ADAPTABILIDADE E ESTABILIDADE FENOTÍPICA DE GENÓTIPOS DE SOJA ESTIMADAS VIA MODELAGEM MISTA

Abstract

The objective of this study was to evaluate the genotype x environment (G x A), adaptability and stability interaction of 46 soybean genotypes in eight environments, using the mixed model methodology, and using Selegen REML / BLUP statistical software. The experiments were evaluated during the 2017/18 harvest season in Rio Verde 1-GO, Rio Verde 2-GO, Goiatuba - GO, Rondonópolis - MT, Paraúna - GO, Alto Garças - MT, Jataí - GO and Montividiu - GO . The following traits were evaluated: grain yield (PROD) and one hundred grain mass (MCG). The joint selection for the evaluated characteristic, stability and adaptability was performed by the HMRPGV statistic. The heritability obtained in this study is considered average for yield (0.557) and low for one hundred grain mass (0.247). The low value of genotypic correlation between environments indicates the low predictability of genotype responses. For productivity, the sites of Rio Verde 1-GO (4521.44 kg.ha-1), Paraúna-GO (4397.14 kg.ha-1), Montividiu-GO (4376.43 kg.ha-1), Rondonópolis -MT (4364.89 kg.ha-1) and Rio Verde 2-GO (4046.97 kg.ha-1) obtained the best averages. For mass of one hundred grains the same environments were repeated, however, with different ranking. Genotype 46 was the one that obtained the best yield for productivity. Genotype 35 was the one that obtained the best yield for mass of one hundred grains. However, genotype 31 is the only one among the top 5 genotypes in the genetic gain and stability and adaptability evaluations for the yield and one hundred grain mass variables.