Some anti-convulsant drugs have a calming effect that may potentially be used to reduce cigarette smoking. A cross-sectional study, including all 100 psychiatric in-service patients receiving various anti-convulsant drugs and their age- and sex-matched controls, was done. The intensity of their daily cigarette smoking and their body weights were recorded. Of the patients, 25% were non-smokers. Of the 22 patients on phenytoin, eight were non-smokers as opposed to three in the control group. Two thirds of the phenytoin group patients were either non-smokers, or were nominal smokers. As opposed to this, of the control patients and those on other anti-convulsants, two thirds were either moderate or heavy smokers. The phenytoin group of patients weighed less than the controls (149 v 163 lbs). Valproic acid therapy was associated with a significantly higher body weight and with more smoking. Clonazepam's effect was similar to valproic acid, but carbamazepine did not show any relationship either with smoking or with body weights. Therapy with phenytoin is associated with a lower prevalence and a lower intensity of cigarette smoking, together with lower body weights in psychiatric in-patient population.