On the average coefficient of dominance of deleterious spontaneous mutations

T. Mukai and co-workers in the late 1960s and O. Ohnishi in the 1970s carried out a series of experiments to obtain direct estimates of the average coefficient of dominance ((h) over bar) of minor viability mutations in Drosophila melanogaster. The results of these experiments, although inconsistent, have been interpreted as indicating slight recessivity of deleterious mutations, with (h) over bar approximate to 0.4. Mukai obtained conflicting results depending on the type of heterozygotes used, some estimates suggesting overdominance and others partial dominance. Ohnishi's estimates, based on the ratio of heterozygous to homozygous viability declines, were more consistent, pointing to the above value, However, we have reanalyzed Ohnishi's data, estimating h by the regression method, and obtained a much smaller estimate of similar to 0.1. This significant difference can be due partly to the different weighting implicit in the estimates, but we suggest that this is not the only explanation. We propose as a plausible hypothesis that a putative nonmutational decline in viability occurring in the first half of Ohnishi's experiment (affecting both homozygotes and heterozygous) has biased upward the estimates from the ratio, while it would not bias the regression estimates. This hypothesis also explains the very high (h) over bar approximate to 0.7 observed in Ohnishi's high-viability chromosomes. By constructing a model of spontaneous mutations using parameters in tho literature, we investigate the above possibility. The results indicate that a model of few mutations with moderately large effects and (h) over bar approximate to 0.2 is able to explain the observed estimates and the distributions of homozygous and heterozygous viabilities. Accounting for an expression of mutations in genotypes with the balancer chromosome qi does not alter thr conclusions qualitatively.