We used 31P magnetic resonance spectroscopy to measure maximal mitochondrial function in 12 obesity-prone women before and after diet- induced weight reduction and in 12 matched, never-obese, and 7 endurance- trained controls. Mitochondrial function was modeled after maximum-effort plantar flexion from the phosphocreatine recovery time constant (TC(PCr)), the ADP recovery time constant (TC(ADP)), and the rate of change in PCr during the first 14 s of recovery (OxPhos). Weight reduction was not associated with a significant change in mitochondrial function by TC(PCr), TC(ADP), or OxPhos. Mitochondrial function was not different between postobese and never-obese controls by TC(PCr) [35.1 ± 2.5 (SE) vs. 34.6 ± 2.5 s], TC(ADP) (22.9 ± 1.8 vs. 21.2 ± 1.8 s), or OxPhos (0.26 ± 0.03 vs. 0.25 ± 0.03 mM ATP/s), postobese vs. never-obese, respectively. However, TC(ADP) was significantly faster (14.5 ± 2.3 s), and OxPhos was significantly higher (0.38 ± 0.04 mM ATP/s) in the endurance-trained group. These results suggest that maximal mitochondrial function is not impaired in normal-weight obesity-prone women relative to their never-obese counterparts but is increased in endurance-trained women.