Rats, isolated at mating (Day 1 of pregnancy), were submitted to either 8 h (8L:16D, Exp. I) or 14 h (14L:10D, Exp. II) of light daily with lights on from 12:00 h to 20:00 h and from 06:00 to 20:00 h respectively. In Exp. I, a single dose of RU 486 (10 mg in 0.2 ml ethanol) was given cutaneously at 08:00 h (Group A1), 12:00 h (Group B1), 19:00 h (Group C1) on Day 21 and at 08:00 h (Group D1) and 12:00 h (Group E1) on Day 22. In Exp. II, the same dose of RU 486 was given at 08:00 h (Group A2), 12:00 h (Group B2) and 19:00 h (Group C2) on Day 21. The solvent was given once at each of the preceding times to the control groups (T1 and T2) in both experiments. Groups T1 and T2 gave birth at two periods, the first on Day 22, the second on Day 23; the proportion of births during each of these periods depended on the light regimen (66.3% in 8L:16D; 50% in 14L:10D on Day 22). The distribution of births in Groups D1 and E1 treated on Day 22 were similar to their controls (T1). Rats treated on Day 21 (Groups A1, A2, B1, B2, C1, C2) gave birth over single periods on Day 22 after an interval correlated with the time of RU 486 administration. The earlier the treatment was given, the higher was the number of dead young and the lower the weight of live young 1 day after birth. These effects of prematurity did not impair further survival rates or weight at weaning.(ABSTRACT TRUNCATED AT 250 WORDS)