The objective of this study was to assess the risk of multiple recurrences after an initial seizure recurrence in childhood. In a prospective study, 407 children were followed for a mean of 9.6 years from the time of their first unprovoked seizure. Data regarding each seizure recurrence were obtained and analyzed using statistical methods for survival analysis. The cumulative risk of a second seizure was 29%, 37%, 43%, and 46% at 1, 2, 5, and 10 years, respectively. Of the 182 children who experienced a second seizure, 131 (72%) experienced a third seizure, 105 (58%) have had 4 or more seizures, and 52 (29%) have experienced 10 or more seizures. The cumulative risk of a third seizure was 57%, 63%, and 71% at 1, 2, and 5 years, respectively, after the second seizure. After a third seizure, the cumulative risk of another seizure was 69%, 72%, and 81% at 1, 2, and 5 years, respectively. After a second seizure, factors associated with an increased risk of additional recurrences included a remote symptomatic etiology (rate ratio = 1.7) and the occurrence of a second seizure within 6 months of the first seizure (rate ratio = 1.7). After a second seizure, the risk of subsequent seizures was greater than 50% even in the lowest risk group. With the exception of etiology, factors associated with an increased risk of multiple recurrences after the initial seizure were different than those associated with multiple recurrences after a second seizure. Factors associated with multiple recurrent seizures may be different than those associated with an initial recurrence. As most patients who experience a second seizure experience further seizures, these data suggest that two seizures are a sufficient epidemiological criterion for the definition of epilepsy.