Non-parametric approach to the analysis of phenotypic stability of two half-sib maize populations

One of the most important phases in commercial maize breeding programs is the assessment of the value of newly-developed progeny by testing in hybrid combinations. In this study, non-parametric stability measures were applied to analyze the genotype × environment interaction and to assess phenotypic stability of two half-sib maize populations, each consisting of 40 genotypes, across 9 variable environments. Non-parametric tests of significance determined the presence of qualitative interaction for grain yield in both observed populations. Results of the stability analysis showed no significant differences between the two progeny groups indicating that the used testers did not bring significant increase in stability in either of the analyzed half-sib populations. Individual genotypes were also compared based on grain yield stability within both progeny groups using the stability parameters Si(1), Si(2), Si(3) and Si(6). Association between the grain yield and stability indices Si(1) and Si(2) of the analyzed genotypes was presented graphically enabling the identification of genotypes which can be recommended for further breeding process as the most promising ones. The correlations between grain yield and stability parameters were tested by Spearman’s rank correlations. Both progeny groups (HS1 and HS2) showed no significant correlations between the grain yield and stability parameters Si(1) and Si(2), but the rank correlations between Si(1) and Si(2) values were very strong and highly significant. Highly significant negative correlations were found between grain yield and stability indices Si(3) and Si(6) in both progeny groups, and very strong and highly significant correlations were found between Si(3) and Si(6) values. [Project of the Serbian Ministry of Education, Science and Technological Development, Grant no. TR31073: Improvement of maize and sorghum production under stress conditions]