Hypertension is the most important known risk factor for stroke. Clinical, experimental, and epidemiologic evidence suggests that a high dietary intake of potassium is associated with lower blood pressure. In hypertensive rats, a high intake of potassium is reported to protect against stroke, even though blood pressure is not affected. We examined the relation between the 24-hour dietary potassium intake at base line and subsequent stroke-associated mortality in a population-based cohort of 859 men and women (aged 50 to 79 years) in Southern California. After 12 years, 24 stroke-associated deaths had occurred. The relative risks of stroke-associated mortality in the lowest tertile of potassium intake, as compared with that in the top two tertiles combined, were 2.6 (P = 0.16) in men and 4.8 (P = 0.01) in women. In multivariate analyses, a 10-mmol increase in daily potassium intake was associated with a 40 percent reduction in the risk of stroke-associated mortality (P less than 0.001). This effect was independent of other dietary variables, including the intake of calories, fat, protein, fiber, calcium, magnesium, and alcohol. The effect was also apparently independent of known cardiovascular risk factors, including age, sex, blood pressure, blood cholesterol level, obesity, fasting blood glucose level, and cigarette smoking. These findings support the hypothesis that a high intake of potassium from food sources may protect against stroke-associated death.