Yield stability evaluation of bread wheat promising lines using multivariate methods

To determine yield stability of 16 bread wheat promising lines along with two commercial check cultivars,Chamran and Aflak, were studied in six experimental field stations for two successive cropping seasons (2012-2014). The experiments were conducted using randomized complete block design (RCBD) with four replications. The analysis of variance showed that 80.9, 3.2 and 15.9 percent of total variation were related to the environment (E), genotype (G) and G×E interaction effects, respectively. The results of additive main effects and multiplicative interaction (AMMI) model showed that the first four principal componentsof AMMI were significant and described 78.4% of the variance of G×E interaction. The AMMI1 model analysis indicated that S-91-9, S-91-14, S-91-13 and S-91-15, with an average grain yield higher than total mean, were the most stable lines. Based on AMMI2 model, S-91-13 was the most stable line. The shifted multiplicative model (SHMM), classified the environments in three groups. The first group consisted of Darab (first and second years), Iranshahr (first year), Khorramabad (first and second years) and Dezful (first year), while Dezful (second year), Ahwaz (first and second years), Zabul (second year) and Iranshahr (second year) were classified in the second group andZabul (first year)was only into the third group.The presence of Darab and Khorramabad in the same group indicates the relative similarity of these stations and existence of an additive or non-crossover interaction. Finally, the S-91-13 line with high yield and broad adaptability was selected as superior line for further investigation to introduce the new commercial bread wheat cultivar.