LAUSR

Genetic information on a variety of traits contributing to grain yield is required to improve many crops. However, this information is lacking in bambara groundnut (Vigna subterranea (L.) Verdc.). To this end, 24 African bambara groundnut landraces were investigated for 11 quantitative traits in three contrasting planting seasons. Analysis of variance showed significant variation for almost all of the traits. High broad-sense heritability together with relatively high genetic advance obtained for number of pods per plant, 100-seed weight and seed growth rate revealed the role played by additive genetic components in the expression of these traits. Principal component analysis demonstrated the roles of grain yield, number of pods per plant, biomass, harvest index and yield system traits as strong determining traits controlling the overall variation. Correlation analysis showed that biomass production and harvest index had positive and highly significant correlation with grain yield and most of the traits, indicating that they may be of use as additional indices to improve yield. Cluster analysis, which revealed a partial relationship between diversity pattern and ecogeographical origin, proved to be an efficient tool for separating the landraces and therefore provides a foundation for emergence of improved bambara groundnut cultivars.