Of 10,119 men aged 35 to 57 years studied in the Multiple Risk Factor Intervention Trial (MRFIT), ventricular premature beats (VPBs) were more frequent among the 112 men who reported having previously had a myocardial infarction (MI) (22.3%) as well as among the 143 men with a history of diabetes mellitus (13.3%) than among the remaining 9864 (7.6%). There was no demonstrable association between VPBs and cholesterol or blood pressure levels, either overall or in various age groups. There was, however, a strongly positive relationship between VPBs and age (p less than 0.001). A positive association between VPBs and smoking was present which was statistically significant overall (risk ratio = 1.21, 95% confidence limits from 1.03 to 1.43), as well as in the age groups 35 to 39 years (1.74, 1.13 to 2.68) and 55 to 57 years (1.50, 1.01 to 2.22). This association, if confirmed, raises the question of whether VPBs offer a mechanism for the increased morbidity and mortality from coronary heart diseases among cigarette smokers compared with nonsmokers.