The present study was undertaken to determine whether, in patients with a focal nerve lesion, the axonal hyperpolarization produced by conduction of brief trains of impulses would result in conduction block in cutaneous afferents, thus indicating a site of impaired safety margin for impulse transmission. In 25 patients with focal conduction slowing across the carpal tunnel segment of the median nerve, a conditioning train of 10 supramaximal stimuli at 200 Hz resulted in a reduction in amplitude and an increase in latency of the test volley set up by a supramaximal stimulus. These changes exceeded those seen in control subjects, but followed a similar time course, with full recovery within 150 ms. There was a significant correlation between these changes and the severity of the compression neuropathy as indicated by the degree of focal conduction slowing in routine nerve conduction studies. Control data suggested that the measured changes in amplitude could be explained by temporal dispersion of the compound sensory volley. This view was supported by measurements of the changes in amplitude (and latency) in normal subjects during acute compression before conduction block had developed. In addition, there were similar linear relationships between the activity-dependent amplitude reduction and the corresponding change in latency for both the patients and the control subjects, indicating that there was no need to invoke factors additional to those operating in the control subjects to explain the greater amplitude depression in the patients. It is concluded that, although the depression in amplitude was greater in patients than in healthy subjects, the magnitude of this change can be explained by temporal dispersion of the abnormal compound sensory action potential associated with greater conduction slowing. Activity-dependent conduction block may play little role in the pathophysiology of carpal tunnel syndrome.