Peripheral nerve biopsies when processed with conventional techniques for paraffin embedding usually do not provide sufficient data for the diagnostic conclusion. However, if the nerve is processed for resin embedding for semi and ultra-thin sections and teasing of fibres, several aspects can be analysed including quantitative and morphometric data. We studied the sural nerve biopsy of 78 patients examined at the Antonio Pedro University Hospital, Niterói RJ, applying those techniques and we found that in 55 cases (70.5%) the pathologic diagnosis was conclusive, in 11 (14.1%) although the nerve had abnormalities it was not possible to establish a diagnosis, and in 12 (15.4%) the nerve was normal. In 68 cases there was a clinical diagnosis which was confirmed in 49 but not in the remaining 19, since 8 had non-specific changes and 11 were normal. From the 10 cases which did not have a clinical diagnosis the biopsy was conclusive in 6, showed non-specific changes in 4, and was normal in 1 case. The pathologic conclusion in most of our cases was possible because not only we had the clinical data but all the nerves were processed for resin embedding.