Epidemiologic investigations into the natural history of airway obstructive disease have focused on both the growth and decline of lung function measurements. Although a general picture has emerged as to overall patterns of growth/decline, uncertainty remains with regard to important details of these patterns. Pulmonary function data from subjects aged 5 yr and and older who participated in any or all of the first 10 annual surveys of the East Boston population-based study have been used to investigate patterns of growth and decline of lung function derived from maximal forced expiratory maneuvers. At each annual survey, subjects completed a standardized respiratory illness and smoking questionnaire and provided forced expiratory volume-time curves from which forced vital capacity (FVC) and forced expiratory volume in one second (FEV1) were derived. A nonparametric curve smoothing method was used to provide an overall summary of growth and decline of each measure of function. Asymptomatic, nonsmoking males demonstrated a pattern of growth of FEV1 that included a prolonged plateau phase or period of slow, continued growth from ages 23 to 35 yr. Decline in lung function began after this period and occurred in two phases that averaged about -20 and -30 ml/yr. In contrast, no plateau phase was observed for male current smokers. In these smokers, decline in FEV1 began in the early part of the third decade at a rate only slightly greater than that observed for nonsmokers (-25 to -30 ml/yr). Similar patterns and rates of decline were observed for females, although the presence of a plateau phase was less clearly discernible. These data suggest that a major effect of cigarette smoking on lung function decline involves the premature onset of a "normal" rate of decline in function and, to a lesser extent, more rapid rates of decline later in life and that the pattern and magnitude of decline is similar in males and females.