The arctic fox (Alopex lagopus): an opportunistic specialist Bodil Elmhagen, Magnus Tannerfeldt, Paolo Verucci and Anders Angerbjo Èrn Department of Zoology, Stockholm University, S-106 91 Stockholm, Sweden (Accepted 7 July 1999) Abstract Reliable and abundant resources are likely to favour specialization, while unpredictable environmental variation should favour a generalist strategy. The rodent population cycles of northern latitudes can be seen as both predictable and unpredictable, depending on the scale in time and space. The arctic fox Alopex lagopus is an opportunistic carnivore, but paradoxically, it seems to function as a specialist on ¯uctuating rodent Arvicolinae populations in most inland areas. We have studied the dietary response of arctic foxes in Sweden during 5 years of varying abundance of Norwegian lemming Lemmus lemmus, and how these changes in¯uenced the reproductive success of the foxes. The arctic fox population on mainland Fennoscandia is threatened by extinction and the situation has deteriorated during the 1980s and 1990s because of an absence of lemming peaks. Our results showed that in all years, lemming was the main prey for arctic foxes, with 85% frequency of occurrence in summer faeces (scats). Bird remains (mainly Passeriformes) were present in 34% of the scats, reindeer Rangifer tarandus in 21%, voles and shrews in 4% and hares Lepus timidus in 2% of the scats. The occurrences of lemming, bird and larger mammal (reindeer and hare) remains in the scats varied signi®cantly between years. Temporal variations within summer seasons and dietary differences between sub-areas, indicated that arctic foxes fed opportunistically on the alternative prey types. Den occupancy rates were positively correlated with lemming population densities during the previous winter, indicating a strong numerical response. We conclude that from a functional aspect, the arctic fox in Sweden is a lemming specialist, since lemming is the main prey and their abundance is the best predictor of arctic fox reproductive success. Other prey are used opportunistically in relation to their availability. Key words: Alopex lagopus, diet, predation, numerical response, functional response INTRODUCTION Food availability is the most important environmental factor in¯uencing reproduction in mammals, func- tioning both as a proximate and ultimate cause for annual variations in reproductive output (Bronson, 1989). A predator showing strong preference for one prey species does not change its diet when the abun- dance of that species decreases, while a predator with weak preferences can switch to another type of prey (Murdoch, 1969). The degree of specialization can also be de®ned with regard to the function of a predator in a predator guild. The specialist predator can thus be restricted in its diet not only by preference, but also through lack of ability or possibility to use other prey species. The generalist predator is less dependent on the availability of one particular prey species and abandons a declining prey type earlier than the specialist. As a consequence, the specialist will have a more pronounced numerical response to changes in abundance of a particular prey species than the semi-generalist or generalist (Andersson & Erlinge, 1977). The specialist± generalist concepts can thus be used to analyse the strategies or functions of different predators within a prey community (see e.g. Reid, Krebs & Kenney, 1997; Wiklund et al., 1999). The response of a predator to changes in the avail- ability of prey depends on the predator's strategy. Andersson & Erlinge (1977) suggested that a specialist would have a type II functional response, i.e. the number of killed prey would increase rapidly with prey density. A generalist would show a slower increase, resulting in the sigmoid curve of a type III response (Holling, 1959). This could be the result of changes in search image, foraging habitat or diet preferences (Taylor, 1984). The functional (dietary) and numerical responses of a predator to changes in prey abundance may also exert a strong in¯uence on the population J. Zool., Lond. (2000) 251, 139±149 # 2000 The Zoological Society of London Printed in the United Kingdom