Estratégia para seleção de progênies de milho e validação de grupos heteróticos

Abstract

This work was divided into two topics. The first regards the selection strategies of corn progenies, and had the objective of verifying which strategies applied to the S 0:3 generation resulted in better performing hybrids in the next generation. We crossed 13 progenies, from generations S0:3 and S 0:4, in a complete diallel scheme. The hybrids and progenies of generation S 0:3 were evaluated for grain productivity in two environments. In another two environments, we evaluated the hybrids of generation S 0:4. Posteriorly, we conducted analysis of variance on the progenies from generation S 0:3 and on hybrids of generations S 0:3 and S 0:4. The diallel analysis was performed considering the complete diallel with the progenies in generation S 0:3, also considering all possibilities of partial diallels, in which one of the groups was comprised of only two progenies. Finally, we estimated the selection differential in generation S0:4, considering the selection of two better performing progenies in generation S0:3, by using the strategies: per se evaluation of the progenies; hybrid performance; general combination ability (GCA), obtained by means of complete diallel; and GCA, obtained from the partial diallels. Among the evaluated strategies, GCA presented the best performance for obtaining corn hybrids. The second topic regards the allocation of corn progenies in heterotic groups, and had the objective of validating the efficiency of two simple cross hybrids, with high estimate of specific combination ability (SCA), in classifying progenies into heterotic groups. We divided 13 progenies into two groups, according to their performance in hybrid combinations, with two testers (simple cross hybrids). The 13 progenies were crossed in a complete diallel scheme. The hybrids generated were evaluated for grain production, and the results submitted to analysis of variance and diallel analysis. The progenies were genotyped and, with the results obtained with the markers, the genetic distances between progenies were estimated. Aiming at confirming the groups formed by means of the testers, another two groupings was performed. One group was formed using the genetic dissimilarity obtained by the markers, and the other, using the SCA estimates obtained by the diallel analysis. We verified that the simple hybrids were not efficient in classifying the progenies into heterotic groups.