In a double-blind controlled clinical trial 62 patients with recurrent idiopathic renal calcium oxalate stone formation were allocated either to treatment with bendroflumethiazide, 2.5 mg three times a day, or placebo. In each group the rate of stone formation during medication (average follow-up period 36 months) was compared with the rate of stone formation before medication (average control period 36 months). In both groups a similar striking fall in the rate of stone formation was found, indicating that thiazides in this study did not alter the spontaneous course of idiopathic renal calcium oxalate stone formation. It is doubtful whether life-long prophylaxis with thiazide is justified in patients with a moderate rate of stone formation.