This study investigated in eight healthy male volunteers (a) the gastric emptying pattern of 50 and 100 grams of glucose; (b) its relation to the phase of interdigestive motility (phase I or II) existing when glucose was ingested; and (c) the interplay between gastric emptying or duodenal perfusion of glucose (1.1 and 2.2 kcal/min; identical total glucose loads as orally given) and release of glucose-dependent insulinotropic peptide (GIP), glucagon-like peptide-1(7-36)amide (GLP-1), C-peptide, insulin, and plasma glucose. The phase of interdigestive motility existing at the time of glucose ingestion did not affect gastric emptying or any metabolic parameter. Gastric emptying of glucose displayed a power exponential pattern with a short initial lag period. Duodenal delivery of glucose was not constant but exponentially declined over time. Increasing the glucose load reduced the rate of gastric emptying by 27.5% (P < 0.05) but increased the fractional duodenal delivery of glucose. Both glucose loads induced a fed motor pattern which was terminated by an antral phase III when approximately 95% of the meal had emptied. Plasma GLP-1 rose from basal levels of approximately 1 pmol/liter of peaks of 3.2 +/- 0.6 pmol/liter with 50 grams of glucose and of 7.2 +/- 1.6 pmol/liter with 100 grams of glucose. These peaks occurred 20 min after glucose intake irrespective of the load. A duodenal delivery of glucose exceeding 1.4 kcal/min was required to maintain GLP-1 release in contrast to ongoing GIP release with negligibly low emptying of glucose. Oral administration of glucose yielded higher GLP-1 and insulin releases but an equal GIP release compared with the isocaloric duodenal perfusion. We conclude that (a) gastric emptying of glucose displays a power exponential pattern with duodenal delivery exponentially declining over time and (b) a threshold rate of gastric emptying of glucose must be exceeded to release GLP-1, whereas GIP release is not controlled by gastric emptying.