Biometric approach applied to soybean genotypes cultivated in Rio Grande do Sul, Brazil

The study had the purpose to evidence agronomic performance, inter-relations of characters and multivariate differentiation soybean genotypes cultivated in preferential season, state Rio Grande do Sul, Brazil. In crops season 2017/2018, experimental design was completely randomized blocks, being 25 with three replicates. data obtained submitted presuppositions based on normality homogeneity residual variances, variance analysis, Tocher method, Euclidian algorithm, linear correlations, relative contribution by Singh artificial neural networks. performance presents superior seeds yield per plant through elevated magnitude reproductive nodes, legumes plant. height is positively associated number total nodes main stem branches, where they are directly linked productive potential. most polymorphic correspond mass thousand seeds, possible differentiate a way though similarity profiles.