In 25 middle-aged infarction survivors and 76 corresponding controls, representative for a well-defined community, multivariate analysis was used to evaluate whether serum apolipoproteins were better discriminators of infarction survivors than serum lipids and other risk factors. Levels of serum cholesterol and triglycerides, alphalipoprotein cholesterol, apolipoproteins A-I, A-II, B, and D, as well as tobacco smoking and other risk factors, were included. In descending order, serum apo A-II levels (t(b) = -3.12, p = 0.002), tobacco consumption (t(b) = 2.64, p = 0.010), and serum triglycerides (t(b) = 2.06, p = 0.042) contributed significantly to the multiple regression on myocardial infarction (R = 0.53, p = 0.00001). When entered into ka discriminant function, these three variables gave a good separation between survivors and controls. Of the survivors, 50% were above the 90th percentile inthe control group. The relative prevalence of infarction increased continuously with increasing values of the function from zero to more than 6 times the average. Serum apo A-II levels alone were almost as good in separating cases and controls. From this study, we concluded that, among apolipoproteins, apo A-II seems to be a more sensitive discriminator of infarction survivors than other risk factors.