A cross-classification (North Carolina Design II) mating design was used to produce half-sib and full-sib progenies within each of three groups of Iowa Stiff Stalk Synthetic (BSSS) maize (Zea mays L.) lines. The groups of lines represented the resynthesized BSSS population, the current BSSS population maintained at Iowa State University; and the BS13(S)C1 population that has undergone seven cycles of half-sib and one cycle of S1-S2 recurrent selection for improved grain yield. Progenies were evaluated in experiments at three central Iowa locations in 1986 and 1987. Data were collected for grain yield, grain moisture, ear length, ear diameter, days to anthesis, plant height, ear height, root lodging, stalk lodging, and dropped ears;Compared with the current BSSS population, the resynthesized version of BSSS showed mean increases in grain yield, ear length, plant height, ear height, root lodging, stalk lodging, and dropped ears. This was accompanied by mean decreases in grain moisture, ear diameter, and days to anthesis. It seems that the resynthesized BSSS population is more vigorous than the current BSSS population. Recurrent selection resulted in mean increases in grain yield, ear length, days to anthesis, and root lodging as compared to the current BSSS population. There were decreases in grain moisture, ear diameter, plant height, ear height, stalk lodging, and dropped ears;Comparisons of the variance component estimates showed the dominance variance estimates associated with the resynthesized BSSS population were often smaller than for the current BSSS population, particularly for grain yield, ear diameter, and plant height. For most traits, estimates of additive genetic variation were similar for the two populations. Recurrent selection for improved grain yield did not result in a decrease in additive genetic variance for grain yield. The additive genetic variance associated with grain moisture, ear length, plant height, and ear height decreased after selection. All traits exhibited a decrease in the amount of dominance variance;It seems that recurrent selection has increased the frequency of favorable alleles in the improved population resulting in decreases in additive genetic and dominance variation for most traits. However, the additive genetic variance associated with grain yield did not decrease.