In the last few years, many studies have been carried out concerning the effects of fumes from stainless steel (SS) welding on the health of welders. The respiratory effects of exposure to SS welding fumes have already been studied, but the results of lung function investigations have not been consistent. However, the main factor of risk for the welders' health seems to be related to the great concentration of chromium and nickel contained in fumes coming from SS welding. The aim of this study was to detect the chronic effects of SS welding exposure on pulmonary symptoms and ventilatory function tests. Respiratory symptoms and lung function tests were studied in 134 SS welders and 252 controls (C). Welders and controls were of similar average age, height, and duration in employment. The smoking habits of the groups were also similar. The medical questionnaire on respiratory symptoms was a version of the Medical Research Council questionnaire, modified by the British Occupational Hygiene Society. The flow-volume curves were performed with a calibrated pneumotachograph spirometer before each subject started working. After adjustment for tobacco habits, the SS welders presented a higher prevalence of bronchial irritative symptoms such as cough (P = 0.01) or sputum production (P = 0.02) than the controls. On the other hand, chronic bronchitis appeared to be significantly linked to tobacco consumption. The pulmonary function analysis underscored no significant difference between stainless steel welders and controls (forced expiratory volume in one second, observed/predicted: SS = 0.99 vs C = 0.98; maximal midexpiratory flow, observed/predicted: SS = 0.90 vs C = 0.92; maximal expiratory flow at 50 % of the forced vital capacity, observed/predicted: SS = 0.95 vs C = 0.95). On the other hand, by the mean of the two-ways analysis, a significant tobacco effect was found, without exposure or interaction of tobacco-exposure effects. There was no influence of the specific welding processes on the spirographic parameters, but a decrease in spirographic values after 25 years of welding activity was evident. The results of multiple regression indicated that age was not a confounding factor.