Bacterial biomass and functional diversity in four marine and four freshwater samples, collected from Resolute Bay, Nunavut, Canada, were studied using fluorescent nucleic-acid staining and sole-carbon-source utilization. Viable microbial counts using the LIVE/DEAD BacLight Viability Kit estimated viable marine bacterial numbers from 0.7 to 1.8×106 cells/l, which were lower than viable bacterial numbers in freshwater samples (2.1–9.9×106 cells/l) (RCBD-ANOVA). Calculations of the Shannon-Wiener diversity index and average well colour development were based on substrate utilization in ECO-Biolog plates incubated at 4°C and 20°C for 38 and 24 days, respectively. The Shannon-Wiener diversity of the marine water samples was significantly greater ( x H'=2.40±0.08, P <0.005; RCBD-ANOVA) than that of freshwater samples ( x H'=1.20±0.00, P <0.005; RCBD-ANOVA). Differences in microbial diversity between fresh and marine water samples at 4°C ( x 4°C =2.01) and 20°C (x20°C =2.31) were also detected by RCBD-ANOVA analysis. Interactions between water type and incubation temperature were not significant ( F =1.926, F c=5.12). Principal component analysis revealed differences in metabolic substrate utilization patterns and, consequently, the microbial diversity between water types and samples.