Dispersal patterns in a growing population of Przewalski’s horse (Equus ferus przewalskii) in the semi-reserve Pentezug (Hortobágy National Park/Hungary).
Zoolog. Institut, Universität Köln
In this thesis, patterns of natal and maternal dispersal in a growing population of Przewalski’s horses in the semi-reserve Pentezug / Hungary were investigated. Based on ethological observations of 19 focal horses (10 subadult males and 9 subadult females), paternity data of all of these horses and data about the population dynamics (1998-2006), proximate causes of male and female dispersal were identified. Dispersal patterns concerning dispersal ages and seasonality of dispersal were determined. On the basis of the results two hypotheses for the function of dispersal in phylogenesis were discussed: the inbreeding avoidance hypothesis and the resource competition hypothesis. It was shown that subadult horses of both sexes generally dispersed before first breeding in the first three years of their lives. The only exceptions were noted for the early years of the project, when the sex distribution in the population and the social structure were far from natural. At that time, only few adult stallions were present and all adult mares were organized in one big band, so that young females had no bands or suitable bachelor stallions they could disperse to. Under these circumstances three mares failed to disperse and remained in their natal bands for breeding.
Male and female dispersal occurred most frequently in summer and autumn. Females dispersed about half a year earlier than males at an age of about 20 months, although males were confronted with higher frequencies of agonistic behaviour by band members in their natal bands. Females left their bands without being forced to do so, whereas several males were expelled from their natal bands by the band stallion.
All dispersed females joined other bands; dispersed males joined the bachelor group.
The proximate causes of dispersal differed for males and females. Females left after the time of first oestrus, as their band stallions refused to mate with them. Matings with stallions from outside the band were prevented by the band stallion, so that females had to leave the natal band to reproduce. Only in three cases females mated before dispersal. In all three cases they did so with a stallion from outside the band.
Male dispersal occurred due to a high level of aggression by both, adult mares and the band stallion of the natal band. It was noted that males dispersed later when peers were present. The assumption of different authors, that this delay of dispersal might be due to the availability of playmates in the natal band, could not be confirmed for the males in this population. Although most of the focal males had peers in their bands, none of them was observed playing. As soon as these males dispersed to the bachelor group, they started to play with their peers and other stallions. It is assumed, that males in the natal bands are hindered from playing by the aggression of band members. The presence of peers effected that the aggression of band members was distributed to several individuals. If one of the males dispersed, the remaining one suffered more aggression than before. Additionally, criteria for the choice of a new band by dispersing females were analysed. Females tended to disperse to bands that were smaller than their original bands and contained fewer resident females. Dispersed females received little or no agonistic behaviour from their new band members; the only exception was Gréta, who joined the biggest band of the population together with her grandmother Sima.
The choice of a small band was more important than the choice of a band stallion, which was successful in fights. Whereas adult mares chose successful stallions when they changed bands in breeding dispersal, young females rather dispersed to less successful stallions with smaller bands.
Highly significant was the choice of an unfamiliar stallion as new band stallion: all dispersed females chose a stallion with which they had not been associated in the same band previously. In most cases the choice of an unfamiliar stallion involved the avoidance of inbreeding. Only in two cases females joined unfamiliar stallions, which were closely related to them (father and paternal half-brother). All dispersed females showed affiliative interactions with members of their new bands exclusively. The preffered partners were in one case the grandmother (Gréta) and in the other case the only foal of the band (Helka). The third female (Hóvirág) dispersed to a bachelor stallion and had no other partners for affiliative interactions than him.
The dispersal patterns found for this population of Przewalski’s horses indicate that both the inbreeding avoidance hypothesis and the resource competition hyothesis can explain the function of dispersal. Dispersal assured inbreeding avoidance as both females and males dispersed before first reproduction and females joined bands with unfamiliar stallions only. But for both females and males, resource competition appeared to be another crucial factor. Males avoided aggressive behaviour of band members by dispersing to the bachelor groups, and females dispersed before first foaling