There is an adverse cross-sectional association between sedentary time, physical activity (PA) and adiposity, but weak and inconsistent estimates raise question to the direction of associations. The present study aims to examine whether the prospective association between sedentary time, different PA intensities and indicators of adiposity is bi-directional. The Active Smarter Kids Study obtained data from 869 ten-year-old children with valid measurements for sedentary time, PA, and adiposity at baseline and follow-up. Time spent sedentary and PA was measured by accelerometry, adiposity was assessed by three different measures: body mass index (BMI), waist circumference (WC) and sum of four skinfolds (S4SF). Neither overall PA nor time spent sedentary predicted lower BMI or WC at follow-up, but the time spent in moderate-and-vigorous PA (MVPA) and vigorous PA (VPA) predicted lower S4SF at follow-up among boys (MVPA β - 0.066 [95% CI -0.105, -0.027] p = 0.001). Baseline BMI and WC predicted less overall PA, MVPA and VPA in boys. All adiposity measures predicted more time spent sedentary at follow-up in boys. The results suggest that overall PA and sedentary time do not predict future adiposity. Baseline adiposity may rather predict more sedentary time and less higher intensity activity.