Selection by wild birds foraging on artificial prey in heterogeneous backgrounds

Abstract

Many animals that are prey to animals that hunt by sight tend to rest in places that are similar to them in colour, texture, and other visual properties and they thus minimise the probability of being detected. Such an animal is said to be 'cryptic'. Within prey species living in heterogeneous backgrounds this could result in the evolution of several different colour morphs with the fitness of each morph being related to its degree of crypsis in the background. At equilibrium the frequencies of the morphs will be proportional to the frequencies of their matching background elements and this relationship could contribute to the maintenance of the polymorphism. The experiments in much of the thesis attempted to determine (1) whether birds are more likely to eat prey in unmatching backgrounds that make them conspicuous rather than prey in matching backgrounds that make them inconspicuous and (2) whether birds select in a more strongly apostatic fashion when presented with different proportions of two colours of prey in backgrounds that make them inconspicuous in comparison to when the same prey are presented in backgrounds that make them conspicuous. In the last section of the thesis I try to determine whether proximity of a prey to stones has an effect on its risk of predation and, if so, why. I present the results of five sets of experiments with wild birds and artificial prey. In two experiments (Chapter III) two prey types were presented to predators in equal frequencies (to test for frequency independent selection). Three experiments (Chapter IV) used different prey frequencies (testing for frequency dependent selection). In all cases the experiment trials were stopped when half of the prey had been eaten and selection then was measured. Different components of these experiments were carried out in Culiacan, Mexico and Southampton, England. In backgrounds of co loured stones that made one of the prey inconspicuous, birds tended to concentrate on the conspicuous (unmatching) prey. This selection against the unmatching form was stronger when the prey were presented in high density backgrounds (600 stones) than in low density backgrounds (200 stones). In backgrounds where both prey were equally cryptic but presented in different frequencies the predators tended to concentrate on the most common type and overlooked the rare ones. This 'apostatic selection' was more accentuated when the stones were denser, perhaps because the prey were more difficult to detect and the birds therefore concentrated their efforts on the colour they encountered more frequently. When prey were presented in equal frequencies selection was dependent on the background composition. These results lend support to the idea of parallelism given by Endler in 1978, because this kind of selection should result in a neutral equilibrium, with morph frequency depending of the background composition. In the experiments described in Chapter IV apostatic selection was stronger when prey was in backgrounds that made them inconspicuous (the backgrounds used were natural grass and soil and stones that matched the prey colours on hessian sheets). This implies the adoption of search images. In Chapter V, three experiments (in Southampton, England) tested the effect of distance to the stones on the risk of predation and showed that stones offered more protection to the prey that matched the closest stone. Resting close to the stones offered protection from predation regardless of the colour of the stone. In summary the results of my work from both Mexico and the UK, strongly support the general idea that apostatic selection is most effective when prey match the colour of the background.

More information

Identifiers

Catalogue record

Export record

ASCII CitationAtomBibTeXData Cite XMLDublin CoreDublin CoreEP3 XMLEndNoteHTML CitationHTML CitationHTML ListJSONMETSMODSMPEG-21 DIDLOpenURL ContextObjectOpenURL ContextObject in SpanRDF+N-TriplesRDF+N3RDF+XMLRIOXX2 XMLReferReference ManagerSimple Metadata