Rationale: Beyond the risks of smoking, there are limited data on factors associated with change in lung function over time. Objectives: To determine whether cardiorespiratory fitness was longitudinally associated with preservation of lung health. Methods: Prospective data were collected from 3,332 participants in the Coronary Artery Risk Development in Young Adults study aged 18-30 in 1985 who underwent treadmill exercise testing at baseline visit, and 2,735 participants with a second treadmill test 20 years later. The association between cardiorespiratory fitness and covariate adjusted decline in lung function was evaluated. Measurements and Main Results: Higher baseline fitness was associated with less decline in lung function. When adjusted for age, height, race-sex group, peak lung function, and years from peak lung function, each additional minute of treadmill duration was associated with 1.00 ml/yr less decline in FEV1 (P < 0.001) and 1.55 ml/yr less decline in FVC (P < 0.001). Greater decline in fitness was associated with greater annual decline in lung function. Each 1-minute decline in treadmill duration between baseline and Year 20 was associated with 2.54 ml/yr greater decline in FEV1 (P < 0.001) and 3.27 ml/yr greater decline in FVC (P < 0.001). Both sustaining higher and achieving relatively increased levels of fitness over 20 years were associated with preservation of lung health. Conclusions: Greater cardiopulmonary fitness in young adulthood, less decline in fitness from young adulthood to middle age, and achieving increased fitness from young adulthood to middle age are associated with less decline in lung health over time. Clinical trial registered with www.clinicaltrials.gov (NCT 00005130).