Skip to main content
Dryad

Data from: Allopatric divergence and secondary contact within a single river system in a freshwater fish group

Data files

Feb 03, 2025 version files 31.04 KB

Abstract

Because riverine fishes can disperse along linear river corridors and even between tributaries, allopatric divergence within river systems is limited to large rivers. Within small rivers, such as island rivers, allopatric divergence rarely occurs, and no cases of secondary contact have been reported to date.

Two congeneric riverine ricefishes, Oryzias dopingdopingensis and O. landangiensis, are found in the Doping-doping and Cerekang Rivers, respectively, on Sulawesi Island, which are connected via estuarine waters. These two species are suggested to have diverged in allopatry within this single river system.

Phylogenetic and population genetic structure analyses using genome-wide single nucleotide polymorphisms revealed that Oryzias individuals recently discovered from the downstream Cerekang River are phylogenetically distinct not only from O. dopingdopingensis in the Doping-doping River but also from O. landangiensis in the upstream Cerekang River. Species tree analysis and demographic model inference revealed that the downstream Cerekang Oryzias diverged from upstream Cerekang Oryzias in allopatry approximately 57,000 generations ago, and they came into secondary contact with each other approximately 22,000 generations ago. Thereafter, the upstream Cerekang population split into two populations, O. landangiensis and O. dopingdopingensis, approximately 14,000 generations ago.

We are not aware of any freshwater fish groups that may have experienced both allopatric divergence and secondary contact within such a small single river system. The allopatric divergence and secondary contact within the Cerekang–Doping-doping River system can be explained by assuming that a lake once existed in this region, where many tectonic lakes have formed and disappeared because of the collision and/or juxtaposition of tectonic subdivisions since the Pliocene/Pleistocene.

Gene flow from O. landangiensis to the downstream Cerekang Oryzias is also evident. Although the secondary contact between the upstream and downstream Cerekang populations was estimated to have occurred more than 40,000 years ago, they are still not completely homogenized within the linear corridor of the Cerekang River. This stagnation of homogenization can be explained by their weak dispersal ability, which may reflect past adaptation to lacustrine environments.