High-resolution flow-cytometric measurements of ethidium bromide/mithramycin-stained human fibroblast cultures reveal discrepancies between fluorescence pulse-height distributions and genome size as a function of cell strain, culture conditions, culture age, and proliferative stage. These discrepancies were quantitatively assessed by (1) evaluation of the variation of the widths of the 2c fluorescence pulse-height histogram peaks, (2) comparisons between sample to standard fluorescence ratios of 45, XO and 49, XXXXY cell strains, and (3) comparison of fluorescence intensities among cell populations of identical genome size but differing degrees of chromatin condensation (mitotic vs. nonmitotic cells of diploid and tetraploid cell strains). As in our previous studies with lymphocytes, our results suggest caution in equating fluorescence intensity with “DNA content” in flow measurements of nonhomogeneous cell populations. Conditions of cell culture and sample preparation must be standardized in order to compare “DNA content” differences by flow techniques. Remaining sources of variation presently limit detection to differences in DNA content of greater