Occlusion of the abdominal aorta of the rabbit by inflating the balloon of a Swan-Ganz catheter positioned in the aorta is a simple and reliable method of producing spinal cord ischemia. The electrophysiological, neurological, and neuropathological correlates of ischemia with progressively longer durations and of ischemia after drug interventions were studied with the goal of developing an easily monitored, reproducible model for central nervous system ischemia. The percentage of animals developing paraplegia after varying periods of ischemia was zero after 15 minutes, 30% after 17 minutes, 33% after 20 minutes, 38% after 25 minutes, and 100% after 60 minutes of ischemia. After 25 minutes of ischemia the percentage of animals developing paraplegia was 87% when they were awake and not ventilated during ischemia and reperfusion, but dropped to 38% in animals that were paralyzed, sedated with ketamine, and ventilated, and when the metabolic acidosis that follows aortic occlusion was corrected during reperfusion. Pretreatment with thiopental, hypothermia, naloxone, methylprednisolone, and verapamil changed the percentage of animals developing paraplegia after 25 minutes of ischemia to 0%, 0%, 25%, 40%, and 100%, respectively. The component waves of the spinal somatosensory evoked potential (SSEP) disappeared sequentially during ischemia in the following order: P2, N4, N3, N2, and N1. After reperfusion, the SSEP components returned in reverse order of their disappearance. In the untreated animals, absence of the N3 wave for more than 10 minutes during ischemia was always followed by a neurological deficit. Pretreatment with thiopental or hypothermia permitted longer periods of electrophysiological silence without permanent neurological deficit. The ratio of the amplitude of N3 to N1 (N3/N1) was at least 70% of the control level, and N4 and P2 amplitudes were at least 30% of the control level at 120 minutes after reperfusion in all animals that had a normal outcome. Return of the N3/N1 amplitude to at least 90% of the control level or return of N3/N1 to 70% to 89% of control and P2 to at least 60% of control at 120 minutes after reperfusion reliably correlated with a normal 48-hour motor examination in animals with and without drug interventions.