Department of Genetics and Plant Breeding, G.B. Pant University of Agriculture and Technology, Pantnagar - 263 145, India.
Forty one genotypes of sorghum (Sorghum bicolor (L.) Moench) consisting of tillering and non tillering types, were evaluated for classification using Principal Component Analysis. The genotypes were sown in randomized block design and observations were taken for days to 50% flowering, plant height. number of leaves per plant, leaf length, leaf width, total leaf area, number of nodes per plant, internodal length, stem diameter, T.S.S. %, shootfly infestation protein %, dry matter yield and green fodder yield. The fourteen principal components were extracted and cluster analysis was performed to classify the genotypes. As per the cluster analysis based on this principal component analysis the fortyone genotypes were grouped in five non-overlapping clusters. The spatial arrangement of clusters on the basis of mean principal component (PC) scores of each genotypes within each cluster, showed maximum distance between cluster number one and four. The bar diagram prepared on the basis of cluster mean for various characters indicated highest mean for green fodder yield in cluster one, whereas cluster one and two recorded highest dry matter yield. This provide convenient selection of superior clusters for various yield contributing and quality characters.
