Systolic and diastolic blood pressures were compared as predictors of death due to coronary heart disease using data on the 10 year mortality outcome from the 18 403 male civil servants, aged 40-64, in the Whitehall study. There were 727 deaths due to coronary heart disease. At entry to the study the systolic pressure in these men was significantly higher than the diastolic pressure, and a standardised index of relative risk for death from coronary heart disease was greater for systolic blood pressure. After adjustment for age the top quintile of systolic pressure (greater than 151 mm Hg) identified 5% more men at risk of death from coronary heart disease than for the top diastolic quintile (greater than 95 mm Hg). The findings suggested that clinicians should pay more attention to systolic levels as a criterion for making diagnostic and therapeutic decisions.