Objective: Physical activity appears effective in reducing the risk of depression in adults however associations between physical activity and depression in youth remain unclear. We have previously demonstrated physical activity domain specificity in associations between physical activity and depression in young adults participating in the Childhood Determinants of Adult Health study. The current analysis examined associations between physical activity in different domains (active commuting, school physical education, discretionary sport) and depressed mood when cohort participants were children and adolescents. Method: Participants were 6070 school-children (aged 9–15 years) participating in a national health and fitness survey in 1985. Physical activity and depressed mood were assessed via self-report. Analyses were conducted separately by gender and school-level, and prevalence ratios adjusted for scholastic level, health status, smoking, alcohol consumption, residential arrangements and familial language. Results: Associations between physical activity and depressed mood were mixed however dose–response relationships were observed whereby increasing durations of total physical activity and discretionary sport in secondary boys were associated with decreasing prevalence of depressed mood (both Ptrend < 0.001). Further, increasing durations of physical education were associated with decreasing prevalence of depressed mood in primary girls (Ptrend = 0.02). No significant associations were observed for total physical activity or discretionary sport in primary and secondary girls. Associations were not mediated by overweight/obesity in girls. Conclusions: Associations between physical activity and depressed mood varied by gender, school-level and the domain in which activity was assessed. Further research utilizing more accurate assessment of childhood physical activity and depression is required to clarify potential associations.