A prospective study was carried out at a tertiary care hospital on 100 pregnant women admitted for induction of labour to compare the effect of misoprostol and dinoprostone on the induction of labour. The patients were divided randomly into two groups of 50 each. Group I received 25 microg misoprostol intravaginally every 3 h (maximum dose 200 microg), and Group II received 0.5 mg PGE(2) gel (dinoprostonev) intracervically every 6 h (maximum three doses in 24 h) until good uterine contractions started. The primary outcome measure was vaginal delivery occurring within 24 h of administration of the first dose of either study drug (successful induction). Statistical analysis were conducted using chi(2) test, Fisher exact test, Student's t-test and relative risk (RR) with 95% confidence interval (CI). In the misoprostol group, more patients achieved successful inductions as compared with the dinoprostone group, 80% vs. 62% (P = 0.0473, RR 1.63, 95% CI 0.95-2.81). The mean induction to delivery interval (IDI) was shorter in the misoprostol group, 13.30+/-8.74 (3-40.15) hours, as compared with the dinoprostone group, 18.53+/-11.33 (2-48.07) hours (P = 0.011). Misoprostol was associated with significantly less oxytocin use (18% vs. 50%) as compared with dinoprostone (P = 0.001 RR 0.36, 95% CI 0.19-0.69). In conclusion, although both misoprostol and dinoprostone appear to be effective agents for labour induction, misoprostol is cheaper, stable at room temperature, has shorter IDI and requires less oxytocin. These results make misoprostol superior to dinoprostone for induction of labour especially in developing and tropical countries.