African Crop Science Journal African Crop Science Society ISSN: 1021-9730 EISSN: 2072-6589 Vol. 8, Num. 1, 2000, pp. 63-76

L. R. Ndlovu, L. Mlambo¹ and B. H. Dzowela¹

Department of Animal Science, University of Zimbabwe, P. O. Box MP167, Harare, Zimbabwe
¹Southern African Development community (SADC)- International Centre for Agroforestry (ICRAF) Project, Department of Research and Specialist Services, Causeway, Harare, Zimbabwe

Code Number: CS00006

ABSTRACT

Leaves from Leucaena species L. esculenta, L. diversifolia, L. pallida, L. pulverulenta, L. salvadorensis, L. shannonii, L. trichodes and the interspecific hybrid L. Leucocephala x L. diversifolia that are resistant to the psyllid Heteropsylla cubana were harvested at the end of the rainy season (April) and the dry season (November) from two sites. The leaves were either sun or oven ( 50°C) dried and subsequently analysed for concentrations of dry matter (DM), organic matter (OM), crude protein (CP), neutral and acid detergent fibres (NDF and ADF), lignin, extractable proanthocyanidins (PAs), soluble phenolics (SPs) and protein precipitating potential (PPP). In vitro gas production was measured over 72 hours. Season of harvest and method of drying had no significant (P>0.05) effect on the variables measured. Species significantly affected concentrations of CP, lignin, PAs, SPs and PPP. Cluster analysis based on content of fibre or content of polyphenolics or total gas produced identified L. shannonii, L. salvadorensis, L. esculenta and the intraspecific hybrid L. leucocephala x L. diversifolia as potentially good quality forages.

Key Words: Gas production, Heteropsylla cubana, leucaena, nutritive value, phenolic, tannin

Les feuilles des espèces Leucaena L. esculenta, L. diversifolia, L. pallida, L. pulverulenta, L. salvadorensis, L. shannonii, L. trichodes et les hybrides interspécifiques L. Leucocephala x L. diversifolia qui sont résistant aux psyllids Hétéropsylla cubana ont été récoltées à la fin de la saison de pluies (Avril) et à la fin de la saison de sécheresse (Novembre) à partir de deux sites. Les feuilles ont été séchées soit au soleil ou dans un four à (50°C) et par la suite, analysées pour les concentrations de la matière sèche, la matière organique, les protéines brutes, les fibres neutres et détersifs de l’acide, la lignine, les proanthocyanidines extractibles, les phénols solubles et la potentialité de précipitation des protéines. La production de gaz (in vitro) a été mesurée au cours de 72 heures. La saison de récolte et la méthode de sèchement (P>0.05) n’avaient aucun effet impotant sur les variables mesurés. Les espèces ont eu un effet significatif sur les concentrations de la protéine brute, la lignine, les proanthocyanidins, les phenols solubles et la potentialité de précipitation des protéines. L’analyse par groupement basée sur le contenu de fibre ou le contenu de polyphenols ou encore le gaz total produit a identifié L. shannonii, L. salvadorensis, L. esculenta et l’hybride intraspécifique L. leucocephala x L. diversifolia comme potentiels d’être des fourages de bonne qualité.

Mots Clés: Production de gaz, Hétéropsylla cubana, leucaena, valeur nutritive, phenolique, tannin

Leucaena leucocephala is widely used as a source of high quality forage in tropical animal production systems because of its rapid juvenile growth, high protein concentration, high yield of digestible forage, lack of thorns and coppicing ability (Brewbaker and Macklin, 1990). The presence of mimosine and its goitrogenic metabolite dihydro-xypyridone (DHP) has not limited its use in systems where it is used as a supplement (Jones, 1994). However, its value has been challenged by the outbreak of Leucaena psyllid (Heteropsylla cubana), an aphid-like insect that produces large-scale defoliation. Some other Leucaena species and their crosses with L. leucocephala are resistant to psyllids (Wheeler and Brewbaker, 1990). It has been hypothesised that the reason for the resistance to psyllid defoliation is the presence of condensed tannins (Telek, 1992), although Wheeler et al. (1994) did not find a significant correlation between condensed tannins and psyllid resistance.

The Southern African Development Community- International Centre for Agroforestry (SADC-ICRAF) project has introduced several accessions of these psyllid-resistant Leucaena species into Zimbabwe in order to assess their adaptability and potential to supply high quality forage. The experiments described here evaluated the chemical composition of these species, including condensed tannin concentration and in vitro gas production. In vitro gas production provides a quick assessment and initial screening of fodder trees (Nsahlai et al., 1994) and has been shown to be positively related to intake (Blummel and Orskov, 1993) and microbial protein synthesis (Hillman et al., 1993).

Study sites. The study was conducted at two sites, Domboshava and Makoholi. Domboshava (31°13' E, 17°30'S altitude 1530 m) has a mean annual rainfall of 895mm that falls predominantly from November to March. Summer temperatures rise to a maximum monthly mean of 27.9° C in October and winter temperatures fall to a minimum monthly mean of 5.5° C in July. Evapotranspiration is highest (32 mm day^-1) in October and lowest in (14.3 mm day^-1) in June (Dzowela et al., 1995a; 1995b). The soils are ferrallisic cumbisols derived in situ from granodionite of relatively low fertility and a predominantly sandy loam texture (Nyamapfene, 1991). Makoholi ( 30°45'E, 19°48'S; altitude 1204 m) has a mean annual rainfall of 688 mm that falls mainly from Nove-mber to March. The hottest month is October with a mean temperature of 28.6°C and the lowest temperatures occur in July with a mean minimum of 6.6°C. Evapotranspiration ranges from 34.1 mm day^-1 in October to 14.1 mm day^-1 in June. The soils are sandy fersiallitic with inherently low available water holding capacity (Whingwiri et al., 1987).

The stands of the Leucaena species were defoliated by harvesting at the end of the dry season (November) and the regrowth harvested at the end of the rainy season (April). At each harvest, leaves were separated from stems and either sun-dried for 5 days on an open concrete floor or oven-dried at 50°C for 48 hr. At each site there were 3 replicates for each accession and 4 trees were harvested and bulked per replicate. In Domboshava 3 accessions of L. diversifolia, 2 of L. esculenta, 1 of L. pallida, 1 of L. pulverulenta, 1 of L. savadorensis, 2 of L. shannonii, 1 of L. trichodes and 2 of an interspecific hybrid L. leucocephala X L. diversifolia ( L.l. X L.d.) were harvested whilst in Makoholi there were 3 accessions of L. esculenta, 2 of L. diversifolia, 1 of L. pallida, 2 of L. pulverulenta, 1 of L. shannonii and 1 of inter specific hybrid L.l. X L.d. Details of the accessions are given in Tables 1 and 2.

TABLE 1. Chemical composition of some psyllid-resistant Leucaena accessions grown at Domboshava

TABLE 2. Chemical composition of some psyllid-resistant Leucaena accessions grown at Makoholi

Laboratory analysis. The dried samples were ground to pass through a 1 mm screen and were analysed for concentration of dry matter (DM), organic matter (OM), crude protein (CP) (AOAC, 1990), neutral detergent fibre (NDF), acid detergent fibre (ADF), acid detergent fibre lignin (ADL) (Robertson and van Soest, 1981), extractable proanthocyanidins (PAs) (Porter et al., 1986) and soluble phenolics (Reed et al., 1985). The protein-precipitating potential (PPP) of the phenolics was estimated by the radial diffusion method (Hagerman, 1989) and expressed as square of diameter of cleared sphere (mm²). In vitro gas production was carried out in triplicate as described by Nsahlai et al. (1994).

Statistical analysis. All the statistical analysis were performed using the general linear model (GLM) procedure of SAS (1990). Data for each site were analysed separately. The chemical analysis data were subjected to analysis of variance (ANOVA) using a model that accounted for effects of season of harvest, drying method, species, accession within species and two- and three-way interactions among them. None of the interactions were significant and they were dropped from the final model.

The gas production data were fitted to the non-linear equation: Y= a + b(1-e ^-ct), where Y= gas production at time ‘t’, ‘a’ = gas evolved within 1 hour and ‘b’ = the volume of gas evolved with time at a fractional rate of ‘c’ (Blummel and Orskov, 1993). The gas production constants (a, b and c) were then subjected to ANOVA as above to compare differences between accessions within species and between species.

Relationships between chemical composition, phenolic content (PAs, SPs and PPPs) and gas production constants were established by correlations and multiple regression using a stepwise procedure in which the variable entry criterion was set at 0.15 probability level of significance. Cluster analysis was used to classify the accessions based on gas production constants, fibre constituents (ADF, ADL and NDF) and phenolic content.

The drying method had no significant effect (P>0.05) on the concentration of macro constituents (DM, OM, ADF, NDF, ADL, CP), phenolics and gas production constants for either site. Season of harvest did not significantly affect concentration of phenolics and gas production constants at either site but affected (P<0.05) the concentration of macro constituents at Makoholi. Results on effects of species and accessions within species are given (Tables 1 and 2).

Concentration of organic matter did not differ across species but CP varied between species on the same site and, for some species, across sites. For example L. shannonii had the second lowest CP at Domboshava (223 g kg^-1) but the highest at Makoholi (329 g kg^-1). However, within each site the ranges of CP concentration, though statistically significant (P<0.05), were narrow 181-292 g kg^-1 DM at Domboshava and 264-329 g kg^-1 DM at Makoholi. The content of NDF was high for L. shannonii and L. trichodes and low for L. pallida, L. esculenta and L. pulverulenta at Domboshava but species had no significant effect at Makoholi. The concentration of PAs showed intraspecific variation for L. diversifolia, L. esculenta and L.l. X L.d. at Domboshava and only for L.esculenta at Makoholi (P<0.05). For L. esculenta, the accession Oxford Forestry Institution (OFI) 52/87 had 300 % less concentration of PAs than accession OFI 47/87 at Domboshava but it had only 32 % more at Makoholi. Overall, species had higher concentrations of SPs at Makoholi than at Dombo-shava.

There were large intra-specific variations between L. esculenta and L. pulverulenta species at Makoholi and among L. diversifolia, L. esculenta and L.l. X L.d. at Domboshava in gas production constants (Tables 3 and 4). In general, all species produced very little gas within 1 hour of incubation. L. shannonii produced the highest gas volume (circa 27 ml) and L. pulverulenta had the least (circa 11 ml) at both sites.

Correlations between chemical composition and gas production constants are shown in Tables 5 and 6. For the Makoholi site, only PAs and DM were negatively related to total volume of gas produced (A + B) whilst for the Domboshava site ADL, PAs, SPs and PPPs were all negatively correlated to total volume of gas produced. Mul-tiple linear regression equations relating chemical composition to gas production are shown in Tables 7 and 8 for Domboshava and Makoholi, respectively. There were no significant seasonal effects (P>0.05) for the Domboshava site but there were significant seasonal effects (P<0.05) for the Makoholi site. PPP and PAs were poorly predicted from fibre concentration of the browses for the Domboshava site. For the Makoholi site, PPP was poorly predicted (R²=0.38) from fibre concentration for the herbage harvested in November but could be better predicted (R²=0.83) from fibre concentration for the herbage harvested in April. PAs could be predicted accurately (R²>0.9) for both harvests from concentrations of macro constituents.

TABLE 3. Gas production constants of some psyllid-resistant Leucaenas grown at Domboshava

TABLE 4. Gas production constants of some psyllid-resistant Leucaenas grown at Makoholi

TABLE 5. Correlation coefficients (P<0.05) between chemical composition^@ , phenolic content and gas production constants of psyllid resistant Leucaenas grown in Domboshava

TABLE 6. Correlation coefficients (P<0.05) between chemical composition, phenolic content and gas production constants of psyllid resistant Leucaenas grown in Makoholi

TABLE 7. Relationships⁺ between chemical composition, phenolic content and gas production of some psyllid resistant Leucaenas harvested at Domboshava

TABLE 8. Relationships+ between chemical composition, phenolic content and gas production of some psyllid resistant Leucaenas harvested at Makoholi

Grouping of the accessions using cluster analysis suggested 3 clusters whose membership differed with the variables used in classification (Tables 9 and 10). When fibre constituents (ADF, ADL and NDF) were used for classification, cluster 1 contained accessions low in ADL (circa 100 g kg^-1). Cluster 2 had accessions with medium ADL concentration (circa 120 g kg^-1) and cluster 3 had those with high ADL concentration (circa 150 g kg^-1). When clustering was based on phenolic content (PA, PPP and SP) those accessions with low PPP (circa 46 mm²) were found in cluster 1, those with medium PPP (circa 68 mm²) were in cluster 2 and those with high PPP (circa 108 mm²) were in cluster 3. For gas production parameters (a, b and c) cluster 1 had high total gas production (mean 26 ml + 1.56), cluster 2 had medium total gas production (mean 16.2 + 2.64 ml) and cluster 3 had low gas production (mean 11.2 + 0.53 ml).

TABLE 9. Classification of some psyllid resistant Leucaenas from Domboshava based on fibre content, phenolics and gas production

TABLE 10. Classification of some psyllid resistant Leucaenas from Makoholi based on fibre content, phenolics and gas production

Production of secondary compounds in response to herbivory and/or pathogenic organisms is a common defense mechanism in plants (Harborne and Grayer, 1994). Polyphenolic compounds such as tannins (proanthocyanidins and hydrolysable tannins) are a major class of secondary compounds which are involved in these interactions (Swain, 1979). The content of tannins in psyllid-resistant Leucaena species was thus of primary interest in this research. These compounds bind to proteins, starch, cellulose and minerals (Reed, 1995) thereby affecting nutrient release and supply to animals consuming forages high in these compounds.

Current chemical assays for polyphenols do not always reflect their biological effects (Makkar et al., 1993) and thus in this study 3 assays were used in order to encompass as wide an array of this diverse group of compounds as possible. Biological effects were assessed by the production of gas in vitro since tannins have been shown to interfere with the process (Khazaal et al., 1994).

Considerable intra- and inter- specific variations in CP, ADL and polyphenolic concentration were found. Similar results have been reported in Indian (Makkar and Singh, 1991), West African (Rittner and Reed, 1992), Greek (Khazaal et al., 1993) and Hawaiian browse (Wheeler et al., 1994). Levels of CP were high (180-329 g kg^-1 DM) and comparable with those reported by Wheeler et al. (1994) and Dzowela et al. (1995a) for psyllid- susceptible L. leucocephala. At both sites, L. pallida and L. shannonii species were low in PAs, SPs and PPP compared with L. pulverulenta. In general, however, the browses had more SPs when grown at Makoholi than when grown at Domboshava, reflecting a more stressful environment in the former site possibly due to lower soil fertility and moisture levels. The SPs were determined using the ytterbium precipitation method of Reed et al. (1985) which has been criticised as unsuitable for the determination of total phenolics (Lowry and Sumpter, 1990). Although ytterbium does not precipitate all phenolics, it does precipitate all phenolics with free vicinal hydroxyl groups which is a characteristic of most tannins (Rittner and Reed, 1995). Moreover other researchers have found that the amounts precipitated correlate well with PAs determined by both the butanol-HCl and Vanillin-HCl method (Reed, 1995; Giner-Chavez et al., 1997).

Differences in chemical composition have a strong bearing on the nutritive value of forages and it was hypothesised that these will be reflected in gas production parameters. The low but significant negative correlations (0.38 - 0.6) between phenolic content and total gas production obtained in this study (Tables 5 and 6) are indicative of the negative effect of polyphenols on fermentation of forages. However, in multiple regression equations that included fibre fractions, only soluble phenolics, among the tannin assays used, met the criteria for inclusion in equations predicting total gas production (Tables 7 and 8) except for the November harvest at Makoholi. The failure of phenolics to predict gas production in regression equations that included fibre fractions highlights the complexity of the interactions between fibre and polyphenols in influencing the nutritive value of forages. There is a need for detailed analysis of chemical structural differences in polyphenolics before these interactions can be well understood.

The gas production constants obtained in this study were lower than those reported by Khazaal et al. (1993; 1994), Siaw et al. (1993) and Nsahlai et al. (1994), probably due to the fact that, nitrogen (N₂ ) gas instead of carbon dioxide (CO₂) was used to achieve anaerobiosis during incubations. Whilst N₂ achieved this objective, it did not lower the pH of the incubation medium as much as CO₂ would have. This could have influenced microbial activity (Bryant, 1973) and/or complexing properties of tannins (Makkar et al., 1995).

Cluster analysis was carried out to ascertain if phenolics and gas production parameters would group the accessions similarly. However, there was minimal overlap in the clusters obtained from the three classification variables at both sites. This failure to obtain clusters with largely similar membership when the classification criterion variables were chemical components, phenolics and gas production has been reported by Siaw et al. (1993) and Nsahlai et al. (1994) for Sesbania species. These observations therefore, further emphasise the need to improve the understanding of the relationships between these components as they affect nutritive value of browses.

Psyllid-resistance in Leucaena species grown in Zimbabwe was not associated with polyphenolic concentrations substantially different from those published for psyllid-susceptible L. leucocephala. The relationship between polyphenolic conce-ntration and in vitro gas production constants indicated a complex interaction with fibre concentration. At Domboshava, the interspecific hybrid L.l. X L.d. (accession ILCA 15009) , L. shannonii species, L. salvadorensis and L. esculenta (accession OFI 47/87) rated high in indices indicative of good nutritive value. The corresponding accessions at Makoholi were L. shannonii OFI 19/84 and L. esculenta OFI 52/87. These accessions require further evaluation in vivo to determine palatability and digestibility.

The financial support of the Germany Academic Assistance programme DAAD through a scholarship to L. Mlambo and the logistical support of the SADC-ICRAF Agroforestry Project are gratefully acknowledged.

• Association of Official Analytical Chemists (AOAC), 1990. Official Methods of Analysis. 214^th Edn. Association of Official Analytical Chemists: Washington, DC.
• Blummel, M. and Orskov, E. R. 1993. Comparison of in vitro gas production and nylon bag degradability of roughages in predicting feed intake in cattle. Animal Feed Science Technology 40:109-119.
• Brewbaker, J.L. and Macklin, B. 1990. Nitrogen fixing trees for fodder in Agroforestry systems. In: Agroforestry Land-Use Systems. Moore, B. (Ed.), pp. 53-61. Nitrogen Fixing Tree Association: Waimanalo, Hawaii.
• Bryant, M. P. 1973. Microbiology of the rumen. In: Dukes’ Physiology of Domestic Animals. 8^th Ed. Swenson, M. J. (Ed.), pp. 484-515. Comstock Publishing Associates, Ithaca, New York.
• Dzowela, B. H., Hove, L. and Mafongoya, P. L. 1995a. Effect of drying method on chemical composition and in vitro digestibility of multi- purpose tree and shrub fodders. Tropical Grasslands 29:263-269.
• Dzowela, B. H., Hove, L., Topps, J. H. and Mafongoya, P. L. 1995b. Nutritional and anti-nutritional characteristics and rumen degradability of dry matter and nitrogen for some multi- purpose tree species with potential for agroforestry in Zimbabwe. Animal Feed Science Technology 55:207-214.
• Giner-Chavez, B. I., Van Soest, P. J., Robertson, J. B., Lascano, C., Reed, J. D. and Pell, A. N. 1997. A method for isolating condensed tannins from crude plant extracts with trivalent ytterbium. Journal of the Science of Food and Agriculture 74:359-368.
• Haslam, E. 1979. Vegetable tannins. In: Biochemistry of Plant Phenolics (Recent Advances in Phytochemistry). Swain, T., Harborne, J. B. and Van Sumere, C. F. (Eds.), pp. 475 -523. Plenum Press, New York,
• Hagerman, A. E. 1979. Radial diffusion method of determining tannin in plant extracts. Journal of Chemical Ecology 13: 437-449.
• Hagerman, A. E., Robbins, C. T., Weerasuriya, X., Wilson, T. C. and McArthur, C. 1992. Tannin chemistry in relation to digestion. Journal of Range Management 45:57-62.
• Harborne, J. B. and Grayer, R. J. 1994. Flavonoids and insects. In: The Flavonoids: Advances in Research since 1986. Harborne, J. B. (Ed.), pp. 589-618. Chapman and Hall, London.
• Hillman, H. K., Newbold, C. J. and Stewart, C. S. 1993. The contribution of bacteria and protozoa to ruminal forage fermentation in vitro, as determined by microbial gas production. Animal Feed Science and Technology 42:193-208.
• Jones, R. J. 1994. Management of anti-nutritive factors with special reference to Leucaena. In: Forage Tree Legumes in Tropical Agriculture. Gutteridge, R. G. and Shelton, H. M. (Eds.), pp.216-231. CAB International Wallingford, UK.
• Khazaal, K., Dentinho, M. T., Ribeiro, R. and Orskov, E. R. 1993. A comparison of gas production during incubation with rumen contents in vitro and nylon bag degradability as predictors of the apparent digestibility in vitro and voluntary intake of hays. Animal Production 57:105-112.
• Khazaal, K., Boza, J. and Orskov, E. R. 1994. Assessment of phenolics-related antinutritive effects in Mediterranean browse: a comparison between the use of the in vitro gas production technique with or without insoluble polyvinylpolypyrrolidone or nylon bag. Animal Feed Science and Technology 49: 113-149.
• Khazaal, K. and Orskov, E. R. 1994. The in vitro gas production technique: an investigation on its potential use with polyvinylpyrrolidone for the assessment of phenolic related anti-nutritive factors in browse. Animal Feed Science and Technology 47:305-320.
• Lowry, J. B. and Sumpter, E. A. 1990. Problems with ytterbium precipitation as a method for determination of plant phenolics. Journal of Science of Food and Agriculture 52:287-288.
• Makkar, H. P. S., Singh, B. and Negi, S. S. 1989. Relationship of rumen degradability with microbial colonization, cell wall constituents and tannin levels in some tree leaves. Animal Production 49:299-303.
• Makkar, H. P. S. and Singh, B. 1991. Distribution of condensed tannins (proanthocyanidins) in various fibre fractions in young and mature leaves of some oak species. Animal Feed Science and Technology 32:253-260.
• Makkar, H. P. S., Blummel, M., Borowy, N. K. and Becker, K. 1993. Gravimetric determination of tannins and their correlations with chemical and protein precipitation methods. Journal of the Science of Food and Agriculture 61:161-165.
• Makkar, H. P. S., Blummel, M. and Becker, K. 1995. Formation of complexes between polyvinylpolypyrrolidones or polyethylene glycols and tannins, and their implications in gas production and true digestibility in in vitro techniques. British Journal of Nutrition 73:817-913.
• Nsahlai, I. V., Saiw, D. E. K. A. and Osuji, P. O. 1994. The relationships between gas production and chemical composition of 23 browses of the genus Sesbania. Journal of the Science of Food and Agriculture 65:13-20.
• Nyamapfene, K. 1991. The soils of Zimbabwe. Nehanda Publishers: Harare. pp. 5-10.
• Porter, L. P., Hrstich, L. N. and Chan, B. G. 1986. The conversion of proanthocyanidins and prodelphinidins to cynidin and delphinidin. Phytochemistry 25:223-230.
• Reed, J. D. 1995. Nutritional toxicology of tannins and related polyphenols in forage legumes. Journal of Animal Science 73: 1516-1528.
• Reed, J. D., Horvarth, P. J., Allen, M. S. and Van Soest, P. J. 1985. Gravimetric determination of soluble phenolics including tannins from leaves by precipitation with trivalent ytterbium. Journal of the Science of Food and Agriculture 36:255-261.
• Rittner, U. and Reed, J. D. 1992. Phenolics and in vitro degradability of protein and fibre in West African browse. Journal of the Science of Food and Agriculture 58:21-28.
• Robertson, J. B. and van Soest, P. J. 1981. The detergent system of analysis and its application to human foods. In: The Analysis of Dietry Fibre in Foods. James, W. P. T. and Theander, O. (Eds.), pp 123-158. Manel, Dekker, New York.
• SAS. 1990. Procedures Guide for personal computers. Version 6 . SAS Institute Inc, Cary.
• Siaw, D. E. K. A., Osuji, P. O. and Nsahlai, I. V. 1993. Evaluation of multipurpose tree germplasm: The use of gas production and rumen degradation characteristics. Journal of Agricultural Science Cambridge 120:319-330.
• Swain, T. 1979. Phenolics in the environment. Recent Advances in Phytochemistry 12: 617-640.
• Telek, L. 1992. Preparation of leaf protein concentrates from Leucaena leucocephala. Leucaena Research Reports 13: 93-94.
• Wheeler, R. A. and Brewbaker, J. L. 1990. An evaluation of results from the Leucaena psyllid trial network. Leucaena Research Reports 11:23-31.
• Wheeler, R. A., Chaney, W. R., Butler, L. G. and Brewbaker, J. L. 1994. Condensed tannins in Leucaena and their relation to psyllid resistance. Agroforestry Systems 26:139-146.
• Whingwiri, E. E., Natarajan, M., Kanyanda, C. W. and Van Lindert, H. J. A. 1987. Agro-climate and soil characteristics of the research centres in the Department of Research and Specialist Services. Zimbabwe Agricultural Journal 84:37-54.

©2000, African Crop Science Society