The Garren-Edwards Gastric Bubble (GEGB) was introduced in 1984 as an alternative to surgery (jaw wiring, gastrointestinal bypass, vertical banded gastroplasty) for the treatment of morbid obesity in patients who had failed behavior modification therapy or dietary management for weight reduction. Its mechanism of action is unclear and previous reports have not demonstrated any significant consistent alteration in gastric emptying (GE) as measured by radionuclide techniques. Other proposed mechanisms include: placebo, hormonal, mechanical "satiety", behavioral modification, and neuronal. In order to determine the effect of the GEGB on GE, ten obese (mean % overweight = 89%) patients, 27-50 yr old (mean = 36 yr), had solid GE scans before and 5 wk after endoscopic placement of the bubble. GE scans were performed in six patients after removal (12 = wk residence time). The meal consisted of 300 microCi [99mTc]sulfur colloid in the form of a 300 kcal egg sandwich (egg white 248 g, white bread 40 g, butter 6 g; composition = CHO 40:PR 40: FAT 20) with 180 ml deionized water. Images were obtained in the anterior and posterior projections at 15-min intervals for 1 hr (four patients) or 2 hr (six patients) and the %GE (decay corrected geometric mean) was calculated. Unlike other studies involving the GEGB, adjunctive therapy in the form of dieting and behavior modification were not employed in this study. The effect of the GEGB alone in the treatment of obesity has not been previously evaluated. There was a significant (p less than 0.025) delay in gastric emptying at 1 hr (pre-bubble mean % gastric retention = 46%; bubble mean = 57%; n = 10). After removal, GE returned toward baseline (mean % gastric retention = 51%; n = 6) (p less than 0.05) (Student's t-test). The average weight loss was 5.5 lb (n = 10; p less than 0.025). One mechanism of action of the GEGB may be delayed gastric emptying resulting in early satiety and decreased food intake with resultant weight loss.