Since obese patients with orthopaedic disabilities are often advised to undertake swimming as a part of a weight loss program, the effect of swimming on body weight was systematically studied. Minimally to moderately obese, otherwise healthy young women seeking to lose weight through a program of exercise without dietary restrictions were randomly assigned to one of three groups in which only the type of daily exercise was different. The three types of exercise were brisk walking, riding a stationary cycle, and swimming laps in a pool. All women slowly but progressively increased the time spent in daily exercise to 60 minutes. After 6 months or slightly longer, the women assigned to walking lost 10% of initial weight, the women who cycled lost 12%, but the women who swam lost no weight. The thickness of the subcutaneous panniculus over the middle of the extensor surface of the upper arm was measured using a Lang skin-fold caliper (Graham Field Co, New York, NY) and showed equivalent substantial reductions in the walkers and cyclists, but no change in the swimmers. The results of this study show that both walking and cycling are effective methods of reducing body fat, but that swimming is not.