PHT has multiple effects in adult humans and animals, and there is no reason to assume that it will not have multiple effects in embryos and fetuses. Although one of the first associations between anticonvulsant therapy and an adverse development effect in humans was noted in 1964 (104), the mechanism(s) whereby these adverse effects occur has thus far eluded research efforts. In this review, I have focused on three possible mechanisms. Overall, the evidence does not appear to implicate folate deficiency in PHT-induced embryotoxicity. A role for glucocorticoids or interaction between PHT and the glucocorticoid receptor has not been ruled out. However, a significant amount of work remains to be done to examine the involvement of the arachidonic acid cascade in PHT-induced embryotoxicity in vivo. The bulk of the experimental evidence would seem to favor a role for the generation of a reactive intermediate and its subsequent binding to embryonic macromolecules. This metabolite(s) has not been identified. Additionally, the association between covalent binding of metabolites and embryotoxicity remains simply an association; a causal relationship has not been established. Much work remains to be done to determine whether any of these possibilities, some other possibility, or a combination of several mechanisms will explain the adverse development effects of this very important, therapeutically useful anticonvulsant.