Soil Organic Carbon Fractions in a Long-Term Experiment and the Potential for their Use as a Diagnostic Assay in Highland Farming Systems of Central Kenya
J. J. Kapkiyai, N. K. Karanja^1, P. Woomer^1 and J. N. Qureshi^2
Uasin Gishu District Agricultural Office, P.O. Box 95, Eldoret, Kenya
(Received 19 September, 1996; accepted 13 November, 1997)
Soil organic matter fractions provide insight into soil quality and the maintenance of crop productivity in smallhold cropping systems. Soil organic carbon (SOC) pools were measured and compared to crop performance during the eighteenth year of a long-term field experiment established on a Humic Nitisol at the Kenya Agricultural Research Institute. Maize stover retention (+S), cattle manure (+M, 10 t ha^-1 yr^-1) and fertilizer inputs (+F, 120 kg N and 23 kg P ha^-1 yr^-1) were compared in a 2 x 3 factorial arrangement in a manner that is broadly reflective of farmer resource management options. Maize (Zea mays) and beans (Phaseolus vulgaris) were cultivated in rotation during the long (March-June) and short (October-December) rains of each year, respectively. Maize yields ranged between 1.21 (+S) and 5.21 t ha^-1 (+SMF). Beans produced between 0.14 (-S) and 0.77 t ha^-1 (+SMF). Total crop yields were significantly affected by manure and fertilizer addition (P < 0.001) but not stover retention (P = 0.13). Soil microbial biomass carbon (MBC, r = 0.63), particulate organic carbon (POC, r=0.82) and its density fractions varied significantly with combined maize and bean yields in plots not receiving fertilizers. Only MBC was significantly correlated with crop yields in the inorganically fertilised treatments (r = 0.50). Livestock manure contributed to POC to a greater extent than did retention of maize stover or the addition of fertilizers. Of the different Ludox density fractions of POC, the lightest ( < 1.13 g cm^-3) was best correlated with crop yields (r = 0.73) and the heaviest fraction ( > 1.37 g cm^-3) least correlated with yield (r = 0.37). When 190 Central Kenyan smallhold 'households' were surveyed, 79% were classified as maize-bean farmers and 79% of these relied on combinations of stover, manure and fertilizers to maintain soil fertility. Thus, the long-term experiment at Kabete is broadly applicable to the surrounding farming community. Soil organic matter fractions were more favourably influenced by addition of livestock manure than the retention of maize stover, suggesting that the widespread farmer practice of harvesting maize stover as animal feed and returning animal wastes may effectively contribute to soil fertility management.
Key Words: Particulate organic carbon, Ludox fractions, soil microbial biomass, East African Highlands, Kenya. long-term experiment
Les fractions de la matiere organique du sol constituent un element important pour la qualite des sols et pour le maintien de la productivite des cultures dans les sysemes de production de petits fermiers. Les poold de carbone organique du sol etaient determines et correle avec la performance des cultures a la huiteme annee d'un essai de longue duree installee sur un Nitosol restitution des chaumes de mais (+S), l'apport du fumier de fer,e (+M), 10t ha^-1 an^-1)etaient testes dans un essai factoriel 2 x 3 conccu de maniere a ressembler le plus aux options d'amenagement de ressource par l'agriculteur. Le mais (Zea mays) et la haricot (Phaseolus vulgaris) etaient cultives en rotation chaque anne, respectivement pendant la longue (mars-juin) et la petite (octobre-decembre) saisons de pluie. Les rendements de mais ont varie de 1.21 (+S) et 5.21t ha^-1 (+SMF). Le haricot a produit entre 0.14 (+S) et 5.21 t ha^-1 (+SMF). Les rendements totaux des cultures etaient tres significativement affectes par l'application du fumier et des engrais mineraux (P < 0.001); les chaumes de maiz pa contre n'avait pas d'effeet significatif. Le carbone dans la biomasse microbienne du sol (MBC, r=0.63), le carbone organique dans les particules de sol (MBC, r=0.82) et ses fractions a base de densite variaient significativement avec le rendement du mais et du haricot dans les parcelles n'ayant pas recu d'engrais mineraux. Seul, le MBC etaint significativement correle avec le rendement des cultures dans les parcelles fertilisees aux ebgrais mineraux (r=0.50). Le fumier de ferme a miux contibue au POC que les chaumes de mais ou les engrais mineraux. Parmi les differentes fractions Ludox du POC, la fraction la plis legere ( < 1.13g cm^-3) etait la mieux correlee avec le rendement des cultures (r=0.73) tandis que la plus lourde ( > 1.37 g cm^-3) montraint la plus faible correlation avec le rendement (r-0.37). Lors de l'enquete de 190 petits exploitants du Kenya Centre, 79% de fermiers ont ete identifie comme planteur de Mais-haricot, 79% de ces fermiers utilisaient a la fois les chaumes de mais, le fumier et l'emgrais, mineral pour ,aintenir la fertilite des terres. Ainsi, l'essai de longue duree de Kabete est largement applicable au systeme de production des agriculteurs d'aux environs. Les fractions de la matiere organique du sol etaient plus favorablement influencees par l'application du fumier que la restitution des chaumes de mais. Ceci monte que la pratique repandue des agriculteurs de recolter les chaumes de mais pour les utiliser comme fourrage et de collecter ensuite le fumier pour fertiliser les champs peut effectivement contribuer a la gestion de la fertilite des sols.
Mots Cles: Carbone organique en particules de sol, fractions Ludox, biomasse microbienne du sol, hautes terres d'Afrique de l'Est, Kenya, essai de longue duree.
Soil organic matter (SOM) content is a critical component of soil productivity and its maintenance is a sound approach to maintaining productivity of continuously cropped soil (Follet et al., 1987). Changes in SOM results from imbalances between organic inputs and losses and declining SOM is frequently observed when lands are converted from natural vegetation to agriculture (Woomer et al., 1994). The maintenance of SOM is particularly difficult within smallholder cropping systems of the East African Highlands where crop residues are regarded as an important component of yield as livestock feed and farm size usually preludes opportunities for fallow. One of the most important attributes of SOM is nutrient cycling through the decomposition of active fractions (Duxbury et al., 1989).
One approach to better understanding of organic matter function is the identification of biologically meaningful-pools based on either particle size (Cambardella and Elliott, 1992) or density (Meijboom et al., 1995). This approach is fundamentally different from chemical fractionation procedures that characterise less active and more recalcitrant humic substances (Stevenson and Elliott, 1989) in that it quantifies the organic matter fractions that are most closely associated with nutrient turnover (Barrios et al., 1996). Because different pools of organic matter have separate functional roles within the soil (Woomer et al., 1994), one avenue toward improved land husbandry is to differentially manage SOM pools through a combination of inputs which limits SOM loss while at the same time providing plant nutrients through decomposition and mineralisation processes (Brown et al., 1994).
An field experiment which examines the effects of various soil fertility management strategies in a maize-bean crop rotation was established at the National Agricultural Research Laboratory of the Kenya Agricultural Research Institute (KARI) at Kabete, Kenya in 1976. This experiment seeks to improve soil management in the Kenya Highlands by exploring management options of farmer-available organic inputs and chemical fertilizer resources on the two principle food crops of the area, and is situated on the predominant high potential agricultural soil of the Central Kenya Highlands. Swift et al. (1994a) reported that soil organic carbon contents declined during the first 14 years of the experiment regardless of inputs regime and this trend has continued (Kapkiyai, 1996). After eighteen years of continuous cropping, the treatment receiving no external inputs expresses compound nutrient deficiency symptoms and resembles the fields of the least productive, nearby smallholdings. The purpose of this study was to examine the soil organic matter fractions present in the various soil fertility management strategies and to relate those soil properties to differences in crops performance.
Materials and Methods
Site and experimental description. Kabete is situated 36 degrees 41'E and 1 degrees 15'S and 7 km N.W. of Nairobi. The soil is a well-drained, very deep, dark reddish, brown to dark red, friable clay classified as a Humic Nitisol (UNESCO, 1974) but locally referred to as the Kikuyu Red Clay Loam (Sombroek et al., 1982). The precipitation is bimodal, allowing for two cropping seasons per year. The first and longer rain of the year fall from March to June and the second rainy season lasts from October to December. The site is sub-humid and isohyperthemic with an average annual precipitation of 970 mm. Mean annual temperature is 23 C with the coolest temperatures occurring in July (16 C) and the warmest in January (28 C).
The experiment examines the maintenance of soil fertility through the use of fertilizers (120 kg nitrogen as calcium ammonium nitrate and 23 kg phosphorus as triple super phosphate ha yr^-1) and farmyard manure (10 t ha^-1 yr^-1) with and without the retention of maize stover. Maize and beans are grown in rotation during the first and second rains of each year, respectively. The experimental design is a randomised complete block with four replicates and eight treatments arranged as a 2 x 3 complete factorial (+/- fertilizer, manure and maize stover). Ten additional treatments occur within the long-term experiment but were not sampled in this investigation. Fertilizer and manure were applied at planting of maize while maize stover, when applied, follows maize harvest prior to cropping of beans. The maize variety was based upon KARI recommendations for the area and is currently hybrid 512 while the bean variety was Rose Coco. Crops were grown to grain maturity, harvested, air dried, shelled and weighed.
Soil sampling. Soil samples were collected to a depth of 15 cm with a soil auger in April 1994, 18 years after the experiment was initiated and 11 months after the most recent application of fertilizer and manure. Nine soil cores were collected from each plot, bulked, mixed in polythene bags and transported to the laboratory. The soil sample was stored at 4 C prior to analyses and passed through a 0.25mm sieve prior to carbon analysis. Soil bulk density was assumed to be 1.1 g cm^-3 (E. Muchena, personal communication).
Soil organic matter fractionation. Soil organic matter was fractionated by wet sieving through mesh sizes 2 mm, 250 um and 53 um (Okalebo et al., 1993). A fresh soil sample of known moisture content (100 g d.w.) was dispersed in 1% sodium hexametaphosphate, shaken overnight and then passed through a nest of sieves on a sieve shaker. Particulate organic matter (POM) was collected from the samples collected on the 53 and 250 um sieves (Cambardella and Elliott, 1992). Density separates were recovered by the "Ludox" procedure modified from Meijboom et al. (1995). Ludox is a colloidal silica suspension that may be adjusted to different densities with water. Fresh soil (250 g dry weight) was dispersed and fractionated as for POM above. The > 53 um fraction was then placed in ludox solution of density 1.13 g cm^-3. The material which was collected from the surface with a 53 um net made was considered the ludox light fraction (Lul). The material which settled at the bottom of the container was blot dried and placed in ludox solution of density 1.37 g cm^-3. The material which floated is referred to as the ludox medium fraction (Lum) and that which settled is the ludox heavy fraction (Luh). All fractions were thoroughly washed and oven dried at 50 C for 48 hours and then weighed. Colorimetric determination of total C from whole soil and the size and density fractions was conducted as described by Anderson and Ingram (1993) and Okalebo et al. (1993), respectively.
Soil microbial biomass. Fresh soil samples of known weight and moisture content were fumigated with chloroform and then extracted with 0.5 M K2SO4 as described by Anderson and Ingram (1993). Soil microbial biomass carbon was determined as for total carbon above.
Statistical analysis. Results were compiled on a commercially-available spreadsheet programme with the 32 (8 treatments x 4 replicates) cases entered as rows and measurements as columns. These were then imported into SYSTAT (Wilkinson, 1990) for statistical analyses including linear regression, Pearson pairwise correlation analysis and analysis of variance (ANOVA). The ANOVA was conducted as a randomised complete block design with three factors where, effect = constant + block + fertilizer addition (F) + manure addition (M) + stover retention (S) + (FxM) + (MxS) + (FxS) + (FxMxS). This statistical approach allows for the individual management practices to be partitioned within the ANOVA table (Peterson, 1994). The Least Significant Difference approach of Little and Hills (1978) allowed for direct mean comparisons.
Maize grain and dry bean yields resulting from different nutrient management strategies are presented in Table 1. Beans are grown during the second (short) rains and produce much less than maize grown earlier in the year. Withholding external inputs resulted in significantly lower yields in both maize and beans regardless of stover management. Stover retention in conjunction with manuring reduced yields of maize but not beans. Manuring and fertilisation as combined managements produced the greatest yields, even with beans which receive only residual inputs from the proceeding maize crop.
Significant differences between land management strategies were observed in total soil organic carbon, microbial biomass carbon and particulate organic carbon (Table 2). Total soil organic carbon contents ranged from 23.57 to 28.69 t ha^-1 depending on land management. Soil microbial biomass carbon increased more than two-fold, from 90 kg ha^-1 in the complete control and fertilised, stover removed treatment to 200 kg ha^-1 in the treatment receiving fertilizer, manure and maize stover. Particulate organic carbon contents increased from 1.15 to 2.66 t ha^-1 with those treatments receiving manure having the greatest POC content.
The carbon pool sizes of the three ludox separates are presented in Figure 1. Significant differences were observed in the individual carbon pools between land managements and between the pools within managements. The heavy fraction ( > 1.37 g cm^-3) is the largest of the three density fractions and the one least responsive to management. The lightest fraction ( < 1.13 g cm^-3) varied from 220 to 690 kg ha^-1 and at its greatest concentration was less than the lowest levels of the heavy fraction. The medium fraction (1.13-1.37 g cm^-3) was the smallest of the three density separates in the soils with the lowest total carbon contents but increased in size in response to inputs of manure. The sum of the ludox fractions presented in Figure 1 corresponds closely to the particulate organic carbon fraction in Table 2 (r = 0.81, P < 0.001, data not presented) because of the similar manner in which these two fraction recoveries involve collection upon a 53 um sieve.
Total soil organic matter and all of its fractions were significantly correlated with the combined yield of the maize-bean rotation (Table 3). Total soil organic carbon and the ludox heavy fraction demonstrated the weakest relationship (P = 0.05), followed by microbial biomass (P = 0.01) and then particulate organic carbon and light and medium density fractions (P = 0.001). The correlation of yield and soil organic carbon fractions was much greater in the plots not receiving fertilizer. An example of the strengthed relationship between particulate organic matter and total yield is presented in Figure 1. Note that when all cases are considered, a positive intercept is obtained along the yield (y) axis and the regression accounts for 34% of the observed variation (R2). A similar regression of only plots not receiving fertilizers reveals a negative intercept, suggesting greater dependence, and a strengthened linear trend (R2 = 0.67). No regression line or equation is presented for the fertilised cases because the relationship between yield and particulate carbon was not significant but also note that the yield of all fertilised treatments are greater than the linear trend describing those that were not chemically fertilised.
Of the 190 households interviewed, 168 (79%) were classified as maize-bean farmers based on their rankings of crop importance. Among these maize-bean farmers, 79% reported the combined use of maize residues, animal manure and fertilizers. Additional information on these farming practices is presented in Table 4 including the high frequency of stover fed to livestock (95% of farms) and the confinement of cattle (91%).
Soil organic matter fractions were more strongly correlated with crop yields when fertilizers were not applied. Particulate and the least dense Ludox density fraction of particulate soil organic matter (Cambardella and Elliott, 1992; Meijboom et al., 1995) were the SOM fractions most closely associated with crop productivity after 18 years of continuous cultivation. Every ton of POM resulting from organic inputs to soils increased maize yield by 873 kg ha^-1 (P = 0.03) and dry beans by 236 kg ha^-1 (P < 0.001). The addition of fertilizers to maize in the maize-bean rotation obscures these relationships (Fig. 2) as only microbial biomass was significantly correlated with combined crop yields (P < 0.05) in the fertilised plots (Table 2).
Addition of 10 t ha^-1 yr^-1 cattle manure showed the greatest effects on soil particulate organic matter content (p > 0.001) but the effects of stover retention, chemical fertilisation and all interactions were not significant. Recovery of the Ludox light fraction was more sensitive to land management with significant effects of all three inputs (S, M, F) observed with ANOVA (P = 0.001). The addition of fertilizer as a sole management practice did not contribute to soil organic matter content, unlike the reports of others who link inorganic fertilisation to SOM formation through the increase in below-ground (root) inputs (Gregorich et al., 1997).
Retention of maize stover significantly increased total soil organic matter and soil microbial biomass but not the particulate fractions closest associated with crop productivity (Fig. 1, Table 3). Maize stover appears to be a readily available substrate to soil biota but relatively low in mineralisable nutrients. The retention of maize stover as a sole management practice reduced maize yields by 453 kg ha^-1 and offered only small benefit to the bean rotation to which stover was applied (+29 kg ha^-1) after 18 years of continuous cultivation. This probably indicates stover's contribution to the microbial immobilisation of plant nutrients as soil microbial biomass increases with stover retention (Table 2).
Farmer practice in Central Kenya support the findings from the long-term experiment at Kabete. The least productive land managements, such as cultivation with no external inputs or exclusive reliance upon maize stover were not reported by any of the 168 farms classified as maize-bean systems. When asked to identify the major constraints to their cropping systems, fewer farmers specified soil conditions (33%) than those identifying farm input costs (79%), and damage by wildlife (77%) and pests and diseases (65%). Yet these farmers have not overcome the limitations of soil fertility. Average maize yields are less than 1700 kg ha^-1 (per crop) and the widespread difficulty in purchasing farm inputs is reflected in low rates of fertilizer application. Most farmers raise cattle under confinement, which facilitates the recovery of manure, but herd size is small (6.1 heads household^-1) to meet the manure requirements of the average farm (3.9 ha). Strobel (1987) reported that cattle produce up to 1500 kg of manure annually, so the annual manure recovery would be insufficient to manure 1 ha at the rate of the Kabete experiment. These farmers' appear to have identified the management practices that contribute to soil fertility but in many cases their resource base are too restricted to optimally manage their soils.
Soil organic matter fractionation is rapid (Okalebo et al., 1993), provides quantitative results (Meijboom et al., 1995) sensitive to land management (Barrios et al., 1996) and measurements that are well correlated with crop productivity at the Kabete site, but much less so when fertilizers are applied. Given the widespread testing and use of fertilizers, SOM fractionation is unlikely to become a relevent diagnostic tool within commercial agricultural systems practicing recommended fertilizer application in the Central Kenyan Highlands (see FURP, 1994). However, average farmer yields reported in our survey resemble those of the least productive treatments at the Kabete long-term experiment (Table 1) suggesting that, despite the diverse soil management strategies (Table 4) and farmer-recognised constraints, nutrient inputs continue to limit crop productivity. Furthermore, the frequency of fertilizer use in Central Kenya is much greater than that in other parts of the country (Woomer et al., 1997) or in neighbouring Uganda (Turkahirwa, 1992; Bekunda and Woomer, 1996).
Diagnostic assays relying upon key soil organic matter fractions offer potential as assay tools to measure soil fertility depletion and will be most appropriate to smallhold agriculture where there is little opportunity to apply recommended levels of fertilizer, or where future improvements to those systems are anticipated to occur through organic resource management (see Swift et al., 1994b). The results of this study suggest that crop failure occurs when particulate organic C levels drop to 0.33 t ha^-1 and that little benefit from combined N and P fertilizers are realised at 4.0 t ha^-1 POC (Fig. 2). One irony is that as more chemical fertilizers are applied to overcome nutrient limitations in the future, an assay based on key SOM fractions will lose its ability to predict crop performance (Fig. 2). Yet in the interrum, SOM fractionation will improve mechanistic understanding of organic matter turnover in soils, which organic inputs contribute to various SOM pools and which of these pools are closest associated with crop performance.
The authors are grateful to the staff of the ICRAF Machakos Research Station especially Dr. Smithson,V. Mbugua, R. Chacha and P. Mutisya and to the Rockefeller Foundation Forum for this research through the Forum on Agricultural Resource Husbandry.
Anderson, J.M. and Ingram, J.S.I. (Eds.). 1993. Tropical Soil Biology and Fertility: A Handbook of Methods. CAB International, Wallingford, UK. 221 pp.
Barrios, E., Buresh, R.J. and Sprent, J.I. 1996. Organic matter in soil particle size and density fractions from maize and legume cropping system. Soil Biology and Biochemistry 28:185-193.
Bekunda, M.A. and Woomer, P.L. 1996. Organic resource management in banana-based cropping systems of the Lake Victoria Basin, Uganda. Agriculture, Ecosystem and Envronment 59:171-180.
Brown, S., Anderson, J., Woomer, P.L., Swift, M.J and Barrios, E. 1994. Soil biological processes in tropical ecosystems. In: The Biological Management of Tropical Soil Fertility. Woomer, P. and Swift, M.J. (Eds.), pp. 15-46. John Wiley & Sons, Chichester, UK.
Cambardella, C.A. and Elliot, E.T. 1992. Particulate soil organic matter across a grassland cultivation sequence. Soil Science Society of American Journal 56:777-783.
Duxbury, J.M., Smith, M.S. and Doran, J.W. 1989. Soil organic matter as a source and sink of plant nutrients. In: Dynamics of Soil Organic Matter in Tropical Ecosystems. Coleman, D.C. Oades, J.M. and Uehara, G. (Eds.), pp. 33-67 University of Hawaii Press, Honolulu.
Fertilizer Use Recommendation Project (FURP). 1994. Fertilizer Use Recommendations; Volumes 1-22. Kenya Agricultural Research Institute, Nairobi, Kenya.
Follet, R.F., Gupta, S.C. and Hunt, P.G. 1987. Conservation practices: Relation to the management of plant nutrients for crop production systems. Pages 19-51. USA. Soil Science Society of America, Madison. Special Publication No. 19.
Gregorich, E.G., Drury, C.F., Ellert, B.H. and Liang, B.C. 1997. Fertilization effects on physically protected light fraction organic matter. Soil Science Society of American Journal 61: 481-484.
Kapkiyai, J. 1996. Dynamics of soil organic carbon, nitrogen and microbial biomass in a long-term experiment as affected by inorganic and organic fertilization. M. Sc. Thesis, University of Nairobi. 102 pp.
Little, T.M. and Hills, F.J. 1978. Agricultural Experimentation: Design and Analysis. John Wiley & Sons. New York. 350 pp.
Meijboom, F.W., Hassink, J. and Van Noordwijk, M. 1995. Density fractionation of soil macro organic matter using silica suspensions. Soil Biology and Biochemisty 27:1109-1111.
Okalebo, J.R., Gathua, K.W. and Woomer, P.L. 1993. Laboratory Methods of Plant and Soil Analysis: A Working Manual.TSBF Programme. Nairobi, Kenya. 88 pp.
Peterson, R.G. 1994. Agricultural Field Experiments: Design and Analysis. Mercel Dekker Inc. New York. 409 pp.
Sombroek, W.G., Braun, H.M.H. and Van der Pouw, B.J.A. 1982. Soil Map and Agro-climatic Zone Map of Kenya. Kenya Soil Survey, National Agricultural Laboratories, Nairobi.
Stevenson, F.J. and Eliott, E.T. 1989. Methodologies for assessing the quality and quality of soil organic matter. In: Dynamics of Soil Organic matter in Tropical Ecosystems. Coleman, D.C., Oades, J.M. and Uehara, G. (Eds.), pp. 173-199. University of Hawaii Press, Honolulu.
Strobel, H. (Ed.). 1987. Fertilizer Use Recommendation Project Final Report: Annex II.3: Maintaining Soil Fertility. Ministry of Agriculture. National Agricultural Research Laboratory, Kenya Agricultural Research Institute, Nairobi. 20 pp.
Swift, M.J., Seward, P.D.,Frost, P.G.H., Qureshi, J.N. and Muchena, F.N. 1994a. Long-term experiments in Africa: developing a database for sustainable land use under global change. In: Long-term Experiments in Agricultural and Ecological Sciences. Leigh, R.A. and Johnson, A.E. (Eds.), pp. 229-251. CAB International, Wallingford, UK.
Swift, M.J., Bohren, L., Carter, S.E., Izac, A.M., and Woomer, P.L. 1994b. Biological management of tropical soils: Integrating process research and farm practice. In: The Biological Management of Tropical Soil Fertility. Woomer, P. L. and Swift, M.J. (Eds.), pp. 209-207. John Wiley & Sons, Chichester, UK.
Turkahirwa, E.M. 1992. Uganda: Environmental and Natural Resource Policy and Law: Issues and Options. II. Documentation. Institute of Environment and Natural Resoures. Makerere University, Kampala, Uganda.
United Nations Educational, Scientific and Cultural Organisation (UNESCO). 1974. FAO-Unesco Soil Map of the World: Volume VI. Africa. UNESCO, Paris. 299pp.
Wilkinson, L. 1990. SYSTAT: A System for Statistics. Systat Inc., Evanston, Illinois, USA.
Woomer, P.L., Palm, C.A., Qureshi, J.N. and Kotto-Same, J. 1997. Management of carbon sequestration in soil. Advances in Soil Science. Lal, R., Kimble, J.M., Follett, R.F., and Stewart, B.A. (Eds.), pp. 153-173. CRC Press, Boca Raton, USA.
Woomer, P.L., Martin, A., Albrecht, A., Resck, D.V.S. and Scharpenseel, H.W. 1994. The importance and management of soil organic matter in the tropics. In: The Biological Management of Tropical Soil Fertility. Woomer, P.L. and Swift, M.J. (Eds.), pp. 47-80. John Wiley & Sons, Chichester, UK.
Copyright 1998, African Crop Science Society
The following images related to this document are available:
Line drawing images[cs98003c.gif] [cs98003a.gif] [cs98003f.gif] [cs98003d.gif] [cs98003e.gif] [cs98003b.gif]