There is a scarcity of population-based epidemiological investigations concerning the prevalence of epilepsy among alcoholics, and of alcoholism among epileptic patients. Available data seem to suggest that the prevalence of epilepsy among alcoholics is at least triple that in the general population, and that alcoholism may be more prevalent among epileptic patients than in the general population. The term "alcoholic epilepsy" has been used with varying definitions in different investigations. It is suggested that a uniform definition be adopted so as to minimize confusion when comparing data from different laboratories. Although there is general agreement that excessive alcohol intake can increase the frequency of seizures in epileptic patients, limited available data suggest that light to moderate social alcohol drinking may not affect seizure frequency. However, epileptic patients should be warned about the possible adverse effects of alcohol, especially those who have refractory forms of epilepsy. Except for a few anomalous cases, evidence for the direct seizure-provoking effect of alcohol is not strong. This is because it is difficult to pinpoint alcohol as the only etiology; more likely, alcohol is only one factor among others (e.g., head trauma, cerebral infarct, alcohol withdrawal, and metabolic effects of alcohol) in provoking seizures. Because seizures are a symptom and not a disease, it is often difficult to distinguish epileptic seizures from alcohol-withdrawal seizures. Patients with only the latter kind of seizures should not need chronic antiepileptic medication.