Assessment of non-parametric methods in selection of durum wheat (Triticum turgidum L. var. durum) stable genotypes

To evaluate of non-parametric methods for identification of high yielding and stable durum wheat genotypes, 18 durum wheat promising lines along with two commercial durum (Behrang) and bread wheat (Chamran) as check cultivars, were evaluated in four warm and dry locations of Iran including Darab, Ahvaz, Khoramabad and Dezfoul stations during two cropping seasons (2013-2014). The experiments were conducted in a randomized complete block design with three replications. Results showed that the effect of environment (E), genotype(G) and interaction of GE were significant in Bredenkamp (except for genotype) and Hildebrand for non-crossover interaction (without change in rank) and de Kroon/van der Laan (except for interaction of GE) for non-crossover interaction (with change in rank) methods. Results of SIIG (Selection Index of Ideal Genotype) method, based on all non-parametric methods, showed that DW-92-17 and DW-92-20 genotypes (with 6095 and 6215 kg.ha-1 grain yield) were the most stable genotypes. According to SIIG method, the majority of non-parametric methods may identify the stable genotypes considering the static (biological) concept, so the selected genotypes may not necessarily produce high grain yield. Genotypes DW-92-17 and DW-92-20 were selected as stable lines with higher yields (compared to control) and DW-92-4 and DW-92-5 genotypes were selected as high yielding lines for complementary experiments, especially for resistance to diseases and quality characteristics evaluation