Objective: To present data on neurologic recovery gathered by the Model Spinal Cord Injury (SCI) Systems over a 10-year period. Design: Case series. Setting: Twenty-one Model SCI Systems. Patients: A total of 3,585 individuals with traumatic SCI admitted between January 1, 1988 and December 31, 1997. Main Outcome Measures: Neurologic impairment category; Frankel grade; American Spinal Injury Association (ASIA) Impairment Scale (AIS) grade; motor score. Results: SCI caused by violence is more likely than SCI from nonviolent etiologies to result in a complete injury. Changes in severity of injury were similar using the older Frankel scale and the newer ASIA Impairment Scale. Individuals who were motor-complete with extended zones of sensory preservation but without sacral sparing were less likely to convert to motor-incomplete status than those with sacral sparing (13.3% vs 53.6%; p < .001). Motor score improvements at 1 year were related to severity of injury, with greater increases for better AIS grades except grade D, because of ceiling effects. Individuals with AIS grade B injuries have a mixed prognosis. Conclusion: Neurologic recovery after SCI is influenced by etiology and severity of injury. Multicenter studies on prognostic features such as preserved pin sensation in grade B injuries may identify subgroups with similar recovery patterns. Identification of such groups would facilitate clinical trials for neurologic recovery in acute SCI.