We studied the white residents of Lebanon, Connecticut, a rural community, cross-sectionally in 1972 and a follow-up survey in 1978. In this community, we examined prospectively the growth and decay of lung function and the prevalence, attack, and remission of respiratory symptoms in 1,262 persons. These longitudinal findings were related to sex, age, and smoking habits. Respiratory symptoms and 3 lung function variables were examined: forced expiratory volume in one second (FEV1) and maximal expiratory flow rates at 50 and 25% of vital capacity (Vmax50 and Vmax25). The representativeness of the follow-up population was examined by an extensive analysis of the respondents and nonrespondents based on their cross-sectional characteristics. These analyses did not, in general, reveal any consistent results indicating that a healthier population was followed. The longitudinal data showed that lung growth continued into early adulthood (25 to 34 yr of age) with a leveling off and decline beginning at 35 yr and older. This is in contrast to cross-sectional studies of growth and decay of lung function, which show less growth among the young and more decay among adults. The rates of growth in both males and females were affected by their smoking status, with smokers having less growth and greater decline than nonsmokers. In particular, lung function growth was impaired in young smokers as early as 15 to 24 yr of age, and an actual decline was seen in the females of this age group.