The overall prevalence of salt sensitivity was studied in 75 men stratified by diagnosis (hypertensive v normotensive) and race (black v white). All were studied in a crossover design employing a 200 mEq and 10 mEq Na/day. High salt led to a decrease in diastolic pressure for all groups (P less than .002). For systolic pressure, there was no salt effect on blood pressure across the whole group; however amongst the hypertensives, particularly the black hypertensives, high salt led to increases in systolic pressure (P less than .022). Obese patients were more likely to increase their systolic pressure in response to salt loading (P less than .05). The patients whose pressure increased on high salt were those who manifested less of a decrease in plasma levels of norepinephrine and renin in response to salt loading (both P less than .05). Systolic salt sensitivity was predicted with high statistical power (R = 0.689, P less than .00001) by a multiple regression equation employing: race; diagnosis; the change in renin and norepinephrine levels with diet; and the change in BP sensitivity to infused norepinephrine across the two diets. In view of the findings of increased norepinephrine, renin and diastolic pressure on low salt and in view of the particular physiological and epidemiological setting associated with systolic salt sensitivity, one wonders about the advisability of across-the-board recommendations of low salt diets for all hypertensive patients.