Social interaction and dispersal patterns of golden snub-nosed monkeys (Rhinopithecus roxellana) living in multi-level societies

This study aimed to explore the benefits that individuals gain from group living and the role of kin and nonkin affiliation and cooperation in the formation of social networks in primates by investigating the multi-level social structures exhibited by Sichuan snub-nosed monkeys (Rhinopithecus roxellana). The multi-level social network of snub-nosed monkeys composed of over 100 individuals, in which individuals form one-male breeding units (OMUs, which include one adult male, several adult females, and their offspring), all male units (AMUs), and bands (several OMUs that travel, feed and rest together). Given the fact that the majority of Asian colobines exhibit a harem social organization, multi-level societies of R. roxellana are proposed to have evolved through the aggregation of individual one-male groups. The specific objectives of this study are to explore 1) the social factors that help to maintain the stability of multilevel societies, 2) the benefits to individuals of forming a higher level social structure, 3) the presence and complexity of kinship networks and dispersal patterns in R. roxellana based on genetic data, and 4) the behavioral mechanisms regulating social interactions within multi-level social networks, and whether these are most consistent with kin selection theory, reciprocity theories, or biological market theory. Behavioral observations for this study were conducted at Zhouzhi National Natural Reserve, Shaanxi, China. A habituated band of snub-nosed monkeys was followed from September 2007 to August 2008. Along with behavioral observations, fecal samples were collected from the focal band and two neighboring bands. DNA was extracted from the fecal samples. The d-loop region of the mitochondrial DNA was amplified and sequenced for each sample. The behavioral data indicate that OMUs were socially and sexual independent since the majority of social and sexual interactions were restricted to members of the same OMU. Both direct affiliative and agonistic interactions between members of different OMUs were infrequent. Compared to the harems formed by other Asian colobines, the OMUs of R. roxellana were more cohesive. Leader males played a critical role in maintaining the cohesion of his OMU by actively threatening or chasing both adult and juvenile members of other OMUs that were within 5 meters of his harem. It is likely that the formation of multi-level societies in R. roxellana is the result of social and spatial tolerance among harem males in response to the foraging requirements associated with the exploitation of highly seasonal and low productive habitat. Three distinct haplotypes were found among 99 samples collected from the three neighboring bands. Based on the assumption that individuals with less frequent haplotypes represent immigrants from other bands, it was estimated that approximately 17-21% of females and 8-15% of males immigrated from neighboring bands. The genetic data also indicated that females transfer between OMUs within the same band since females with the same haplotype were present in different OMUs. In other words, OMUs appear to be not grouped by maternal lineages. Behavioral data suggested that female choice played a critical role in dispersal decision because females were voluntarily leave their natal OMUs. Within OMUs, both kin and non-kin dyads formed long-term grooming partners. Therefore, kin selection is not sufficient to explain the formation of social bonds in snub-nosed monkeys. In contrast to the prediction of biological market theory, dyads did not evenly exchange grooming within short-time frames such as within bouts or during the same day. There is no evidence that females or males exchanged grooming for sex. However, dyads balanced grooming exchanged over lone-time period. In conclusion, long-term reciprocity appears to offer the strongest explanation for the social interactions of snub-nosed monkeys