Tangled banks: A landscape genomic evaluation of Wallace's Riverine barrier hypothesis for three Amazon plant species

Abstract

Wallace's Riverine Barrier hypothesis is one of the earliest biogeographic explanations for Amazon speciation, but it has rarely been tested in plants. In this study, we used three woody Amazonian plant species to evaluate Wallace's Hypothesis using tools of landscape genomics. We generated unlinked single‐nucleotide polymorphism (SNP) data from the nuclear genomes of 234 individuals (78 for each plant species) across 13 sampling sites along the Rio Branco, Brazil, for Amphirrhox longifolia (8,075 SNPs), Psychotria lupulina (9,501 SNPs) and Passiflora spinosa (14,536 SNPs). Although significantly different migration rates were estimated between species, the population structure data do not support the hypothesis that the Rio Branco—an allopatric barrier for primates and birds—is a significant genetic barrier for Amphirrhox longifolia, Passiflora spinosa or Psychotria lupulina. Overall, we demonstrated that medium‐ sized rivers in the Amazon Basin, such as the Rio Branco, are permeable barriers to gene flow for animal‐dispersed and animal‐pollinated plant species.