We investigated the social effects of the removal of two adult females, the founder of a matriline and a non-reproductive female, in a group of captive Campbell’s monkeys (Cercopithecus campbelli campbelli). Social behaviour in the group was recorded one month before, one week after, and seven months after removal of the females. Each observation period lasted 12 days. We collected data on vocal (contact calls) and non-vocal (affiliative, play, avoidance, aggression, directed gaze) interactions using focal animal sampling. We also collected nearest neighbour data using instantaneous scan sampling. The effects of removal on: (1) the global frequency of interactions for each behavioural category (Wilcoxon tests); (2) the networks of significant preferential partners (drawing sociograms based on Chi-square and Binomial tests); (3) the hierarchy of dominance (Zumpe & Michael’s index), were analysed. Before removal, social interactions were mainly affiliative and occurred mostly within matrilines. Moreover, play was rare and no kin bias was observed in the hierarchy. Immediately following removal, social interactions (play in particular), increased among matrilines and this increase was still present seven months later. The role of the younger individuals was important in this increase of play among matrilines. Moreover, following removal, a clear inter-matriline dominance emerged in the hierarchy as members of the disturbed matriline received aggression more often than members of the other matriline. The results of this study add to the existing literature on the effects of the removal of adult females in primates. The removal would have to be planned according to species and social networks. Individuals who are less involved in social interactions can probably be removed without too much effect on the group but removing a mother can expose her progeny to aggression. The presence of younger animals however may favour the maintenance of stable social networks.