OBJECTIVE To compare the effects of oxytocin and amniotomy or vaginal prostaglandin E2 (PGE2) for induction of labour. METHODS We conducted a randomized clinical trial. Eligible for the trial were women with normal pregnancy, parity 0-3, with intact membranes, >40 weeks of gestation documented by ultrasound examination before 20 weeks gestation, observed in a network of 13 general and teaching hospitals in Italy. Inclusion criteria were cervical Bishop's score 5-7, less than six uterine contractions per hour, single pregnancy, cephalic presentation, no history of cesarean section and uterine surgery. Eligible women were randomly assigned by phone to oxytocin plus amniotomy (163 women) or vaginal PGE2 2 mg, two doses at 6-h intervals (157 women). RESULTS Overall, 50 women (15.6%) delivered by cesarean section, 22 (13.5%) randomized to oxytocin, and 28 (17.8%) randomized to PGE2 (not significant). Twelve hours after randomization, induction had failed in 26 women of the 163 randomized to oxytocin plus amniotomy (21.6%) and 34 out of the 157 randomized to PGE2 (15.9%): the difference was not significant. Neonatal outcome was similar in the two groups. CONCLUSIONS This study did not find marked differences in labour and neonatal outcome between women randomized to oxytocin plus amniotomy or vaginal PGE2. A shorter induction delivery interval in the group receiving amniotomy and oxytocin after PGE2 priming was observed.