Unexploited populations of Arctic char (Salvelinus alpinus) sampled in autonomous lake ecosystems in northern Svalbard demonstrate extraordinary catch curves with age and size frequency distributions characterized by discrete bimodality. Analyses of size‐age relationship, summer diet and food‐related intestinal parasite intensities of modal char groups revealed a pattern of discrete ontogenetic niche shifts. Life‐history changes at age 10–15 and size 200–300 mm/50–300 g involved shifting from an initial mode of small‐sized, slow‐growing and sexually mature individuals feeding on micro‐crustaceans and aquatic insects (Chironomidae, Trichoptera), to a terminal mode of large‐sized and fast‐growing cannibals. Cannibalism, however, was found to result in accumulation of cestodan parasites, of which Diphyllobothrium ditremum increases age‐related mortality rates and may be lethal at 1500–2000 plerocercoids. Genetically allopatric populations with cannibalism demonstrated a female‐biased sex ratio, primarily in the initial mode, suggesting sexual asynchrony in their ontogeny. By contrast, a small population of large‐sized, non‐cannibalistic Arctic char feeding exclusively on the large amphipod Gammaracanthus lacustris, demonstrated unimodal size and age frequency distribution, faster growth, an excess of males and lower parasite burden. Seasonal prey shortage and slow juvenile growth in association with fitness components favoring large body size is a suggested mechanism for inducing cannibalism. Although not the basic cause of bimodality as such, it is concluded that ontogenetic niche shift by cannibalism reinforces discrete age modal divergence resulting in the numerical preponderance of large‐sized individuals in these marginal char populations. Cannibalism is thus considered an important strategy for survival of landlocked Arctic char in the High Arctic. As a conflicting cost to the more efficient use of available energy by larger individuals, the accumulation of cestodan parasites in cannibals, however, will reduce the survival rate of old individuals and accelerate their termination within this modal group.

Among-lake variation in mercury (Hg) concentrations in landlocked Arctic char was examined in 27 char populations from remote lakes across the Canadian Arctic. A total of 520 landlocked Arctic char were collected from 27 lakes, as well as sediments and surface water from a subset of lakes in 1999, 2002, and 2005 to 2007. Size, length, age, and trophic position (delta(15)N) of individual char were determined and relationships with total Hg (THg) concentrations investigated, to identify a common covariate for adjustment using analysis of covariance (ANCOVA). A subset of 216 char from 24 populations was used for spatial comparison, after length-adjustment. The influence of trophic position and food web length and abiotic characteristics such as location, geomorphology, lake area, catchment area, catchment-to-lake area ratio of the lakes on adjusted THg concentrations in char muscle tissue were then evaluated. Arctic char from Amituk Lake (Cornwallis Island) had the highest Hg concentrations (1.31 microg/g wet wt), while Tessisoak Lake (Labrador, 0.07 microg/g wet wt) had the lowest. Concentrations of THg were positively correlated with size, delta(15)N, and age, respectively, in 88, 71, and 58% of 24 char populations. Length and delta(15)N were correlated in 67% of 24 char populations. Food chain length did not explain the differences in length-adjusted THg concentrations in char. No relationships between adjusted THg concentrations in char and latitude or longitude were found, however, THg concentrations in char showed a positive correlation with catchment-to-lake area ratio. Furthermore, we conclude that inputs from the surrounding environment may influence THg concentrations, and will ultimately affect THg concentrations in char as a result of predicted climate-driven changes that may occur in Arctic lake watersheds.

Three ecologically and morphologically distinct forms of Arctic charr (Salvelinus alpinus L.) have been identified in Loch Rannoch, Scotland, whose evolutionary status and origins are incompletely understood. A study was made of restriction fragment length polymorphism (RFLPs) detected variation in the D-loop, ND1 and cytochrome b regions of the mitochondrial genome, encompassing [3500 bp. Eight RFLP haplotypes were identified that clustered into three distinct clans based on restriction differences and into four clans based on sequence differences. Significant differences in RFLP frequencies were found among all morph groups. The pelagic morph was highly divergent from the two benthic forms, with the benthic forms having variants from only one genetic clan while the pelagic was dominated by a single variant from another clan. The relative divergence observed among benthic and pelagic forms is *10 fold greater when nucleotide divergence among the haplotypes, as well as haplotype frequency differences, is taken into account. Sequence divergence between haplotypes in the two main clans is of a similar order to that between haplotypes in these clans and a charr from North America. In contrast, divergence among the two benthic morphs relates entirely to differences in haplotype frequencies. The study confirms the genetic distinctiveness of the pelagic and benthic forms as well as of the two benthic forms. It strongly supports previous evidence that the genetic divergence between the pelagic and benthic populations is allopatric in origin. Additionally, the results strongly suggest that the two benthic populations have undergone peripatric divergence through the sequential colonisation of the two basins by one lineage, followed by their spatial separation and reproductive isolation.

Electrophoretic analysis of successive samples of Salvelinus collected in the Fraser River watershed of northern Labrador in 1984 and 1986 revealed the existence of natural hybrids between Arctic char (Salvelinus alpinus) and brook trout (S. fontinalis). We examined 11 electrophoretic loci in liver tissue; 10 loci were informative and six were diagnostic of the two species and their hybrids. The electrophoretic phenotypes of some hybrid specimens sugested that they were second-generation hybrids and/or backcrosses to the parental species. Hybrids were represented by six different year classes, indicating that crossing between the species in the system is a regular occurrence. The hybrids, although visually difficult to distinguish from brook trout, were meristicaîly intermediate or more closely resembled Arctic char. The observations raise a number of questions regarding the ecological circumstances promoting hybridization in the evolution of Arctic char and brook trout populations. The results stress the need for restrictions on introducing brook trout into systems with original populations of Arctic char. Further, Arctic char from the Fraser River are commonly used for aquaculture research and development; therefore, there is a clear need to screen fish carefully to determine whether or not they represent introgressed gene pools.

No abstract