Anadromous, resident, and landlocked Arctic char (Salvelinus alpinus) differentially experience drift and gene flow, making them ideal for studying incipient divergence. We investigated genetic divergence within and among char occupying landlocked and sea-accessible sites in Labrador, Canada, using 11 microsatellites. Unlike anadromous char, landlocked char were highly genetically differentiated. Genetic subgroups were detected within landlocked and sea-accessible sites. Within Ramah Lake (a sea-accessible site containing two subgroups), one subgroup matured at a small size, and both subgroups had equal proportions of males to females. These findings refute residency as a sneaker male tactic and instead suggest the presence of reproductively isolated resident and anadromous char. Subgroups demonstrated equal frequencies of Atlantic and Arctic lineage mtDNA haplotypes, suggesting their genetic differences were not due to allopatry during the last glacial maximum. Our results are therefore consistent with the sympatric genetic divergence of resident and anadromous Arctic char morphs.

The genetic underpinnings of incipient speciation, including the genomic mechanisms which contribute to morphological and ecological differentiation and reproductive isolation, remain poorly understood. The repeated evolution of consistently, phenotypically distinct morphs of Arctic Charr (Salvelinus alpinus) within the Quaternary period offer an ideal model to study the repeatability of evolution at the genomic level. Sympatric morphs of Arctic Charr are found across this species' circumpolar distribution. However, the specific genetic mechanisms driving this morph differentiation are largely unknown despite the cultural and economic importance of the anadromous morph. We used a newly designed 87k SNP chip to investigate the character and consistency of the genomic differences among sympatric morphs within three recently deglaciated and geographically proximate lakes in Labrador, Canada. We found genetically distinct small and large morph Arctic Charr in all three lakes consistent with resident and anadromous morphs, respectively. A degree of reproductive isolation among sympatric morphs is likely given genome‐wide distributions of outlier SNPs and high genome‐wide FSTs. Across all lakes, outlier SNPs were largely nonoverlapping suggesting a lack of genetic parallelism driving morph differentiation. Alternatively, several genes and paralogous copies of the same gene consistently differentiated morphs across multiple lakes suggesting their importance to the manifestation of morphs. Our results confirm the utility of Arctic Charr as a model for investigating the predictability of evolution and support the importance of both genetic parallelism and nonparallelism to the incipient speciation of Arctic Charr morphs.

Summary Freshwater systems are ideal for landscape genetics studies; they generally exhibit asymmetry in gene flow with populations arranged in a hierarchical, dendritic fashion. Such gene flow asymmetry has the potential to influence the distribution of genetic variation, with downstream populations typically exhibiting higher genetic diversity than headwater populations. Lake trout (Salvelinus namaycush) are widely distributed throughout temperate North America. We examined the relationship between landscape attributes, molecular genetic diversity and differentiation among lake trout populations inhabiting a hierarchically structured freshwater system in northern Labrador (the Kogaluk River drainage). Lake trout (N = 567) collected from ten lakes in this drainage were examined for polymorphism at 12 microsatellite loci. The lakes differed in connectivity, size, elevation and position within the Kogaluk catchment. We observed relatively high levels of population structure and little migration among lakes. Waterfalls were identified as a factor contributing to the observed differentiation. Although a preliminary analysis suggested an isolation‐by‐distance pattern, a subsequent decomposed pairwise regression analysis allowed the identification of outlier lakes. Removal of these outlier lakes and of one other headwater lake, isolated from the rest by several waterfalls, led to the disappearance of the isolation‐by‐distance pattern, suggesting very little or no migration among lakes despite migration being physically possible. Estimates of effective population size correlated with lake area but did not differ between headwater and downstream populations, nor was there a relationship with elevation, a result consistent with the evidence of very little to zero gene flow among lakes. Our study is the first to describe patterns of genetic diversity among lake trout populations inhabiting a spatially fragmented system in an ecologically pristine and sensitive area of northern Canada, the barren grounds of northern Labrador.

Aim The Pleistocene glaciation event prompted the allopatric divergence of multiple glacial lineages of Arctic char ( Salvelinus alpinus ), some of which have come into secondary contact upon their recolonization of the Holarctic. While three glacial lineages (Arctic, Atlantic, and Acadian) are known to have recolonized the western Atlantic, the degree of overlap of these three lineages is largely unknown. We sought to determine the distribution of these three glacial lineages in Labrador and Newfoundland at a fine spatial scale to assess their potential for introgression and their relative contribution to local fisheries. Location Labrador and Newfoundland, Canada. Methods We sequenced a portion of the D‐loop region in over 1,000 Arctic char ( S. alpinus ) samples from 67 locations across Labrador and Newfoundland. Results Within Labrador, the Arctic and Atlantic lineages were widespread. Two locations (one landlocked and one with access to the sea) also contained individuals of the Acadian lineage, constituting the first record of this lineage in Labrador. Atlantic and Acadian lineage individuals were found in both eastern and western Newfoundland. Multiple sampling locations in Labrador and Newfoundland contained fish of two or more different glacial lineages, implying their introgression. Glacial lineage did not appear to dictate contemporary genetic divergence between the pale and dark morph of char present in Gander Lake, Newfoundland. Both were predominately of the Atlantic lineage, suggesting the potential for their divergence in sympatry. Main conclusions Our study reveals Labrador and Newfoundland to be a unique junction of three glacial lineages which have likely hybridized extensively in this region.

Dendritic metapopulations have been attributed unique properties by in silico studies, including an elevated genetic diversity relative to a panmictic population of equal total size. These predictions have not been rigorously tested in nature, nor has there been full consideration of the interacting effects among contemporary landscape features, colonization history and life history traits of the target species. We tested for the effects of dendritic structure as well as the relative importance of life history, environmental barriers and historical colonization on the neutral genetic structure of a longnose sucker (Catostomus catostomus) metapopulation in the Kogaluk watershed of northern Labrador, Canada. Samples were collected from eight lakes, genotyped with 17 microsatellites, and aged using opercula. Lakes varied in differentiation, historical and contemporary connectivity, and life history traits. Isolation by distance was detected only by removing two highly genetically differentiated lakes, suggesting a lack of migration-drift equilibrium and the lingering influence of historical factors on genetic structure. Bayesian analyses supported colonization via the Kogaluk's headwaters. The historical concentration of genetic diversity in headwaters inferred by this result was supported by high historical and contemporary effective sizes of the headwater lake, T-Bone. Alternatively, reduced allelic richness in headwaters confirmed the dendritic structure's influence on gene flow, but this did not translate to an elevated metapopulation effective size. A lack of equilibrium and upstream migration may have dampened the effects of dendritic structure. We suggest that interacting historical and contemporary factors prevent the achievement of the idealized traits of a dendritic metapopulation in nature.