The cortical somatosensory evoked potentials (CSEPs) were recorded to determine if the administration of nimodipine improves axonal function after spinal cord injury. Animals receiving a 52 g compression injury (a moderately severe injury) for 5 minutes were randomly allocated to one of five treatment groups. Each group was given an infusion of one of the following nimodipine regiments over 2 hours, commencing 1 hour before compression: placebo (n = 20), 0.5 micrograms/kg (n = 10), 0.25 micrograms/kg (n = 20), 0.125 micrograms/kg (n = 10), and 0.25 micrograms/kg + Hetstarch (n = 10). In the control group, 65% of animals lost the CSEPs immediately after the injury with almost all (95%) of these regaining the CSEPs within 15 minutes after decompression of the spinal cord. In the treated groups, the rate of the CSEP loss was highest in the 0.5 micrograms/kg group. This group also had the lowest CSEP recovery. The proportion of the CSEP loss was essentially the same for the other nimodipine-treated groups, although it seemed that there was an increasing number of nonresponses with increasing the nimodipine dose. Our data indicate lack of any beneficial effects of nimodipine on axonal function as measured by evoked activities in experimental spinal cord injury.