Abstract
Aim Sea‐level changes and coastal geomorphology (extension of continental shelves) have affected the pattern of diversification of brackish/freshwater fish populations due to the confluence of basins/estuaries during low sea‐level periods (opportunity) and isolation during high sea‐level periods. It is hypothesized that populations from estuaries in extensive continental shelves, which have greater potential for confluence, should have more opportunities for gene exchange than those in narrow continental shelves. To test this, we assessed the distribution of the genetic diversity and the patterns of diversification of the estuarine silverside, Atherinella brasiliensis, from the Brazilian coast. Location Eight estuaries along the southern and north‐eastern portions of the Brazilian coast. Materials and methods Sequences of the mitochondrial control region (D‐loop) and the cytochrome B were used to assess the distribution of genetic diversity, and these mitochondrial fragments and two nuclear introns were used to assess the patterns of diversification of A. brasiliensis . The palaeochannels that were exposed along the Brazilian coast during the Last Glacial Maximum were reconstructed using bathymetry data and the GIS method. Results The populations of A. brasiliensis from the estuaries of the north‐eastern Brazilian coast (narrower and shallower continental shelf) showed higher levels of genetic differentiation and larger (more haplotypes) endemic lineages than the populations in the southern coast (broader and deeper continental shelf). Furthermore, south coast lineages showed more evidence of genetic exchange among populations, which have resulted in a reticulated relationship between lineages and estuaries. The reconstruction of palaeochannels revealed more connections among southern than north‐eastern basins. Main conclusions Our results support the hypothesis that continental shelf extension and sea‐level changes have influenced the diversification of A. brasiliensis . By extension, they also support the hypothesis of diversification by taxon pulse, which makes predictions based on the opportunities for expansion during periods of low sea level and the width of the continental shelf. We suggest this as a model for the genetic diversification and distribution of estuarine species along the Brazilian coast and other regions of the world.