The aim of the present study is to understand the basic relationship between swimming exercise and natural course of epilepsy in animals by performing an electrophysiological study. For this purpose, male Wistar rats were submitted to daily swimming exercise program of three different durations. Animals were swim-exercised for 90 days with either 15 minutes, 30 minutes or 60 minutes/day. Thereafter, the epileptiform activity was induced by a single microinjection of penicillin (500 units) into the left somatomotor cortex. Short-duration swimming exercise (15 min per day for 90 days) decreased the mean frequency and amplitude of penicillin-induced epileptiform activity in the 70 and 90 minutes after penicillin injection compared to penicillin administered group, respectively. Moderate-duration (30 min per day for 90 days) and long-duration (60 min per day for 90 days) swimming exercise did not alter either the frequency or amplitude of epileptiform activity. The results of the present study provide electrophysiologic evidence that short-duration swimming exercise partially inhibits penicillin-induced epileptiform activity. These data also suggest that moderate and long-duration swimming exercise do not increase either the frequency or severity of seizure in the model of penicillin-induced epilepsy.