The generation of knock-out and transgenic mice offers a promising approach to the identification of novel biochemical factors that contribute to persistent pain conditions. To take advantage of these mice, however, it is important to demonstrate that the traditional models of persistent pain, which were largely developed for studies in the rat, can be used in the mouse. Here, we combined behavioral and anatomical methods to characterize the pathophysiology of a partial nerve injury-evoked pain condition in the 'normal' mouse. In male C57BL6 mice we tied a tight ligature around 1/3 to 1/2 of the diameter of the sciatic nerve and evaluated the time-course and magnitude of the ensuing mechanical and thermal allodynia. We also used immunocytochemistry to analyze nerve injury-induced changes in substance P (SP) and NK-1 (SP) receptor expression in the spinal cord. As in the rat, partial nerve injury markedly decreased paw withdrawal thresholds to both mechanical and thermal stimuli on the injured side. We detected threshold changes one day after the injury. The thermal allodynia resolved by 49 days, but the mechanical allodynia persisted for the duration of the study (70 days). We found no changes contralateral to the nerve injury. Sympatholytic treatment with guanethidine significantly reduced both the thermal and mechanical allodynia. We observed a reduction of SP immunoreactivity in the superficial dorsal horn on the injured side at 7 and 14, but not at 3 or 70 days after the nerve injury, and we observed an increase of NK-1 receptor expression at 3, 7, 14 and 42, but not at 70 days after the injury. We conclude that partial injury to the sciatic nerve produces a comparable allodynia and neurochemical plasticity in the rat and mouse. These results establish a valuable model for future studies of the biochemical basis of neuropathic pain in mice with specific gene modifications.