Estimates of Genetic Variance in Open‐Pollinated Varieties of Cornbelt Corn 1

THE choice of method to be used for the genetic improvement of crop plants is dependent upon the type of gene action involved. In the case of a quantitatively inherited character such as grain yield, only the mass action of all the genes involved can be studied. However, generalizations can be drawn as to the important types of gene action taking place through a study of genetic variation. Total genetic variance can be divided theoretically into additive genetic, dominance, and epistatic variance (Fisher, 3; Wright, 15). The relative magnitude of these three genetic variances is of importance to the plant breeder in order to plan the most effective breeding scheme. Comstock and Robinson (1) presented theoretical derivations together with an experimental procedure applicable to corn which permits the estimation of additive genetic and non-additive genetic variance (including dominance variance) portions. Robinson et al. (9, 10) and Gardner et al. (5) presented results from advanced generations of hybrids from inbred lines of southern prolific origin and in general obtained estimates of appreciable dominance variance relative to additive genetic variance for grain yield. Genetic linkages as well as epistasis were considered to be possible sources of bias in these results. Robinson et al. (11) in studying open-pollinated southern prolific varieties estimated that additive genetic variance constituted the major portion of the total genetic variance for grain yield, with the conclusion that overdominant loci were not the single important source of genetic variability in the material studied. Sprague and Tatum (14) obtained estimates of the variances associated with combining ability for grain yield in corn. In previously selected material, the variance for specific combining ability (related to dominance variance) was found to be larger than the variance for general combining ability (related to additive genetic variance). In unselected