Aerobic fitness and adiposity are each independently associated with health outcomes among children, although the relationship between these two variables is unclear. Our objectives were to evaluate (i) the association of adiposity with aerobic fitness using objectively measured levels of percent body fat, compared to BMI as a percentile proxy for adiposity while controlling for genetic admixture, and (ii) the congruence of BMI categories with high and low body fat categories of objectively measured percent body fat. Participants were 232 African-American (AA), European-American (EA), and Hispanic-American (HA) children aged 7-12 years (Tanner stage < 3). Aerobic fitness was measured via a submaximal indirect calorimetry treadmill test (VO 2-170), and physical activity levels with accelerometry. Genetic admixture estimates were obtained using 140 genetic ancestry informative markers to estimate European, African, and Amerindian admixture. Fat mass was determined using dual-energy X-ray absorptiometry (DXA). Children were classified into a low body fat group ( < 25% in males, < 30% in females) or a high body fat group based on their percent body fat; children were also categorized according to BMI percentile. Children in the low body fat group had significantly higher aerobic fitness (P < 0.05) regardless of BMI percentile classification. Higher African genetic admixture was associated with lower aerobic fitness (P < 0.05), while physical activity was positively associated with fitness (P < 0.01). In conclusion, aerobic fitness levels differ by percent body fat and genetic admixture irrespective of BMI classification, and such differences should be taken into account when evaluating outcomes of health interventions. © 2011 The Obesity Society.