The use of antidepressant drugs (ADs) in patients with epilepsy still raises uncertainties because of the widespread conviction that this class of drugs facilitates seizures. A detailed knowledge of this issue in its various aspects may help in optimal management of patients suffering concurrently from epilepsy and depression. This article reviews the available data in vitro in animals and humans concerning the known potential of various ADs to induce epileptic seizures. Emphasis has been placed on those variables that may generate confusion in interpreting the results of the various studies. Most ADs at therapeutic dosages exhibit in nonepileptic patients a seizure risk close to that reported for the first spontaneous seizure in the general population (i.e., <0.1%). In patients taking high AD doses, seizure incidence rises markedly and may reach values up to 40%. With a patient history of epilepsy and/or concomitant drugs that act on neuronal excitability, low or therapeutic AD doses may be sufficient to trigger seizures. Experimental data are in partial conflict with human data on the relative potential seizure risk of the various ADs. Therefore, a reliable scale for assigning a relative value to an individual AD or to single AD classes cannot be made. It appears fair to say that maprotiline and amoxapine exhibit the greatest seizure risk, whereas trazodone, fluoxetine, and fluvoxamine exhibit the least. Some ADs may also display antiepileptic effects, especially in low doses, in experimental models of epilepsy and in humans, but the mechanism of this action is largely unknown. The available data suggest that ADs may display both convulsant and anticonvulsant effects and that the most important factor in determining the direction of a given compound in terms of excitation/inhibition is drug dosage. It is probable that drugs that increase serotonergic transmission are less convulsant or, even, more anticonvulsant than others. Because of mutual pharmacokinetic interactions between antiepileptic drugs and ADs, with consequent marked changes in plasma concentrations, it remains to be established whether or not plasma AD levels that are effective against depression also facilitate seizures. Finally, exploring the mechanisms through which ADs modulate neuronal excitability might open new possibilities in antiepileptic drug development.