The clonal assay was used to measure frequencies of 6-thioguanine-resistant (HPRT) T-lymphocytes in 111 donors from the following 5 control populations: 55 adult healthy volunteers; 20 untreated cancer patients; 8 healthy hospital workers serving as controls for 9 hospital workers sterilizing equipment with ethylene oxide; 15 factory workers serving as controls for 15 workers occupationally exposed to high doses of ethylene oxide; 13 pretreatment samples from donors undergoing a diagnostic test with Technetium-99m for an analysis of heart function. With respect to mutant frequency (MF), cloning efficiency (CE) and age distribution, the first 4 populations were identical. The Technetium group had significantly higher MFs and lower CEs but this can be attributed to the higher mean age of this group. Using the total data base, we calculated the following relationships between MF, CE, age and smoking: (1) ln MF = 4.23-0.63 x ln CE indicating that a doubling of the CE has the effect of decreasing the MF by 37%, (2) ln MF = 0.71 + 0.03 x age meaning that the MF increases by 3% from one year to the next, (3) ln CE = 4.87-0.04 X age indicating that the CE decreases by 0.98% from one year to the next, (4) ln MF = 3.25-0.52 x ln CE + 0.02 X age being the equation quantifying the interrelationship between MF, CE and age, (5) ln MF = 3.32-0.56 x ln CE + 0.01 x age + 0.31 s (where s = 1 for smokers and s = 0 for nonsmokers). Using the latter equation, which allows for effects of CE and age on the MF, a statistically significant effect of smoking could be established. For any combination of CE and age smoking has the effect of increasing the MF by 36%. The above conclusions and calculations remain essentially the same when donors with cloning efficiencies lower than 10 or 20% are excluded from the data base.