The effect of ethanol on recovery of neural conduction after spinal cord compression was evaluated in an isolated rat spinal cord preparation. Controlled compression of 50 to 75 percent of the cord cross-section was delivered using a piezoelectric translator. Postcompression compound action potential (CAP) amplitude, latency, and refractory periods were measured relative to pre-compression values. Recovery of CAP's was compared for spinal cords exposed to ethanol in vitro (100 mg/dl bath concentration, started one hour prior to compression) versus those maintained in normal artificial CSF. The in vitro effects of ethanol were evaluated on spinal cords from rats maintained on a normal diet and from those repeatedly intoxicated with ethanol for 15 days prior to the acute experiment. Compression of the cord resulted in an immediate 68% decrease in CAP peak amplitude and an increase in latency (171%) and refractory period (256%). In normal bathing medium, CAP amplitude recovered to 83% of pre-compression values 180 minutes after compression. The addition of ethanol to the artificial CSF did not directly affect CAP parameters, but combined with compression, CAP amplitude recovered to only 42% of pre-compression values 180 minutes after impact (p less than .01). Recovery was less affected by acute ethanol exposure in cords from ethanol pretreated animals. CAP amplitude recovered to 83% of pre-compression levels and was not different from compression-only recovery (p less than 0.10). The data suggest that direct effects of ethanol on axonal membranes may affect the sensitivity of axons to mechanical trauma or their capacity to recover normal function. Since spinal cords from repeatedly exposed animals are less sensitive to the acute effects of ethanol, ethanol may be acting to "fluidize" the axonal membrane.