Addressing a gap in study of reciprocal relations between marital conflict and children, this project is concerned with examining children's agency, or intentional influence, in the marital relationship. The processes by which marital conflict leads to child agency, the impact of children's agency on marital conflict, and how children's agency relates to child adjustment or maladjustment over time are the focus of this project. Consistent with emotional security theory (EST, Davies & Cummings, 1994) and clinical theory and research, this research advances previous work by operationalizing and examining the independent contributions of agentic behavior, behavioral dysregulation, and perceptions of agency to subsequent marital conflict; examining links between behavioral responses, marital conflict, and mental health over time; and testing developmental change in children's behavioral responses to conflict. Participants were 236 families, tested in a three-wave, multi-method, multi-informant design. Results indicated that marital discord predicted high levels of children's agentic behavior and behavioral dysregulation, and that children's negative emotional reactivity fully mediated relations between marital discord and agentic behavior and behavioral dysregulation. Tests of effects on subsequent marital discord supported the notion that agentic behavior predicts low levels of subsequent marital discord. In contrast, behavioral dysregulation predicted high levels of subsequent discord, and perceived agency did not predict later marital functioning when tested in a model that included agentic behavior and behavioral dysregulation. Tests of correlations between children's concurrent behavioral responses, perceived agency, and mental health suggested that agentic behavior is positively associated with prosocial behavior, whereas behavioral dysregulation appears to be positively linked with adjustment problems, and perceived agency is unrelated to mental health. Longitudinal analyses were consistent with these relations, and revealed that behavioral dysregulation predicts subsequent adjustment problems, but that adjustment problems do not predict subsequent behavioral dysregulation. Moreover, agentic behavior was not linked with prosocial behavior over time. Finally, analyses of developmental change suggested that children's behavioral and emotional responses to conflict decrease with increasing age, and predictors of interindividual differences in change were inconclusive. Results are discussed in terms of the emotional security hypothesis and the functionalist perspective on emotions, with implications for research on family relations.