Evaluation of soybean genetic backgrounds under different growing environments of Ethiopia is necessary to identify the most important traits in soybean genetic improvement. The present study were aimed to evaluate and identify the best performing soybean variety (s) for yield, oil content and estimate the association of yield, oil and other agronomic traits. Nineteen released soybean varities were evaluated in randomized complete block design with three replications in 2015/16 main cropping season. Analysis of variance showed highly significant differences among varieties. Varieties wello, belesa-95 and TGX-1332644 at Mankush and davis, awassa-95 and gishama at Pawe were the best-performing varieties in terms of grain yield. The highest oil content was recorded for varieties TGX-1332644, belesa-95 and cocker-240 at Pawe and varieties hawassa-04, belesa-95 and gishama at Mankush. The genotypic (rg) and phenotypic (rp) correlation analysis revealed that grain yield had positive and highly significant association with days to flowering (rg=0.63, rp=0.28), plant height (rg=0.73, rp=0.47) and oil content (rg=0.44, rp=0.25). Oil content had a positive and significant phenotypic and genotypic correlation with days to flowering (rg=0.71, rp=0.27), days to maturity (rg=0.48, rp=0.30) and plant height (rg=0.53, rp=0.22). In general, this study showed the existence of significant genetic variation among varieties implying better opportunity for further improvement through selection. Days to flowering, plant height, and oil content could be used as target traits to improve soybean yield because of their significant phenotypic and genotypic correlations with grain yield.

Keywords: Correlation; Grain yield; Oil content; Soybean; Variety