Differences in social behaviour between late pregnant, post-partum and barren mares in a herd of Icelandic horses

Document Type


Publication Date



Group dynamics and the social behaviour of mares were investigated in the periods before and after foaling during two non-consecutive years, in a herd of Icelandic horses consisting of adult mares and geldings and juvenile fillies, geldings and colts. The horses were observed 24 h a day for 5–6 weeks; i.e. a total of 1483 h. There were significant changes in the social behaviour and interactions of mares after foaling, compared both to their own behaviour pre-partum and to that of barren control mares. While the frequency of allogrooming dropped during the study period in all mares, it dropped significantly more in foaling than in control barren mares. However, while the absolute allogrooming frequency was strongly correlated with reproductive state, it was still remarkably consistent within an individual adult mare between the study years. After foaling, mares with foals separated off into a distinct subgroup, whereas barren control mares increased interaction with the rest of the herd; it is presumed that separation served to keep the foals at a safe distance from the more boisterous geldings and sub-adults; alternatively, it could have been the result of mutual attraction between the foals. In general, the linear dominance order was correlated strongly with age, and the top-ranking mares were older mares that had not yet begun a senescent loss of physical condition which resulted in a later slight drop down the hierarchy. There was no consistent relationship between dominance rank and the pattern of preferred recipients of affiliative interactions, while familiarity was a more important determining factor than kinship in the selection of a preferred partner for affiliative interactions. The preferred partner for proximity was often the same individual as that for allogrooming, suggesting that proximity was a result of an active process rather than of passive acceptance. The presence of adult geldings in the herd did not seem to change the social behaviour of mares as compared to that described for feral horses, and did not appear to negatively affect behaviour during parturition, mare-foal bonding or subsequent maternal care. In short, maintenance of breeding mares in a herd of mixed age and sex did not interfere with normal species-specific behaviour or with the development of the stable, long-lasting mare–mare bonds that are central to maintaining a stable herd. It is concluded that such a system could be a welfare-friendly alternative to typical modern husbandry systems, as long as adequate space is available.