Characterization and comparative analyses of genetic divergence for identification of diverse parents on napier grass germplasm (Pennisetum purpureum L. Schumach)

  • S. Anandhinatchiar, C. Babu, K. Iyanar, T. Ezhilarasi and V. Ravichandran Tamil Nadu Agricultural University


The present study was carried out to characterize 56 Napier grass (Pennisetum purpureum L. Schumach) accessions using 17 qualitative traits and to assess the magnitude of genetic diversity among the accessions using 13 biometrical traits in order to select diverse genotypes for using them as parents in future breeding programme. Characterization of the germplasm revealed the presence of discernible variation for majority of the traits with an extremity in the following traits viz., presence of prop roots, leaf sheath pigmentation and bristle length ranging from 8.93 % - 67.86 %, 8.93 % - 64.29 % and 3.57 % - 58.93 %, respectively. Genetic diversity among the genotypes was assessed based on Mahalanobis’s D2 statistics and agglomerative hierarchical clustering approach. Based on Tocher’s method, the genotypes were classified into 11 clusters. The highest inter cluster distance was observed between cluster X and VI (67.82), VI and IV (57.63) and cluster VIII and VI (52.48). Therefore, the genotypes from the above clusters viz., FD 437, FD 470, FD 474, FD 471, FD 461, FD 473, FD 476, FD 462, FD 426 could be used as parents for the development of high yielding Pearl millet Napier hybrids. Clusters V, X and XI exhibited the highest mean values for majority of the yield attributing traits. Therefore, genotypes from the clusters V, X and XI viz., FD 459, FD 426, FD 468 could be utilized in future hybridization programme for enhancing the appropriate traits. In agglomerative hierarchical clustering method also, the genotypes were classified into 11clusters. Outcome from Tocher’s method and agglomerative hierarchical approach relatively complement with each other and satisfactory to delineate the genetic divergence among the germplasm.

Research Article