(1871) proposed the theory of sexual selection, whereby the choice of mates by females could result in the evolution of elaborate secondary sex traits. Various patterns in nature suggested sexual selection to Darwin. He proposed that when morphological characters were sexually dimorphic, developed in males only at sexual maturity, and appeared during the breeding season, then such traits functioned to enhance a male's chance of reproducing (see Andersson 1994). While recent research has contributed considerably to the development of sexual selection theory, empirical studies of physiological mechanisms responsible for the expression and development of the selected traits have lagged behind (see Cronin 1991; Andersson 1994).The evolution of bright coloration in male birds has been a frequent, albeit controversial, topic since Darwin's time (Cronin 1991). Of particular recent interest has been the role of carotenoid pigments in sexual selection. These compounds are responsible for most bright reds and yellows in vertebrates (Brush 1990), but they also are important physiological modulators and so have a range of health-related functions (Lozano 1994;Rock, Jacob & Bowen 1996). A number of studies of birds (Zuk et al. 1990;Hill & Montgomerie 1994;Hudon 1994;Dufva & Allander 1995;Sundberg 1995)
Summary1. Sexually selected traits are expected to vary seasonally, with the maximal expression of the character being evident during mate choice; however, the mechanisms controlling or regulating such traits are generally poorly known. 2. Carotenoid pigments responsible for bright red or yellow coloration in the feathers, skin or other integumentary structures of birds are generally believed to vary seasonally because of diet. 3. Variation in carotenoid-dependent skin colour between winter and spring (mating season) was investigated, as was variation in plasma carotenoids across the breeding season in captive American Kestrels, Falco sparverius, fed a uniform diet. 4. Kestrels were more brightly coloured in the mating period than in winter, and plasma carotenoid concentrations declined from the time of mating to the rearing of young. 5. Although carotenoid levels were highly sexually dimorphic during mating and laying, males and both breeding and non-breeding females all had similar levels by the incubation period, and the pattern of variation over time suggests rheostatic regulation. 6. These results suggest kestrels may have the ability to regulate (rather than merely control) their colour physiologically, the variation in colour and carotenoids is consistent with that expected of a sexually selected trait, and the loss of colour after breeding may suggest a trade-off between the show and health functions of carotenoids.