Working with low density, low biomass material can be challenging, especially when working near the detection limit. Although background contamination is a universal consideration in microbiological research, its impact is increased when the cells under assessment approach the same concentration as the background contamination. The aim of this work was to identify and remove laboratory sources of background contamination in the cell mounting process for epifluorescence microscopy to improve the reliability of cell counting for low biomass samples. Microscope slides and coverslips were assessed before and after autoclaving, washing with detergent and rinsing with ethanol solution. The solutions used in sample mounting
4',6-diamidino-2-phenylindole, phosphate buffered saline and immersion oil, were tested before and after autoclaving as well as both single and triple filtering with a 0.2 µm membrane filter. Using a combination of detergent and ethanol rinses of glassware and triple filtering of all solutions, we were able to reduce the background contamination by almost two orders of magnitude, down from 1×104(±4.3×103) cells to 302(±312) cells per filter paper. This method was then validated with low biomass glacial sediment samples from Renardbreen, Svalbard with cell concentrations of 1.8×105(±2.9×104) cells g-1, close to the reported detection limit of epifluorescence microscopy.