Reperfusion injury is a pathophysiological entity distinct from the primary ischaemic injury; the oxygen arriving with blood recirculation, although necessary for alleviating the ischaemic status, may be harmful and provoke additional injury in the already damaged tissue. This study aims to analyse whether nimodipine reduces cerebral dysfunction after transient global cerebral ischaemia, using our previously described experimental model’ which permits the impregnation of cerebral tissue during the periods of ischaemia and reperfusion. Some aspects of this study contribute to our understanding of the reperfusion injury concept Three groups of rats were used. Animals in Croup 1 (n = 13) served as normal controls for neurophysiological recordings. Rats in Croups 2 (n = 7) and 3 (n = 7) were subjected to global cerebral ischaemia and either isotonic saline (Croup 2) or nimodipine solution (Group 3; 40 μg/kg) was intra-arterially injected through the external carotid artery during ischaemia and reperfusion and distributed to the circle of Willis. Seventy-two hours after global cerebral ischaemia somatosensory evoked potentials were evaluated and P1 wave latency was used to compare the three groups of animals. The peak onset of this wave was 8.13 ± 1.5 msec, 18.63 ± 3.1 msec and 13.17 ± 2 msec for Croups 1, 2 and 3 respectively. P1 latency was significantly higher in Group 2 than in Groups 1 and 3 (p < 0.01). Histopathological findings showed that the level of injury in the hippocampus and striatum in Group 3 was more limited than in Group 2, although no statistical significance could be found. There was correlation between neurophysiological findings and the neuropathological damage observed in the striatum for Groups 2 and 3. It is concluded that the intra-arterial injection of nimodipine lessens brain damage caused by transient global cerebral ischaemia in rats. [Neurol Res 1993; 15: 395-400]