This study was conducted to investigate the genotype × environment interaction effects on yield and other agronomic traits of wheat. Ten cultivars of wheat were evaluated in 8 environments (combinations of two levels of nitrogen and four irrigation regimes) in growing season of 2006-2007 at the Research Farm of Isfahan University of Technology. The results of combined analysis of variance showed significant differences among the environments for fertile panicles per m 2 , seeds per panicle, 1000-grain weight and grain yield. High genetic differences were found among the cultivars for days to heading, plant height, fertile panicles per m 2 , seeds per panicle, 1000- grain weight, grain yield and harvest index. Pishtaz, Moghan1, Falat, Darab and Khazar cultivars had higher seed yield (9.27, 9.1, 8.6, 8.55, 8.48 t/ha, respectively). Genotypes had different responses to changes of environments and there was significant genotype × environment interaction for all studied traits, except days to heading and plant height. Different methods of stability analysis including Wrick ecovalance, Shukla stability variance, Eberhart and Russel regression, AMMI and path analysis (Tai's method) were applied to investigate the genotype × environment interaction and to identify the stable genotypes. Wrick ecovalance and Shukla stability variance presented similar were more adapted cultivars. However, Khazar and Alvand cultivars showed specific adaptation to more rich environments and Falat, Darab and Roshan had specific adaptation to poor environments. Analysis of additive main effects (analysis of variance) and multiplicative interaction effects (principal component analysis) revealed that the effects of genotype, environment and genotype × environment interaction were highly significant and the first three principal components explained 89% of sum of squares for genotype × environment interaction. The biplot of the first interaction principal component (IPC 1) with each of mean yields of genotypes and environments, and also the results of cluster analysis based on scores of the first interaction principal component showed that Pishtaz, Moghan1 and Khazar cultivars had lower IPC 1 , and Pishtaz and Moghan1 with higher seed yields were recognized as the most stable cultivars, however, Alvand and Sabalan cultivars were unstable. Also, based on the biplot of IPC 2 and yield means of genotypes, Pishtaz cultivar with the lowest IPC 2 and the highest mean yield was the most stable cultivar. Cluster analysis based on stability parameters of AMMI 3 model (SIPC 3 and EV 3 ) and mean yield of genotypes showed