The trophic structure of pelagic communities in lakes of glaciated regions is highly variable due to restricted dispersal of glacial relict taxa and recent species introductions. Much of the enormous between-lake variability in PCB levels in lake trout flesh (15–10 000 ng/g) from the St. Lawrence system results from differences in the length of pelagic food chains. Ontario Ministry of the Environment data (1978–81) on PCB concentrations in lake trout flesh indicate that PCB concentrations increased with the length of the food chain and tissue lipid content, and decreased with distance north of urban-industrial centres. Each trophic level contributed about a 3.5-fold biomagnification factor to the PCB concentrations in the trout, and the lipid content of the trout flesh increased by a factor of 1.5 for each additional trophic level. An empirical model capable of predicting PCB levels in pelagic salmonids and forage fish (smelt and coregonids) indicated that biomagnification of small atmospheric inputs of persistent lipophilic contaminants can explain the frequent occurrence of high levels of contaminants in some biota from remote areas, and that species introductions that lengthen food chains will lead to significant increases in levels of atmospherically dispersed persistent organic contaminants in top predators.

No abstract

Hydrophobic organochlorine contaminants were surveyed in net plankton from 33 lakes in southern Ontario. Organochlorines were detected in all lakes. The relative concentrations of 12 common organochlorines were generally similar, suggesting that they derive from a common source, probably atmospheric transport. Variability in relative abundances of these contaminants increased with contaminant concentration, while contaminant concentration was related to plankton biomass, total phosphorus, and conductivity. Lakes with low plankton biomass had high organochlorine concentrations in that biomass, indicating that a large portion of the variation among lakes in the concentration of contaminants in plankton can be ascribed to variable planktonic biomass. This negative relationship with plankton biomass, or biomass dilution effect, was strongest for the more hydrophobic and refractory compounds.

Tissue levels of chlorinated dioxins and furans and the physiological and biochemical responses of fish were measured upstream and downstream of a bleached kraft mill in the St. Maurice River, Quebec. Relative to fish 10 km upstream of the source, white suckers (Catostomus commersoni) downstream showed evidence of chemical exposure, chemical metabolism, stress, and altered energy use. Levels of 2,3,7,8–tetrachlorodibenzo‐p‐dioxin and tetrachlorodibenzofuran in gutted whole fish averaged 23 and 260 pg/g immediately downstream of the plant and declined to 18 and 112 pg/g at 95 km downstream. The activity of hepatic aryl hydrocarbon hydroxylase (AHH) was 10X higher immediately downstream and 5X higher 95 km downstream. There were corresponding increases in relative liver size, hematocrit, serum glucose, serum protein, and fin‐ray asymmetries, with a pattern of response that matched enzyme induction. These responses pointed to a strong effluent effect and followed the patterns of tissue contamination by dioxins and furans and of water contamination by chlorinated phenolics and guaiacols. There were strong correlations between AHH induction and tissue contamination. Whereas lipid levels increased at all downstream sites, condition factor decreased at 95 km from the source, indicating that there may be disturbances in energy metabolism. Effects on sexual maturation, as shown by levels of serum hormones and gonad somatic index, were inconclusive.

Concentrations of polychlorinated biphenyls (PCBs), dibenzo-p-dioxins, and dibenzofurans, organochlorine pesticides, and their metabolites were measured in eggs of the common snapping turtle (Chelydra s.serpentina) collected from four wetlands on the shorelines of Lakes Ontario, and Erie, and one control location in central Ontario, Canada. Snapping turtle eggs from these sites were also artificially incubated to determine hatching success, and incidence of deformities in embryo and hatchling turtles. The hypothesis that elevated incidences of egg death and/or deformities of hatchling turtles would occur in populations with high concentrations of organochlorine contaminants in eggs was tested. The results were elevated using epidemiological criteria. Unhatched eggs and deformities occurred at significantly higher rates in eggs from Lake Ontario wetlands. Two of three sites from Lake Ontario had substantially higher levels of PCBs, dioxins, and furans compared to eggs from Lake Erie and the control site. It could not be shown that contamination of eggs preceded the occurrence of poor development of eggs, although excellent hatching success and low numbers of deformities in eggs from the control site were considered representative of development in healthy eggs. The statistical association between contaminant levels in eggs and poor development of these eggs supported the hypothesis that eggs from sites with the greatest contamination had the highest rates of abnormalities. PCBs were the most strongly associated chemicals, although possible effects due to the presence of other chemicals in eggs was a confounding factor. The deformities and rates of unhatched eggs were similar to those occurring in other vertebrates collected from highly contaminated areas of the Great Lakes. There were several chemicals present in the eggs that can cause similar reproductive effects in other species; therefore a specific chemical effect was not identified. Results were coherent with known statistical and biological information. Theoretical and factual evidence of PCB contamination in wild-caught snapping turtles supported and hypothesis. However, lack of controlled studies of reproductive effects of polychlorinated hydrocarbons upon this species hindered the agreement of all factual and theoretical evidence with the hypothesis.