A history of depression or depressive symptomatology has been reported in up to two-thirds of patients with medically intractable epilepsy, whereas community studies have demonstrated affective disorder only in a quarter of these patients. Depression has been reported peri- and interictally. However, differentiation may be difficult in patients with frequent seizures. Most authors have found no correlation between depression and epilepsy variables. However, complex partial seizures, especially of temporal lobe origin, appear to be etio-logic factors, particularly in men with left-sided foci. Depression is also more common in patients treated with polytherapy especially with barbiturates, phenytoin, and vigabatrin. Depression has also been described de novo after temporal lobectomy. Psychosocial factors also play a part, but underlying risk factors (e.g., genetic, endocrine and metabolic) may explain the increased rates of depression in people with epilepsy compared to those with other neurologic and chronic medical conditions. The depression appears to be endogenous. Patients tend to exhibit fewer neurotic traits and more psychotic symptoms such as paranoia, delusions, and persecutory auditory hallucinations. Treatment approaches include psychotherapy, rationalization of antiepileptic drug medication, antidepressant treatment, and ECT. The tricyclic and related antidepressants appear to be epileptogenic, especially in people at high risk (personal or family history of seizures, abnormal pretreatment EEG, brain damage, alcohol or substance abuse/withdrawal and concurrent use of CNS-active medication). Seizures tend to occur early in treatment or after dose increments, especially if rapidly titrated. There is little evidence that the newer antidepressants, e.g., selective serotonin reuptake inhibitors, moclobemide, venlafaxine, or nefazodone are more epileptogenic than placebo.