To evaluate the safety and efficacy of intravaginal misoprostol for cervical ripening in the third trimester, a randomized, double-blind, placebo-controlled trial was conducted in 85 patients indicated for induction of labor and with unfavorable cervices. They were randomly assigned to receive either intravaginal misoprostol (100 mg) or placebo placed in the posterior vaginal fornix. The Bishop score, fetal heart rate and Doppler blood flow velocity waveforms were measured before and 12 h after drug administration. Placenta and decidu were histopathologically observed in some cases. Among 85 patients enrolled, 43 received misoprostol and 42 received placebo. Whereas the mean initial Bishop scores were not significantly different between the two groups, the mean Bishop score in misoprostol group was significantly better than those in placebo group. The mean change in Bishop score was also significantly different (4.4 for misoprostol versus 1.0 for placebo, P < 0.01). The prevalence of spontaneous onset of labor within 12 h after drug insertion in misoprostol group (67.4%, 29/43) was significantly higher than that in placebo group (14.3%, 6/42), P < 0.01. The average Doppler velocity systolic to diastolic (S/D) ratios of umbilical artery, middle cranial artery, renal artery were not significantly different before and 12 h after drug insertion between both groups. There was no significant difference in frequency of abnormal fetal heart rate tracings or fetal distress and in the mean Apgar scores between the two groups. Except the presence of vasodilation in villi vessels in the misoprostol group, the placental and decidual histopathological changes had no significantly difference between two groups. It is concluded that intravaginal misoprostol may be an effective and safe cervical ripening agent in the third trimester of pregnancy.