Fission and troop size in a mountain baboon population

Baboon, Papio cynocephalus, populations show a circumscribed range of troop sizes, dependent on features of the habitat that set limits on the ability of animals to balance their time budgets. To assess the proposal that a rising probability of troop fission is the mechanism underlying this it is first shown that (1) the distribution of troop sizes in two very different baboon populations, in the Drakensberg and at Amboseli, are not described by the null model of random allocation of individuals to troops (the Poisson process); and (2) data on troops undergoing or attempting fission support the hypothesis that the probability of fission increases with increasing troop size. A model was constructed that generated a distribution of troop size from an underlying, 'pre-fission' distribution by varying two parameters: the probability that a troop of given size will split, and the relative sizes of the two daughter troops that result if fission occurs. Both the Drakensberg and the Amboseli populations were adequately described by the model and it is concluded that the distribution of troop size in both populations is governed by fission in the manner proposed by Dunbar (1992, Behav. Ecol. Sociobiol., 33, 35-49). The model also indicates that, relative to Amboseli, the probability of fission in the Drakensberg is very high, with all troops of more than 28 members eventually splitting. The hypothesis that low food availability accelerates fission in mountain baboons through the decreased foraging efficiency of larger troops was tested, Troops of a wide range of sizes allocated very similar amounts of time to foraging, foraged with the same efficiency, and travelled approximately the same distances to do so. It is concluded that fission in mountain baboons, being exacerbated neither by predation nor by inefficient foraging, may be caused directly by the inability of individuals to service social relationships in growing troops in the face of small amounts of time available for social interaction. (C) 1997 The Association for the Study of Animal Behaviour.