The pharmacokinetics of antiepileptic drugs have been investigated extensively during the last two decades. It has become evident that the therapeutic management of epileptic patients would be significantly improved with the elimination of the drawbacks associated with existing antiepileptic drugs. Based on the information accumulated to date, the following set of principles is proposed: (1) The requirement of central nervous system (CNS) penetration automatically creates vulnerability to CNS toxicity. (2) The narrower the therapeutic range of a given drug, the more limiting its pharmacokinetic properties become. (3) The requirements of chronic administration and compliance place narrow limits on the optimum half-life (24 h). (4) The practice of polytherapy and the potential for pharmacodynamic and pharmacokinetic interactions make it necessary to simplify metabolic schemes (i.e., use of metabolically directed design approaches). (5) New antiepileptic drugs have shown that widely different structures are associated with anticonvulsant properties--it becomes important to minimize toxicity and optimize pharmacokinetic characteristics early in the development of future antiepileptic drugs. (6) Yet-to-be-discovered antiepileptic drugs will probably include biotechnology-based pharmaceuticals. Such biologic agents as neuropeptides present a new set of pharmacokinetic requirements because they are generally receptor targeted, they do not cross absorption barriers easily, and they are labile and subject to degradation.