Kant, RaviKASHYAP, YOGESH2024-09-262024-09-262023M/PBG/217/2021-22https://krishikosh.egranth.ac.in/handle/1/5810215026The present experiment was conducted at the research farm of TCA, Dholi, RPCAU, Pusa, during the rabi season of 2022-23. The study involved twenty-five different grass pea lines, including the reference check DLY-13-7. The experimental setup employed a Randomized Block Design (RBD) with three replications. Data was collected from five randomly selected plants in each replication, and the mean values were subjected to statistical analysis using the R software. The outcomes of the analysis revealed significant genetic variability across all traits studied. Notably, the Phenotypic Coefficient of Variation (PCV) and Genetic Coefficient of Variation (GCV) were notably high for traits such as harvest index, seed yield, and the number of pods per plant. These findings suggest that efficient selection of desired genotypes could be performed for these particular traits. Traits like seed yield per plant, harvest index, and biological yield per plant exhibited both high heritability and significant genetic advance. This indicates the prevalence of additive gene action in the expression of these attributes, implying their suitability for direct selection to enhance genetic traits. Seed yield per plant exhibited significant positive correlations with traits like the number of pods per plant, 100-seed weight, harvest index, pod length, and the number of seeds per pod. These findings underscore the importance of prioritizing these traits in the selection process for crop improvement, given their positive associations with seed yield. Several traits such as days to maturity, plant height, 100-seed weight, number of primary branches, number of seeds per pod, number of pods per plant, and pod length exerted direct effects on seed yield per plant. In contrast, other traits indirectly influenced seed yield through their impact on these key traits. Therefore, selection strategies centered around these influential traits are likely to be successful in boosting grass pea production. Correlation and path analysis of characters such as no. of pods per plant,100 seed weight, harvest index, pod length & no. of seed per pod would be dependable and efficient as they illustrated positively strong alliance with seed yield and positive interrelation amongst themselves and also through high indirect effects of many of the traits on seed yield. The twenty-five genotypes were categorized into five groups, with cluster III containing the most genotypes. Cluster IV and V were composed of single genotype. Maximum inter-cluster distance was observed between cluster III and II, whereas the minimum inter-cluster distance occurred between clusters II and IV. This finding implies the potential for obtaining transgressive segregants by crossing genotypes from these clusters. Based on molecular diversity, the twenty-five genotypes were classified into four clusters. Clusters I and III encompassed the highest number of genotypes, while cluster II contained the fewest. Genotype LAT-2019-5 displayed superiority in terms of seed yield per plant, plant height, number of primary branches, and pod length, although its 100-seed weight was comparatively lower. For 100-seed weight, LAT-2019-12 emerged as a promising genotype. Genotype LAT-2019-14 show excellence in the number of branches per plant, whereas DLY-13-7 stood out for the number of pods per plant. These findings emphasize the necessity for extensive testing of genotypes across different years and locations for potential use as varieties or integration into future grass pea improvement programs. Notably, strong similarity coefficients were observed between LAT-21-3 and LAT-21-6, followed by LAT-21-4 and LAT-2019-12.EnglishThesis