Several studies have been performed to determine if plasma insulin concentration bears a relationship with blood pressure and may be a causative factor in the genesis of hypertension. The results reported were discrepant between studies: insulin was observed to be independently correlated with blood pressure in some studies, whereas no such relationship was observed in others. Serum insulin, whole blood glucose, weight, height, and diastolic and systolic blood pressure of 839 fasting individuals, randomly selected from the adult population of Mahé Island, The Seychelles, were recorded; this population has a high hypertension prevalence rate. The proportion of hypertension was higher in subgroups of individuals with diabetes mellitus (52.4 percent) and with overweight (34.5 percent) than in the general population. Subgroups of individuals with diabetes, excessive weight or high blood pressure had higher mean fasting plasma insulin (13.6 +/- (s.d.) 7.9, 14.0 +/- 8.0, 11.1 +/- 9.7 respectively) than individuals not affected with these conditions (8.1 +/- 4.6 microU/ml). Both systolic and diastolic blood pressure were positively correlated with age, body mass index and fasting blood glucose in a multivariate model. No correlation with insulin was observed. In contrast, in a subgroup of 223 overweight individuals, age, blood glucose and plasma insulin were each positively correlated with blood pressure. It is concluded that, in this cross-sectional study, insulin was not independently correlated with blood pressure except for overweight individuals.