Spirometry was performed on 165 subjects with spinal cord injury (84 with quadriplegia and 81 with paraplegia). Subjects were characterized by level of lesion as: high quadriplegia (HQ, C4 and above not requiring mechanical ventilation), low quadriplegia (LQ, C5-8), high paraplegia (HP, T1-7), and low paraplegia (LP, T8-L3). Thirty-nine subjects had complete motor lesions, and 126 had incomplete motor lesions. Nonsmokers (54 with quadriplegia and 53 with paraplegia) were defined as those who had never smoked or those who had stopped smoking for 1 year. Current smokers (28 with quadriplegia and 28 with paraplegia) were defined as those who currently smoked cigarettes, cigars, and/or pipe or those who had quit for < or = 1 year. We found by linear regression analysis that forced vital capacity (FVC), forced expiratory volume in 1 s (FEV1), peak expiratory flow (PEF), and maximal voluntary ventilation (MVV) were inversely correlated with the level of injury (i.e., the higher the level of injury, the lower the parameter). In the HQ group, those with complete injury had significantly lower pulmonary function parameters than those with incomplete lesions. In contrast, there were no significant differences in pulmonary function parameters between complete and incomplete lesions in subjects in the LQ, HP, and LP groups. In the LQ, HP, and LP groups, the FEV1 and PEF were significantly lower in smokers than in nonsmokers. Thus, this study demonstrates the effects of completeness of injury and smoking on pulmonary function in a large group of subjects with spinal cord injury.