The mathematical model that described the relationship between cell-count decay and storage time in fixed bacterioplankton samples from three Antarctic lakes of differing trophic status was determined after a one-year experiment. Bacterial density was estimated by epifluorescence microscopy. Cell count data fitted a negative exponential model in all three cases (p < 0.00001). However, the slopes of their curves were significantly different (p < 0.01), as well as the percentage of bacterial loss after a period of two months. This fact might be related to the limnological characteristics of the water bodies, though the individual genetic variability of their bacterioplankton should not be left aside. Original bacterial numbers in the samples could also be a reason of the differences observed in the pattern of decay in cell counts. Thus, applying a general decay function to any sample and assuming the idea that freshwater bacterioplankton samples can be stored for a two month-period before the bacterial counts decay, can lead to an erroneous estimation of bacterial numbers with direct consequences in ecological investigations.