- Y Higashiura: School of Life Sciences, Tokyo University of Pharmacy and Life Sciences, Tokyo, Japan. yasu@ls.toyaku.ac.jp
We explored the origin of all-female broods resulting from male death in a Hokkaido population of Lymantria dispar through genetic crosses based on the earlier experiments done by Goldschmidt and by testing for the presence of endosymbionts that are known to cause male killing in some insect species. The mitochondrial DNA haplotypes of the all-female broods in Hokkaido were different from those of normal Hokkaido females and were the same as those widely distributed in Asia, including Tokyo (TK). Goldschmidt obtained all-female broods through backcrossing, that is, F1 females obtained by a cross between TK females (L. dispar japonica) and Hokkaido males (L. dispar praeterea) mated with Hokkaido males. He also obtained all-male broods by mating Hokkaido females with TK males. Goldschmidt inferred that female- and male-determining factors were weakest in the Hokkaido subspecies and stronger in the Honshu (TK) subspecies. According to his theory, the females of all-female broods mated with Honshu males should produce normal sex-ratio broods, whereas weaker Hokkaido sexes would be expected to disappear in F1 or F2 generations after crossing with the Honshu subspecies. We confirmed both of Goldschmidt's results: in the case of all-female broods mated with Honshu males, normal sex-ratio broods were produced, but we obtained only all-female broods in the Goldschmidt backcross and obtained an all-male brood in the F1 generation of a Hokkaido female crossed with a TK male. We found no endosymbionts in all-female broods by 4,'6-diamidino-2-phenylindole (DAPI) staining. Therefore, the all-female broods observed in L. dispar are caused by some incompatibilities between Honshu and Hokkaido subspecies.