Methods of Constructing Core Set Using Agro-morphological Traits in Foxtail Millet [Setaria italica (L.) Beauv]

The two different methods were employed to construct core sets of foxtail millet germplasm using data on geographical distribution and agro-morphological traits and the variation generated for different traits in core sets developed by two methods was compared with the entire collection. First method was grouping the accessions under distinct morphological traits followed by clustering and Principal Component Analysis which resulted in C₁ having 156 accessions; second was by using a software called Power Core which yielded C₂ consisting of 78 accessions. In both the core sets formed, maximum accessions were from Asian continent. Newman-Keuls test for means inferred that both the core sets were true representatives of entire collection. Similarly, Levene's test for homogeneity of variances revealed non-significant differences between entire set, C₁ and C₂. The range for the traits studied in both C₁ and C₂ were almost similar to the range of the traits in the entire collection. Likewise, the chi-square test for frequency distribution analysis for different morphological traits indicated that the variation available in the entire collection was preserved in both the core sets. The Shannon-Weaver diversity indices of II quantitative and 12 qualitative traits of entire collection, C₁ and C₂ indicated the presence of diversity of entire collection in the core sets.