A large number of soybean (Glycine max L.) genotypes of diverse growth habit and adaptive characters were used in the experiment. Soil salinity-induced changes in nine morpho-physiological characters of 30-day-old seedlings of 170 soybean genotypes were compared in the study. The first and second principal components (PC) of principal component analysis (PCA) results accounted for 97 and 2.5%, respectively, of the total variations of soybean genotypes. The variation for the first PC was composed mainly of relative total dry weight (DW), relative shoot dry weight, as well as petiole dry weight. There were four clusters distinguished in the cluster analysis. The genotypes in cluster IV performed better in respect to relative total dry weight and relative shoot dry weight and hence having salt tolerance. The genotypes clusters III performed very poorly and those of clusters II and I were moderate to poor. $D^2$ analysis indicated that the clusters differed significantly from each other. Discriminant function analysis (DFA) again asserts strongly that more than 92% of the genotypes were correctly assigned to clusters. Both PCA and DFA confirmed that the relative total DW followed by shoot and petiole DW were the major discriminatory variables, and the root DW were the secondary important variables to distinguish genotypes into groups. In this study, multivariate analyses were used in identifying the soybean genotypes of desirable traits for salt tolerance.