations, 329-330 pH, 329 China, hybrid rice technology in accomplishments, 41 1-412 economic analysis of, 440-44 1 grain quality improvements under development, 425 improvements to seed production, 434 Chloris gayana, nitrate depletion in Vertisols and, 332 Chromic Vertisols in classification of Vertisols, 319 description, 302 Chromosomes, rice restorer gene loci, 397 wide compatibility gene loci, 401-403 Citrus Grove at Pineland, see Pineland archaeological site case study Clam-shell mound, see Snyder’s Mound Classification systems, Vertisol, 3 16-322 Clay-organic complexes in Vertisols, 335-336 Clays, see also Argillic horizons; Vertisols at archaeological sites A.B.’s Midden site, 62 Maya, 1 1 Savanna River Valley distribution curves, 28 exchangeable potassium and, 254 nitrogen mineralization and, 205 Vertisols COLE values and content of, 339 content required for formation of, 295-298 liquid limit versus, 341 microstructure and shrink-swell phenomena, 307-310 phyllosilicates, 322-326 plastic limit versus, 342 as potassium source, 330-331 relative age versus parent materials, 300 resilience to degradation, 290, 363 texture, 301 zinc buffer power and uptake in, 273-227 Clay-size mineral species at Seminole Rest site, 45, 46 Climate, see also Paleoclimate buried paleosols and in characterization of, 22 in studying paleopedological processes, 23
468
INDEX
Climate (continued) Vertisols and clay-organic complex aggregates, 335336 in formation of, 298 global distribution and, 292 in West Africa, 353 Clover interlayer potassium and red, 258 nitrogen mineralization after ploughing, 198199, 221 PR dissolution versus application method, 105-106 Clovis clay, 25-28 Coastal environments A.B.’s Midden archaeological site case study, 57-67 Pineland archaeological site, 48-57 Seminole Rest site case study, 42-48 Cobweb Swamp of Belize, I I Coefficient of Linear Extensibility, in Vertisols, 337-339 COLE, see Coefficient of Linear Extensibility Colha, Mayan cultural center of, I 1 Colombia, hybrid rice technology transfer in, 443 Color, see ulso Morphology, soil forage sorghum plant mineralogical properties and, 327 versus tannin content, 166 of paleosols versus present day soils, 22-23 Vertisol deemphasis in classification of, 320 in early classification of, 319 organic matter content and, 334 types compared, 302 Combining ability rice heterosis and, 385-386 in testing value of parental lines, 390-392 Composting C:N ratio of manure, 200 PRs with organic manures, 143 Conductivity, see Electrical conductivity, soil; Hydraulic conductivity of Vertisols Conservation tillage farming of clayey soils, 361 Consistence, Vertisol, 339-342 Copper, potassium buffering capacity and, 277 Core sampling methods of archaeological soils,
44 Corn, EUF study of P kinetics in, 269 Cotton, potassium uptake of, 258
Coulomb-Mohr theory of shear failure, 310315 P-Coumaric acid in brown midrib mutants, 168 CP, see Crude protein Cracks, see Dessication cracks Cretaceous Era, Vertisol parent materials formed in, 300 Crops, see also Cultivation; Plants; specific crops availability of P from PR to versus planting time, 109-1 11, 140 versus species of, 109 bicarbonate in Fe and Zn nutrition of, 278 forage, plant fiber in, 176; see also Forage sorghums nitrogen and concentration in residues of, 201-202 mineralization measurements based on, 208-209 removal of, in harvested, 188 returned to soil after harvest, 198 sources of total uptake by, 221 nutrients in sustainable production of, 239240 potassium availability to perennial, predicting, 254259 commercial significance of K buffer power determination for perennial, 259-260 soil organic matter levels and rotation of, 189-190 types suited for phosphate rocks, 146 on Vertisols cracking patterns and, 300 in high-input cultural systems, 357-363 in low-input cultural systems, 353-357 Crop yields, see also Heterosis in rice in measuring agronomic effectiveness of PR, 120- 121 MYR aPhosphate rocksoach in maximizing, 262 rice economic analysis of, 438-440 in evaluation of experimental hybrids, 408 sink in hybrid versus inbred, 421 stability of heterotic, 415-419 Crude protein in forage sorghums genetic variability for, 162 phophorus content and, 176 tannin versus digestibility of, 164 Cryert Vertisols, 320-321
INDEX Cultivars forage sorghum leaf and stem variations among, 163 selection of, 176 hybrid rice, production of seed from, 432434 Cultivation, see also Crops; Crop yields nitrogen mineralization and earthworm mortality resulting from, 196 effects of, 206-207 soil organic matter levels resulting from, 189-191 as source of organic matter for mineralization, 198 of Vertisols bulk density, effect on, 339 consistence, effects on, 340-341 high-input cultural systems, 357-363 low-input cultural systems, 353-357 pedality and porosity, effects on, 346-352 sorghum production in West Africa, 353355 Cultures diversity of anthrosols produced by, 6 Vertisol management case studies of high-input, 357-363 low-input, 353-357 Cyanogenic glycosides in forage sorghums, 166 Cytoplasmic-genetic male sterility lines basmati rice under development from, 425 bred by IRRI, 412-414 future outlook of, 445-446 hybrid rice technology accomplishments in China, 41 1-412 in parental line development, 405 in seed production of experimental hybrids, 406 seed production practices, 43 1-434 as tool for developing rice hybrids, 393-395 transfer into elite breeding lines, 408, 409 Cytoplasm, rice agronomic traits and, 387-388 male sterility inducing, 393-394
D Data, archaeological interpretation of soil landscape analyses, 20-21 micromorphology, 2 1 soil laboratory analyses, 17-20
469
soil morphology, 14-17 soil surveys and maps, 12-14 DCPD, see Dicalcium phosphate dihydrate Decomposition of soil organic matter mathematical models, 191 of microbes, in nitrogen immobilization, 193 organisms involved, 196 Degradation, soil by humans, 238-239 Vertisols clay-organic complexes in, 336 factors influencing, 290-29 I in high-cultural input systems, 360-361 high pH conditions causing, 329 massive structure formation, 304 in West Africa, 357 Delineations, soil map, 13 Deoxyribonucleic acid, transformation of rice, 408 Depleted soil, see also Degradation, soil as anthrosol, 7 caused by inadequate replenishment of nutrients, 241 Depositional history of archaeological soils particle-size analyses in determining, I8 quantified by soil morphological features, 21 soil morphology in, 16 Depositional surfaces in archaeological reconstruction of landscapes, 20 Depressions in Vertisol gilgai structures, 304305 Dessication cracks in slickenside genesis, 314-315 in Vertisol suborder classification, 319 Dhurrin in forage sorghum leaves, 166-167 germplasm for sudangrasses low in, 173 Diagnostic and Recommendation Integrated Systems, 241-242 Diapirs, 304, 307 Dicalcium phosphate dihydrate, model for dissolution of, I13 Dicalcium phosphate in partially acidulated phosphate rocks, 145 Differential loading model of Vertisol genesis, 310 Diffusion gas, in Vertisols, 343 nitrogen mineralization and, microsites and, 207-208 phosphate fertilizer, 249
470
INDEX
Diffusion (continued) potassium fertilizer, 255 of zinc to plant roots, 270-271, 272 Digestibility, forage sorghum lignin content and, 167, 176 microanatomical features of leaves and stems and, 163 tannin content and, 164-166 temperate versus tropical, I62 Disease/insect resistance of hybrid rice description, 42 1-424 parental line selection for, 446 Dissolution of phosphate rocks availability to plants crop characteristics. 109 management practices, 109- I 1 I overview, 107 soil factors, 107-109 in economic evaluation of PR use, 131 measurement of in soils in acid soils, 86-90 in calcareous soils, 91 overview, 86 modeling rate of, in field soil Kirk and Nye Model, I 13-1 15 overview, I 1 1-1 13 Watkinson model, 115-1 18 partially acidulated, 145-146 rate of P release from method and rate of application, 105-106 plant effects, 103- 105 PR properties affecting, 91-94 site factors affecting, 102-103 soil properties affecting, 94- 102 reactivity and Dissolution Rate Function, 82 as measure of, 79 Dissolution Rate Function, 82-84 Distribution of Vertisols in United States, 292-294 world, 292, 293s Dolomite dissolution of phosphate rock containing, I13 as phosphate rock impurity, 85-86 Domains, Vertisol clay-size crystallite structures, 307 in measuring clay-size distribution, 301 Drainage in archaeological use of soil maps, 14 Vertisol types occuring with good, 299 DRF, see Dissolution Rate Function
DRIS, see Diagnostic and Recommendation Integrated Systems Dry matter, rice hybrids versus inbred, 419, 420 sink formation in hybrids and, 382 DTPA extraction in zinc availability studies, 271
E Earth science, archaeological studies and, 2-3 Earthworks, see Mounds Earthworms cultivation and mortality of, 196 in nitrogen mineralization, 197 in Vertisols, 334-335 Ecology, human anthrosols in interpretation of, 7 as unifying theme in archaeology, 4-5 Economics hybrid rice technology, viability of, 438-442 of phosphate rock fertilizers, 130-132 of potassium fertilizer, 261-263 Egypt, hybrid rice technology transfer in, 443 Electrical conductivity, soil of El Mirador Bajo soils, 31, 35r, 36 solute movement and, 274 Electromagnetics in soil science and archaeology, 17 Electro-ultrafiltration labile nitrogen measurements, 212 P and K intensity measurements, 267-270 Elements, see also Chemical properties, soil; Heavy metals; specific elements at Hadrian’s Villa archaeological site, 36-37 from soils near Mount Vesuvius, 37, 39f, 40f. 42 Elerraria cardamomurn M, potassium availability to, 255-256 Elite breeding lines, rice CMS transfer into, 408, 409 disease and insect resistance, 422 yield advantage of IRRI-bred, 418r El Mirador Bajo archaeological site, 30-36 Environment, see also Paleoenvironments; Stress environments, hybrid rices and controls exerted over mineralization by, 203206 fertilizers and nitrogen loss to atmosphere and, 189 phosphate rocks, benign effect of, 78
INDEX reducing nitrogen mineralization impact on, 222-224 human impact on anthrosols in analysis of, 7-8 in archaeological studies of Holocene, 2-3 as new focus in archaeology, 4-5 Pineland site case study, 48 Environmental archaeology, see GeoarchaeologY Environment-sensitive genic male sterility in rice description, 398-401 seed production practices using, 431-434 Epicuticular waxes on forage sorghum plants, 169-170 Erosion in archaeological reconstruction of landscapes, 20 exhumed paleosols resulting from, 22 human-induced as anthrosol, 7 at Mayan Colha site, I I quantified by soil morphological features, 21 replenishment of nutrient loses caused by, 239 Shell midden sites retarding shoreline, 62 of Vertisols topography, effect on, 299 in West Africa, 357 EUF, see Electro-ultrafiltration Exhumed soils as paleosols, 22 Experimental rice hybrids evaluation of, 408 seed production for, 406-407 Extractants, see also Chemical extraction failure to precisely define roots’ chemical environments, 243 for soil P tests, 133
F F, rice hybrids CMS lines in development of, 393-395 grain quality, 424 parental line development, 405 Facultative female sterile rice seeds, 447 Fatty acids in forage sorghums, 169-170 Fauna, soil, see also Soil biomass in nitrogen mineralization, 196- 197 at Snyder’s Mound at Seminole Rest site, 45, 47 in Vertisols, 334-335
47 1
Feces in nitrogen mineralization, see also Manure N returned by grazing, 198 versus stored manures, 200 Feldspars as potassium source in alkali Vertisols, 33033 I as Vertisol constitiuents, 327 Fertility indices, 241-242 Fertility, rice, see also Sterility in rice hybrids environmental effects on, 398-401 restoration of, description, 395-398 temperature and, 398-401 Fertilizers, see also Manures cost-effective recommendations, importance of, 279-280 equilibration in soil of soluble, 249 hybrid rice responsiveness to, 420 nitrogen, see also Nitrogen mineralization additions to soil via, 188 improving advice for, 224 management models, 218 lSN-labled in mineralization field studies, 2 15-216 nitrification rates and, 195 organic N levels and, 190 overtillering of rice and, 420 phosphate rocks as direct application, 78-79; see also Phosphate rocks phosphorus as archaeological indicator of human activity, 42 PRs versus soluble, 125 potassium commercial significance of buffer power determination, 259-260 MYR and MEY in buffer power measurements, 261-263 predicting mobility of, in soil, 244-246 FFV, see Forage feeding value Fiber in forage crops, 176; see also Cellulose; Lignin; Polysaccharides Fiddle Crab Mound archaeological site results of soil analyses at, 45 soil sampling methods used at, 44 Field measurements and experiments of nitrogen mineralization applicability of methods for, 215 changes in mineral N, 213-214 field incubations, 214-215 lsN-labeled fertilizer, use of, 215-216
472
INDEX
Field measurements and experiments (conrinued) phosphate rocks, determining agronomic effectiveness of, 120 Flavanones in forage sorghums, 165- 166 Flavonoids in forage sorghums, 164-165 Floral traits, hybrid rice, 427-429 Florida, see A.B.’s Midden archaeological site case study; Pineland archaeological site case study; Seminole Rest archaeological site Flowering process, hybrid rice in guidelines for seed production, 431 outcrossing and, 427 in seed production practice, 431-432 Fluorapatite, lack of solubility of, 84-85 Fluorine, phosphate rock properties versus amount of, 85 Footslope landscape position, in archaeological reconstruction of landscapes, 20 Forage feeding value, definition, 162 Forage sorghums breeding, I7 I - I73 genetic parameters FFV criteria, 162 nutritional quality, 163-171 germplasm, 173-175 types of, 161-162 Fractals in Vertisol studies, 352 Fractionation procedures phosphate rock in soils ANAOH-P and ACA, dissolved from, 90 inorganic P, 86-90 soil organic matter chemical structures, 191 Fulvic acid in Vertisols, 335 Functions for phosphate rock response patterns, 124- 125
G Gas diffusion in Vertisols, 343 General combining ability rice heterosis and, 385-386 in testing value of parental lines, 390-392 Genes forage sorghum for bloom, 170 for carbohydrate accumulation, 176 for cyanogenic glycosides, 166- 167 dwarf, 175-176 for LAC content, 165
in modifying protein, 177 needed for improving quality, 177-178 for stem sweetness, 171 rice in CGS systems, 393 of inbred, 404 for recessive tall plant type, 427 tagging and transfer of, 408, 409 TGMS and PGMS, 399-400 wide compatibility, 401-403 for yield, limitations of, 385-386 Genetic diversity, rice in breeding for heterosis, 388-392 wide compatibility genes and, 401 Genetic parameters of forage sorghums FFV criteria, 162 nutritional quality antiquality factors, 164-167 lignin and brown midrib mutants, 167-169 lipids in bloom versus bloomless, 169-170 overview, 163 protein, 163- 164 stem sweetness, 170-171 Genetics as basis for heterosis in rice, 382-388 of fertility restoration in rice, 395-398 Genetic tools for developing rice hybrids apomixis, 403-404 cytoplasmic-genetic male sterility, 393-395 environment-sensitive genic male sterility, 398-401 fertility restoration, 395-398 wide compatibility genes, 401-403 Geoarchaeology archaeopedology as component of, 1 1- 12 description, 5 Geographical Resources Analysis Support System, 13 Geographic information systems soil maps for archaeological purposes and, 9 Vertisol structure and porosity, in study of, 352 Geologic eras in Vertisol formation, 300-301 Geomorphology differential erosion at A.B.’s Midden altering, 62 geoarchaeology in analysis of, 5 Seminole Rest site, 42-48 in validating paleosols, 23 Geophysical tools for soil science and archaeology, 17
INDEX Georgia, see Savannah River Valley archaeological sites Germplasm, forage sorghum, 173-175 Gilgai structures, see also Microtopography in differential loading model, 310 structures and types of, 304-306 topographic position of Vertisols and, 299 vegetation on, 299-300 in Vertisol classification, 320 CIS, see Geographic Information System Global distribution of Vertisols, 292, 293f Glossiness of forage sorghum seedlings, 170 Gossypium hirsutum L, potassium uptake of, 258 GPR, see Ground-penetrating radar Grains forage sorghum silage types, 176 rice in estimating genetic distance between parents, 389 quality of hybrid, 424-425, 446 GRASS, see Geographical Resources Analysis Support System Grasses, see Forage sorghums Grasslands, nitrogen mineralization of impact on, 219-221 mineral N measurements, 214 nitrification in soils of, 195 nitrogen turnover after cutting and grazing, 198 seasons and, 204, 205 Green leafhopper infestations in rice, 422 Gross nitrogen mineralization basing mechanistic models on, 194 definition, 193 difficulty of measuring, 208 measurement of, 216-217 Ground-penetrating radar in archaeometry, 5-6 for detection of buried archaeological sites, 17 Groundwater contamination by soil nutrient build-up, 240-241 Grumic Vertisols, 318 Grumusols, 317-318; see also Vertisols Guatemala, see El Mirador Bajo archaeological site
H Hadrian’s Villa archaeological site, 36-37 Halloysite, 323
473
Hand-crossing in hybrid rice seed production, 407 Haploidization by argillipedoturbation, 3 10 Heavy metals buffering of major elements and contamination by, 277 possible effect on plant acquisition of, 277278 buffer power and availability of, 270-277 at Hadrian’s Villa archaeological site, 3637 reduction in nitrogen mineralization by, 206 Herculaneum archaeological site, 37-42 Heterobeltiosis in rice, 380 Heterosis in rice, see also Agronomic traits, rice demonstration of commercial exploitation, 378 description, 379-382 genetic basis for, 382-388 prediction of, 388-392 stress environments and, 444-445 Heterotic rice hybrids development of inside of China, 41 1-412 outside of China, 414-415 yield stability of, 415-419 HIS, see Hydroxy-interlayered smectite Hiwasse River archaeological site case study, 29-30 Hog-wallows, formation of, 310 Holocene period archaeological studies of, importance of, 2-3 paleosols as boundary markers with Pleistocene, 6-8 time period covered by, 2 Hordeum vulgare L , 104; see also Barley Horizons, soil, see also Morphology, soil archaeology and A.B.’s Midden site, 62 anthrosols, 6-7 of buried soils, 22 chronology at archeological sites, 3, 4f at Pineland site, 50-51, 53-54, 56-57 in soil morphological descriptions, 14- 15 Vertisol organic matter turnover rates, 334 structure, 302-304 structure and porosity variations, 346 vertic, proposal of, 319-320 wet-dry cycles and, 328
474
INDEX
Horticulture in archaeological interpretation, 1 1 archaeopedological analysis of Mayan, 11 Humic acid in Vertisols, 335 Humus, Vertisol topsoil carbon:nitrogen ratio, 334 in clay-organic complexes, 335 Hybrid rice accomplishments in China, 411-412 outside China, 412-419 adaptability to stress environments, 425-426 agronomic management, 419-421 breeding procedures for developing biotechnological applications, 408-41 1 evaluation of hybrid combinations, 408 parental line development, 405 seed production of experimental, 406-407 conclusions, 448-449 disease/insect resistance, 421-424 economic analysis, 438-442 future outlook, 444-448 genetic tools for developing apomixis, 403-404 fertility restoration, 395-398 male sterility cytoplasmic-genetic, 393-395 environment-sensitive genic, 398-401 wide compatibility genes, 401-403 grain quality, 424-425 heterosis in genetic basis for, 382-388 overview, 379-382 prediction of, 388-392 overview, 378-379 seed production, hybrid guidelines, 430-43 1 outcrossing floral traits and, 427-429 flowering behavior and, 427 natural mechanisms for, 429-430 plant characteristics and, 426-427 performance of, 434-438 practices, 43 1-434 technology transfer and policy issues, 442-
444 Hybrids, see also Hybrid rice crop breeding and influence of maize, 378 forage sorghum developing, 17 1- 173 as type of, 161
Hydraulic conductivity of Vertisols, 343-345 Hydroxy-interlayered smectite, 322
I ICAP, see Inductively coupled argon plasmaatomic emission spectrometer Igneous soils as Vertisol parent material, 295 Mite shrink-swell potential, 326 as Vertisol constituent, 323 Image analysis techniques in Vertisol porosity and structure analysis, 346 Immobilization, nitrogen, see also Nitrogen mineralization carbon from manures and, 200 description, 192-194 Inbred rices breeding procedures versus hybrids, 404 comparing to hybrids to estimate heterosis, 379 in development of hybrid parental lines, 405 Inclusions, archaepedological interpretation of soil, 7 Incubation, soil nitrogen mineralization field techniques, 214-215 indices of potential, 209-21 1 INDF, see Indigestible neutral detergent fiber India as potential largest supplier of hybrid rice technology, 441-442 yield advantage of heterotic rices, 419 Indica rices environment-sensitive genic male sterility, 398 production improvements in China, 412 recessive gene for tall plant type, 427 restorer genes in, 397-398 Indigestible neutral detergent fiber, 167 Indonesia commercial hybrid rice seed industry, 443 economic analysis of hybrid rice production in, 441 Inductively coupled argon plasma-atomic emission spectrometer, 36 Insect resistance of hybrid rice, 421-424 International Rice Research Institute initiation of work with hybrid rice, 378 revival of hybrid rice research by, 412
475
INDEX In v i m dry matter digestibility, forage sorghum advantages of improving, 163 bloom versus bloomless plants, 170 genetic variability of, 162 parental lines of hybrids, 175 tannin content and, 165 Ion exchange resins, 136- I37 IRAE, see Isotopic Relative Agronimic Effectiveness Iron buffer power of, 276 root response strategies to deficiencies of, 277-278 in Vertisols color due to oxides of, 302 smectites rich in, 326 Iron oxide-impregnated paper, 137- 139 IRRI, see International Rice Research Institute Irrigation hybrid rice, economic analyses of, 441 of Vertisols in West Africa, 355, 357 Islands, see A.B .'s Midden archaeological site case study Isotopes in measuring gross mineralization, 208, 217 PR fertilizer effectiveness measurements with radioisotopes, 121 Isotopic ion exchange test, 139 Isotopic Relative Agronimic Effectiveness, 121I24 Italy, see Hadrian's Villa archaeological site; Vesuvius archaeological sites, Mount IVDMD, see In v i m dry matter digestibility
J Japan, technology transfer of hybrid rice in, 443 Japonicalindica hybrid rice expanded leaf area and heterosis, 380-382 wide compatibility genes for overcoming sterility in, 401-403 Japonica rices consumer preference and higher price of, 440-44 I environment-sensitive genic male sterility, 412 genetic diversity from indica rices, 389-390 lack of restorer genes, 398 recessive gene for tall plant type, 427
K Kaolinite in interpretation of palemtimate, 9 shrink-swell potential, 326 in Vertisols, 323 Kinetics EUF in study of P, 268 nitrogen mineralization in incubation studies, 21 I overview, 201 phosphate rock dissolution models compared, 1 1 1- 1 13 Watkinson model, 116 Kirk and Nye model compared to other models, 1 1 1- I 13 description, 113-1 15 experimental results, 115 Korea, technology transfer of hybrid rice in, 443
L Laboratory analyses, soil of archaeological soils, 17-20 of potential mineralization indices chemical methods, 21 1-213 incubation methods, 209-21 1 LAC, see Leucoanthocyanidins LAI, see Leaf area index for rice Landscapes anthrosols and modification of, 7 archaeological analyses of, 20-21 archaeological case studies El Mirador Bajo site, 31 Hiwasse River site, 29-30 Pineland site, 48-57 human-modified, 8-9 soil formed on past, see Paleosols Land use systems Nambillo ridgetop archaeological site, 8 nitrate loss from soil profiles in Vertisols, 332 Vertisol bulk density differences among, 339 organic matter content, maintenance of, 336 structure and porosity variations among, 346 Layer charge potassium anion and cation behavior in Vertisols, 331 in smectite minerals, 326
476
INDEX
Leaching nitrogen loss through, 188, 204 nitrogen mineralization controlling nitrate losses by, 225 Lead in archaeological soils at Hadrian’s Villa, 36-37 at sites near Mount Vesuvius, 39, 40f Leafstem ratio of forage sorghums, 163, 175 Leaf area index for rice, 380 Leaves forage sorghum brown midrib mutants, 167-169 cyanogenic glycosides in, 166- 167 digestibility rates in ruminants and, 163 tannin content, 164 nitrogen loss from, 209 rice, outcrossing and, 426 Leucoanthocyanidins in forage sorghum description, 164-165 isolating genes for, I77 versus dhurrin levels, 167 Lignin in forage sorghums mutants, 167- 169 in predicting nitrogen mineralization, 213 Lime in economic evaluation of PR use, 132 Limestone at El Mirador Bajo archaeological site, 31 Linear shrinkage phase of Vertisols, 342; see also Shrink-swell phenomena, Vertisol Linear xylan, 168 Lipids in clay-organic complexes in Vertisols, 336 in forage sorghums, 169-170 Liquid limit in Vertisols, 341 Lolium perenne L, see also Ryegrass EUF study of P kinetics, 268-269 potassium uptake, 257-258 Lupinus angustifolius L, 104
M Magnesium in chemical analyses of archaeological soils, 19
Vertisol cation exchange sites, 330 Male sterile lines, rice, see also Cytoplasmicgenetic male sterility lines; Temperaturesensitive genic male sterility lines in seed production of experimental hybrids, 406-407
seed production practices when using, 431434 seed yield, factors influencing, 430-43 1 Management agronomic, of hybrid rice description, 4 19-42 1 versus inbred rices, 445 of nitrogen mineralization, models for, 218219 improving, 224 plant nutrient, as key to sustainable soil management, 239-242 of PR application versus P availability to plants, 109- 11 1 soil biomass, impact on, 196 of Vertisols high-input cultural systems, 357-363 low-input cultural systems, 353-357 organic matter content, maintenance of, 336 overview, 352-353 Manganese buffer power of, 276 in Vertisols, 327 Manures, see also Feces; Urine composting of PRs with organic, 143 in nitrogen mineralization C:N ratio, 202-203 laboratory extractions for N in, 213 long term implications, 221-222 nitrogen content, 199-200 Maps, soil, see Soil maps Maps of subsurface element distributions, 5-6 Marbut’s classification system, 3 16 Marshes, see Seminole Rest archaeological site Mathematical functions for phosphate rock response patterns, 124- 125 Maximum Economic Yield, 261-263 Maximum Yield Research, 261-263 Maya clay, I I Mayan civilization agroecological evolution of Colha site, 11 El Mirador Bajo case study, 30-36 Mazic Vertisols, 318 Measurement of agronomic effectiveness of phosphate rock, 120-121 of K buffer power, 261-263 of nitrogen mineralization field measurements, 213-216
INDEX of gross, 216-217 laboratory determination of potential indices of, 209-213 from N balance/cropping data, 208-209 overview, 208 prediction models, 217-219 of nutrient buffer power and effect on concentrations on root surfaces, 244-246 p h o s p h o ~ s ,248-250 of phosphate rock reactivity methods, 81-84 precautions, 79-80 principles of, 80-81 types of, 80 of Vertisol properties, bulk density, 339 Mercato paleosol, 39, 40f, 42 Meso-micromorphology techniques, 346 Metals, see Heavy metals; specific metals Metamorphic soils as Vertisol parent material, 295 MEY, see Maximum Economic Yield; Maximum Yield Research Micromorphology, 21; see also Morphology, soil Micro-organisms in nitrogen mineralization, see also Soil biomass, microbial in clay soils, 205 temperature and, 204 Microsites, soil diffusional constraints, 207-208 distribution, 193- 194 in prediction models, 218 Microtopography, topographic position of Vertisols and, 299 Middens A.B.’s Midden case study, 57-67 as anthrosols, 6, 7 at Seminole Rest site, 44, 47 Mineralization, see Nitrogen mineralization Mineralization-Immobilization-Turnover, 193 Mineralogy and phosphate rock reactivity carbonate apatites, 85-86 fluorapatite, 84-85 Vertisol properties other minerals, 326-328 phyllosilicates, 322-326 topography, effect of, 299
477
Minerals; see also specific minerals phosphate rock classes based on, 92 at Seminole Rest site Synder’s Mound, 4647 Vertisol, versus pigmentation of, 302 MIT, see Mineralization-Immobilization-Turnover Mitochondria1 complementation, 392 Models nitrogen mineralization mechanistic, of gross, 194 need for extending database, 2 I9 predicting, 217-2 I9 of phosphate rock dissolution, I 1 1-1 18 for Vertisol genesis, 310-315 Moisture, soil nitrogen mineralization and changes in availability of, 204 distribution of, 204-205 microbial activity and, 203 phosphate rock and diffusion versus, 112 dissolution rate versus, 102 in Kirk and Nye model testing, 115 Vertisols and in formation and distribution of, 298 mineral phase sensitivity to, 328 retention characteristics, 338r, 343 suborders based on, 319 Molybdenum buffering power and uptake of, 275 in determining agronomic effectiveness of PR, 119 Monocalcium phosphate, 145 Montmorillonite, 9; see also Smectite; Vertisols Morphology, soil, see also Horizons, soil; Micromorphology in archaeological interpretation A.B.’s Midden case study, 62-64 description, 14- 17 El Mirador Bajo case study, 32, 33t, 35t Savannah River Valley case study, 27t Vesuvius case study, Mount, 37-42 Vertisol properties color, 302 overview, 301 structure and special physical features, 302-307 texture, 301 Mounds burial, at Pineland archaeological site, 48
478 Mounds (continued) shell at Pineland archaeological site, 48 at Seminole Rest site, 42-48 in Vertisol gilgai structures, 304 Mukkara, see Gilgai structures Mutants forage sorghum, 167-169 rice, 404
N Nambillo ridgetop site, 8 Net mineralization and immobilization, 193I94 Nitrate nitrogen mineralization and cultivation timing versus release of, 206 incubation in determining increase of, 209211 in nitrification process, 195 toxicity of, in sorghum forages, 177 in Vertisols anion behavior of, 331-332 cation and anion behavior, 331-332 Nitrification description, 194 in Vertisol anion behavior, 331-332 Nitrobacteracae, 194 Nitrogen, see also Carbon:Nitrogen ratio; Nitrate chemical extractants in quanitfying release from labile SOM, 21 1-212 concentration in plants and crop residues, 201-202 field measurements of changes in mineral, 213-214 forage sorghums and correlation with height, 176 increased utilization in ruminants by forage tannins, 164 hybrid rice uptake of, 420-421 labile, electro-ultrafiltration in measuring, 212 measurement of crop removal of, 208-209 in nitrification process, 194- 195 as pollutant, 241 P release from phosphate rocks and effect of, 103
priming effect on mineralization, postulated, 205, 216 Nitrogen balance in determining mineralization, 208-209 in measuring gross mineralization, 216-217 Nitrogen mineralization conclusions and future progress, 224-225 impact of, 219-224 measurement and prediction of field measurements, 213-216 gross mineralization, 216-217 laboratory determination of potential indices, 209-213 N balance/cropping data, 208-209 overview, 208 prediction models, 217-219 overview, 188- 189 pools and processes mineralization/immobilization, 192-194 nitrification, 194- I95 soil biomass, 195-197 soil organic matter, 189- 191 process controls effects of cultivation, 206-207 microsites, diffusional constraints, and soil architecture, 207-208 overview, 201 resource quality, 201 -203 Nontronite, 323-324; see also Smectite; Vertisols Nucleus, rice cell, 394 Nuram gilgai, 306 Nutrients, plant buffer power of soils effect on availability of, 242-246 quantifying and testing effect on phosphorus, 247-254 potassium, 254-266 increasing rice yields, 421 management of overview, 239-240 soil tests and availability of, 240-242 PR agronomic effectiveness and, 118-119 in Vertisol cation and anion behavior, 330333 Nutritional quality, forage sorghum antiquality factors, 164- 167 lignin and brown midrib mutants, 167-169 lipids in bloom versus bloomless, 169-170 overview, 163
479 protein, 163- 164 stem sweetness, 170-171 Nye model, see Kirk and Nye model
0 Oklawaha River, underwater archaeological studies of, 21 Olsen bicarbonate test, 133- 134 Organic matter, soil in chemical analyses of archaeological soils, 19 in nitrogen mineralization cultivation modifying, 206-207 effects of additions of, 197-201 gross mineralization measurements, 217 immobilization and, 192-194 and kinetics of, 201 as pool, 189-191 in predicting potential to supply N, 212213 quality of, as resource for, 201-203 restricting potential for, 203 soil texture and, 205 phosphate rock dissolution rate and, 100-102 in Vertisols, distribution and biology of, 333335 Oryza longistimanafa,429 Oryza perennis, 393-394 Oryza rujipogon, 393-394 Oryza sativa f. spontanea male sterility inducing cytoplasm in, 394 transfer of long stigma from 0.safiva to, 429 Ottaviono quarry archaeological site, 37-42 Outcrossing in hybrid rice seed production for experimental hybrids, 407 floral traits influencing, 427-429 flowering behavior and, 427 natural mechanism for, 429-430 plant characteristics influencing, 426-427 Oxygen flux density in Vertisols, 343 in nitrogen mineralization, 203
P P, see Phosphorus Paleoclimate paleosols in predicting Vertisol behavior, 298 of southwestern France, interpretation of, 9
Paleoenvironments, 20 Paleo-Indian sites alluvial sequence at Thunderbird site in Virginia, 25-28 paleosols at Savannah River flood plain, 24 Paleolithic era in southwestem France, 9 Paleomorphology of Nambillo ridgetop archaeological site, 8-9 Paleosols in archaeological interpretations characterization of, 6 evolution of Nambillo ridgetop site, 8 importance of buried, 22-23 landscape positions of, 20 Mt. Vesuvius site case study, 37-42 overview, 21-22 Savannah River site case study, 24 soil morphology at sites, 16 in predicting Vertisol behavior, 298 Panicles, rice in estimating genetic distance between parents, 389 rice, 426 PAPRs, see Partially acidulated phosphate rocks Paralelepipeds as morphogenic markers, 304 Parastarte triquetra, 45, 47 Parental lines, hybrid rice bred by IRRI, 412-414 developing, 405 in grain quality, 424-425 Parent materials, Vertisol description, 295-298 geologic timescale, 300-30 1 Parents, rice plant characteristics influencing outcrossing, 426-427 in predicting heterosis, 388-392 Parthenogenesis in rice, see Apomixis in rice Partially acidulated phosphate rocks, 143 Particle-size analyses, soil of archaeological soils, 18 A.B.'s Midden site, 63t, 64t at Pineland site, 49-50 of Snyder's Mound at Seminole Rest site,
44 Particle size, soil phosphate rock dissolution rate versus, 94 Vertisol clay aggregates, 301 Pedality of Vertisols chisel tillage and, 346-352
480
INDEX
Pedality of Vertisols (continued) continuous cultivation and, 361 Pedoarchaeology, A.B.’s Midden case study of, 57-67 Pedogenic processes, Vertisol structure and, 302-304 Pedology, see also Archaeopedology; Pedoarchaeology; Soil science in archaeological interpretation of paleosols, 23 paleopedology subdiscipline, 2 1 Pedoturbation model, 3 10 Pellic Vertisols in classification of Vertisols, 319 description, 302 inclusion of all Vertisol great-groups into, 322 PGMS lines, see Photoperiod-sensitive genic male sterility lines Phenols, see also Tannins, forage in forage sorghums, 165 hybrid rice gene for reaction to, 447 Phenotypes forage sorghum bmr mutant, 169 RRPS selection technique, 173 transposon tagging to isolate genes for, 177 Philippines commercial hybrid rice seed industry, 443 yield advantage of heterotic rices, 416-419 Phosphate determining availability of as nutrient, 248254 molybdenum availability and, 275 phosphate rocks and diffusion into soil from, I 1 I- 1 13 in soil solution and buffer capacity, 98-99 in Vertisols anion behavior, 332-333 mineralogical properties, 328 Phosphate rocks agronomic effectiveness of determining, 118-121 quantifying comparative performance of, 121- 125 residual effectiveness, 125- 130 amendments to assemblages with sulfur, 143- 145 composting with organic manures, 143 partially acidulated, 145-146 conclusions, 146
dissolution in soil acid soils, measurement in, 86-90 availability of P to plants, 107-1 I 1 calcareous soils, measurement in, 91 modeling rate of, 1 1 1-1 18 rate of P release, 91-106 economics of use as fertilizer, 130-132 future research needs, 139-142 soil testing where used, current research, 133- 139 EUF study of P kinetics, 268-269 overview, 78-79 reactivity of definition, 79-80 measurement of, 80-84 mineralogy and, 84-86 soil testing current research, 133- 139 future research needs, 139- 142 Phosphoric acid in phosphate rock acidulation, 145 Phosphorus in archaeological soils A.B.’s Midden site, 62-64, 65t in anthropic epipedons, 6 Pineland site, 53 quantification of, 9-10 from sites near Vesuvius, 42, 43f in soil chemical analyses, 18-19 buffer power of depeletion and replenishment at root surface, 246 EUF in constructing power curves, 267270 quantifying availability of, 247-254 hybrid versus inbred rice use efficiency, 421 phosphate rocks and, see also Phosphate rocks in chemical extraction tests, 133- 136 content versus other fertilizers, 78 in empirical measurement of reactivity, 8182 future research needs for estimating available, 139-132 inorganic P fractionation, 86-90, 91 ion exchange resins for estimating available, 136- I37 radioisotopes of, in evaluating, 121 rate of release from soil, 91-106 Quality/Intensity relationship and, 250-254
INDEX
48 1
in Vertisols, phosphate anion behavior, 332phosphate rocks and 333 availability to, from dissolved, 107-1 1 1 Photoperiod-sensitive genic male sterility lines effects on P release from, 103-105 description, 398-401 growth of, in agronomic effectiveness, I18 in parental line development, 405 Plasticity index, see also Atterberg constants Photosynthesis in heterosis of hybrid rice, 382 description, 342 pH values, soil values reported for, 3381 A.B.’s Midden archaeological site, 62-64 Plastic limit, 342; see also Atterberg constants El Mirador Bajo archaeological site, 32, 35r, Pleistocene period 36 landscape changes resulting from glaciation, nitrogen mineralization and, 205 23 phosphate rock and paleosols as boundary markers with Holodissolution rate versus, 94-98 cene, 6-8 in Kirk and Nye model testing, 115 Ploughing, nitrogen mineralization and in pot experiments, 119 increase after, 198- 199, 22 1 reactivity versus, 80 versus drilling, 206 root-induced, 103- 105 Policy issues, hybrid rice, 442-444 Vertisols and Pollen in chemical properties of, 329 in archaeopedological analysis of Mayan horminerals indicating extremes of, 328 ticulture, I I zinc availability and, 272 rice Phyllosilicates, 322-326 abortion of, in cytomplasm and nucleus Physical properties of Vertisols combinations, 394 bulk density and COLE, 337-339 floral traits and outcrossing, 428 consistence and Atterberg constants, 339-342 in natural outcrossing mechanism, 429gas diffusion, 343 430 hydraulic conductivity, 343-345 panicle exsertion and outcrossing, 426-427 overview, 336-337 Pollution ranges of, 338t heavy metal, 273 shrinkage curve and moisture retention, 342by soil nutrients, 241 343 Pol ysaccharides soil structure and porosity, 345-352 in forage sorghums Phytosiderophores lignin limiting digestibility of, 167-169, iron solubilization and, 276 I76 secretion in response to iron deficiency, stem sweetness and, 170-171 278 in Vertisol clay-organic complexes, 336 Pineland archaeological site case study Pompeii archaeological site, 37-42, 43f Citrus Grove, description, 53-56 Pools in nitrogen mineralization methods, 49 dilution techniques in measuring gross minerobjectives, 48-49 alization, 217 results, 49-53 and immobilization, 192-194 Smith Mound, description, 56-57 nitrification, 194-195 summary, 57 organic matter additions, effects of, 197-201 Pi test, 137-139 soil biomass, 195- 197 Plaggen epipedon, 6; see also Anthrosols soil organic matter Plants, see also specific plants description, 189- 191 nitrogen mineralization and in mineralization kinetics, 201 nitrogen components, proportions of, 201Porosity, Vertisol 202 hydraulic conductivity and, 343-344 provision of substrate by, 206 pedogenic processes and, 307-310
482
INDEX
Porosity, Vertisol (conrinued) soil structure and, 345-352 Potash fertilizers, see Fertilizers, potassium; Potassium Potassium ammonium in availability of, 260-261 buffer power of availability to perennial crops and, 254259 commercial significance of determination of, 259-260 EUF in constructing power curves, 267270 Quantity/Intensity relationship and, 263-266 in Vertisols cation behavior, 330-331 cation exchange sites, 330 effect on liquid limits, 341 hydraulic condctivity effects on, 344 Pot experiments in mineralization measurements, 210 phosphate rocks and, 119-120 Pozzelle quany archaeological site, 37-42 Precambrian Era, Vertisol parent materials formed in, 300 Process controls, mineralization cultivation, effects of, 206-207 environmental controls, 203-206 microsites, diffusional constraints, and soil architecture, 207-208 overview, 201 resource quality, 201 -203 Profile descriptions, soil alluvial system case study methodology, 25 El Mirador Bajo case study, 30-31 Pineland case study, 50 Protein in forage sorghums forage quality and level of, 163-164 genetic engineering for, 177 genetic variability for, 162 phosphorus content and, 176 tannin versus digestibility of, 164 PRs, see Phosphate rocks Pueraria javonica, acid secreted by, 103
Q Quantity/Intensity relationship K buffer power and, 254, 263-266 P buffer power and, 250-254 Quartz in Vertisols, 327
Quasicrystals, Vertisol clay-size crystallite structures, 307 in measuring clay-size distribution, 301 Quaternary Era, Vertisol parent materials formed in, 300
R Radioisotopes in measuring PR fertilizer effectiveness, 121 Radiometric '4C dating of archaeological site materials anthrosols, 7 paleosols, 23 in soil chemical analyses, 19 Rape plants EUF study of P kinetics, 268-269 phosphate uptake and acid secreted by, 104105 Reactivity of phosphate rocks definition, 79-80 inorganic P fractionation and, 89-90 measurement of, 80-84 mineralogy and, 84-86 Recurrent restricted phenotype selection, 173 Redzinas, see also Vertisols 1938 Soil Classification System definition, 317 as early name for Vertisols, 316 Regeneration of Vertisols, structural in high-input cultures, 361-362 in north Cameroon, 357, 359f Relative response in quanitfying PR agronomic effectiveness, 121-123 Relict soils as paleosols, 22 Research models, nitrogen mineralization, 218219 Residual shrinkage phase of Vertisols, 342; see also Shrink-swell phenomena, Vertisol Residues, plant and crop in laboratory extraction of N, 213 nitrogen concentration in, 20 1-202 Resistivity in soil science and archaeology, 17 Resources for nitrogen mineralization, see Pools in nitrogen mineralization; Process controls, mineralization Restorer lines, rice basmati rice under development from, 425 in CMS system deployments, 395-398 in parental line development, 405
INDEX Restriction fragment length polymorphism maps, 408 Rhizosphere, see also Root systems potassium concentration levels, 255 PR dissolution and pH values in, 103-105 zinc availability and, 271-272 Rhodes grass, nitrate depletion in Vertisols and, 332 Rice, see also Hybrid rice cultivation of in developed countries, 357 on West African Vertisols, 355 as world premier food crop, 378 Roman archaeological sites, heavy metal distributions in, 36-37 Root systems, see also Rhizosphere buffer power effect on nutrient availability basic concepts, 242-244 measuring, 244-246 differential uptake of heavy metals by, 277278 as nitrogen mineralization source, 198 phosphate rock and availability of P from PR to versus extent of, 109 effect on dissolution rates, 114-1 15 structure versus K+lsup feeding capacity, 258, 259 Ruminants forage sorghum digestibility leaf and stem features, 163 lignin and, 167-169 forage tannins in, beneficial effects of, 164 molybdenosis in, 275 Ryegrass in chemical extraction studies of N mineralization, 212 EUF study of P kinetics, 268-269 nitrogen mineralization after ploughing, 198199 phosphate rock dissolution and P uptake versus, 107 rhizosphere soil pH, 103-104 potassium ion uptake, 248, 257-258
S Salt deposits at El Mirador Bajo archaeological site, 31, 32, 351 in Vertisol classification, 321
48 3
Samples, soil in nitrogen mineralization studies, incubation methods, 210-211 Snyder’s Mound auger, 44, 45 Sand at A.B.’s Midden site, 62-65 in Seminole Rest site Synder’s Mound sediments, 46 Sand mounds at Pineland archaeological site, 48 San Luis Archaeological Site soil morphological features, 15 total versus extractable phosphorus distribution, 19 Savannah River Valley archaeological sites alluvial soils, 27r, 28 paleosols, 24 Scale, soil map, 13 Scleroglucan in Vertisols, 336 Seasons nitrogen mineralization and impact on field measurements, 215 rate of versus, 204, 205 Vertisols and formation of, 298 rainy and dry, in West Africa, 353 Sedimentary deposits chronology of, at Savannah River Valley site, 271 in interpretation of paleoclirnates, 9 marine, at Seminole Rest site, 46 at Pineland archaeological site, 54 Vertisols inherited versus neoformed clays and, 298 as parent material, 295 Seedlings forage sorghum, bloom versus bloomless, 170 rice hybrid low temperature tolerance, 425 mutants of twin seedlings per seed, 404 transplantation, 419 Seeds, rice apomixis and, 403-404 commercial aspects of, 442-444 discoloration by fungi, 447 production of, see also Outcrossing in hybrid rice seed production for experimental hybrids, 406-407 guidelines, 430-43 1 overview, 426 performance of, 434-438 practices, 431-434
484
INDEX
Seeds, rice (continued) recommended amount of hybrid versus inbred, 419 Selection techniques, forage sorghum breeding program, 173 Self-mixing models of Vertisol genesis, 310 Self-mulching disadvantages of proposed definitions for, 345 Vertisol surface horizon structure, 302-304 Seminole Rest archaeological site case study description, 46-47 methods, 44-45 overview, 42-44 results, 45 summary and conclusions, 47-48 Settlement patterns in human effects on soils and landscapes, 8-9 soil phosphorus quantification and, 9-10 Sewage sludges heavy metal contamination and, 277 nitrogen content of, 200-201 Shear failure in Vertisol genesis model, Coulumb-Mohr theory of, 310-316 Shell midden, see A.B.’s Midden archaeological site case study Shelves in Vertisol gilgai structures, 305 Shoulder, landscape, 20 Shrink-swell phenomena, Vertisol bulk density variation with, 337-338 clay-size particles and, 301 macroscopic and microscopic features, 307310 phyllosicates, effect of, 322-323 shrinkage curve and moisture retention, 342343 in smectites of higher specific surface area, 326 theoretical phase relationship and, 336-337 in Vertisol genesis, 307-315 Siderophores iron solubilization and, 276 secretion in response to iron deficiency, 278 Silage grains, forage sorghum, 176 Silicates, see also Phyllosilicates high pH conditions and dissolution of Vertisols containing, 329 in Vertisols, 326-327 Silt at A.B.’s Midden archaeological site, 62, 65 at Seminole Rest site, 45, 46
Slickensides countering differential loading model, 3 10 gilgai as indicators of presence of, 320 microscopic features, 307, 308, as Vertisol morphogenic markers, 304 Slurry, cattle, see Fertilizers; Manure SMB, see Soil biomass, microbial Smectite at Seminole Rest site, 45, 46, 47 in Vertisols clay-organic complex aggregates, 336 in formation of, 299 as major component of, 323-326 potassium cation and anion behavior, 33 1 Smith Mound, see Pineland archaeological site case study Snyder’s Mound at Seminole Rest site, 4248 Sodic soils, exchange complex values determining, 330 Sodium carbonates in Vertisols, 328 ASodium hydroxide, measurement of phosphate rock dissolved from, 90 Sodium in Vertisols cation exchange sites, 330 effect on liquid limits, 341 Soil biomass microbial, see also Micro-organisms in nitrogen mineralization in nitrogen immobilization, 193 N turnover rate, 196 in soil organic matter pool, 190, 191 in Vertisols, 335 nitrogen content of plant below-ground, 209 in soil N cycle, 195-197 Soil maps, archaelogical uses of description, 12-13, 13-14 El Mirador Bajo case study, 31, 32f Soil nitrogen cycle, 188-189; see also Nitrogen mineralization Soils, see also specific types acid, see also Acid Vertisols; pH values, soil chemical extraction tests on PR-treated, 136 forage sorghum germplasm for, 175 PR dissolution measurement in, 86-90 PR dissolution rate versus moisture levels in, 102 residual effect of PRs in, 125-126
INDEX aerobicity of factors influencing, 203 in mineralization measurements, 210 anthropogenic, see Anthrosols archaeological versus pedological, 15 calcareous, PR dissolution measurements in, 91 detection of human effects on natural, 8-9 distinguishing anthropogenic from natural, 48-57 exchange reactions in, 260-261 as medium of plant growth, 237-239 organic matter in, see Organic matter, soil properties of in archaeological investigations, 13- 14 buried paleosols versus present day soils, 22-23 PR dissolution rate versus, 94-102 shrinkage of, see Coefficient of linear extensibility in Vertisols; Shrink-swell phenomena, Vertisol sodic, 330 solute movement and conductivity of, 274 testing, see Tests, soil Soil science archaeological applications of, 8- 10 contribution to archaeopedology, 12 geophysical tools, applicable, 17 interdisciplinary approach with archaeology summary, 67-68 value of, 2-4 Soil series, characteristics defining and differentiating, 13 Soil solution, 242 Soil surveys as new application of soil science, 2 soil maps as component of, 12-13 Solid phase, soil, 243 Solubility of phenolic acids in forage sorghums, 165 of phosphate rocks carbonate apetites, increasing for, 92 composting with organic manures to increase, 143 Diffusion Rate Function and, 83-84 factors affecting, 81 Buorapatite, reactivity of, 84-86 Solum thickness of volcanic soils near Vesuvius, 39-42 Solution phase, soil, 243; see also Soil solution
48 5
SOM, see Organic matter, soil Sorghum bicolor, 332 Sorghums, see also Forage sorghums cultivation on Vertisols in West Africa, 353355 nitrate depletion in Vertisols and, 332 South Carolina, see Savannah River Valley archaeological sites Spodic Quartzipsamments, 50, 56-57 Spodosols at Pineland site burial mound, 5651 Stem borer infestions in rice, 422 Stems, forage sorghum digestibility rates in ruminants and, 163 sweetness of, 170-171 Sterility in rice hybrids, 401-403; see also Cytoplasmic-genetic male sterility; Environment-sensitive genic male sterility in rice; Fertility, rice; Male sterile lines, rice Stratigraphic markers paleosols as, use of, 22 soil morphology as, 16-17 Stratigraphy of A.B.’s Midden site, 60-65 archaeopedology and reverse, 6 Stress environments, hybrid rices and adaptability to, 425-426 heterosis and, 444-445 heterotic rice yields, 416-419 Structural shrinkage phase of Vertisols, 342; see also Shrink-swell phenomena, Vertisol Structure, Vertisol, see also Morphology, soil description, 302-304 porosity and, 345-352 Subsoils, see Horizons, soil Substitution value in determining most profitable fertilizer, 130 in quanitfying PR agronomic effectiveness, 121-123 Sudangrass sterility-inducing pollinator lines, 172 as type of forage sorghum, 161 Sulfates molybdenum availability and, 275 in Vertisols anion behavior, 333 mineralogical properties, 328 Sulfides in Vertisols, 328 Sulfur admixtures with phosphate rocks, 143I45
486
INDEX
Sulfuric acid in phosphate rock acidulation, 145 phosphate rock-sulfur assemblages and, 144, 145 Summit landscape position in archaeological reconstruction of landscapes, 20 Superphosphate in economic evaluation of PR use, 132 soluble fertilizers and, 126 Surfaces in archaeological reconstruction of landscapes, 20 Sustainability agricultural, of Vertisols, 355 nitrogen fixation and attempts to increase, 221 of soil, plant nutrient management in, 239242 SV, see Substitution value
T Tannins, forage as antiquality factor, 164-166 forage sorghum germplasm low in, 175 reduction of protein digestibility by, 164 Taxonomy, Vertisol, 316-322 Technology, transfer of hybrid rice, 442-444 Temperature forage sorghums, effect on growth of, 161 male sterility in rice versus, 398-401 mineralization and soil microbial activity and, 203 range controlling, 204 phosphate rock and soil availability to plants versus, 108- 109 diffusion versus, 112 dissolution versus, 102- 103 Vertisols and as basis for cryert suborder, 320-321 in formation and distribution of, 298 Temperature-sensitive genic male sterility lines bred by IRRI, 414 description, 398-401 in parental line development, 405 Tennesee, see Hiwasse River archaeological site case study Terra preta soils in Amazon basin, 8 Tertiary Era, Vertisol parent materials formed in, 300 Tests, soil, see also Electro-ultrafiltration fertility indices derived from, 241-242
K Buffer power measurement versus, 261263 nutrient availability, 240-241, 241-242 P buffer power measurement versus, 248250 phosphate rocks and, 133-142 requirements for accuracy of, 278-280 Texas Vertisols bulk density of surface horizons, 340f extent of, 294 hydraulic conductivity studies, 344-345 management of, 357-363 Texture, soil nitrogen mineralization and, 205 Vertisol, description, 301 TGMS lines, see Temperature-sensitive genic male sterility lines Thermodynamics in defining chemical environment of roots, 242 Thiobacillus spp, phosphate rock-sulfur assemblages and, 142, 143-144 Thunderbird archaeological site case study, 2528 Tikal, Mayan city of, Vertisols at, 30 Tillage of Vertisols conservation systems, 361 pedality and porosity, effects on, 346-352 Timescale, geologic, see Geologic eras; speciJic eras Titratable acidity, phosphate rock dissolution rate versus, 94-98 Toeslope landscape position in archaeological reconstruction of landscapes, 20-2 I Topography, Vertisols and color, effect on, 302 description, 299 Torrerts, 319; see also Vertisols Toxicity cyanide in forage sorghums, 166- 167 heavy metals in crop production, 277 molybdenosis in ruminants, 275 nitrate in forage sorghums, 177 Traits, genetic for commercial sorghum silage breeding improvement, 172-173 of rice floral and outcrossing of, 427-429 for heterosis, 385-387 Transplantation of hybrid rice seedlings, 419
487 of sorghum plants on West African Vertisols, 355 Trend Surface Analysis, 10 Trifolium pratense, 258 Trifolium repens L, 199 Trifolium subterraneum, 105- 106 Triticum aestivum EUF study of P kinetics, 269 source of K + , 255
U Uderts, 3 19; see also Vertisols Ultisols, P adsorption from PRs versus soil temperature, 108 Underwater archaeology, 21 United States of America technology transfer of hybrid rice production to farms, 443 Vertisol distribution in, 292-294 Urine in nitrogen mineralization, see also Manure N returned by grazing, 198 versus stored manures, 200 Ustic Vertisols, 318
v Vegetation buried paleosols in characterization of, 22 in pedogenesis and distribution of Vertisols, 299-300 Vertisols biological properties clay-organic complexes, 335-336 organic constituents, 333-335 chemical properties cation and anion behavior, 330-333 cation exchange capacity and exchangeable cations, 329-330 pH, 329 classfication of, 3 16-322 conclusions and summary, 363-364 distribution in United States, 292, 293f world, 292 formation of classical soil forming factors involved, 294-295 climate, 298
parent material, 295-298 time, 300-301 topography, 299 vegetation, 299-300 management of with high-input cultural systems, 357-363 with low-input cultural systems, 353-357 at Mayan El Mirador Bajo site, 30, 31-32 mineralogical properties other minerals, 326-328 phyllosilicates, 322-326 morphological properties color, 302 structure and special physical features, 302-307 texture, 301 overview, 290-292 pedogenic processes, 307-3 16 physical properties bulk density and COLE, 337-339 consistence and Atterberg constants, 339342 gas diffusion, 343 hydraulic conductivity, 343-345 shrinkage curve and moisture retention, 342-343 soil structure and porosity, 345-352 Vesuvius archaeological sites, Mount, 37-42, 43f Vietnam NPK fertilizer rate responses of hybrid versus inbred rices, 421 yield advantage of heterotic rices evaluated in, 419 Villa Oplontis archaeological site, 37-42, 43f Volcanic soils PR dissolution and moisture levels, 102 pH of, 95-96 Volcanoes, archaeology and, 37-42 Volusia County, Florida, see Seminole Rest archaeological site case study
W Wales, archaeological soil phosphorus distribution patterns in, 10 Water, see also Irrigation; Moisture, soil in nitrogen mineralization field measurements, 215
488
INDEX
Water (continued) rice hybrid use efficiency of, 426 in Vertisols conservation of, in West Africa, 357 gilgai structures, 305-306 hydraulic conductivity and, 343-345 plastic limit and, 342 Watkinson model of phosphate rock dissolution compared to other models, I 11-1 13 description, 115-1 16 testing, 117-1 18 Wax of forage sorghum plants, epicuticular, 169-170 West Africa, case study of Vertisols in, 353357 Wheat EUF study of P kinetics, 269 N response after ploughing out swards, 22 1
PR dissolution versus application method, 105-106 Wide compatibility genes, rice, 401-403
x Xererts, 319; see also Vertisols Xylan, linear, 168
Y Yields, crop, see Crop Yields; Heterosis in rice
2 Zinc buffer power and availability of, 270-272, 273-275 at Hadrian’s Villa archaeological site, 37
This Page Intentionally Left Blank