Genotoxic effects of occupational exposure to cytostatics were investigated in 20 nurses and physicians working in various departments of one hospital. The group was divided into two equal subgroups one of which was involved in the administration of cytostatics (exposed subgroup) and the other was not (unexposed subgroup). The whole group and the two subgroups were compared with a control group of 11 healthy blood donors. Two differently labeled whole chromosome painting (WCP) probes specific for the chromosomes 1 and 4 were used simultaneously. Chromosome aberrations were classified in terms of the Protocol for Aberration Identification and Nomenclature (PAINT) nomenclature. The results obtained by the painting method were compared with findings of conventional unbanded chromosome analysis. Significant differences in the numbers of translocations (FG/100 = 2.25 +/- 1.50 vs. 0.66 +/- 0.21, p < 0.01) and unstable chromosome aberrations determined by the conventional method (AB.C/100 = 2.70 +/- 2.31 vs. 1.63 +/- 1.59, p < 0.05) were found between the exposed subgroup and controls. The unexposed subgroup differed from the controls only in the number of translocations (FG/100 = 2.93 +/- 2.79 vs. 0.66 +/- 0.51, p < 0.05). No significant differences in the number of stable and unstable aberrations were found between the exposed and the unexposed subgroups. On the other hand, highly significant differences (p < 0.01) were demonstrated by the two methods between the whole group (all medical personnel) and the controls. All differences which were found to be significant when translocations were compared were also found to be significant when total stable chromosome exchanges, i.e., the sum of translocations and insertions, were considered. Multicolour chromosome painting is apparently a more sensitive method than the conventional metaphase-based analysis.