Principal Component Analysis for Yield and Yield Related Traits in Pearl Millet Cultivars

Abstract

A set of 40 pearl millets genotypes were evaluated with a view of studying genetic parameters for eight quantitative characters. In the experimental materials, analysis of variance revealed significant differences for all analyzed traits. The study identified 3 Principal Components (PCs) with Eigen value greater than 1.00 which accounted for 82.3% of the total variation for discriminating the lines. From principal component analysis, PC1 showed highest amount of variance (36.2%) with mostly related to traits like yield per plant, productive tillers per plant, and fodder yield per plot. As a result, the first component mainly identifies the characters responsible for yield. PC2 showed second highest amount of variance (29.9%) with cumulative variance (66.1%) with mostly related to traits like days to maturity, days to 50% flowering, and panicle diameter. Third highest variance (16.2%) with cumulative variance (82.3%) with mostly related to traits plant height and panicle length was observed in PC3. Elbow type line is seen after PC3 showing little variation after PC3. Highest Eigen value is shown by PC1. From biplot, PC1 components showed negative relationship with plant height and Days to 50% flowering, days to maturity showed slightly negative with PC1. PC2 had negative relationship with traits like productive tillers per plant and slightly negative yield per plant and both components showed positive relation with panicle diameter, panicle length and fodder yield per plot.