A randomised, placebo-controlled, double-blind crossover study was conducted in 14 patients with mild essential hypertension comparing different regimens of administration of the "short-acting" diuretic, chlorothiazide and the "long-acting" diuretic, chlorthalidone. There were 6 randomised treatment phases each of 5 weeks duration. For blood pressure recorded both at the Clinic and at home the most prominent effects were seen with standing blood pressure, for which significant reductions of mean blood pressure compared to the placebo phase were observed with chlorthalidone 25 mg once daily (-5 ± 1 (SE) mm Hg - p < 0.05), chlorthalidone 50 mg once daily (-7 ± 1 mm Hg - p < 0.01) and chlorothiazide 500 mg twice daily (-4 ± 1 mm Hg - p < 0.05). Blood pressure reductions with chlorothiazide 500 mg once daily and 1000 mg once daily were not significant. About half of the patients completing the trial could be described as "non-responders" The observed biochemical changes (reduced plasma potassium and chloride concentrations and increased plasma bicarbonate and urate concentrations) were most marked in the phases with the most prominent blood pressure effects. The results support the suggestion that a sustained diuretic effect is desirable when diuretics are used in the treatment of hypertension. This can be obtained by selection of the appropriate dosage regimen for a particular diuretic, such as once daily administration for "long-acting" and twice daily for "short-acting" thiazide-type diuretics.