A sample population of 67 males and 71 females with longitudinal lung function and other anthropometric measurements from all non-Hispanic white children in the Tucson Epidemiological Study of Airway Obstructive Diseases (AOD) was studied to evaluate biological determinants of the rate of lung growth. Groups within gender were defined by the following factors: 1) maximum height, 2) age at maximum forced expiratory volume at 1 second (FEV1), 3) % predicted initial FEV1 or FEV1/FVC ratio. Only groups defined by low initial function (FEV1 or FEV1/FVC less than or equal to 85% predicted versus greater than 85% predicted) showed statistically significant differences from those with more "normal" function by comparing their maximum % of predicted FEV1s and FEV1/FVC ratios. The longitudinal FEV1 data by age for the latter groups were characterized by a mathematical model (polynomial smoothing spline) yielding optimal fitted curves and an estimate of each group's growth velocity curve. The statistical comparisons between these fitted curves indicate that subjects with low initial pulmonary function continued to have significantly lower FEV1 values for males older than 13.12 years and for females between 8.23 and 15.3 years. At post-hoc analysis persistent wheezing was more likely in the initially more impaired group. Disease at the end of follow-up was not related to initial functional status.