In our study population, dispersal was most likely opportunistic, as indicated by the absence of spatial genetic structure and the lack of geographic clustering in the mtDNA haplotype network (Fig. 2, 3). There was no obvious difference in dispersal distances between sexes, as both mtDNA haplotype diversity and mean relatedness were similar in females and males, suggesting that both sexes migrated over varying distances. As geographic scale of our study was confined, with the maximum distance between home-range centers of only 3200 m, these results should be treated as preliminary. However, they are in line with the direct observation of dispersal and the kin structure of the study population. In one directly observed case of dispersal, a subadult male (the oldest offspring of Group 1), moved to an unoccupied area adjacent to his natal group and later formed a pair (Group 11) with an unknown female. The female did not have any first-degree kin among the sampled animals, indicating that she, unlike her mate, had not dispersed from any of the neighboring groups. The closest relative of this female was the adult male of Group 4 with r = 0.156 (corresponding to a relatedness level between unrelated and half-sibling), with whom she also shared the same mtDNA haplotype (Supplementary Table S1). The kinship structure of the study population (Fig. 1) further suggested that while some dispersing individuals stay in the area (indicated by adult first-degree kin occupying home ranges that are either adjacent or separated by 1–2 home ranges), others migrate further (as many individuals in the study area are unrelated).