Insulin hypoglycaemia provides a glycopenic stimulus producing a vagal gastric secretory response (Jogi, Strom, and Uvnas, 1949). Earlier investigators (Hollander, Jemerin, and Weinstein, 1942;Davis, Brooks, and Robert, 1965) believed that the gastric acid response to insulin hypoglycaemia represented an 'all-or-none' phenomenon which was (1) initiated when the blood glucose fell to a given threshold value and (2) not related to the degree of hypoglycaemia below such a threshold. The recent finding, however, of a quantitative relationship between insulin dose, hypoglycaemia, and acid secretion (Baron, 1970;Spencer and Grossman, 1971) does not support the 'all-or-none' hypothesis.A single rapid intravenous injection of insulin, such as used by the previously mentioned workers, produces a nadir of hypoglycaemia with gradual restoration of normoglycaemia. In human subjects (Carter, Dozois, and Kirkpatrick, 1972) an infusion of insulin provides a steady level of hypoglycaemia which, if reproducible in the dog, should facilitate correlation studies between blood glucose changes and gastric acid changes.The present investigation was undertaken to study the effect of a continuous infusion of insulin on the blood glucose response in the dog and to establish the relationship between insulin dose, degree of hypoglycaemia, and acid response.