Abstract
In social species, costs and benefits of group living are numerous and dynamic, and the balance between them is a key determinant of group size. Group size is a component of sociality which can have important effects on the behaviour of individuals, their demographic parameters and their social dynamic (intergroup movements of individuals, associations between groups). However, these effects remain poorly understood. In this thesis, I explore the effect of group size on the behaviour, demography and social dynamic of plains zebras Equus quagga in the wild populations of Hwange National Park (HNP, Zimbabwe) and Hluhluwe-iMfolozi Park (HiP, South Africa). In the first chapter, I show that, in HNP, group size has little effect on individual vigilance and foraging, and has no significant effect on the survival of individuals, suggesting that dilution and detection effects do not matter behaviourally and demographically in this population. In the second chapter, I show that the probability of stallion turnover increases with the number of reproductive females in harems in HNP, but that these stallion turnovers do not affect foal survival. This suggests that, if infanticide by males occurs in wild plains zebras, this should be extremely rare and has no important consequences for foal survival at the population level. Finally, in the third chapter, I show that the number of reproductive females in harems has no effect on the probability for harems to be alone or with other harems. However, harems with many reproductive females seem to be found less often near bachelors (although significant uncertainty remains). This suggests that harems including many reproductive females may seek to avoid bachelors, and actually seem to have more success in doing it in HiP where vegetation is overall more closed than in HNP. Accordingly, the probability of stallion turnover does not increase with the number of reproductive females in harems in HiP. Overall, this thesis questions the functioning of the dilution effect and highlights the importance of considering the potential regulatory effect of social dynamics on populations, through their effect on the spatial distribution of groups.