In the cross between Drosophila melanogaster females and D. simulans males, hybrid males die at the late larval stage, and the sibling females also die at later stages at high temperatures. Removing the D. simulans allele of the Lethal hybrid rescue gene (Lhr (sim) ) improves the hybrid incompatibility phenotypes. However, the loss-of-function mutation of Lhr (sim) (Lhr (sim0) ) does not rescue the hybrid males in crosses with several D. melanogaster strains. We first describe the genetic factor possessed by the D. melanogaster strains. It has been suggested that removing the D. melanogaster allele of Lhr (Lhr (mel) ), that is Lhr (mel0) , does not have the hybrid male rescue effect, contrasting to Lhr (sim0) . Because the expression level of the Lhr gene is known to be Lhr (sim) > Lhr (mel) in the hybrid, Lhr (mel0) may not lead to enough of a reduction in total Lhr expression. Then, there is a possibility that the D. melanogaster factor changes the expression level to Lhr (sim) < Lhr (mel) . But in fact, the expression level was Lhr (sim) > Lhr (mel) in the hybrid irrespectively of the presence of the factor. At last, we showed that Lhr (mel0) slightly improves the viability of hybrid females, which was not realized previously. All of the present results are consistent with the allelic asymmetry model of the Lhr gene expression in the hybrid.

Theoretical considerations predict that balancer chromosomes in Drosophila melanogaster should accumulate numerous deleterious mutations with time. We counted the number of recessive lethal mutations on two balancer chromosomes from the In(2LR)SM1/In(2LR)Pm strain maintained in our lab, after making the balancers heterozygous with deficiencies from second-chromosome Kyoto Deficiency kit strains. We detected 10 recessive lethal mutations in the balancer In(2LR)Pm, which is consistent with the mutation rate estimated previously. However, we detected only three mutations, a significantly smaller number, in the balancer In(2LR)SM1, although this may be an artifact. In conclusion, we observed genetic decay over an estimable timescale by using balancers with historical records. Thus, balancers of any strain may have accumulated many unidentified recessive lethal mutations.