The antiepileptic effect of lamotrigine (Lamictal) was assessed in a double-blind, placebo-controlled, crossover trial in 56 adult patients with refractory partial seizures. Lamotrigine or placebo was added to the patients' existing antiepileptic drugs (AEDs). The dose of lamotrigine varied from 75 to 400 mg daily. Thirty-eight patients completed the trial and 7 withdrew because of adverse experiences. There was a statistically significant reduction in seizure counts on lamotrigine compared with placebo for total seizures (30.3% reduction, 95% CI 8.4%, 47.0%), complex partial seizures (29.2% reduction, 95% CI 3.8%, 47.9%) and secondary generalised seizures (37.9%, CI 18.9%, 52.4%). The analysis of total seizure days showed a similar significant reduction during lamotrigine treatment for the same seizure categories. There was no statistically significant difference in reporting of adverse events between lamotrigine and placebo except for dizziness which was reported more frequently on lamotrigine than on placebo. There were no differences in abnormal haematological or biochemical findings between lamotrigine and placebo, and lamotrigine had no effect on plasma concentrations of concomitant AEDs.