Sexual signal exaggeration affects physiological state in male barn swallows

A prevailing view in sexual selection theory is that costly physiological processes underlie the development, maintenance and expression of sexual signals, and that the costs of these signals enforce their honesty [1,2]. However, this unidirectional view of how physiology governs signal expression is narrow, because many of the putative physiological underpinnings of signals, such as health status, are themselves dynamic [3]. As such, we predicted that physiological parameters should be affected by sexual signal expression. We therefore manipulated a known sexual signal — plumage coloration — in male barn swallows (Hirundo rustica erythrogaster) and measured circulating androgen levels and body mass before and after the manipulation. We found that androgen concentrations increased in color-enhanced males, but decreased in control males, as expected due to typical seasonal androgen declines [4,5]. Colorenhanced males also lost body mass, whereas control males gained weight between successive captures one week apart. These results indicate the existence of feedbacks between an individual’s morphological signals and physiology — a finding that is not currently explained by honest signaling theory. Androgens, such as testosterone, are likely candidates for mediating bidirectional feedbacks between sexual signals and physiological condition, as they can play a role in promoting the development, expression and honesty maintenance of ornamental traits in male vertebrates [2,5–8]. In Correspondences addition, they also can respond to an individual’s social experiences [4]. To understand whether signal expression alters the physiological state of individuals, we targeted androgens as plausible physiological respondents to sexual signal exaggeration in male barn swallows. To explicitly test the effect of signal expression on circulating sex-steroid levels, we conducted a two-year study in which we examined a male’s plasma androgen concentrations both before and after the experimental enhancement of ventral feather coloration, a known sexual signal in North American populations of this species [9]. For comparison, we used a randomly selected group of control males the plumage color of which was not manipulated but which were otherwise treated identically as males assigned the experimental treatment. All males were captured in the wild using mist nets placed around areas of high activity at breeding sites and randomly assigned to one of the two treatments. Following a well known seasonal decline in circulating concentrations of sex steroids, males captured later in the breeding season showed lower androgen concentrations than males caught only a week earlier (estimate = –0.03; F1,4.62 = 34.96; P = 0.003; R2 = 0.10; n = 130). Before experimental manipulation, darker males had higher androgen concentrations (estimate = –0.24; F1,93 = 11.32; P = 0.001; R2 = 0.22; n = 97; lower color scores indicate darker feather color), and heavier males had lower concentrations of androgens (estimate = –0.26; F1,57 = 5.15; P = 0.03; R2 = 0.14; n = 60). Prior to experimental manipulation, treatment groups did not differ significantly in plumage color (F1,55.66 = 0.13; P = 0.71; n = 63), androgen concentrations (F1,59.18 = 0.66; P = 0.41; n = 63), or body mass (F1,55.47 = 0.14; P = 0.70; n = 63). The color-enhancement treatment was successful in that males in the colorenhanced group had significantly darker color throughout the ventral region after the experimental manipulation (Unequal Variance t-test: t = –10.53; P < 0.0001; df = 48.74; n = 52). After one week, males from the color-enhanced group showed increased concentrations of circulating androgens, whereas males in the control group showed decreased androgen concentrations (F1,27 = 6.12; P = 0.02; n = 30; Figure 1A). Additionally, males in the color-enhanced treatment group lost body mass, whereas males in the control groups gained weight between the two sampling periods (F1,24.29 = 13.03, P = 0.001, n = 30; Figure 1B). In this study, we found that the usual seasonal trajectory of androgen decline observed in male barn swallows was reversed by changing a male’s visual appearance, an observation that counters the view of a one-way relationship between androgens and sexual signals. Males with experimentally enhanced color and, accordingly, increased androgen concentrations also lost body mass. That androgen concentrations and mass were coupled is not surprising, given