In a nested case-control design the Body Mass Index at 18 years of age was related to 32-year mortality from coronary heart disease (ICD-codes: 410-414) and cancer (ICD-codes: 140-208) among the 1932 Dutch male birth cohort. By means of a logistic regression model the risk ratios were estimated for four BMI-categories: less than or equal to 18.99, 19.00-19.99, 20.00-24.99, greater than or equal to 25.00. Information on blood pressure, resting heart rate, initial health status, education and region of origin was available and these factors were taken into account as possible confounders. The mortality rate for coronary heart disease was highest for the men in the fattest group, while the leanest group showed an excess cancer mortality. The adjusted coronary heart disease mortality risk ratio was elevated for both BMI-categories above the reference category of 19.00-19.99, with an estimate of 2.5 for the men with a BMI greater than or equal to 25.00 at the age of 18. An increased cancer mortality risk ratio of 1.4, mainly due to lung cancer, was observed during the last 20 years of the 32-year follow-up period for the men with a BMI less than or equal to 18.99. The importance of age and differentiation of causes of death in interpreting the results of prospective studies is discussed.