Impact of Quaternary Amazonian river dynamics on the diversification of uakari monkeys (genus Cacajao)

Abstract Aim The central and western Amazonia underwent several landscape changes during the Quaternary. Whereas the Riverine Barrier Hypothesis is traditionally used to explain the influence of rivers on speciation, processes such as river rearrangements have been overlooked to explain the geographic distribution and evolutionary history of Amazonia biota. Here, we tested how river rearrangements influenced the evolutionary history of uakari monkeys, genus Cacajao , a primate genus primarily associated with seasonally flooded forests in central and western Amazonia. Location Central and Western Amazonia. Taxon The genus Cacajao , including the black uakaris ( C. melanocephalus , C. ayresi , C. hosomi ); and the bald‐headed uakaris ( C. calvus , C. amuna , C. rubicundus , C. ucayalii , C. novaesi ). Methods We performed a continuous phylogeographic analysis using 77 cytochrome b sequences to identify the origin and dispersal of Cacajao lineages. We used genome‐wide SNP variation (ddRADseq) to investigate population structure, gene flow and demographic history in Cacajao populations and used digital elevation models to identify landscape and riverscape characteristics that may have influenced the geographic distribution of Cacajao . Results Our continuous phylogeographic reconstruction pointed out that the ancestral Cacajao lineage occupied the flooded forests of the Solimões River, in central Amazonia, at ~1.7 Mya and descendant lineages dispersed throughout central and western Amazonia more recently. We identified gene flow in both black and bald‐headed uakari populations, even across rivers considered barriers (e.g. the Negro River). Landscape analysis showed that river rearrangements influenced the geographic distribution and population structure in Cacajao . Historical demographic analyses suggest varied scenarios of population size changes among Cacajao monkeys consistent with periods of intense dynamism in flooded habitats and the formation of non‐flooded upland forests. Main Conclusion Our results support that the river rearrangements have shaped the geographic distribution and divergence of recently diverged Cacajao lineages. Landscape and riverscape changes, along with retractions of the flooded forests, isolated some Cacajao populations in floodplain areas. Our study also suggests that these events led to the recent changes in demographic histories in species with a restricted geographic distribution.