This randomized double-blind placebo-controlled study was undertaken in a series of 179 patients to determine whether phenytoin administered soon after head injury lessens the incidence of late posttraumatic epilepsy. When delayed hypersensitivity to phenytoin developed, the patient was switched to phenobarbital. The patients were followed for 18 months to detect the occurrence of seizures and to serially measure plasma phenytoin concentrations. There was no significant difference in the percentage of patients having late seizures in the treated and placebo groups (p = 0.75). The time between injury and seizures did not significantly differ between the two groups. The results provide no support for the continued use of phenytoin in the low therapeutic range for prophylaxis against late posttraumatic seizures. It cannot be concluded that higher phenytoin plasma concentrations and higher compliance rates than obtained in this study would not have significantly decreased the occurrence of late posttraumatic epilepsy. The finding that no patient with a phenytoin plasma concentration of 12 microgram/ml or higher had a seizure raises the question of whether phenytoin in blood concentrations in higher therapeutic ranges might lessen the occurrence of posttraumatic epilepsy, and should be studied further. Posttraumatic epilepsy is a major public health problem deserving a large cooperative trial to determine if phenytoin at higher blood levels than obtained in this study, or other currently available or newly developed drugs, can prevent the occurrence of posttraumatic epilepsy.