Ibis (2003), 145, 472– 483 © 2003 British Ornithologists’ Union Blackwell Science, Ltd The decline of Common Kestrels Falco tinnunculus in a forested area of northern England: the role of predation by Northern Goshawks Accipiter gentilis S. J. PETTY, 1 * D. I. K. ANDERSON, 1 M. DAVISON, 2 B. LITTLE, 3 T. N. SHERRATT, 4 C. J. THOMAS 4 & X. LAMBIN 5 1 Woodland Ecology Branch, Forest Research, Northern Research Station, Roslin Midlothian EH25 9SY, UK 2 37 Beech Grove, Blackhall Mill, Newcastle upon Tyne NE17 7TD, UK 3 37 Stella Hall Drive, Blaydon, Tyne and Wear NE21 4LE, UK 4 Department of Biological Sciences, University of Durham, Science Laboratories, South Road, Durham DH1 3LE, UK 5 Department of Zoology, University of Aberdeen, Tillydrone Avenue, Aberdeen AB24 2TN, UK We have previously documented the decline of the Common Kestrel Falco tinnunculus over a 23-year period in a large coniferous forest in northern England. Kestrels fed predominantly on Field Voles Microtus agrestis, which were most abundant in young plantations (1–11 years old). Over the 23 years, voles remained abundant in the study area, but their numbers fluc- tuated cyclically. Here we consider whether the decline of Kestrels was linked to predation by Northern Goshawk Accipiter gentilis. Goshawks first bred in the study area in 1973 and increased until 1989, after which numbers stabilized. We use a number of approaches to explore the role of Goshawk predation, all of which are correlative, but independent. First, there was a significant negative relationship between Kestrel and Goshawk numbers after controlling for a decline in vole habitat. Short-eared Owls Asio flammeus, which also hunt by day, declined over the same period as Kestrels. Second, numbers of Tawny Owl Strix aluco and Long-eared Owl Asio otus did not decline as Goshawk numbers increased. These two species are also vole-dependent, but active by night, and less vulnerable to Goshawk attack. Third, six species of raptor comprised 4.5% of 5445 Goshawk prey items during the breeding season, but more Kestrels were killed than the combined total of all other raptors. Goshawks not only killed many adult Kestrels in early spring, prior to breeding, when it would have most impact on population levels, but there was also a temporal trend for predation on Kestrels to be inversely density-dependent. Finally, we estimated that Goshawks removed more Kestrels than were recorded each spring in the study area. We interpreted this as indicating that immigrant Kestrels were being removed continually, mostly before they could breed. We conclude that the decline of Kestrels (and possibly Short-eared Owls) was mainly due to predation by Goshawks. This study provides some of the strongest evidence yet of the role of predation in the hierarchical structuring of raptor communities. Intraguild predation is the killing of species that use similar resources and are thus potential competitors (Polis et al. 1989, Polis & Holt 1992, Holt & Polis 1997, Morin 1999). Such a process may be fundamental in the structuring of many wildlife communities in pris- tine habitats, but one that often decays progressively with habitat fragmentation, including the removal of top predators (Caro 1987, Mech et al. 1988, Terborgh 1988, Wright et al. 1994, Estes 1995, 1996). In raptor guilds, there is often a hierarchical struc- ture that appears to be maintained by larger or more aggressive species displacing smaller or less aggres- sive species rather than by direct killing (Newton 1979), although larger raptors are known to predate smaller ones (Uttendörfer 1952). Within European raptor communities, the Northern Goshawk Accipiter *Corresponding author. Dr S.J. Petty, Craigielea, Kames, Tighnabruaich, Argyll PA21 2AE, UK. Email: StevePetty@compuserve.com