Little is known about the mechanisms of action that link perinatal risk and the development of aggressive behavior. The aim of this study was to examine whether perinatal risk and parenting interacted to specifically predict reactive aggression, as opposed to general aggressive behavior, and to examine cortisol reactivity as a mediator of this developmental risk process. In a community sample of 99 elementary school-aged children, prenatal risk was measured by a count of minor physical anomalies (MPAs), reactive aggression was measured by laboratory observations of aggression in response to provocation, and general aggression was measured by parent report. Cortisol reactivity was not found to mediate the association between MPAs and reactive aggression or general aggression. However, MPAs were found to interact with parenting behaviors to predict reactive aggression and general aggression, as well as cortisol reactivity. Specifically, as the deficits in parenting increased, MPAs became more strongly and positively associated with reactive aggressive and general aggressive outcomes. Similarly, in cases of poor parenting behaviors, MPAs were positively associated with higher cortisol reactivity. Implications for theory and prevention are discussed.