Evaluation and classification of Indian mustard (Brassica juncea L.) genotypes using principal component analysis

Abstract

Principal component and hierarchical cluster analyses were carried out with 25 quantitative and qualitative
traits in 60 genotypes of Indian mustard (Brassica juncea L.). Principal factor analysis identified 11 principal
components (PCs) which explained about 75% variability. PC1 had 13.19% of total variation in agromorphological
traits, PC2 depicted 10.07% of total morphological variability, and PC3 accounted for 8.56%
of the total variation. Varimax rotation enabled loading of similar type of variables on a common principal
factor permitting to designate them as seed yield, maturity, leaf and siliqua characters, and oil content factors.
The genotypes JMM-937, RC-199, RH-0401(YS), Pusa Bold, Pusa Bahar, and KM-888 were found to be
superior on the basis of principal factor scores with regard to seed yield, its main components, and oil content,
when both the principal factors were considered together. These genotypes may further be utilized in
breeding programmes for developing Indian mustard varieties with high seed yield and superior oil content.
Hierarchical cluster analysis categorized all the 60 genotypes into 10 clusters containing one to 23 genotypes.
Based on the inter-cluster distances, maximum genetic diversity was observed between clusters I and IV
(221.4), followed between CII and IV clusters (200.5), C IV and C IX (191.8) and C IV and C X (181.5)
indicating that genotypes from these clusters can usefully be hybridized for getting superior recombinants in
segregating generations. The results of cluster and principal factor analyses confirmed each other.