Fig. 5. Evoked potentials. (A; from top to bottom) Averaged traces of evoked potentials are shown, all obtained from the same animal, from the pre-injury electrical recording to those obtained 1 month post-injury. Note the complete absence of somatosensory evoked potentials (SSEPs) following spinal cord injury. This was characteristic in 100 % of the control population at all times tested. A median nerve control stimulation was also carried out at these times, but only the 1 month recording is shown in the bottom trace, the arrow pointing to a strong early-arriving evoked potential (for an explanation of P1 and P2, refer to Fig. 3). (B) Recovery of SSEP conduction in a PEG-treated animal. The characteristic double SSEP peaks are present in the uninjured animal. Note the complete loss of these peaks following injury and the positive median nerve control procedure carried out at this time point. One day post-injury to 1 month post-injury recordings show the recovery of SSEP conduction. The dotted line marks the approximate peak magnitude of the early-arriving SSEP. Note that the latency to peak contraction is reduced over time (refer to Fig. 6). Such recovering SSEPs were characteristic of 100 % of the PEG-treated animals and are in contrast to the complete absence of such conduction in all control animals.