Evaluation of wheat genotypes under different environments is essential for testing stability of their performance and range of adaptations. Where, development of any crop genotypes with adaptation to changes is one of most important goal of breeding program. This study examined fifteen durum wheat accessions over different eight environments; two planting dates: Nov. 30th (favorable) and Dec. 30th (unfavorable) at two locations: Assiut and Qena during two seasons (2010/2011 and 2011/2012). The experiment was grown in a randomized complete blocks design with three replications in each location. The objective of this investigation was to study the stability parameters of the fifteen durum wheat genotypes under different environments. The combined analysis of variance for environment (E), genotype (G) and (GxE) interaction was highly significant for all studied traits, suggesting differential responses of the genotypes and the need to stability analysis. The stability parameters are useful in characterizing genotypes by showing their relative performance in various environments. Results revealed that highly yielding genotypes can also be stable. The genotype No. 12 had desired performance (grain weight/spike) compared to the grand mean, regression coefficient (bi) did not differ significantly from unity and least deviation from regression (S2d), indicating the role of linear portion of GxE interaction in the performance of this genotype. The value of regression coefficient (bi) of genotypes No. 12 for No. of grains/spike, No. 8 and 13 for grain weight/spike, No. 1, 2, 5, 8, 9, 14 and 15 for grain yield/plant were less than one (bi<1), indicating that these genotypes were considered specially adopted to unfavorable environments. Meanwhile, the value of regression coefficient of genotypes No. 9 for plant height, No. 13 for No. of grains/spike, No. 3,12 and 14 for grain weight/spike and No. 10 for grain yield/plant had bi values more than unity (bi>1) and could be adapted to optimum environment.