JOHN, KDURGA NANDINI, GURRAM2019-05-032019-05-032018http://krishikosh.egranth.ac.in/handle/1/5810101803D5714The experiment material consisted of forty genotypes of groundnut which were evaluated at dry land farm of S.V. Agricultural College, Tirupathi during kharif 2017 so as to identify the best genotypes with high pod yield per plant. The analysis of variance revealed significant differences for 14 characters except specific leaf area at 60 DAS, specific leaf weight at 60 DAS, relative water content at 60 DAS, number of primary branches per plant and harvest index indicating the existence of sufficient variability in the material. The genotype, TCGS-1847 registered higher mean values for pod yield per plant, kernel yield per plant, harvest index, shelling per cent, number of mature pods per plant, number of primary branches per plant and desirable low specific leaf area at 60 DAS. The genotype, TCGS-1889 recorded high pod yield per plant, high kernel yield per plant, high dry haulms yield per plant, high 100-pod weight and high 100-kernel weight. Similarly, TCGS-1887 recorded higher pod yield per plant mainly due to high kernel yield per plant, high harvest index, high dry haulms yield per plant, and more number of primary branches per plant. Hence, the studies revealed that the genotypes viz., TCGS-1847, TCGS-1889, TCGS-1877, TCGS -1861 and TCGS-1838 could be used for pod yield improvement in groundnut. xiv Moderate heritability coupled with moderate genetic advance as per cent of mean was recorded by number of primary branches per plant, 100-pod weight, 100-kernel weight and dry haulms yield per plant. Moderate heritability coupled with high genetic advance as per cent of mean was observed for pod yield per plant, kernel yield per plant, number of immature pods per plant and number of secondary branches per plant indicating additive gene action and selection is effective for these characters. The D2 analysis revealed the presence of considerable diversity among 40 genotypes and were grouped into nine clusters. The maximum intra- cluster D2 (674.96) and D (25.98) distances were recorded by cluster II. The maximum inter cluster D2 value was observed between cluster III and IX (12973.21), III vs IV, IV vs VIII, II vs IV, VIII vs IX, II vs V, III vs VI, II vs IX, I vs III, III vs VII and VII vs IX were found to be divergent in the decreasing order of their magnitude. Hence, genotypes in these clusters could be utilized as parents and crossing among them would results in heterotic expression for yield components. Based on mean performance and genetic divergence studies the crosses: TCGS-1847 × TCGS-1864, TCGS-1877 × TCGS-1864, TCGS-1859 × TCGS-1912, TCGS-1861 × TCGS-1864, TCGS-1859 × TCGS-1864, TCGS-1847 × TCGS-1912 and TCGS-1912 × TCGS-1889 would create desirable segregants by combining all physiological, yield and yield components in groundnut. Critical analysis of results obtained from character association and path analysis indicated that kernel yield per plant, dry haulms yield per plant and harvest index had strong positive association with pod yield per plant which also had high magnitude of positive direct effect on pod yield per plant.en-USnullThesis