The present study was conducted to evaluate the genetic divergence among 32 black gram genotypes using Mahalanobis D2 statistic for ten biometric characters. The genotypes were grouped into nine discrete clusters. Among them cluster I grouped with a maximum of 11 genotypes, followed by cluster IV with 7 genotypes, clusters II and III each with 4 genotypes and cluster VIII with 2 genotypes, while the remaining clusters were solitary. The inter-cluster distance was higher than intracluster distance, confirming the presence of high genetic variability among the genotypes. Based on the inter-cluster distance, cluster V and IX and clusters V and VI were identified as the most diverse clusters and hence the genotypes from these clusters can be used in the hybridization programme for further genetic improvement. The maximum contribution to genetic divergence was by the number of branches per plant followed by plant height and hundred seed weight. Hence selection and utilization of genotypes based on these traits would be more effective in the selection for hybridization programme.