As part of the collaborative Dorset Research Project investigating mercury and energy fluxes in fresh-water lakes, I measured mercury in the seston (Chapter 1) and studied the biomanipulation impacts on the phytoplankton (Chapter 2) of two Precambrian Shield lakes. Sestonic-mercury (HgT) was measured in the metalimnion and hypolimnion of each lake throughout the summer of 1995 to determine seasonal fluctuations and the relationship with algal productivity. In each lake, sestonic-HgT (pg Hg/L) did not significantly change in the metalimnion but significantly increased in the hypolimnion by season's end. Combined influences of external HgT inputs, seston sedimentation and increased methylmercury production in the hypolimnia over the season may have contributed to these trends. In comparison to other variables measured, algal productivity was highly correlated with sestonic mercury concentrations in both lakes at each limnetic depth. Although there were no significant differences between lakes with respect to average weight-specific HgT (pg HgT/mg D.W.), chlorophyll a exhibited the best correlations with HgT in MouseL. whereas algal biomass was more highly correlated with HgT in Ranger L. This disparity between lakes may be the result of apparent inter-lake differences in light availability and algal community structure. It was also apparent that changes in the proportions of large and small cells over the season affected the magnitude of sestonic mercury measured. With respect to the potential for trophic transfer of mercury, I suggest that small edible algal cells may bioconcentrate more mercury per unit weight than larger, inedible ones. The data also indicate that seston samples should be collected throughout the season at discrete depths if sestonic-mercury measurements are to be used in trophic transfer models. I also examined the effects of fish biomanipulation on the phytoplankton community of these study lakes. Prior to the biomanipulation, Ranger L. had a top-piscivore community whereas Mouse L. had a top-planktivore community. The biomanipulation involved the removal of top-piscivores from Ranger L. and adding top-piscivores to Mouse L. Trophic Cascade theory predicts that algal biomass in these lakes, with their similar morphometries and resource characteristics, should be ultimately controlled by top-consumer abundance. In addition, model predictions expect "edible" algal size-classes and groups in the community to experience the greatest changes in abundance. Therefore in Ranger L., it was expected that the removal of piscivores would result in higher algal biomass (particularly edible algae), whereas the addition of piscivores in Mouse L. would result in lower algal biomass (particularly edible algae). However, for those years following the biomanipulation, algal biomass significant increased in both lakes compared to pre-manipulation years. This suggests that variables other than direct trophic forces were controlling algal biomass from year to year, regardless of changes in the fish communities. When algal size-classes were tested, only edible cells varying from 10-30 μm increased in Mouse L., contrary to what was predicted. In Ranger L., large cells and colonies > 30 μm unexpectedly increased when all other size-classes did not significantly change. With respect to algal group composition, both Greens and Cryptomonads significantly increased in Mouse L. whereas only Greens significantly increased in Ranger L.. Both of these groups were considered to be edible and thus these results were not consistent with the model predictions. As such, I suggested that "bottom-up" influences were important in controlling both size-class and taxonomic abundances. However, when individual size-classes of representative algal genera were compared between pre-and post-manipulation years, there were some effects which may be attributed to the biomanipulation. In particular, large Green colonies became prevalent in Mouse L. during post-manipulation years as a probable response to increased grazing pressure. Conversely, "edible" Greens became prevalent in Ranger L. after the biomanipulation, supporting the prediction of reduced zooplankton grazing pressure. These results have revealed the necessity to test specific algal genera of varying size-classes in order to detect the effects of biomanipulation. They also showed that the majority of algal genera, regardless of size, were not affected by the biomanipulation. Limitations to my interpretation of the data are discussed and vary from time-scale issues to consumer and resource availability unknowns. Along with recommendations for further studies in this area, I hypothesized that the trophic transfer of sestonic-mercury to zooplankton could be intensified if small, edible algal genera (shown to be impacted by Top-Down forces), have relatively higher weight-specific mercury concentrations. However, considering that the phytoplankton community as a whole has shown resilience to herbivory, I also suggest that the majority of mercury measured in the seston is not available for trophic transfer to zooplankton consumers. / Thesis / Master of Science (MSc)