We evaluated felbamate (FBM) monotherapy in 111 patients with uncontrolled partial-onset seizures in a multicenter, double-blind, parallel-group trial. During the 56-day baseline period, patients had at least eight partial-onset seizures and received one standard antiepileptic drug (AED) at a therapeutic level; a second AED was allowed if at a subtherapeutic level. Patients received either FBM 3,600 mg/d or valproate (VPA) 15 mg/kg/d. The baseline AED at therapeutic levels was discontinued by one-third decrements on study days 1, 14, and 28 and the sub-therapeutic AED, if any, was discontinued completely on study day 1. Study endpoints were completion of 112 study days or fulfilling one or more escape criteria. Criteria for escape relative to baseline were (1) twofold increase in monthly seizure frequency, (2) twofold increase in highest 2-day seizure frequency, (3) single generalized tonic-clonic seizure (GTC) if none occurred during baseline, or (4) significant prolongation of GTCs. The primary efficacy variable was the number of patients in each treatment group who met escape criteria. Thirty-seven patients on VPA and 18 on FBM met escape criteria (p < 0.001). Even when we considered FBM dropouts to have fulfilled escape criteria and VPA dropouts to have completed the 112-day trial, the treatment difference remained statistically significant (p = 0.039) in favor of FBM. Adverse experiences with FBM were all mild or moderate in severity. The frequency of adverse experiences was much lower during monotherapy. FBM monotherapy was effective in the treatment of partial-onset seizures with or without secondarily generalized seizures and demonstrated a favorable safety profile.