Fifty-five patients with mild to moderate, renal or essential hypertension were admitted to a double blind cross-over trial of 18 weeks, involving treatment periods with placebo, the thiazide bendrofluazide (15 mg daily) and the beta blocker atenolol (600 mg daily). Compared to the placebo period (190/117 mm Hg) the hypotensive effect of atenolol (-24/16 mm Hg) was more pronounced than the hypotensive effect of bendrofluazide (-17/6 mm Hg). Arguments in favor of initiating antihypertensive drug therapy with beta blocker were its more powerful hypotensive effect, the quicker onset of its action, less metabolic disturbance, decreased frequency of complaints and patient's preference. On thiazides body weight and the frequency of swollen ankles decreased. Plasma renin concentration was not found to have a strong predicting power for the hypotensive effect of atenolol or bendrofluazide but low renin patients showed a more pronounced blood pressure decrease on bendrofluazide and high renin patients, especially essential hypertensives, on atenolol. While these points can be a guide to therapy today, the preference of one drug over the other must eventually be based on their relative efficacy in decreasing morbidity and mortality from the hypertensive disease.