Genetic Stability of Grain Yield and Principal Component an analysis in Pearl Millet (Pennisetum glaucum L)

An experiment was conducted to study Principal Component A analysis, yield potential and yield stability of thirty four pearl millet genotypes at Gezira Research Farm (GRF) and Rahad Research Farm (RRS) in the autumn of 2009. The genotypes were of different genetic backgrounds and origin. The experiment was arranged in randomized complete block design with three replications and was carried out during the rainy season. Data were collected on days to 50% flowering, plant height, panicle length, number of productive tillers, head weight, grain yield and 100 seed weight (TSW).Days to50% flowering, head weight, panicle length, number of productive tillers and TSW were significantly and positively associated with grain yield. The first three components of principal component analysis (PCA) accounted for 69% of the total variability attributable to grain yield. The PC scores were associated with days to 50% flowering, panicle length, head weight, number of productive tillers and TSW. Results from yield stability analysis showed that Baladi white was the most stable genotype in terms of grain yield followed by Dembi yellow and Ugandi across sites, while ICMV 155 was the most unstable genotype. At least three of the thirty four pearl millet genotypes, Baladi white, dembi yellow and Ugandi, were stable across the tow site. The three genotypes had grain yield and TSW slightly higher the overall means and could be used in breeding program to further exploit yield potential and stability.