In the Amsterdam Growth and Health Longitudinal Study (AGAHLS), a cohort of about 400 boys and girls (mean age 13 years) were followed over a period of 20 years. Over that period repeated measurements were done of body dimensions (height, weight, skinfolds), physical fitness (eight motor performance field tests: plate tapping, bent arm hang, 10 x 5 m sprint, arm pull, sit and reach, standing high jump, 10 leg lifts, 12-min endurance run, and one laboratory test to measure maximal aerobic power), and physical activity (by a cross-check interview). Three research questions were studied: (1) Is there a positive relationship between adolescent fitness (age 13-17 years) and adult physical activity (age 33 years)? (2) Do physical fitness and physical activity track from adolescence into adulthood? (3) What is the longitudinal relationship between physical fitness and physical activity? Multiple linear regression analysis showed that of the 9 physical fitness tests, only the 12-min endurance run and the maximal aerobic power during adolescence are significant (P < 0.05) predictors of adult physical activity. The effects are not influenced by biological age but by sex: only in females are the predictions significant (P < 0.05) Tracking over the period of 20 years estimated from stability coefficients showed values for physical fitness varying between 0.83 (plate tapping) to 0.38 (standing high jump and maximal aerobic power). Physical activity shows lower stability coefficients (0.35-0.29). A longitudinal linear regression technique was used to analyse the relationship between physical activity and physical fitness over the 20-year period; in this analysis corrections were made for both time-dependent (time, biological age, and cardiovascular factors) and time-independent variables (sex). All physical fitness tests show positive and significant (P < 0.05) standardized regression coefficients with physical activity, but the explained variance is less than 1%. Only maximal aerobic power has a higher explained variance of 1.8%. It can be concluded that: (1) Physical fitness in adolescence is only weakly related to adult physical activity; (2) between age 13 and 33 years, physical activity has low stability and physical fitness was higher stability; and (3) the longitudinal relationships between physical fitness and physical activity are only meaningful with maximal aerobic power.