Child maltreatment increases the risk for impaired social functioning and cortisol regulation. However, the longitudinal interplay among these factors is still unclear. This study aimed to shed light on the effect of maltreatment on social functioning and cortisol regulation over time. The sample consisted of 236 children (mean age 7.64 years, SD = 1.36; 125 maltreated children and 111 nonmaltreated children, 128 boys and 108 girls) who attended a week-long summer camp for 2 consecutive years. Saliva was collected during 5 days at 9:00 a.m. and 4:00 p.m. Means of morning and afternoon cortisol levels and cortisol change (difference between morning and afternoon levels, controlled for morning levels) were used to group the children into low-, medium-, and high-cortisol groups. Prosocial, disruptive/aggressive, and withdrawn behaviors were assessed using information from peers and counselors. Maltreated children showed less prosocial and more disruptive/aggressive and withdrawn behavior. Results of structural equation modeling analyses indicated that there were indirect effects of maltreatment on Year 2 morning cortisol via prosocial and disruptive/aggressive behavior: Lower levels of prosocial behavior and higher levels of disruptive/aggressive behavior were related to lower morning cortisol levels 1 year later. Withdrawn behavior was related to higher afternoon cortisol values 1 year later. Results of this study suggest that maltreated children are more likely to experience difficulties in social functioning, which in turn is related to cortisol regulation 1 year later. This altered hypothalamic-pituitary-adrenocortical (HPA) axis functioning may put children at risk for later psychopathology. (Contains 5 figures, 5 tables, and 1 footnote.)