A series of experiments was performed to examine the utility of using the properties of artifically induced seizures as indices of the toxic effect of CO upon the CNS. The two behavioral seizure models tested, 6 Hz ECS and pentylenetetrazol, were unable to differentiate between exposed and unexposed animals, even at severe concentrations of CO (55% HbCO). On the other hand, the hippocampal afterdischarge (AD) proved to be at least as sensitive as the visual evoked potential method. There was a CO concentration-dependent shift in the AD type. CO increased the probability of occurrence of post-ictal depressions (PIDs) without rebound ADs, an event never seen in control animals. CO also decreased the spike frequency within ADs in a dose-dependent manner.