Persons whose body fat is distributed predominantly in the abdomen compared with the hips are at increased risk of several chronic diseases. This study examined the cross-sectional relation of percent body fat, computed from skinfold thickness, and fat distribution, measured by the waist-to-hip girth ratio, to physiologic cardiovascular risk factors in a biracial sample (blacks and whites) of young adults aged 18-30 years. The subjects were persons who were examined at baseline (1984-1986) in the Coronary Artery Risk Development in Young Adults Study in four US metropolitan areas. The two hypotheses tested were that 1) after adjusting for percent body fat, waist-to-hip girth ratio is associated with several physiologic risk factors, and 2) fasting concentrations of serum insulin partly explain such association. Percent body fat was significantly associated with all measured blood lipids, lipoproteins, apolipoproteins, uric acid, and blood pressure. Waist-to-hip girth ratio was significantly, although more weakly, associated in multivariate models with blood concentrations of triglycerides, high density lipoprotein (HDL) cholesterol, HDL2 cholesterol, apolipoproteins A-I and B, low density lipoprotein cholesterol (in women only), uric acid, and systolic blood pressure, but was not associated in either sex with total cholesterol, HDL3 cholesterol, or diastolic blood pressure. Fasting serum insulin concentrations were significantly associated with percent body fat (Pearson r = 0.45-0.53), waist-to-hip girth ratio (Pearson r = 0.18-0.27), and most of the physiologic risk factors. Inclusion of fasting insulin in multivariate models reduced, but rarely eliminated, associations between waist-to-hip girth ratio and the physiologic risk factors. These findings suggest that obese young adults, especially those with abdominal fat preponderance, carry a physiologic profile that places them at higher risk of cardiovascular disease, and that fasting insulin concentrations are only partly explanatory.