Three methods of estimating dominance were applied to F1 data on plant height, panicle number, heading date, number of spikelets per panicle, number of panicles per plant, and grain yield from the 15 diallel crosses among 6 oat varieties. Genotypic correlations and heritability percentages, in the broad sense, were calculated for the F2.In general, the dominance values from the 3 methods of estimation were in good agreement with one another and with the observed effects. Overdominance was indicated for number of panicles per plant in the F1 whereas estimates for yield ranged from partial to overdominance. Partial dominance was indicated for short panicles and plant height. Earliness of heading was partially dominant in both the F1 and F2 and few spikelets per panicle were partially dominant.The heritability percentages were 33, 53, 54, 61, 74, and 87 for number of panicles per plant, grain yield, panicle length, plant height, number of spikelets per panicle, and heading date, respectively. The genetic correlations between plant height, panicle length, number of spikelets, heading date, and grain yield were positive, and most of them were relatively high.The heritability percentages and high levels of dominance for yield and panicles per plant observed in the F1 indicate that selection for these attributes should be delayed until later generations. Selection for plant height, panicle length, number of spikelets and heading date should be feasible in early generations.