To determine whether or not the previously reported association between alcohol intake and high blood pressure is influenced by differential intake of calcium and potassium in drinkers compared with nondrinkers and to assess the magnitude of the independent contributions of alcohol, calcium, and potassium to blood pressure, these associations were evaluated in 7,011 men of Japanese descent. Categorical analyses and multiple linear regression techniques were used to test the hypotheses that alcohol, calcium, and potassium were independent predictors of blood pressure. Alcohol consumption above a threshold of approximately 20 ml/day was found to be positively, strongly, and independently correlated with systolic and diastolic pressures, and this effect was completely independent of the effects of calcium and potassium. Calcium and potassium intake were highly correlated (r = 0.59) and were inversely related to blood pressure, and their combined effect was greater than the effect of either alone. However, in the subgroup of moderate and heavier drinkers, only potassium was inversely related to blood pressure. This finding is compatible with previous reports of malabsorption and increased excretion of calcium at higher levels of alcohol intake, and it indicates that a small portion of the alcohol-induced blood pressure elevation may be mediated through calcium depletion. In the range of dietary intake in this cohort, the effect of alcohol on blood pressure was stronger than was either the separate or combined effects of calcium and potassium.