Rats rendered chronically epileptic by the implantation of cobalt in the right parietal cortex were simultaneously prepared with permanent cortical and temporalis muscle electrodes for longitudinal EEG and EMG recording. Animals treated similary with glass rods served as the controls. Wet dog shakes were quantified by counting the characteristic artifacts produced in the EEG tracings upon their appearance. The number of wet dog shakes exhibited by the control rats remained at a low and constant level over 18 days of recording. In contrast, wet dog shakes in the chronically epileptic rats began to increase by the fourth day after cobalt placement and remained significantly elevated up to the 18th day. Administration of reserpine to naive rats or to cobalt epileptic rats on days 7 and 9 after implantation resulted in an almost complete suppression of wet dog shakes which endured over a period of 3-5 days. These results suggest that the abnormally elevated wet dog shake response of the cobalt-epileptic rats and the spontaneous wet dog shake behavior of normal rats may be mediated by common neural pathways.