Identification of adaptable, stable and high yielding genotypes under varying environmental conditions prior to release as a cultivar is the first and foremost steps for plant breeding and this has direct bearing on the adoption of the variety, its productivity and total production of the crop. A total of 22 sorghum varieties were evaluated at five locations (Bilo Boshe, Bako, Gute, Mechara and Miesso) in 2017 cropping season using randomized complete block design with the objectives of determining the magnitude and nature of Genotype by Environment interaction for grain yield and related traits, and to identify stable high yielding sorghum varieties for wider and/or specific environments. The combined analysis of variance revealed the significant effect of locations and genotype by location interactions on all the measured traits. This showed that genotypes were inconsistent for grain yield across the testing locations. Birmash was the highest yielder variety with average yield of 3.515ton/ha with better performance across locations. Baji was the second high yielder variety with mean grain yield of 3.344 ton/ha. Eberhart and Russell regression model, AMMI stability value and GGE biplot models revealed that, Emahoy was the most stable variety followed by Baji and Birmash. The first two IPCAs accounted for a total of 88.64% of the interaction sum square. In both the genotype x environment interaction ANOVA, the results indicate that the observed yield variations among varieties were due to the GxE effects rather than main effect of genotypes and environments. Results of ASV parameter showed that six most stable and high yielder genotypes are Gambella_1107, Gobiye, Baji, ESH_1, IS9302 and Emahoy. Emahoy variety is the 3rd top yielder and the most stable variety selected with both stability parameters as well as high yield. Therefore, it is the promising variety from all tested materials.

Keywords: AMMI model, Correlation, GxE Interaction, Sorghum, Stabili