Improvement in the predictive assertion of the micronucleus assay was achieved by treating human malignant melanoma cells (Mewo) with cytochalasin B (CB), generating binucleated cells (BNC) representing cells after a single karyokinesis. Optimal cell binucleation was determined by testing several cytochalasin B concentrations and different incubation times. On average, 56% binucleated cells were found after incubation with 2 to 3 micrograms/ml cytochalasin B for 48 h. Cells with at least one micronucleus (Mn) were defined as fraction of cells with micronuclei and describes the degree of damaged cells. We found in binucleated cells 2.2 fold the fraction of cells with micronuclei than in mononucleated cells (MNC), as expected assuming that an induced micronucleus is associated with only one single daughter cell after mitosis. The mean of micronuclei per binucleated cells, however, was enhanced about 2.9 fold in relation to that of micronuclei per mononucleated cells and is related to the nuclear damage per cell. The application of cytochalasin B did not enhance the fraction of damaged cells although the degree of the injury per cell is intensified. A micronuclei promoting or inhibiting effect of the experimental design due to changes in cell proliferation was excluded by cytofluorometric investigations of DNA content and synthesis after cytochalasin B application. A comparison of the modified with the conventional micronucleus assay shows the superiority of the former.