The relationships between blood pressure and the concentrations of urinary sodium, potassium and creatinine have been examined using casual urine specimens collected from a random sample of 1,556 men aged 40-59. Partial correlations with systolic blood pressure after adjustment for age, body mass index (BMI), alcohol consumption, and town showed that sodium was not significantly associated with BP, but potassium [r = -0.09 (P less than 0.001)] and creatinine [r = -0.07 (P less than 0.01)] were negatively associated. The sodium/creatinine ratio [r = 0.10 (P less than 0.001)] and sodium/potassium ratio [r = 0.11 (P less than 0.001)] were positively correlated with SBP. Comparable results were found throughout for diastolic pressure. Multiple regression analysis suggested that it was creatinine rather than sodium that was the important component of the sodium/creatinine ratio, and raised the possibility that potassium may play a part in this association because of the strong correlation between potassium and creatinine. Similarly, potassium appeared to be the more important component of the sodium/potassium ratio. The regression coefficients for sodium, potassium, and the sodium/potassium ratio in this study of casual urine samples were found to be similar in direction and magnitude to two large studies using 24 hour urine collections. We suggest that causal specimens may be useful for estimating sodium and potassium intake in population studies.