We examined the predictive value of plasma renin activity (PRA) in ambulatory patients in the selection of hypertensive patients for treatment with spironolactone. The patients were classified as frusemide responders or non-responders according to their PRA response to 80 mg frusemide orally. After an initial 4 weeks placebo period forty-five patients with the clinical diagnosis of benign hypertension (WHO stage I-II) entered a 4 X 4 week double blind treatment period during which they received spironolactone (Aldactone) 400, 300, 200 and 100 mg/day, respectively. Five out of the forty-five patients who started the trial were withdrawn on account of side-effects and five patients for other reasons. Of the thirty-five patients who completed the study eighteen belonged to the group of non-responders to frusemide, seventeen to the responders. A gradual and significant decrease in mean blood pressure occurred during the period with the high doses of spironolactone, without any significant difference between the groups. The blood pressure lowering effect obtained with the initial high doses was maintained during the consecutive periods when the doses was reduced to a half or even a fourth, in both groups. We conclude that PRA measurement in ambulatory patients is of little, if any, value for the selection of hypertensives for spironolactone treatment; this in contrast to similar selections performed on hospitalized patients.