Breeding African indigenous crops with increased nutritional value has recently been strongly advocated as a sustainable rural-based strategy of partially meeting the nutritional requirements of a large proportion of Africa's population. This approach requires information on the extent of genetic diversity among available genetic resources. Forty bambara groundnut (Vigna subterranean (L) Verdc) landraces of African origin were evaluated for agronomic attributes, protein, and grain mineral contents in a replicated field experiment at a single location across 2 years at the experimental field of the Federal University of Agriculture, Abeokuta, Nigeria. Principal component analysis (PCA) and cluster analysis (CA) (the Ward's method) were employed to study the variation pattern. Highly significant (P < 0.01) variation was detected among the landraces for all traits. Spearman's rank correlation indicated relatively high G × E interactions for all traits except Fe, K, and protein content, suggesting that single year testing is not sufficient for such traits. The first two PCs, which accounted for 40.81% of the multitrait variation, had all the agronomic traits and Cu as the main contributors. Grain protein content, Zn, Mg, K, Fe, and Ca were the most important loadings for the remaining PCs. Based on nutritional and components of yield traits, cluster analysis stratified landraces into five major groups. Concurrent improvement in pod yield and Zn, Fe, and protein contents could be possible by selecting promising parents from clusters 2 and 3. The diversity found in African landraces of bambara groundnut can be exploited for developing cultivars with improved nutritional value.
ARTICLE HISTORY