Estimation of genetic variances in a maize composite

Abstract

Sixty four randomly selected So plants (used as males) were each mated to four different plants (used as females) to generate 256 full sib families as per NC Design-I and simultaneously selfed to produce 64 S, families for the estimation of genetic components of variance and related parameters. Both full-sib and S, families were evaluated in incomplete block design and randomized block design, respectively in two random environments. Observations were recorded on grain yield per plant, test weight, kernel rows per ear, kernels per row, ear length, ear diameter, plant height, ear height and days to silk. Comparison of the relative magnitude of the additive genetic variance and variance due to dominance deviation revealed that the former was more important for all the traits studied both under NC-Design-I and the one factor design (S1 families) except for kernels per row and ear length In in (NC-Design-I) and days to silk (in both the designs). Variance due to dominance deviations was affected more by the environmental Interaction as compared to additive genetic variance. Estimates of heritability estimates (narrow sense) for grain yield per plant were high In the S, (73.0"10) than Design-I. Expected genetic gain in the reconstituted population Introgressed from the elite families would be more in S1 (29.77"10) than full sib (7.77"10) families.