Across its Holarctic range, Arctic charr (Salvelinus alpinus) populations have diverged into distinct trophic specialists across independent replicate lakes. The major aspect of divergence between ecomorphs is in head shape and body shape, which are ecomorphological traits reflecting niche use. However, whether the genomic underpinnings of these parallel divergences are consistent across replicates was unknown but key for resolving the substrate of parallel evolution. We investigated the genomic basis of head shape and body shape morphology across four benthivore-planktivore ecomorph pairs of Arctic charr in Scotland. Through genome-wide association analyses, we found genomic regions associated with head shape (89 SNPs) or body shape (180 SNPs) separately and 50 of these SNPs were strongly associated with both body and head shape morphology. For each trait separately, only a small number of SNPs were shared across all ecomorph pairs (3 SNPs for head shape and 10 SNPs for body shape). Signs of selection on the associated genomic regions varied across pairs, consistent with evolutionary demography differing considerably across lakes. Using a comprehensive database of salmonid QTLs newly augmented and mapped to a charr genome, we found several of the head and body shape associated SNPs were within or near morphology QTLs from other salmonid species, reflecting a shared genetic basis for these phenotypes across species. Overall, our results demonstrate how parallel ecotype divergences can have both population-specific and deeply shared genomic underpinnings across replicates, influenced by differences in their environments and demographic histories.

The trade-off between cognitive capacity and developmental costs drive brain size and morphology across fish species, but this pattern is less explored at intraspecific level. Physical habitat complexity has been proposed as a selection pressure on cognitive capacity that shapes brain morphology of fishes, but development of brain is also inherently linked to supply of energy and nutrients, particularly of omega-3 long-chain polyunsaturated fatty acids (n-3 LC-PUFA). In this study, we compared brain morphology of brown trout Salmo trutta from stream, lake, and hatchery environments, which differ in physical complexity and availably of dietary n-3 LC-PUFA ranging from low habitat complexity and high n-3 LC-PUFA availability in hatchery to high habitat complexity and low n-3 LC-PUFA availability in streams. We found that brain size, and size of optic tectum and telencephalon differed across the three habitats, being largest in lake fish. We suggest that these differences appeared to associate with diet quality and habitat specific behavioural adaptations rather than physical habitat complexity.