Behav Ecol Sociobiol (1989) 25:49-56 Behavioral Ecology and Sociobiology 9 Springer-Verlag 1989 Coloration in New World orioles 1. Tests of predation-related hypotheses Nancy J. Flood Department of Zoology, University of Toronto, 25 Harbord Street, Toronto, Ontario M5S IA1, Canada Received September 26, 1988 / Accepted February 17, 1989 Summary. Although sexual dichromatism in birds is usually ascribed to sexual selection, some workers argue that avian coloration is better ex- plained by predation-related selection pressures. Supporting evidence for these latter hypotheses comes primarily from broad interspecific compari- sons, which can be biased by a variety of factors. This study examines the predation-based hypothe- ses of Baker and Parker 0979), particularly the Predator Deflection Hypothesis, with reference to two closely-related oriole species: Icterus par- isorum, which is dichromatic, and I. gularis, in which both sexes are brightly coloured. To test the prediction that bright coloration serves to di- vert the attention of predators away from cryptic young, rates of predation on nests of conspicuous (2 years of age or older) and dull-coloured (first- year). L parisorum males were compared. The re- sults showed equal predation on the young of males in both age/plumage classes. The Predator Deflection Hypothesis also predicts that, once a predator has been detected, brightly-coloured birds should attempt to distract it, whereas cryptic indi- viduals should not. Tests using models of avian predators showed that this was not the case: cryptic L parisorum females responded as aggressi- vely toward the model as did conspicuous conspe- cific males. The same was true for L gularis. Over- all, the results did not support the idea that bright coloration has evolved in response to predation pressure in these species. Introduction The evolution of bird coloration Sexual dimorphism, and specifically, sexual di- chromatism, has long been of interest to ornitho- logists. Following Darwin (1871), most work has suggested that sexual colour and pattern differ- ences in birds are related primarily to sexual selec- tion: that the brightly-coloured sex (usually the male) owes its conspicuous appearance mainly to intrasexual and/or epigamic selection. Cryptic col- oration is usually ascribed to a lack or reduction of such sexual selection pressures, which then al- lows selection for predator avoidance to predomi- nate. Recently, however, Baker and Parker (1979) (see also Baker and Hounsome 1983; Baker 1985; Baker and Bibby 1987) have advanced the idea that bird coloration - conspicuous as well as cryptic - has "evolved almost entirely in response to predation-based selection pressures ...not ... in response to sexual selection pressures" (Baker and Parker 1979, p. 65). Developing this idea, they have proposed several different predation-related hypotheses to account for the evolution of both conspicuous and cryptic plumage and, therefore, for differences among species in the extent of sex- ual dichromatism. In this paper, I examine the rele- vance of these hypotheses to an explanation for the bright coloration - and variance in sexual di- chromatism - characteristic of species in the avian genus Icterus (the New World orioles). Baker and Parker are careful to describe the conditions or assumptions under which each of their several predation-related hypotheses are lik- ely to apply. Consideration of these limitations re- veals that in fact, most of their hypotheses are not relevant to the many passerine species (including orioles), (1) that are not noxious in some way (and whose bright coloration cannot therefore be apose- matic), (2) whose conspicuous colors are broadly distributed over the body and always exposed (i.e., do not conform to the pattern Baker and Parker associate with "flash coloration", which is dis-