OBJECTIVE: To investigate in prepubertal children whether physical fitness and/or physical activity are: 1) associated with insulin secretion and sensitivity and 2) account for racial differences in insulin secretion and sensitivity. RESEARCH METHODS AND PROCEDURES: Subjects included 34 African American and 34 white nondiabetic children aged 5 to 11 years. Data were divided into two sets according to the availability of VO2max and physical activity data. Body composition was measured by dual-energy X-ray absorptiometry. Subcutaneous abdominal adipose tissue and intra-abdominal adipose tissue were examined by computed tomography. Insulin sensitivity (S1) and acute insulin response (AIR) were determined by a frequently sampled intravenous glucose tolerance test. An all-out, progressive treadmill exercise test was used for measuring VO2max. Physical activity data were collected by questionnaire. RESULTS: African American children had lower SI and higher AIR than white children, after adjusting for total body fat mass. African Americans reported higher levels of physical activity (hours/wk) than whites, but had a lower VO2max. In multiple linear regression analysis, hours/wk of activity and hours/wk of vigorous activity, but not moderate activity, were independently related to SI and AIR after adjusting for race, total body fat mass or fat distribution, and total lean tissue mass. VO2max was not related to AIR, and was inversely related to SI, after adjusting for body composition. Race remained significantly associated with both SI and AIR, even after adjusting for body composition, fat distribution, and hours/wk of activity or hours/wk of vigorous activity. DISCUSSION: In summary, overall physical activity and, especially, vigorous activity were associated with insulin secretion and sensitivity. However, neither physical activity nor VO2max explained the racial difference in insulin secretion (higher in African Americans) and sensitivity (lower in African Americans). Thus, racial (African American to white) differences in aspects of insulin action seem to be due to factors other than body composition, fat distribution, cardiovascular fitness, and amount of physical activity.