Medidas de associação de variáveis com massa de grãos para cultivares de soja em terras baixas

Abstract

The objective of this work was to study the linear association measures between characters of soybeans cultivated in lowland, based on genotypic and phenotypic correlations and path analysis in different scenarios of explanatory variables. The experiment was conducted at the Federal University of Pampa, Campus Itaqui-RS (29º 09' 21" S and 56º 33' 02" W and 74 m altitude), in the 2014/2015 harvest. The eight treatments were arranged in a randomized complete block design, with three replications, in plots of 11 m², each treatment represented by one of the following cultivars: DM6563; DM7166; BMX POWER; DON MARIO 7.0i; BMX ACTIVE; DM6160; DM5958 and DM6458. We evaluated 100 plants of each plot and,in each plant were measured the following characters: plant height (AP), height of insertion of the first pod (AIPV), number of branches (NR), number of nodes (NN), number of pods (NV), number of grains (NG), number of grains per pod (NGV), plant mass at harvest (MP), pod mass (MV), mass of one hundred grains (MCG) and grain mass per plant (MG). For each variable, the analysis of variance was performed in the randomized block design to verify the effect of cultivar and the subsequent comparison of means, being also calculated the experimental precision, heritability and selective accuracy. The genotypic and phenotypic correlation coefficients between the variables were calculated and the analysis of ridge and traditional path analysis was performed. According to both the genotypic and phenotypic correlation matrix, the variables MV, MP, NG and NN presented the highest correlations with grain mass. Regardless of the correlation matrix used in ridge path analysis (genotype or phenotype), MV, MP and MCG are the variables with the greatest cause and effect relationships on MG and can be used for the indirect selection of more productive plants. In the traditional path analysis, the indirect selection can be performed based on the plant mass.