The associations between blood pressure and urine sodium, potassium and creatinine have been studied among 1,240 men and 1,119 women living in nine British towns in order to assess the contribution of sodium and potassium intake to geographic BP variations within Great Britain. Significant positive associations were found between systolic BP and the urine sodium/potassium ratio for men (P less than 0.05) and for women (P less than 0.001), and for the sodium/creatinine ratio in men (P less than 0.01), after adjustment for age and body mass index. The findings for diastolic BP were similar, but non-significant for men. Associations between BP and sodium concentration were inconsistent and non-significant. The associations between BP and potassium concentration were consistently negative, and significant for diastolic in women (P less than 0.01). The correlations between the mean town systolic BPs and the sodium/potassium ratio were 0.65 (P = 0.058) for men, and 0.60 (P = 0.086) for women. Correlations for diastolic BP were such smaller. The association between BP and the sodium/potassium ratio in this study is consistent both within and between populations, although more so for women than for men. The results are also consistent with the results of other population studies using casual and 24 urine specimens. Although unable to quantify the effects of sodium and potassium with precision, the study suggests that the sodium/potassium ratio is of importance in geographic BP variations in Great Britain, at least for systolic blood pressure.