A prospective study on coronary heart disease (CHD) was initiated in 1960 among 3,524 subjects at risk for its subsequent occurrence. The findings of the first two years of follow-up are presented, during which (mean 2 1/2 years of risk exposure) 70 subjects incurred the advent of manifest CHD. These findings indicate that the presence of (1) abnormalities in lipoprotein pattern, (2) hypertension, or (3) exhibition of a specific overt behavior pattern (type A) each possessed significant prognostic import. The exhibition of the behavior pattern furnished the most important single prognostic entity. Furthermore, the enhanced prognostic risk associated with hypertension or with an abnormal lipoprotein pattern became significant only when either or both of these findings occurred in the subject with type A behavior pattern.