Cluster bean is an economically important legume because of the presence of 'Guar gum' in the endosperm of seeds. However, Bacterial Leaf Blight (BLB) is the major yield constraint of cluster bean, causing severe yield losses in optimal conditions. So, to harmonize the cluster bean demand, it is necessary to identify other attributes which are directly or indirectly involved in yield enhancement which can be achieved by performing correlation and path analysis studies. In our study, path analysis was carried out on the F3:4 population of the interspecific cross (HG563 x Cyamopsis serrata Shinz.) of cluster bean using OPSTAT software. In the growing season of 2016-17, various agronomical characters (plant height, branches per plant, clusters per plant, pods per cluster, pods per plant, pod length, days to 50% flowering, days to maturity, a hundred-seed weight, seed yield per plant, PDI) were recorded under natural BLB infection over two different locations, i.e. RRSB and FS. Path analysis indicated that pods per plant had the highest positive direct effect on yield over both the locations (RRSB, p=0.735; FS, p=0.968) followed by hundred seed weight (RRSB, p=0.640; FS, p=0.187). Thus, both these characters should be kept in mind while selecting the higher-yielding plants for cluster bean improvement. We successfully selected promising lines from both locations and the line no. 6-9 performed better and was resistant to BLB at both locations, suggesting it could be further useful in cluster bean breeding program.