Estimates of additive, dominance and epistatic genetic variances from a clonally replicated test of loblolly pine

Abstract

Nine full-sib families were generated using a factorial mating design consisting of three female and three male loblolly pine (Pinus taeda L.) parents. Full-sib seedlings and clones of the same families were planted in two test sites in Alabama and Florida. Additive, dominance, and epistatic genetic variances were estimated for growth traits and for fusiform rust incidence for ages 1 through 6. Epistatic variances did not have a significant role in growth traits, but additive gene actions were the major source of genetic variance in loblolly pine. Dominance variance for height, diameter, and volume was negligible at early ages, but it was considerable at age 6, particularly for volume. Fusiform rust incidence appeared to be partially under additive and epistatic gene actions and genetic differences among the clones within families accounted for 48.6% of the total variance. Within-plot variance for clones was always smaller than that within-plot variance for seedlings of the same full-sib families. Clonally replicated progeny tests may provide special advantages for loblolly pine tree improvement programs, as they would substantially increase the efficiency of testing by reducing the microenvironmental variance and better estimation of genetic parameters. This may provide greater genetic gains, particularly for fusiform rust incidence. Potential effects of the small parental size and violations of the assumptions of the genetic model are discussed. Negative additive and nonadditive genetic correlations between the growth traits and fusiform rust incidence are encouraging for rapid and simultaneous improvement of the traits during the same cycle selection.