Abstract

Bambara groundnut (Vigna subterranea L. Verdc.) is one of the most promising legume foods in Africa, due to its agronomic and nutritional potentials. To take advantage of these attributes, several research programs gathering agronomic and genetic data throughout Africa are being implemented. In this framework, we tested yield and yield components response to sowing density during 3 years (2005, 2006 and 2007) in field experiment using two landraces of Bambara groundnut. There are bunch and spreading type. Two plant population densities 62, 500 plants ha^-1 (spacing: 40 x 40 cm) and 250, 000 plants ha^-1 (spacing: 20 x 20 cm) were tested. Factorial trial using split-plot design with three replicates was set to analyse five yield components. Both sowing density and landrace influenced significantly (p < 0.05) seed yield. The highest seed yields were observed for high-density plots (means: 3.54 ± 0.21 t ha^-1) and for the bunch type (mean: 3.69 ± 1.63 t ha^-1). Based on the trend of yield response of sowing density, producing the landrace (bunch type) from the collection of the University of Abobo-Adjame at high density was suggested.

Key words: Bambara groundnut, agronomic, yield, sowing density, bunch type, spreading type, landrace.