Multivariate analysis of yield contributory traits for selection criteria in Urdbean (V. mungo L. Hepper)

The present study was consisted of 100 urdbean genotypes grown at two locations i.e. at Punjab Agricultural University, Ludhiana and its Regional Research Station, Gurdaspur during Summer, 2019. In principal component analysis, first four PCs possessed Eigen value more than 1 and revealed 73.29% and 66.43 % variability at Ludhiana and Gurdaspur respectively. PCA analysis suggested the importance of biological yield per plant, plant height, days to 50% flowering, harvest index, yield per plant and pods per plant to speed up the urdbean improvement. The correlation studies suggested significantly positive association of clusters per plant, pods per plant, biological yield per plant and harvest index with grain yield per plant at both locations. Path analysis indicated that harvest index, biological yield per plant, nodes per plant, branches per plant showed the higher and positive direct effect on seed yield at both locations. Direct selection of genotypes with positively correlated characters can be considered useful for increasing seed yield.