In a population of 887 men and 1159 women aged 40 to 89 years in southern California, age adjusted systolic and diastolic blood pressures correlated significantly with sodium/potassium ratios estimated from casual urine samples. The relationship was apparent over the whole range of blood pressures and urinary electrolyte excretions. An age gradient was apparent for systolic pressure in both men and women, the regression slope for blood pressure versus sodium/potassium ratio being steeper with increasing age. The regression slope overall was approximately 4 mm Hg systolic and 1 mm Hg diastolic pressure increase per unit increase in sodium/potassium (mmol/mmol) ratio. The use of casual urine samples in assessing the blood pressure and sodium/potassium relationship should not be prematurely dismissed. Evidence from this and other studies suggests the relationship between blood pressure and electrolyte intake in the population may be more dynamic than previously believed, and also more marked in the old, who are at the greatest risk for cardiovascular disease. A reduction in the age-related increase in blood pressure has considerable implications for the prevention of cardiovascular disease.