DR. GYAN, ANSAH SOLOMON

BREEDING COWPEA (Vigna unguiculata (L). Walp) FOR PHOSPHORUS USE EFFICIENCY IN GHANA, ABSTRACT

A participatory rural appraisal (PRA) survey was conducted to elicit farmers‘ perceptions on soil fertility (especially soil phosphorus) and to gather information on elite cowpea varieties. Later, field studies were conducted to (a) screen 100 cowpea genotypes for their phosphorus use efficiencies, (b) estimate the magnitude of genetic parameters controlling lowsoil P, (c) estimate narrow sense heritabilities of traits, (d) and carry out combining ability estimates in cowpea. Results of the PRA showed that in the Dangbe East District of Ghana, 46 and 54% of farmers, respectively, planted white-seeded and red-seeded cowpea varieties whereas nearly all (90%) of growers preferred white-seeded varieties in Ejura-Sekyedumasi District. Farmers appeared to have a good knowledge of dominant cowpea production constraints, including insect pests, unreliable rainfall and declining soil fertility. Most farmers reported that the dominant soil-related constraints were erosion, ploughing along slopes and continuous cropping. Farmers also showed that they had some knowledge of low soil P symptoms and their solutions, but not their causes. For production attributes, farmers preferred varieties with high yields, ease of cooking and good taste, whereas for marketing purposes, they preferred white-seeded varieties, and high-yielding and disease- and pest –resistant types. Data from the study to screen genotypes for phosphorus use efficiencies was subjected to Principal Component Analysis and clustering. The pruned dendrogram, generated through agglomerative hierarchical clustering based on similarity matrix, revealed two main groups of genotypes. Group A genotypes had 6, 28, 70, 88, 14, 87, 90, 10, 20 and 48 which were judged high in phosphorus use efficiencies whereas group B genotypes had moderate to low phosphorus use efficiencies. Graphical presentation of efficiencies and responsiveness of cowpea genotypes based on shoot dry weight grouped similar accessions (Group A) in the cluster analysis as P-efficient in the first quadrant. Genotypes 70, 87, 90, 28, and 48 were successfully crossed with preferred varieties (AF1, AF2, AF3, AF4, AF5) which are low in P iii efficiencies using North Carolina II mating design. The resulting F1 progeny and their parents, were evaluated in contrasting P conditions in two locations using a split plot design with three replications. Additive genes were responsible for controlling days to physiological maturity and number of pods per plant whereas non-additive genes controlled phosphorus use efficiency, shoot dry weight, phosphorus uptake by seeds, and seed yield under P stress. With regard optimum P environments, high heritability estimates were obtained for days to 50% flowering (50%FLW), shoot dry weight (SDW) and plant phosphorus uptake (PUpS) whereas moderate heritability estimates were obtained for days to physiological maturity, phosphorus use efficiency (PUE) and seed yield. The heritability estimates for number of pods per plant was low under two P contrasting conditions showing that there will be high environmental influence in the transmission of the trait from parent to offspring. High and significant GCA estimates of parents under optimal P conditions indicated that the progenies from crosses among AM3, AM6 and AF3 can be used in a cowpea breeding programme under optimal soil P conditions. High and significant GCA estimates of parents under P stress suggested that the progenies from the crosses among AM3, AM7 and AF3 can be used in a cowpea hybridization programme under low soil P environments. In addition to seed yield, AM3 possessed additional traits such as 50%FLW and SDW under optimum P condition. AM3 also had 50% flowering, phosphorus use efficiency, PUpS and number of pods per plant in addition to seed yield under P stress. The study showed that none of the parents combined all the preferred characteristics. It should be possible to combine these traits relatively easily by carefully selecting the parents to be employed in the crosses. Implications of these studies in cowpea breeding and development in Ghana are discussed.