A Dissertation Submitted in Partial Fulfillment of the Requirement for the Degree of Master’s in Life Sciences of the Nelson Mandela African Institution of Science and TechnologyA set of 41 lablab bean accessions were evaluated based on morphological characteristics and farmers’ participatory selection was performed based on agronomic and sensory traits. An experimental plot was laid down in augmented block design where accessions collected from different gene bank were sown at spacing of 75 cm x 40 cm. Descriptive and multivariate analysis based on 21 quantitative traits revealed high coefficient of variation (CVs) in secondary branches, seed yield, and pods per plant. Principle component analysis (PCA) showed that first 6PCs contributed to 83.3% of the total variation. Days to maturity had significant correlation with days to podding (0.855) and flowering (0.821). Seed yield per plant had strong association with pods per plant (0.793). Cluster analysis based on Un-weighted Pair Group Method Average (UPGMA) grouped 41 accessions into 7 clusters based on traits kinship. Cluster II, VI and VII are genetically different from other clusters. Farmers selected ten best accessions, D163, D137, D88 D27, D85, D155, D7, D159, D151 and D140. The selection criteria perceived were diseases and pest resistance, pod per plant, earliness, bulk leaves, high yield, seed colour, seed size, drought tolerance, plant height and growth type. Farmers’ preferred traits; high yielding, better taste, earliness and short cooking time need to be incorporated in bean breeding programs. In sensory evaluation, panelists chose accessions D137, D85 and D88 due to good sensory potential for home consumption. This study recommends accessions D137, D163, D85, D208 and D88 for further evaluation on different locations for developing lablab varieties in Tanzania.