Reliable quantification of microbial growth in a drinking water environment has typically been difficult, primarily due to the development of thin, patchy biofilms. Therefore, initial sampling and resuspension procedures become critical to the subsequent biomass determination. Biofilm cells attached to polycarbonate coupons of an annular reactor (AR) have typically been removed by aseptically scraping the coupon surface with a sterile utility knife. The advantage of this method is its simplicity, however, scraping often compromises sterility and is highly subject to individual variation. The purpose of this research was to develop a method that could remove and resuspend biofilm cells efficiently, consistently and with a good recovery rate. This paper presents a comparison of removal and resuspension methods. Three methods used to remove cells from the polycarbonate coupons include: scraping with a utility knife, swabbing and stomaching. In addition, four methods were selected for cell resuspension and involved the use of a tissue blender, vortex, stomacher and a sonicator. Of the removal methods examined, stomaching consistently yielded the highest number of culturable and total bacterial cells, ranging from two to four times more cells than scraping procedures. For one experimental set, the number of colonies enumerated by heterotrophic plate count (HPC) from the stomacher ranged from 1.0–1.7×10 6 CFU/cm 2, whereas numbers obtained using the scraping method were 4.6–4.9×10 5 CFU/cm 2. It was found that the number of HPCs recovered with stomaching was significantly greater at the 5% level than that obtained using the scraping method. Similarly, for cell resuspension, stomaching provided the highest enumeration. Once removal was achieved, sonication also provided good resuspension. An analysis of variance showed that, compared to the resuspension step, the removal step is more significant at the 5% level to the recovery of biofilm cells. The stomacher has the unique advantage of combining cellular removal and resuspension into a single step. This method was, therefore, selected as an ideal method for recovery of biofilm cells. Subsequent optimization measures using the stomacher showed that sterile deionized water was a suitable diluent for recovering cells from a drinking water environment. At normal speed (i.e., 230 rpm±5%), the optimal stomacher run length for maximum cell removal was 2 min.