The relationship between free radicals and scavenger enzymes, and the disorders called the neuronal ceroid-lipofuscinoses, has long been an argumentative one. Recent evidence would seem to support the fact that such a relationship might exist but that it is indirect. The relationship does not seem due to an inborn error of free radical scavenger enzyme metabolism. Anticonvulsants play a role, as they influence free radical generating systems. At this juncture, no one has studied the relationship of anticonvulsant therapy, neuronal ceroid-lipofuscinosis, and the free radical-scavenging enzyme system, and their interplay. We have studied a large number of patients with epilepsy who are on either monotherapeutic or polytherapeutic regimens of most of the common anticonvulsants. We have found excessive free radical production in many of these patients, ranging from minor effects in the simpler anticonvulsants when used monotherapeutically, to more complex changes in polytherapeutic combinations. Likewise, we have found subtle and inconsistent findings in the free radical-scavenging enzyme system in a variety of examples of neuronal ceroid-lipofuscinosis. When refractory seizure disorders stimulate the vigorous use of polytherapy with a variety of free radical-facilitating anticonvulsants, free radical production becomes deleterious. Likewise, in certain types of neuronal ceroid-lipofuscinosis, polypharmacy with anticonvulsants, by enhancing the production of free radicals or suppressing scavenging enzymes, tends to be deleterious and induces a worsening in the disease process.