Few quantitative studies addressing immunofluorescence histogram analysis have been published. One study by Overton (Cytometry 9:619-626, 1988) has shown threshold and histogram subtraction methods to be accurate for analysis of well-separated immunofluorescence distributions of positive and negative cells. An evaluation of methods to analyze immunofluorescence histograms when positive and negative immunofluorescence distributions overlap has not, to our knowledge, been reported. In this paper, data obtained from flow cytometry of immunofluorescently stained cells infected with recombinant retroviruses that produce a range of simian virus 40 large T antigen levels were analyzed by threshold, histogram subtraction, and distribution modeling methods. This analysis showed that as the separation between the immunofluorescence distributions of positive and negative cell populations decrease the best methods for histogram analysis are modeling followed, in order, by histogram subtraction, and threshold analysis.