Relationship of Maize Testers Selection and their Genetic Divergences / Relações entre seleção de testadores de milho e suas divergências genéticas

AUTOR(ES)
DATA DE PUBLICAÇÃO

2006

RESUMO

The objectives of this study were to relate the magnitudes of the correlation between testcrosses with the genetic divergences of the testers in order to verify if the genetic similarity of the testers could allow the reduction of testers, and if the level of selection intensity that should be applied is also a function of the genetic similarity of the testers. Five elite testers, previously evaluated in a diallel design, were crossed to 50 inbred lines from different heterotic groups following a factorial mating design, giving rise to 250 testcrosses which were evaluated at 13 environments with two replications per environment in 16 x 16 lattice designs; six commercial hybrids were allocated in the experiments. The traits recorded were: grain yield at 15% grain moisture (GY), silking (SD) and anthesis date (AD), plant (PH) and ear height (EH), plant lodging (PL), prolificacy (PRO), ear length (EL), ear diameter (ED), cob diameter (CD), kernel depth (KD), row number per ear (RN), kernel-row number (KRN), and 500 kernel weight (KW). The testers (inbred lines) were evaluated at four environments with two replications per environment for the same traits following a randomized block design, and they were also genotyped with AFLP markers. The genetic divergence of the testers was computed based on AFLP markers (GD), Mahalanobis distance (MD), and on the specific combining ability (SCA); and heterotic groups was established for these three measures of genetic divergence. The analysis of variance of the factorial model (testcrosses) showed that the general combining ability (GCA) was more important than specific combining ability (SCA) for all traits, but for grain yield the contribution of SCA was almost as important as GCA. Each type of genetic divergence grouped the testers differently, which resulted in different heterotic groups. For grain yield, the magnitudes of Spearman correlation between the correlations of testcrosses with GD and with MD were very low, and were not predictive of any relationship between these measures of genetic divergence and correlation of testcrosses. However, this correlation was negative and high (r=-0.88) with the SCA estimates of the testers, suggesting that the genetic divergence of the testers could predict the correlation between testcrosses; i.e. the higher the genetic similarity of the testers based on SCA the higher the correlation between their testcrosses. For the other traits the correlations between the testcrosses were high, probably because the GCA effects explained a higher proportion of the variation among testcrosses. Thus, these results suggested that the SCA estimates should be used to determine the genetic divergence of the testers accurately, and that the genetic similarity of them could be used to reduce the number of testers to be used when the heterotic groups of a set of lines to be evaluated were not known. For testers from the same heterotic group, the selection intensity should be as high as 30% to assure that the same set of superior testcrosses would be selected irrespective of the tester.

ASSUNTO(S)

seleção genética correlação genética inbred lines genetic divergences linhagens vegetais milho melhoramento genético vegetal correlations testcrosses maize variação genética em plantas

ACESSO AO ARTIGO

Documentos Relacionados