Ganho de seleção em estabilidade de produção de variedades de milho avaliadas em ensaios de valor de cultivo e uso

Abstract

Estimation of genetic progress is of utmost importance in breeding programs to guide the breeder on the efficiency of the selection methods employed for the characteristics of interest in the populations under selection. The present study aimed to estimate the genetic progress in production stability of maize varieties evaluated in Cultivation and Use Value (VCU) trials coordinated by Embrapa Maize and Sorghum in the years 2013/14 and 2014/15 in 46 environments. In total, 47 genotypes were evaluated, consisting of 33 open pollinated varieties, 1 single hybrid, 1 triple hybrid, 1 double hybrid, 3 intervarietal hybrids and 8 hybrid topcrosses, where single, triple and double hybrids were used as controls. The experimental design was a randomized complete block with two replicates, and was evaluated grain yield. Statistical analyzes were performed using the REML / BLUP procedure, where the variance components were estimated by the maximum restricted likelihood (REML), and the genotypic values were predicted by the best non-vitiated linear prediction (BLUP). The Blups were later submitted to the decomposition of singular values to obtain the GGE-Biplot plots. Genetic progress was estimated through Euclidean distances from genotypes to origin. The renewal rate of the cultivars was 36.17%. The genetic introduction and advance gain estimated by the AMMI method was 0.047 and -0.158 respectively, corresponding to a percentage of 6.78% and -24%. When estimated by the GGE method it was observed an introduction and advance genetic progress of -0.657 and 0.114 respectively, representing a percentage of 46% and 5.47% respectively, concluding that the most productive genotypes were less stable. The best performing genotypes in terms of stability were G116 (Sint 10723), G62 (Synthetic 256 L), G108 (Sint 10697), G131 (AL 2010), and G144 (AL2014).