The use of single-plant selection via marker-assisted selection (HAS) has recently been discussed in the literature (Lande and Thompson, 1990). For this reason, six methods of heritability estimation and number of effective factors responsible for a trait (n) were computed and compared for 30 crosses using generation variances. The crosses were created in two six-parent diallels, one containing older inbreds (pre-1960), the other containing newer inbreds (post-1970). Five generations of each cross were evaluated on a single-plant basis in three replications in four environments;Methods of estimating heritability and n differed in some instances, depending on the trait and specific cross. Most differences were detected between broad- and narrow-sense heritability estimates, which would be expected due to the presence of non-additive genetic variance. Lack of significant differences between methods of estimating n were due to, in part, the large confidence intervals associated with the estimates. No changes in narrow-sense heritability estimates were detected comparing older line crosses with newer line crosses, indicating neither a gain or loss in additive genetic variance due to selection over time. A significant loss in effective factors responsible for kernel depth due to selection over time was detected by Warner's method. No change in n was detected from older line crosses to newer line crosses for the remaining traits. Wright's formula, (and variations of Wright's formula) for estimating n seems inappropriate for detecting n in an F2 population of an elite by elite cross. The requirements of parent extremes for the trait of interest (yield, for example) would not be practical in a commercial breeding program. Assumptions for use of the formula are strict and often impractical, and errors are large;Reference. Lande, R., and R. Thompson. 1990. Efficiency of marker-assisted selection in the improvement of quantitative traits. Genetics 124:743-756.