We have investigated the hypothesis that insulin is a causal and independent risk factor for blood pressure elevation in humans by comparing pre- and post-operative blood pressure values of 34 consecutive patients with histologically-confirmed diagnosis of insulinoma and 34 age- and sex-matched control patients. In patients with insulinoma hypoglycaemic symptoms were present for 18 (9-36) months. (Values are given as median and 95% confidence interval or mean and SD). After removal of insulinoma fasting plasma insulin levels decreased from 22 (16-28) mU/l to 11 (6-20) mU/l (p less than 0.003) and minimal fasting plasma glucose concentrations increased from 2.5 (2.0-3.0) to 4.4 (4.2-5.7) mmol/l (p less than 0.002) while blood pressure values remained unchanged. Body mass index before operation was comparable between the groups: 25.5 (5.4) kg/m2 in insulinoma patients and 24.8 (4.7) kg/m2 in control subjects. Pre-operative and post-operative blood pressure values did not differ between the groups, being (systolic/diastolic) 133 (18)/82 (9) mm Hg in insulinoma patients and 128 (15)/78 (10) mm Hg in control subjects before and 129 (19)/80 (10) mm Hg and 125 (11)/76 (7) after surgery. Chronic hyperinsulinaemia in patients with insulinoma is not associated with a detectable elevation of blood pressure values. Correction of hyperinsulinaemia after surgery for insulinoma does not result in blood pressure changes. These results argue against the hypothesis that insulin is an independent causal factor in the development of essential hypertension in humans.