Evaluation of durum wheat genotypes using parametric and nonparametric stability statistics

Abstract

The development of genotypes, which can be adapted to a wide range of environments, is the one of the most important goal of plant breeders in a crop improvement program. In this study, 6 six stability measures consisting of 4 parametric and 2 nonparametric were used to evaluate the genotype by environment interaction (GEI) in 20 durum wheat genotypes. The genotypes were evaluated for grain yield at fourteen environments in the Central Anatolian Region of Turkey for two years. The experimental layout was a randomized complete block design with three replications. Genotypes, environments main effects and GEI were significant at P < 0.01. Both parametric (bi, S2 di, Ri2, Pi) and nonparametric (Si(1), Si(2)) univariate stability statistics were used to determine stability of the durum wheat genotypes. Genotypes 20, 13 and 12 were most stables based on genotypes according to six stability measures. The level of associations among the stability measures was assessed using Spearman's rank correlation. Regression coefficient (bi) was negatively and significantly correlated (P < 0.01) with superiority index (Pi). On the other hand, Si(1), Si(2) and S2 di were positively and significantly correlated with Pi. As a result, these relationships reveal that only one of them could be sufficient to select genotypes of interest in a durum wheat breeding program.