Misoprostol is the drug of choice for medical abortion worldwide but consensus is yet to be reached regarding its preferred route of administration. To compare the outcome of sublingual with vaginal administrations of misoprostol for induction of second trimester abortion. A randomised comparative trial where 300 women at 13-20 weeks gestation, requiring medical abortion, were randomly assigned to sublingual or vaginal route for misoprostol administration with a dose schedule of 400 microg three-hourly, up to a maximum five doses over 24 h. The same doses were repeated for another 24 h in non-responders. Primary outcome measure was complete abortion rate at 24 and 48 h, and the secondary outcome measures were induction-abortion interval, failure rate, side-effects and patients' preference to the route. No statistically significant differences in the complete abortion rates were observed at 24 h (64.03% vs 61.59%, P = 0.767) and at 48 h (79.14% vs 82.01%, P = 0.651) when sublingual and vaginal groups were compared. Mean induction-abortion intervals in sublingual and vaginal groups were 14.1 and 14.5 h, respectively (P = 0.066). Other outcome measures were also more or less similar in both groups. Differences in the incidence of side-effects were also statistically insignificant when both groups were compared. Sublingual administration of the drug was preferred by most of the women as compared to vaginal administration (P < 0.0001). Both sublingual and vaginal administrations of misoprostol are equally effective in inducing medical abortion during second trimester but sublingual route was preferred by the patients.