Multivariate analysis of agro-morphological variation in exotic germplasm of grasspea (Lathyrus sativus L.)

Fifteen exotic accessions of grasspea (Lathyrus sativus. L) were evaluated for 11 quantitative traits including forage and grain yielding attributes over two consecutive years 2008–09 and 2009–2010 following RBD design. The analysis of variance (ANOVA) revealed significant variation for the characters namely plant height, dry matter yield per plant, secondary branches, pods per plant, seeds per pod, 100 seed weight, seed yield per plant, days to flowering and forage crude protein. The analyses for different characters suggested that the pods per plant have the highest standard deviation and the total variance. Principal Component Analysis (PCA) revealed that the first Principal Component (PC1) alone explains about 94% of total variance, and the second Principal Component (PC2) explains only 2%. Cumulatively these two account for 96% of total variance. Eigen vectors depicted highest contribution of pods per plant in PC1. The component pattern profile showed the correlation of different characters with PC1, PC2 and PC3. Cluster analysis performed using PROC CLUSTER option suggested grouping of accessions in 2 clusters. The cluster I consists of EC539025, EC539026, EC 539030, EC 539031, EC 539020, EC 539023, EC 539032, EC 539038, EC 539027, EC 539034 and EC 539035 and cluster II consists of 4 accessions EC 539021, EC539036 EC 539037 and EC 539013. It was also observed that the accessions of cluster I exhibited more variation than the accessions of cluster II. The present study suggests that categorizing and clustering germplasm accessions into morphologically similar and presumably genetically similar groups are useful for selecting parents belonging to different clusters in crossing programme. This could maximize opportunities for transgressive segregation to improve the grain and forage yield of grasspea in the Bundelkhand region of Central India.