Estimates for genetic variance components in reciprocal recurrent selection in populations derived from maize single-cross hybrids.

Matheus Costa Dos Reis,João Cândido De Souza,Rodrigo Vieira Balbi,José Maria Villela Pádua,Fernando Lisboa Guedes,Guilherme Barbosa Abreu

doi:10.1155/2014/540152

Abstract

This study was carried out to obtain the estimates of genetic variance and covariance components related to intra- and interpopulation in the original populations (C0) and in the third cycle (C3) of reciprocal recurrent selection (RRS) which allows breeders to define the best breeding strategy. For that purpose, the half-sib progenies of intrapopulation (P11 and P22) and interpopulation (P12 and P21) from populations 1 and 2 derived from single-cross hybrids in the 0 and 3 cycles of the reciprocal recurrent selection program were used. The intra- and interpopulation progenies were evaluated in a 10 × 10 triple lattice design in two separate locations. The data for unhusked ear weight (ear weight without husk) and plant height were collected. All genetic variance and covariance components were estimated from the expected mean squares. The breakdown of additive variance into intrapopulation and interpopulation additive deviations (σ τ 2) and the covariance between these and their intrapopulation additive effects (CovAτ) found predominance of the dominance effect for unhusked ear weight. Plant height for these components shows that the intrapopulation additive effect explains most of the variation. Estimates for intrapopulation and interpopulation additive genetic variances confirm that populations derived from single-cross hybrids have potential for recurrent selection programs.

Highlights

Reciprocal recurrent selection (RRS) was originally proposed by Comstock et al [1] to improve the hybrid between two populations to exploit additive and nonadditive effects
There was no significant difference (P < 0.05) only for MS22in population 2 (C0) (Table 2). These results confirm the presence of significant variance among the progenies of both populations and indicate that the parent that gave rise to them must have a large number of heterozygous loci
The precision of the experiments measured by the coefficient of variation (CV) indicates that all evaluations have good precision

Summary

Introduction

Reciprocal recurrent selection (RRS) was originally proposed by Comstock et al [1] to improve the hybrid between two populations to exploit additive and nonadditive effects. Junior [18] proposed new genetic variance and covariance components, as well as a new method called testcross half-sib selection (THS). This author suggested obtaining estimates for such components from the different recurrent selection methods. The present study was carried out in order to obtain estimates for genetic variance and covariance components of the original populations (C0) and third cycle (C3) of an RRS program

Materials and Methods

Results and Discussion

Conclusions