Do food-hoarding animals have a cache recovery advantage? Determining recovery of stored food STEPHEN B. VANDER WALL * , JENNIFER S. BRIGGS †, STEPHEN H. JENKINS * , KELLIE M. KUHN * , THEODORE C. THAYER ‡ & MAURIE J. BECK * *Department of Biology and the Program in Ecology, Evolution and Conservation Biology, University of Nevada, Reno yPasadena City College, Pasadena, California zU.S. Forest Service, Pacific Southwest Region, South Lake Tahoe, California (Received 9 August 2005; initial acceptance 8 November 2005; final acceptance 5 January 2006; published online 23 May 2006; MS. number: A10225R) Food stored by animals can either be recovered by the hoarder or pilfered, but it is very difficult to distin- guish between these possibilities under natural conditions. We tested the hypothesis that yellow pine chip- munks, Tamias amoenus, have a recovery advantage (i.e. a greater probability of finding a cached item than does a na € ıve forager). We allowed chipmunks to cache radiolabelled pine seeds and then established arti- ficial caches nearby. Paired artificial and real caches were identical in number of seeds, depth and microsite. Caches (393 pairs at 10 sites) were monitored during autumn 1997, 1999 and 2002. Removal of real caches represented the activity of the hoarder, foraging using spatial memory, plus the activities of pilferers, whereas the removal of artificial caches represented the activities of pilferers. We used these two measures in a competing risks model, which estimated that seed-caching chipmunks retrieved 74% of the food they stored in 1997 and 56% of the food they stored in 1999–2002. Seed-hoarding chipmunks were removed at four of the 10 sites immediately after they prepared caches. Where the seed-hoarders remained, animals took real caches 3.4 to 6.5 times faster than artificial caches, an effect that we attribute to the activity of the hoarder. Where the caching chipmunk was removed, there were no significant differences in the rates of removal of real versus artificial caches. We suggest that artificial caches can be used to estimate the pro- portion of caches retrieved by a seed-hoarding animal versus the proportion discovered by pilferers. Ó 2006 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. Scatter-hoarding birds and mammals make thousands of caches throughout their home ranges each year (Vander Wall 1990). It is generally believed that the hoarder has an advantage over na € ıve foragers when searching for items it has stored (Andersson & Krebs 1978; Stapanian & Smith 1978; Vander Wall & Jenkins 2003). This advantage is probably gained through detailed spatial memory of cache locations (Jacobs & Liman 1991; Vander Wall 1991; Balda & Kamil 1992; Jacobs 1992) or priority of access to space (territories or microhabitats within territories) where items are stored (Moreno et al. 1981; Brodin 1994; Grubb & Pra- vosudov 1994). However, the strength of this recovery ad- vantage has seldom been measured. Furthermore, it is not known whether the recovery advantage persists over time or is short lived (i.e. decays over days or weeks). One reason for this lack of information is that animal caches are usually inconspicuous and much of the exploitation and management of caches occurs when observers are absent. This is especially true of rodents that store food over relatively long periods. As one means of studying the fate of scatter-hoarded food, researchers have used artificial caches, placed in microsites similar to those used by food hoarders, as surrogates for real caches (e.g. Cowie et al. 1981; Stapanian & Smith 1984; Daly et al. 1992; Clarke & Kramer 1994). This technique per- mits researchers to investigate the effects of cache size, depth, spacing, site conditions and other factors on the rate of cache removal. It is generally assumed that the fates of artificial caches accurately represent the fates of animal- made caches. With few exceptions, these studies have reported rapid removal of scatter-hoarded food (Clarke & Kramer 1994; Vander Wall & Jenkins 2003). Although these studies give some insights into cache dynamics, the rate of removal of artificial caches does not provide any information about the recovery advantage of a Correspondence: S. B. Vander Wall, Department of Biology and the Program in Ecology, Evolution and Conservation Biology, University of Nevada, Reno, NV 89557, U.S.A. (email: sv@unr.edu). 189 0003–3472/06/$30.00/0 Ó 2006 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. ANIMAL BEHAVIOUR, 2006, 72, 189–197 doi:10.1016/j.anbehav.2006.01.004