Two studies were performed in the same subjects to explore the relationship between stomach capacity and food intake. In the first study, a balloon was passed orally into the stomach of four lean and four obese subjects before they ingested a liquid lunch meal. The balloon was filled with 0, 200, 400, 600, and 800 ml of water in a random sequence on different days. The balloon was kept inflated during ingestion then deflated and removed. Food intake was significantly reduced (p less than 0.01) by a balloon volume of greater than or equal to 400 ml. In the second study, another balloon was inserted into the stomach of these subjects to estimate stomach capacity. The balloon was gradually filled at the rate of 100 ml/min with 30 sec pauses. The subjects rated their discomfort as 1 to 10, from no discomfort to extreme discomfort. A rating of 10 was the main index for stomach capacity. Mean capacity (ml) for the lean subjects was 1100 +/- 185 (SE) and for the obese 1925 +/- 175 (SE), t = 3.24, p less than 0.02. When stomach capacity from the second study was correlated to spontaneous food intake at 0 balloon volume from the first study, r = .44, n.s. However, the relationship between stomach capacity and the balloon volume needed to suppress 50% of spontaneous intake was significant, r = .66, p less than 0.05. This may have implications for treatment of obesity with a gastric balloon.