Abstract

We assessed the validity of two gekkonid species, Gekko yakuensis and G. hokouensis, in southern Japan. We first assigned all 398 specimens into 18 samples merely on the basis of localities. By conducting significance test for deviations of genotype frequencies from Hardy–Weinberg at 11 allozyme loci, we checked the reproductive unity of constituents in each of those local samples, and where necessary, rearranged them into subsamples on the basis of genetic markers so that we recognized minimum reproductively cohesive units. We then compared allele frequencies among all samples and subsamples examined. Results clearly indicated that all but two can be classified into two groups that can be discriminated from each other by remarkable allele frequency differences at four diagnostic loci, and by large genetic distances even between sympatric subsamples. Observations of morphological features of the samples and subsamples confirmed that the two groups correspond to G. yakuensis and G. hokouensis, supporting validities of these two species. Allele frequency comparisons, however, also revealed that the remaining two samples, both from southern Kyushu, possessed «marker alleles» of bothG. yakuensis and G. hokouensis at all four diagnostic loci. These samples thus were considered to represent populations that have been derived through hybridization of the two species. Detailed analyses for genetic structures demonstrated that all hybrid genotypes in the two samples are post-F1generations with only one individual resulting from the back-cross with a pure line population of G. yakuensis. This finding negates the possibility that the hybrid populations are maintained by a constant supply of newly produced F1hybrids, but suggests that the hybrid genotypes constitute stable breeding populations. This implies that the genealogical independence of G. yakuensis and G. hokouensis in several other sympatric areas has been maintained by operations of some isolation mechanisms at a pre-mating phase. Investigations of the morphological variation in each sample or subsample revealed that although the two species can be externally largely discriminated from each other by slight modifications of the currently used diagnoses, it is difficult to detect their hybrids based solely on the morphological features.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call