Although prospective studies find that CU traits increase risk for later aggression, children with CU traits but without aggression have been identified, implying that the incremental effect of CU traits may be evident only in the presence of aggression. We examined this hypothesis, seeking replication across same and cross-informant designs, and UK and Colombian samples. In a UK epidemiological cohort (WCHADS), CU traits and aggression were assessed at age 3.5, and aggression at 5.0 years by mother (N=687) and partner report (N=397). In a Colombian general population sample, CU traits were assessed at age 3.5 and aggression at ages 3.5 and 5.0 by mother report (N=220). Our analyses consistently showed associations between 3.5 years CU traits and 5.0 years aggression moderated by aggression at 3.5 years, with interaction terms in the UK, same informant, 𝛽=.07 p=.014 cross-informant, β=.14 p =.002 and in Colombia, β =.09 p=.128. In each case, as hypothesised, the interaction arose from a stronger association between CU traits and later aggression in children who were already aggressive. Our findings imply a major role for CU traits in the maintenance and amplification of already established aggression, and cast doubt on their contribution to its origins.