The quantitative genetics of a Non-Stiff-Stalk maize (Zea mays L.) population.

摘要

The genetic relationship among individuals is at the core of nearly all quantitative genetic theory. Dominant gene action has long been either ignored or disregarded as insignificant in many previous genetic models. For grain yield in maize (Zea mays L.), dominance has consistently accounted for a large proportion of genetic variance. We have used previously developed genetic theory that accounts for dominance variance during inbreeding and applied it to a unique breeding design. Our breeding design allowed us to estimate five genetic covariance parameters for six traits. In addition, we developed genetic gain equations that accounted for both dominance and inbreeding. We found that the genetic covariance parameters introduced via inbreeding were significant for five traits. Our estimates of the genetic covariance parameters allowed us to predict genetic gain over a range of selection units and response units. Half-sib selection proved superior to inbred progeny selection when the response was measured in the outbred progeny. In addition, the relative proportions of additive and dominance variance influenced the effectiveness of inbred progeny selection. We also showed that even when dominance constitutes a larger proportion of the total genetic variance than additive variance, the loss of additive effects has a greater influence on the decline associated with inbreeding than the addition of homozygous dominance deviations. Our results also indicated that the reason realized gain often falls short of predicted gain is due to the negative covariance between additive effects and homozygous dominance effects. The effect of a negative covariance is that positive gain via additive effects is offset by negative gain via homozygous dominance deviations.