The aim of this study was to compare the efficacy and safety of 50 µg of sublingual misoprostol with vaginal misoprostol in its currently recommended dose of 25 µg administered at four hour intervals for labour induction at term. A total of 140 women were randomised to receive either 50 µg of sublingual misoprostol with vaginal placebo or sublingual placebo with 25 µg of vaginal misoprostol every 4 hours. Despite the fact that the time from induction to vaginal delivery was shorter, the number of misoprostol doses given was lower and more women delivered vaginally after one dose of misoprostol in the sublingual group, the total number of women and those delivering vaginally in less than 12 hours and less than 24 hours was similar in the both study groups. There was no case of serious maternal morbidity, and maternal side-effects, such as nausea and vomiting, were not common in the two treatment groups. The rate of tachysystole was more than 3-fold higher in the sublingual group, but neonatal outcomes were similar in the both study groups. Logistic regression analyses showed that higher maternal parity, lower BMI, and no need for oxytocine augmentation are associated with the occurrence of spontaneous vaginal delivery; higher initial Bishop score is associated with lower risk of caesarean delivery; there is more likely to require only one dose for labour induction when misoprostol is administered sublingually.