Long-term follow-up of the Framingham cohort for coronary heart disease (CHD) end-points has made it possible to test the hypothesis that those who smoke filter cigarettes are less likely to get clinical manifestations of CHD than those who smoke non-filter cigarettes. Men were classified at the 7th biennial examination (1963--64) according to whether they smoked filter or non-filter cigarettes. 58% of the cigarette-smoking men under age 55 at this examination smoked filter cigarettes. These men had slightly lower prior smoking exposure than smokers of non-filter cigarettes. Despite what seemed to be a favourable cigarette-smoking history, the filter-cigarette smokers did not have lower CHD incidence rates than non-filter smokers. This finding was unchanged even after multivariate logistic regression analysis to adjust for the slight differences in age, systolic blood pressure, and serum cholesterol between the two groups.