Premise of the study: We posed two hypotheses for broad scenarios of postglacial recolonization of Korea by the warm-temperate vegetation: (1) that extant Korean populations are derived from a single refugium, or (2) that they are derived from multiple refugia. We chose a homosporous fern typical of East Asian warm-temperate vegetation, Selliguea hastata, to test which of the two scenarios is more likely and to check whether Japan contained putative glacial refugia.
Methods: Using 16 allozyme loci, we obtained genotypes of 756 individuals from 20 populations, representative of the whole distribution area in Korea (including Jeju Island), Japan, and Taiwan. We assessed genetic variability within and among populations, Wright's F-statistics, and conducted analysis of molecular variance, model-based Bayesian clustering, and bottleneck tests.
Key results: We found no allozyme variation within populations of S. hastata in mainland Korea, whereas genetic polymorphism was detected for populations from Jeju Island, Japan (in particular a population from southeastern Shikoku), and Taiwan. The levels of inbreeding within populations were high, consistent with the potential of S. hastata for intragametophytic selfing.
Conclusions: Data on allelic richness together with Bayesian clustering methods suggest a pattern of postglacial recolonization of mainland Korea from a single refugium, probably located either on Jeju Island or in Japan. Jeju Island should merit the highest priority for conservation biogeography, as it played a role as a Quaternary refugium for arctic-alpine, boreal, temperate as well as warm-temperate plants, as suggested here.
Keywords: Jeju Island; Polypodiaceae; Selliguea hastata; allozyme diversity; conservation; glacial refugium; homosporous fern; re-colonization history.