Clinical studies suggest that serum levels of apolipoproteins A-I and B may be more strongly related to coronary artery disease than are their respective lipoprotein-cholesterol fractions. Therefore, we assessed the association between levels of apolipoprotein B, apolipoprotein A-I, lipids, and lipoprotein cholesterols in children and the reported histories of myocardial infarction in their parents in a survey of 2416 black and white school-age children. As compared with children whose fathers did not report a myocardial infarction, those whose fathers reported having had an infarction (n = 139) had a lower mean level of apolipoprotein A-I (137 vs. 141 mg per deciliter; P = 0.04) and a lower ratio of low-density lipoprotein cholesterol to apolipoprotein B (1.08 vs. 1.11; P = 0.007), along with a higher ratio of apolipoprotein B to apolipoprotein A-I (0.64 vs. 0.61; P = 0.04). These associations existed independently of the children's race, sex, age, and history of obesity, smoking, alcohol intake, and use of oral contraceptives. Children whose mothers reported having had a myocardial infarction (n = 56) had no decrease in the ratio of low-density lipoprotein cholesterol to apolipoprotein B, but they tended to have an elevated ratio of apolipoprotein B to apolipoprotein A-I. In contrast, serum lipoprotein-cholesterol fractions in children were not related to myocardial infarctions in either parent. These results provide further evidence that apolipoproteins are more strongly related to the risk of cardiovascular disease than are lipoprotein-cholesterol fractions.