1. We investigated the toxicity of cyclosporine A (CsA) with reference to the timing of its administration in rats. 2. To elucidate the time-dependent effects of CsA on renal function and survival rate, CsA (75 mg/kg per day) or vehicle was orally administered once daily at four different times (3, 9, 15 and 21 h after lights on; HALO) over a period of 21 days to male Wistar rats (n = 56) kept in rooms with a 12 h light-dark cycle. 3. On the 7th day after treatment, creatinine clearances (Ccr) of groups dosed at 3 and 9 HALO (inactive period) were not reduced in comparison with clearances of time-matched control rats, whereas Ccr significantly decreased in rats dosed at 15 and 21 HALO (active period). Cyclosporine A markedly increased urinary N-acetyl-beta-D-glucosaminidase (NAG) excretion in all dosed groups at the 7th day after treatment, except for rats dosed at 3 HALO. In rats dosed at 3 HALO, Ccr decreased progressively; however, it did not decrease progressively in rats dosed at 9 HALO. In surviving rats treated during the inactive period, urine NAG subsequently returned to control levels. Survival rates were greater in animals dosed during inactive periods than those in groups dosed during active periods. 4. Significant differences in CsA-induced toxicity were obvious as a result of the timing of its administration. A different time course between Ccr and urine NAG excretion was observed during repeated CsA administration. Degenerative changes in proximal tubules were demonstrated after chronic administration of CsA, suggesting that severe and persistent tubular damage cannot be assessed by urinary NAG excretion.