Stability analysis and nutritional quality of soybean (Glycine max (L).Merrill.) genotypes for feed in southwestern...

Soybean is a leguminous crop known for its multiple utilizations both as food and feed for humans and livestock. The objectives of the study were to identify high dry matter yielder and stable genotypes across environments in southwestern Ethiopia. The effect of genotype environment (G x E) interaction on dry matter yield of soybean genotypes were evaluated in two cropping seasons (2019-2020) under rain fed condition. Eight pre tested soybean genotypes with two checks were used as treatment in a randomized complete block design with three replications. Collected data were recorded and analyzed using GGE biplot models using R software. The combined analysis of variance showed that dry matter yield of soybean genotypes was significantly affected by genotype, environment and genotype-environment (G x E) interaction. The genotype, environment, and genotype-environment interaction, respectively, accounted for 11.4%, 49.5%, and 38.8% of the observed variation to the dry mater yield. This indicates that dry matter yield was significantly more affected by environments and G × E interaction than genotypes. The GGE biplot analysis revealed that six environments used in the current study were grouped into four mega-environments. The mega-environments were identified for genotype evaluation. The biplot showed that the vertex genotypes were G4, G10, and G9 and considered as optimum performance in their respective mega-environments and more responsive to environmental changes. The biplot also showed that ENV5 (Kersa 2020) was an ideal and the most discriminating and representative environment. Genotype G4 (TGX1990-114FN) was the ideal genotype and overall winner in dry matter yield and stability in the findings. Therefore, genotype G4 (TGX-1990-114FN) is the better option to be used as forage soybean in Ethiopia. Further demonstration of the feeding values of high yielders and stable genotypes on animal performances should be done.