The role of delayed gastric emptying (GE) in the pathogenesis of gastroesophageal reflux (GER) in infants is controversial at present. GE has been shown to be altered by the composition and osmolality of the feedings. This prospective study was undertaken to assess the changes in the GE and the percentage of time GER was detected by scintigraphy using three different formulas on consecutive days in infants with GER. Twenty-eight infants under 1 year of age diagnosed to have GER by extended intraesophageal pH monitoring (pH being less than 4.0 for greater than 5% of the duration of the test), underwent scintigraphy on three consecutive days using the same volume per single feeding of a casein-predominant, soy, or a whey-hydrolysate formula in a randomized order. The formulas were isocaloric. Gastric emptying and percentage of GER into the esophagus were estimated for 60 min following these feedings. Mean GER percent during the study was 20.39, 17.68, and 16.34 on casein-predominant, soy, and whey-hydrolysate formulas, respectively, and was not significantly different. Mean values of GE were 39.7%, 44.6%, and 48.5% on casein, soy, and whey formula, respectively. No significant difference in GE was also observed between casein-predominant and soy formula. However, a significant difference was observed (p less than 0.05) on GE between casein-predominant and whey-hydrolysate feedings. Our data suggest that formula selection may be important in the treatment of conditions associated with delayed gastric emptying.