Relation among full-sib progenies of maize (Zea mays L.) for different generations of inbreeding

Abstract

<p>Reciprocal full-sib selection is based on testing of S[subscript]0 plant crosses. Because there are some differences of opinion on the value of early testing, objective of this study was to determine the relation between different generations of testing. In the cycle five (C5) of reciprocal full-sib selection, full-sib progenies for five generations of inbreeding (S[subscript]0, S[subscript]1, S[subscript]2, S[subscript]3, S[subscript]4) were evaluated. Analyses of crosses among generations were conducted to determine the relative performances at five levels of inbreeding. Specific objectives of this study were to: (1) estimate the variability among full-sib crosses for each generation of inbreeding to determine changes in variability among full-sib progenies with the advances of inbreeding; and(2) estimate the genotypic correlations between full-sib progenies of different generations of inbreeding to determine if a trend occurs with increased generations of inbreeding of lines used to produce full-sib progenies. The plant materials used were derived from the two 2-eared maize (Zea mays L.) populations, 'Iowa Two-ear Synthetic' (BS10) and 'Pioneer Two-ear Composite' (BS11). One pair of crosses for each of 136 crosses were successful in each of the five generations of inbreeding for testing in three Iowa environments;Reciprocal full-sib selection identified crosses that had yields that were either equal to or superior to three single-cross and three double-cross hybrids included as checks. Average grain yield for the population cross was 9.3 q/ha higher than the average for the three single-cross hybrids and 11.5 q/ha higher than the average for the three double-cross hybrids. The estimates of genetic coefficient of variation (GCV) for grain yield increased with the increased generations of inbreeding; the GCV for the S[subscript]3 x S[subscript]3 generation was nearly twice the estimate for the S[subscript]0 x S[subscript]0 generation. The estimates of genotypic correlations for adjacent generations agreed with the theoretically expected correlations for adjacent generations. Higher correlations between generations were obtained for the adjacent generations for all traits studied. The highest genetic correlation (R[subscript] g = 0.87) for yield was between the S[subscript]0 x S[subscript]0 and S[subscript]1 x S[subscript]1 generations. Heterosis increased with selection, and the method of selection and testing may be used directly for commercial purposes.</p>