1. The effect of osmolality and carbohydrate content on the rate of gastric emptying was assessed by using the double sampling gastric aspiration technique to measure the rate of gastric emptying of isoenergetic and isosmotic solutions of glucose and glucose polymer. Six healthy male subjects were each studied on four separate occasions using a test drink volume of 600 ml. 2. The half-emptying time (t1/2, mean +/- S.E.M.) for a dilute (40 g l-1) solution of glucose (LG, 230 mosmol kg-1) was 17 +/- 1 min. This was greater than that (14 +/- 1 min) for a glucose polymer solution with the same energy content (LP, 42 mosmol kg-1). A concentrated (188 g l-1) glucose polymer solution (HP, 237 mosmol kg-1) emptied faster (t1/2 = 64 +/- 8 min) than the corresponding isoenergetic glucose solution (HG, 1300 mosmol kg-1, t1/2 = 130 +/- 18 min). 3. The dilute (40 g l-1) glucose solution emptied faster than the concentrated (188 g l-1) glucose polymer solution with the same osmolality (LG, 230 mosmol kg-1; HP, 237 mosmol kg-1). 4. The two dilute solutions (40 g l-1) delivered a similar amount of carbohydrate to the small intestine, whereas the concentrated (188 g l-1) glucose polymer solution delivered a greater amount of carbohydrate at 20, 40 and 50 min than the isoenergetic glucose solution. 5. These results indicate that both osmolality and carbohydrate content influence gastric emptying of liquids in man, but the carbohydrate content appears to have greater influence than osmolality.(ABSTRACT TRUNCATED AT 250 WORDS)