The likelihood of seizure recurrence after a first unprovoked seizure has profound social, vocational and emotional implications for the patients. Recurrence rates have varied between 27% and 71% in various studies, and the management of patients with a single unprovoked seizure is a controversial topic. In this prospective study we investigated the influence of age, sex, family history, EEG patterns, and anticonvulsant drug (ACD) therapy on seizure recurrence after a first unprovoked tonic-clonic seizure in adults. For this purpose, between October 1988 and January 1991, we studied adult patients who had experienced their after unprovoked tonic-clonic seizure within last 2 months before neurological consultation, and followed them until June 1993. There were 147 patients who met the criteria for inclusion. Overall cumulative recurrence rates were 31.8% by 6 months, 41.3% by 1 year, 44.1% by 2 years, 42.2% by 3 years, and 45.2% by 4 years. Among the risk factors that were evaluated, the time of the day at which the initial seizure occurred was associated significantly (P < 0.05) with seizure recurrence. In our series, 62 patients received ACD and 85 did not. We did not find a significant difference in recurrence rate with regard to ACD therapy. Our results are comparable with those of studies reported preeviously and suggest that the majority of recurrences after a first unprovoked seizure were seen in the first year (in our series 89% of all recurrences). In our study there was no significant predictor of seizure recurrence, except the time of day at which the initial seizure occurred.