Evaluation of sunflower (Helianthus annuus L.) germplasm using multivariate statistical techniques

Abstract

Evaluation of sunflower hybrids using morphological data is necessary and essential in sunflower breeding programs. The aim of this paper was to evaluate the productive possibilities of some sunflower germplasm using multivariate technique. Fifty sunflower genotypes were characterized using eleven morphological traits. Among the traits studied, high coefficient of variation was observed for duration ofreproductive phase (11.73%), seed yield per plant (11.49%), 100-seedweight (8.98%) and head diameter (8.33%). Agglomerative hierarchical clusteringmethod classified the fifty genotypes into five clusters. A large number of genotypes were placed in cluster III (22) followed by cluster I (14), cluster II (6), cluster V (6) and cluster IV (2). The maximum inter-cluster distance of 47.83 was observed between clusters I and cluster IV indicating the possibility of high heterosis. The first four principal components showed 71.61% of the total variation. The genotype EC-512687 and EC-601746 had found top rank by both the methods. The results of PCA were closely in line with those of composite indices. These results can be used by breeders in sunflower breeding program.