---

Breeding of adapted and variation germplasm can be a main element for strength of research political in stable agricultture system. This is achived by targeting variety selection onto different growing environments under natural heat and drought stresses. To realize this, breeding programs usually undertake a rigorous genotypes performance evaluation across locations and years mostly at the final stage of variety development process. More accurate selection of wheat genotypes requires reducing environmental effects for explaining of their genetic potential with appropriate analysis of genotype × environment. In this research, 18 improved bread wheat genotypes were planted in randomized compelet block design with 4 replications in Gachsaran, Khoramabad, Gonbad, Moghan and Ilam stations during three years (2011-2014). Simple and combined analysis variance were done on grain yield data. Due to significant interaction effects for Year × locatin and genotype × year × locatin, yield stability of considered genotypes were analysed using some parameteric and nonparametric methods. Obtained results showed genotypes no. 10 and 17 with 3107 and 3028 kg/ha had the highest grain yield. Based on parametric statistics: CV_i, , , ،, MS_Y/L and MSPI and non-paramertric statistics: , , , Top, Mid, Low،, R-Sum, and SDR, G10 had more yield stability. More over, G12 and G17 showed suitable yiled stability among high production genotypes. Earliness Genotypes 10 and 17 lines had optimum plant height and partially high thousand kernel weight.