Phylogeography of Ophiorrhiza japonica (Rubiaceae) in continental islands,the Ryukyu Archipelago,Japan

Aim Phylogeographical patterns in the Ryukyu Archipelago have been explained primarily by landbridge formation and the opening of two straits in the Pliocene, namely the Tokara and Kerama gaps. These old straits have been considered to be the barriers most likely to determine genetic boundaries. To test this, we conducted a molecular analysis of the herb Ophiorrhiza japonica. We discuss the causes of and processes involved in its phylogeographical structure and explore aspects of island separation other than the duration of the straits to explain genetic boundaries at the gaps. Location Ryukyu Archipelago, Japan. Methods Plants were collected from 40 localities in the archipelago and vicinity. Non‐coding regions of chloroplast DNA were sequenced. The genealogical relationships among haplotypes were estimated using a statistical parsimony network. To examine the phylogeographical structure, we compared two parameters of population differentiation, namely G_ST and N_ST, and conducted correlation analysis of genetic and geographical distances. Genetic boundaries were identified using Monmonier’s maximum difference algorithm. To test vicariance–dispersal hypotheses, that is, vicariance after migration via the Pliocene landbridge or over‐sea dispersal in the Pleistocene, molecular dating analysis was conducted. Results A statistical parsimony network revealed that the haplotypes from the Ryukyu Archipelago and northwards coalesce to one ancestral haplotype in Taiwan. A clear phylogeographical structure was observed: plants within the same population and populations in geographical proximity were phylogenetically close. A genetic boundary was recognized across the Kerama Gap, but not across the Tokara Gap. Dating analysis suggested that population divergence across the Kerama Gap occurred in the early to late Pleistocene. Main conclusions The statistical parsimony network suggests migration from Taiwan and northward range expansion in the archipelago. Based on the divergence time, over‐sea dispersal in the Pleistocene is likely, although migration via a Pliocene landbridge is not totally rejected. Negligible genetic differentiation across the Tokara Gap suggests recent over‐sea dispersal, possibly facilitated by the small geographical width of the gap. Conversely, the large genetic differentiation across the Kerama Gap is probably explained by the large geographical distance across it. The past splitting of a landbridge would have had a significant influence on population differentiation after a certain geographical distance was reached.