| Summary: | The spatial scale at which populations are genetically structured is of immense interest for the understanding of a species’ ecology and evolutionary biology. This can have important implications for management of resources as well as predicting responses to future change. Rainbow smelt (Osmerus mordax) is an anadromous species with a relatively short freshwater residence time compared to other species with similar life-history strategies. Therefore, while they offer the opportunity to sample distinct spawning aggregations, they also offer an insight into the relative roles of contemporary and historical factors shaping connectivity among marine populations, an area of great interest, and for which further understanding is required. With the use of both mitochondrial DNA and nuclear microsatellite markers, I explored the historical and contemporary factors influencing population structure in smelt. While previous phylogeographic work on this species has resolved two mtDNA lineages dating back to previous glacial episodes, I document the discovery of a zone of contact between these lineages in Newfoundland. This is in addition to the established contact zone in the St. Lawrence estuary, and results in a longitudinal distribution of the races with one race predominating on opposite ends of the species distribution, while the other race is geographically intermediate. Patterns of nuclear genetic variation largely mirror the phylogeographic signals in Newfoundland and suggest a more recent colonization of the Avalon Peninsula as well as implicating a remnant historical signal of colonization of the west coast of Newfoundland from the mainland. In addition, contrasting patterns of genetic diversity and levels of differentiation were apparent between the mainland and Newfoundland and suggest differing scales of dispersal within this species. While the population structure within Newfoundland is most consistent with dispersal restricted to within bays, larger scale biogeographic regions were identified in the mainland range, suggesting dispersal is more common and widespread. In addition, sampling of different run times (i.e. ‘early’ vs. ‘late’) demonstrated the potential for isolation by time when spawning events are separated by a break in activity. Overall, these results shed light into the possible roles of both historical and contemporary factors shaping the dynamics and connectivity among populations.
|
|---|
