The effect of prolonged caffeine administration on blood pressure in hypertensive subjects was assessed in a double-blind placebo-controlled study. Eighteen hypertensive subjects participated, nine of whom received placebo throughout the study and nine of whom received placebo during the first three days, caffeine during the subsequent seven days, and placebo during the final four days of the two-week study. Those who received caffeine were given 250 mg with meals three times daily. There were no untoward reactions in the course of the study, but one subject with unacceptably high blood pressures while receiving placebo had to be discharged from the study to resume antihypertensive therapy. Systolic blood pressure was immediately increased (9.2 +/- 3.4 mm Hg) within 15 minutes after the first dose of 250 mg of caffeine. On the first day of caffeine, systolic pressure was increased a mean of 7.3 +/- 4.0 mm Hg, but this was no longer significant after the initial day of caffeine administration. Diastolic pressure showed a trend toward increasing, but this never reached significance. The minor increases in plasma catecholamine levels and plasma renin activity were not significant on either a short- or long-term basis. After discontinuation of caffeine, no overshoot phenomena were observed. It is concluded that prolonged administration of caffeine is not associated with significant elevation in blood pressure, plasma catecholamine levels, or plasma renin activity in patients with borderline hypertension.