Canonical variate analysis in safflower (Carthamus tinctorius L.)

Seventy five genotypes of safflower representing the broad spectrum of variation were assessed for genetic divergence for eight characters using Mahalanobis D² statistic and principal component analysis. The seed yield contributed maximum towards the total genetic divergence followed by test weight and number of seeds per capitulam. On the basis of clustering method, twelve clusters were obtained for D² statistic. The best clusters with regard to seed yield and oil content were cluster XII and cluster II, respectively. Principal component analysis identified three principal components which explained 83.02% variability. Genotypes GMU 3470, GMU 3484, GMU 3499, A-1, JSF-1 and GMU 3475 (based on PCI axis) were divergent.