The significance of sensory changes determined by pinprick and light touch in individuals with a herniated lumbar disc has been questioned. Discrepancies may be related to the subjectiveness of the test, failure to use dermatome-specific testing sites, overlap of areas that are innervated by different nerve roots, anatomical variations, or lack of sensitivity of the testing technique. For this study, we assessed the results of sensory examinations of twenty-five patients with documented herniation of a lumbar disc. The examinations were done using Semmes-Weinstein monofilaments, vibrometry, pinprick, and light touch in the autonomous skin areas supplied by the fourth and fifth lumbar and first sacral-nerve roots. Right-left differences in Semmes-Weinstein pressure thresholds of more than fifteen milligrams per square millimeter enabled us to localize disc lesions to a specific root in 100 per cent of patients and differences in vibratory thresholds of more than 3.5 micrometers, to localize the correct level in 88 per cent. Lesser differences in thresholds did not help to identify the involved root. The mean sensory threshold on the side of the disc lesion was found to be significantly greater than that on the opposite side by both vibrometry and pressure aesthesiometry (p less than 0.005). These findings were not duplicated using light touch or pinprick testing. Even with the most sophisticated sensibility-testing techniques, correct identification of the nerve root that was compressed by a herniated lumbar disc was accurate in only 50 per cent of patients.