Testosterone Regulates Sexually Dimorphic Coloration in the Eastern Fence Lizard, Sceloporus undulatus
Adult male Eastern Fence Lizards (Sceloporus undulatus) possess blue and black ventral patches that function in sex recognition and intrasexual social signaling, but this ventral coloration is absent or greatly reduced in females and juvenile males. Adult males also exhibit a relatively uniform, reddish brown dorsal coloration, while females and juvenile males are cryptically colored, with two rows of dark brown or black chevrons set against a background of gray and brown. In the present report, we show that sexual divergence in ventral coloration is temporally correlated with sexual divergence in plasma testosterone in free-living juvenile males and females, supporting the hypothesis that sexual dichromatism is regulated at least in part by testosterone. We experimentally tested this hypothesis by (1) removing the primary source of circulating testosterone in juvenile males via surgical castration, (2) restoring testosterone in castrated males with tonic-release implants, and (3) implanting intact juvenile females with exogenous testosterone. As predicted by our hypothesis, the development of blue and black ventral coloration in S. undulatus was (1) inhibited by castration in juvenile males, (2) restored by exogenous testosterone following castration in juvenile males, and (3) promoted by exogenous testosterone in juvenile females. The expression of male-specific dorsal coloration was also (1) inhibited by castration and (2) restored by exogenous testosterone following castration in juvenile males. Our results are consistent with established literature supporting the critical role of androgens in the mediation of sexually dimorphic coloration among phrynosomatid lizards.Abstract
Digital scans of the ventral (top panels) and dorsal (bottom panels) surfaces of individual Sceloporus undulatus, illustrating the typical coloration of each treatment group at 420 d post-treatment. Castrated Males exhibited dorsal and ventral coloration similar to Control Females, while Testosterone Males resembled Control Males. Lettered ellipses indicate the areas that we sampled for our digital analyses of hue, saturation, and brightness (see Fig. 5); (A) blue throat patches, (B) blue abdominal patches, (C) dorsal chevrons, (D) dorsolateral area
Mean ± 1 SE (A) plasma testosterone and categorical scores for (B) blue and (C) black ventral coloration vs. age for free-living juvenile male and female Sceloporus undulatus. Numbers above symbols indicate sample sizes for males and females, respectively. At 2 months of age, males and females did not differ in plasma testosterone (F1,23 = 0.28; P = 0.298), blue coloration (χ2 = 0.79; P = 0.375), or black coloration (χ2 = 0.00; P = 1.00). By 11 months of age, males had significantly more plasma testosterone (F1,19 = 4.67; P = 0.022), blue coloration (χ2 = 18.15; P < 0.001), and black coloration (χ2 = 19.24; P < 0.001) than females. Blue coloration increased with age in both males (χ2 = 35.26; P < 0.001) and females (χ2 = 19.35; P < 0.001), but black coloration increased with age only in males (χ2 = 33.36; P < 0.001)
Mean ± 1 SE plasma testosterone (left panels) and categorical scores for blue and black ventral coloration (right panels) in Sceloporus undulatus treatment groups in our field (A–B; 47 d post-treatment) and lab (C–D; 65 d post-treatment) experiments. Sample sizes are reported for the field and lab experiments, respectively. Statistics are reported for ANOVA (plasma testosterone) and non-parametric Kruskal-Wallis analyses (coloration scores) among male treatment groups (Castrated, Control, Testosterone) or between female treatment groups (Control, Testosterone). Note that Testosterone Females were not included in the field experiment
Patch size (mean ± 1 SE) of blue and black (A) throat and (B) abdominal patches among male Sceloporus undulatus treatment groups at 420 d post-treatment. Statistics are reported for ANOVA of patch size with treatment (Castrated, Control, Testosterone) as the main effect. Lowercase letters denote post hoc statistical separation of treatment groups (Ryan-Einot-Gabriel-Welsch test)
(A–B) Brightness and saturation (mean ± 1 SE) of blue (A) throat, and (B) abdominal patches among male Sceloporus undulatus treatment groups at 420 d post-treatment. Hue of blue throat and abdominal patches did not differ among treatment groups. (C–D) Brightness and hue (mean ± 1 SE) of (C) dorsal chevrons, and (D) dorsolateral area. Saturation of dorsal chevrons and dorsolateral area did not differ among treatment groups. Statistics are reported for ANOVA of hue, saturation, and brightness with treatment (Castrated, Control, Testosterone) as the main effect. Greek letters denote post hoc statistical separation (Ryan-Einot-Gabriel-Welsch test) of treatment groups along the horizontal axis (saturation or hue), and lowercase letters denote post hoc statistical separation along the vertical axis (brightness). Note that Castrated Males (open symbols) differ significantly from Control Males and Testosterone Males (filled symbols) for most traits
Contributor Notes
(RMC, SLS) Rutgers University, Ecology and Evolution, 14 College Farm Road, New Brunswick, New Jersey 08901; and (AL, HBJ-A) Rutgers University, Department of Animal Sciences, 84 Lipman Drive, New Brunswick, New Jersey 08901. (RMC) rmcox@eden.rutgers.edu; and (HBJ-A) henry@aesop.rutgers.edu Send reprint requests to RMC.