Lookup NU author(s): Dr Christina Halpin,
Dr John Skelhorn,
Professor Candy Rowe
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
Many defended prey advertise their defences to potential predators by being conspicuously coloured and/or having distinct bold markings. These colourful warning signals benefit prey since they are easy to learn to avoid, and seem particularly effective as deterrents. However, the evolution of conspicuous coloration is problematic, because if a distinct conspicuous morph arises in a cryptic species, it would be more easily detected and the risk of being attacked would be high. However, this argument assumes that attack probability will be the only factor determining prey survival, but in fact many prey have external defences that predators can detect, allowing prey to survive attacks. This experiment tests whether a rare conspicuously coloured defended morph can have a selective advantage over a cryptic defended morph when defence chemicals can be detected before ingestion by a naïve predator. Using chicks, Gallus gallus domesticus, foraging on coloured chick crumbs, we found that naïve birds can learn to avoid defended prey when they are conspicuous but not when they are cryptic. Crucially, when prey ingestion, and not attack probability, was used as our measure of mortality we found that there can be a selective advantage to being conspicuous in a cryptic population. This advantage can occur even in the first avoidance-learning trial, providing an initial selective advantage for rare conspicuous defended morphs. However, this does not occur in all conspicuously coloured prey populations, and it is evident that colour biases are important and their role in the initial evolution of conspicuous colour signals should be considered. © 2007 The Association for the Study of Animal Behaviour.
Author(s): Halpin CG, Skelhorn J, Rowe C
Publication type: Article
Publication status: Published
Journal: Animal Behaviour
ISSN (print): 0003-3472
ISSN (electronic): 1095-8282
Publisher: Elsevier Ltd
Altmetrics provided by Altmetric