DESCRIPTION: Rationale: Identifying ways to increase physical activity (PA) is paramount to controlling the epidemic of obesity and its co-morbidities, including type 2 diabetes. For unknown reasons, post-obese women and black women appear to be especially prone to weight gain. Data from our current RO1 study suggest that inherent variations in resting energy expenditure and fuel utilization do not distinguish obesity-prone from obesity-resistant women or predict weight gain. By contrast, being of black race and having reduced strength, lower free-living PA, and lower total daily energy expenditure are strongly predictive of future weight gain. Objective: To extend our current studies in order to examine the effectiveness of exercise training to improve free-living PA and, in turn, energy balance and weight control. Specifically, we hypothesize that resistance exercise training will be more effective than aerobic and no exercise training in 1) increasing the physiologic ease of, and spontaneous engagement in physical activities of daily living, and 2) increasing total daily energy expenditure and weight-loss maintenance of obese black and white women. Design & Methods: Obese premenopausal black and white women will be randomized to either diet-only, diet+aerobic or diet+resistance exercise training groups. Diet/behavior intervention, with or without the aerobic or resistance exercise training, will be provided throughout the 18 months of study. Testing will be done in the obese state, 6 months later in the normal-weight post-obese state, and after 1 year of weight-loss maintenance. All testing will be under tightly controlled General Clinical Research Center conditions following 4 weeks of diet-controlled energy balance. Changes in body composition will be assessed by the 4-compartment model, and insulin resistance by the insulin modified, frequently-sampled intravenous glucose tolerance test. Major outcomes will include measures of perceived and physiologic difficulty of exercise (cardiac, ventilatory, electromyographic responses to standardized exercise tasks); aerobic fitness; strength fitness (isometric tests); and spontaneous free-living PA, activity-related energy expenditure, non-exercise activity thermogenesis, and total daily free-living energy expenditure (all derived from doubly labeled water). Significance: The results will provide insight into the effectiveness of, and the mechanisms by which, different types of exercise training can improve physical fitness, spontaneous engagement in physical activities of daily living and, in turn, weight-loss maintenance, especially in obesity-prone black women.