MULTIVARIATE ANALYSIS OF CHILLI (Capsicum frutescens L.)

Abstract

An investigation was carried out during kharif 2014-15 at experimental field of Shere-Bangla Agricultural University, Dhaka, with 13 genotypes of chilli (Capsicum frutescens L.) in a randomized block design with three replications. The study revealed significant differences among genotypes for different characters studied. The GCV for all the characters studied were lesser than the PCV indicating the masking effect of the environment. High heritability coupled with high genetic advance as percent of mean was observed for fruit length, fruit circumference, individual fruit weight and yield per plant indicating the predominance of additive gene action suggesting, direct phenotypic selection may be useful with respect to these traits. Correlation and path analysis revealed that number of primary brunches per plant, fruit length and individual fruit weight had positive significant association and positive direct effects on yield per plant indicating the use of these attributes in selection to evolve high yielding varieties of chilli. The results of multivariate analysis indicated the presence of considerable genetic divergence among the genotypes. Principal component analysis identified two principal components, which contributed 91.26% of cumulative variance. The 13 genotypes were grouped into 3 clusters. The highest inter-cluster distance was observed between cluster I and II (607.55), Therefore, considering the magnitude of genetic distance, contribution of character towards divergence, magnitude of cluster mean and agronomic performance the genotype G10 (GPB0010/14) from cluster I, G6 (GPB0006/14), G7 (GPB0007/14) and G12 (GPB0012/14) from cluster II should be used for the development of heterotic hybrids in future breeding programmes.