This study examined longitudinal relationships between depressive symptoms and use of alcohol, cigarettes, and illicit substances among adolescents, addressing methodological limitations and potential confounding in the extant literature. The sample comprised adolescents who were surveyed in Grades 6 (n = 916), 9 (n = 804), and 11 (n = 791). Cross-lagged path analyses were used to examine bi-directional relationships between substance use and depressive symptoms. Early adolescent depressive symptoms predicted mid-adolescent cigarette, alcohol, and illicit substance use (the latter among females only). Mid-adolescent depressive symptoms predicted late adolescent illicit substance use (females only). There were no statistically significant pathways from substance use to later depressive symptoms. However, these relationships were reduced to non-significance with the addition of covariates associated with the family environment, school, and individual. The findings suggest that the association between adolescent depressive symptoms and later substance use can be explained by common risk factors that produce vulnerability to both depression and substance use.