Search Results

Sigurdur Snorrason, Hrefna Sigurjónsdóttir, Anna Thórhallsdóttir and Machteld van Dierendonck


1. The social relationships in a group of Icelandic horses without a mature stallion were studied. The horses were all familiar to each other. Mutual grooming and play relationships, spatial associations, dominance-subordinate relations and the effect of kinship on these relationships were analysed. 2. The social structure was clearly dominated by the behaviour of the adult mares. The horses preferred to form bonds within their social class (sex/age) and they kept close proximity with their friends. The group was effectively divided into two social subgroups, adult mares as one group and adult geldings and sub-adults as another group. The sub-adults and adult geldings formed associations, which were based on mutual grooming and play, while the adult mares did not play. Differences between the sexes were evident. Males played more than the females, had more playing partners and were more popular as playmates. 3. Aggression rates were low. The dominance hierarchy was linear. Adult mares ranked higher than adult geldings, sub-adults and the foals. Rank was significantly correlated with age. The closer the adult mares were in rank, the more they groomed with each other. Such relationships were not found amongst the other social group. 4. Kinship was calculated between all pairs of animals for up to 4 or 5 generations. Allogrooming and play frequencies and proximity were all positively correlated with kinship. Adult mares, which were close in the dominance hierarchy, were on average more related than those further apart. 5. The social relationships in the Icelandic herd were, to some extent, different from relationships reported from unmanaged and feral horse-herds with mature stallions and bachelors. Our results suggest that adult mares groom more in groups without a stallion. Furthermore, they have more preferred partners than in natural harems and their partners are other adult mares, not their weaned offspring as seems to be the case in feral herds. The sub-adults also seem to be more socially active in the absence of stallions. Interestingly, in the Icelandic group, the adult mares showed stallion like behaviours, like mounting and protecting foals. Only by studying the behaviour and the nature of the relationships of horses in groups of different compositions, can we expect to gain a comprehensive understanding about individual social strategies and cognitive capabilities of the species. Such knowledge is valuable for management and welfare of the horse.

Amos Bouskila, Emmanuel Lourie, Shiri Sommer, Han de Vries, Zef M. Hermans and Machteld van Dierendonck

Relatedness is likely to affect the decisions of animals regarding their affiliations with conspecifics. Social network analysis provides tools to describe the social structure of animals. Here, we investigate the social network of a population of 27 unmanaged Konik horses in the Blauwe Kamer Nature Reserve, in the Netherlands. We test three hypotheses: (1) that related individuals will have stronger associations; (2) that individuals with low values of average relatedness to their neighbors in the network will have more links and (3) homophily, the tendency of individuals to associate with similar others, will lead to stronger associations among individuals of similar sex, reproductive state, age and rank in the social network. We videotaped 22 horses (excluding foals) and their interactions. Relatedness was calculated from the pedigree, which was based on parentage, determined by DNA analysis. The social network was based on spatial proximity data. There was no significant influence of relatedness on strength of associations in the network or an influence of age- or rank-homophily. We argue that the lack of a relatedness effect is not likely to have been caused by an inability to detect kinship. Strength of associations in the social network was significantly affected by the tendency of the horses to associate with individuals of the same sex and the same reproductive state. This social network pattern is not common in mammals, and the study of unexplained variation in choice and strength of associations may have important implications for other equids increasingly confined to reserves worldwide.