Changes in a plasma endogenous digitalislike substance were investigated in relation to the antihypertensive mechanism of mild exercise. Fifteen women with mild essential hypertension and seven normotensive female volunteers were divided into exercised hypertensive (n = 10), nonexercised hypertensive (n = 5), and nonexercised normotensive (n = 7) groups. A 4-week general clinical observation period preceded the study period of 10 weeks. The exercised hypertensive individuals were treated with a lactate threshold exercise that corresponded to approximately 50% of the maximum oxygen consumption three times a week, whereas the nonexercised groups were observed at the outpatient clinic as control groups. In the exercised group, systolic blood pressure fell by 7 mm Hg (p = 0.05), diastolic by 6 mm Hg (p less than 0.01), and mean blood pressure by 7 mm Hg (p less than 0.01) after 10 weeks. The reduction in the plasma endogenous digitalislike substance was significant after 7 (-1.02 ng/ml, p less than 0.05) and 10 (-1.04 ng/ml, p less than 0.05) weeks in this group. It positively correlated with the reduction in diastolic (r = 0.70, p less than 0.05) or mean (r = 0.66, p less than 0.05) blood pressure and with changes in plasma norepinephrine (r = 0.76, p less than 0.05). The mean corpuscular volume of erythrocytes decreased (-1.7 fl, p less than 0.01) after 10 weeks of exercise, and the plasma volume index tended to decrease (-108 ml/m2, p = 0.28). In the control groups, significant changes in blood pressure and plasma endogenous digitalislike substance were not observed.(ABSTRACT TRUNCATED AT 250 WORDS)