This report describes a collaborative study conducted in Montreal, Canada, Portland, Ore., and Winnipeg, Canada, to establish the relationship between the single-breath N2 test and age, sex, and smoking and to determine the prevalence of functional abnormalities in these populations. In nonsmokers, age-related regressions for closing volume, closing capacity, and the slope of phase III obtained from the single-breath N2 test, plus the ratio of the I-s forced expiratory volume to the forced vital capacity had very similar slopes, suggesting that differences in geographic location, climate, air pollution, and occupation had no effect on lung function detectable by these tests. Among the 6 city/six groups there was no systematic difference in the prevalence of functional abnormalities between the cities, but closing capacity expressed as a percentage of total lung capacity was abnormal most often in men and the slope of the alveolar plateau was abnormal most often in women. The prevalence of respiratory symptoms within different smoking categories was similar in the 3 cities. Although the number of cigarettes smoked had a significant effect on every test except the ratio of the I-s forced expiratory volume to forced vital capacity in men, the effect of age was considerably greater than the effect of smoking, and the dose-response relationship was weak. We conclude that additional factors may interact with smoking to place a smoker at risk of developing chronic airflow limitation.