A study was conducted in barley (Hordeum vulgare L.) to compare the relative magnitudes of heterosis to additive × additive epistasis and to compare F2 and F∞, diallel analyses. Both F2 and F∞, progenies were derived from 7 × 7 diallel crosses. Progenies and their parents were evaluated for grain yield, heading date, plant height, and the number of spikes per hill in hill plots with five replications at Elora (Ontario) in 1978. Results suggested that additive × additive epistasis were present for these traits and its magnitude was similar to that of heterosis estimated in F2. Both F2 and F∞ analyses detected the presence of epistasis. Both analyses provided similar estimates of the additive genetic variance for heading date and the number of spikes per hill, but the F2 analysis provided higher estimates than the F∞ analysis for grain yield and plant height. The estimate for grain yield and plant height obtained from the F2 analysis could be biased upward because of the invalid assumption of no epistasis. Estimates of other genetic variance components from the F2 analysis could be biased also. The F∞ diallel analysis not only provided estimates of additive × additive genetic variance for the four traits, it also allowed detection of nonindependent gene distribution in the parents for three of the four traits. Therefore, the limitations of the F2 diallel analysis in the presence of epistasis were apparent in the study. The F2 diallel analysis, however, could be used to detect dominance and maternal effects and thus to complement the F∞ diallel analysisKey words: barley, Hordeum vulgare, diallels, haploids, epistasis, heterosis.
Diallel analysis of scald and net blotch resistance in barley (Hordeum vulgare L.)