Original article Scarcity in the prey community yields anti-predator benefits Matt W. Hayward * Mammal Research Institute, Polish Academy of Science, ul. Waszkiewicza 1, 17-230 Bialowie _ za, Poland 1 Centre for African Conservation Ecology, Nelson Mandela Metropolitan University, PO Box 77000, Port Elizabeth 6031, South Africa School of Biological, Earth and Environmental Science, University of New South Wales, Sydney 2052, Australia article info Article history: Received 28 July 2010 Accepted 16 March 2011 Available online 8 April 2011 Keywords: African large carnivores Density-dependence Predation Predator avoidance strategy Predatoreprey interactions Prey switching Top-down limitation abstract The majority of individuals in a community belong to a small number of abundant species. Under- standing why some species are rare and others are common has been a long-held goal for ecologists. Africa’s large carnivore guild preferentially preys on a small number of species within a limited weight range. Within this weight range however, some species that are expected to be significantly preferred as prey are not. I tested whether these species avoid preferential predation through their low densities. Records of over 40,000 kills from up to 48 different communities were used to test if non-preferred species within the expected prey weight ranges of each large predator avoid preferential predation and why. Species expected to be prey of Africa’s large predators based on their body mass, that are preferred are preyed upon significantly more frequently at low densities than non-preferred prey. This results in a negative relationship between relative abundance and preference for preferred prey, but a positive relationship for non-preferred prey. The non-preferred prey species that are within the expected prey weight ranges of Africa’s large predators are significantly less abundant within the prey community than significantly preferred prey. Rarity in African ungulates may convey an anti-predator benefit in that it was suboptimal for predators to evolve morphological or behavioral strategies to optimally forage on them or in that prey species can avoid predators by existing in habitats with low carrying capacity. Ó 2011 Elsevier Masson SAS. All rights reserved. 1. Introduction In animal and plant communities, the majority of individuals belong to a small number of abundant species, while most species are represented by a small number of individuals (Hughes, 1986; May, 1975). Understanding why some species are rare and others common within local communities has long been a goal of ecolo- gists (Gaston, 1994) and there is a growing body of evidence to suggest that naturally rare species have traits that differ from naturally common species (Kunin and Gaston, 1993). Predators and their prey are often considered to be in an evolutionary arms race to eat or be eaten (Dawkins and Krebs,1979). This analogy implies that predators evolve improved abilities to capture prey, which leads to prey evolving abilities to avoid capture (Abrams, 1986). Theoretical and mathematical models do not support this analogy however (Abrams, 1986; Brodie and Brodie, 1999). One reason for this may be the presence of secondary prey species that provide a buffer to predators in times of primary prey shortages (Pech et al., 1995). Primary prey are likely to be those that a predator has evolved to preferentially and optimally prey upon, whereas secondary prey are likely to be prey species that a predator has not evolved to preferentially prey upon but rather are preyed upon during prey switching (Garrott et al., 2007). Predators exhibit two distinct preference categories of prey: preferred prey species and the preferred weight range of prey (Hayward et al., 2006a). Preferred prey species are those preyed upon by predators significantly more frequently than expected based on the relative abundance of that prey species in the prey community (akin to primary prey). The preferred weight range is a range of prey body masses that a particular predator preys upon more frequently than expected (hereafter termed ‘expected prey’). Amongst this group of species that are expected to be preyed upon, there are species that are significantly preferred (primary prey), but also species that are not preferentially killed by predators (secondary prey). One hypothesis for the presence of these non- preferred prey species within the expected weight range is that they are intrinsically rare in the prey community, which has meant that predators have not evolved morphological or behavioral adaptations that allow preferential predation (Hayward, 2009). Species outside the expected prey weight range are not preferred * Present address: Australian Wildlife Conservancy, PO Box 432, Nichol’s Point 3501, Australia. Tel.: þ613 50245859; fax: þ613 50271200. E-mail address: hayers111@aol.com. 1 Tel.: þ4885 6827781; fax: þ4885 6827752. Contents lists available at ScienceDirect Acta Oecologica journal homepage: www.elsevier.com/locate/actoec 1146-609X/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.actao.2011.03.003 Acta Oecologica 37 (2011) 314e320