This paper reviews chemical models of epilepsy and their relevance in the identification and characterization of anticonvulsants. For each convulsant we discuss possible modes of administration, clinical type(s) of seizures induced, proposed mechanism(s) of epileptogenesis and, where available, responsiveness of the induced seizures to anticonvulsants. The following compounds are reviewed: pentylenetetrazol, bicuculline, penicillin, picrotoxin, beta-carbolines, 3-mercaptopropionic acid, hydrazides, allylglycine; the glycine antagonist strychnine; gamma-hydroxybutyrate; excitatory amino acids (glutamate, aspartate, N-methyl-D-aspartate, quisqualate, kainate, quinolinic acid); monosubstituted guanidino compounds, metals (alumina, cobalt, zinc, iron); neuropeptides (opioid peptides, corticotropin releasing factor, somatostatin, vasopressin); cholinergic agents (acetylcholine, acetylcholinesterase inhibitors, pilocarpine); tetanus toxin; flurothyl; folates; homocysteine and colchicine. Although there are a multitude of chemical models of epilepsy, only a limited number are applied in the routine screening of potential anticonvulsants. Some chemical models have a predictive value with regard to the clinical profile of efficacy of the tested anticonvulsants. Some chemical models may contribute to a better understanding of possible mechanisms of epileptogenesis.