The chronic effects of arc welding on pulmonary function were examined for a cohort of 291 male electric arc welders, mean age 45.0 years. Subjects had an average of 21.3 years welding exposure, primarily at construction sites and power stations. None worked at shipyards or had any known exposures to asbestos (1332214). The subjects completed a questionnaire to obtain information on occupational history, smoking, and respiratory symptoms. Spirometric testing was performed. Chest X-rays were obtained and examined for evidence of asbestosis. Sixty five subjects had radiographic evidence of asbestosis and were excluded from further analysis. Approximately 81% of the subjects were current or former smokers. The overall prevalence rates of chronic bronchitis and asthma were 20 and 11.3%, respectively. The prevalence of chronic bronchitis and asthma in nonsmoking welders was 23 and 13%, respectively. The mean values of forced vital capacity (FVC), 1 second forced expiratory volume (FEV1), midflow (FEF25-75), and terminal flow (FEF75-85) were 101.0, 98.7, 94.0, and 91.8%, respectively. To test for chronic welding effects the data were analyzed by logistic regression techniques using years of welding as the independent variable and FVC, FEV1, FEF25-75, and FEF75-85 as dependent values. The technique predicted that for all welders, 40 years of welding exposure would reduce FVC to 95.2% of the predicted value, FEV1 to 92.2% of normal, FEF25-75 to 79.2% of normal, and FEF75-85 to 81.3% of predicted. For nonsmoking welders, 40 years of welding exposure would not affect FVC but would reduce FEV1 to 93.3% of predicted, FEF25-75 to 77.6% of predicted, and FEF75-85 to 62.0% of predicted.