Six patients with vasopressin-responsive diabetes insipidus (DI) received clofibrate and chlorpropamide, singly and in combination. Decrease in urinary output averaged (mean +/- SEM): (1) clofibrate 2 g/day, 47% +/- 6%; (2) chlorpropamide 250 mg/day 59% +/- 5%; (3) clofibrate 2 g/day plus chlorpropamide 125 mg/day, 54% +/- 7%; (4) clofibrate 2 g/day plus chlorpropamide 250 mg/day 61% +/- 4%. Water deprivation tests before and during treatment showed significantly higher basal, final, and peak urinary osmolalities (Uosm) and lower free water clearance (CH20) on chlorpropamide, singly and in combination: clofibrate raised Uosm less but significantly decreased CH2O. Water load tests before and during treatment showed that chlorpropamide, singly and in combination, markedly decreased maximal urinary flow, maximal CH2O, percentage water load excreted, and increased minimal Uosm; clofibrate significantly decreased maximal urinary flow and CH2O only. One patient responded only to combination therapy. Chlorporpamide caused serious hypoglycemia in three of six patients. Clofibrate had no significant side effects.