The deleterious effects of marital conflict on youth outcomes are well-documented in both cross-sectional and longitudinal studies. To date, longitudinal studies have focused on repeated measures of youths' outcomes and the temporal dynamics of marital conflict have largely been ignored. Marital conflict changes over time as contextual and relationship characteristics change, and these patterns of change may provide unique predictive power in accounting for differences in youth outcomes. This study provides a novel exploration of an old idea by focusing on dynamic patterns of marital conflict in predicting trajectories of adolescents' adjustment. All variables were measured at ages 16, 17, and 18 with 252 adolescents (53% female) enrolled in the longitudinal Family Stress and Youth Development Study. Latent growth curve models with latent variable interactions were used to determine whether marital conflict at age 16 (intercept), change over time in marital conflict (slope), and the intercept--slope interaction predicted change over time in adolescent internalizing and externalizing symptoms and levels of internalizing and externalizing symptoms at age 18. Youth exposed to high and increasing levels of marital conflict reported high and stable levels of internalizing and externalizing symptoms across adolescence. Adolescents exposed to low and decreasing levels of marital conflict had consistently fewer symptoms. Furthermore, exposure to initially low but increasing levels of marital conflict was associated with increases in problems across adolescence, which contrasted with findings for youth with initially high marital conflict exposure that decreased over time. Findings are discussed in relation to both conceptual and methodological advances.