Fig 1.
Variation in female throat color.
Variation in stream resident female throat coloration alongside a typical male.
Fig 2.
Predicted male mating response for female traits.
Fitted mate preference function for each of the 63 males from the sequential choice mating trial assay, as estimated from a random coefficient mixed model (see Methods for details). Predicted values of the male mating response (PC1) are plotted against the composite trait values obtained by scoring females for the linear combination of the three traits (female throat color, spine color, standard length) described by the first eigenfunction of the random effect covariance function of male preference ([−0.210, −0.169, 0.005] for the three traits respectively).
Table 1.
Population-level male mate preferences.
Fig 3.
Average throat reflectance values of females used in simultaneous choice trials.
A comparison of the average throat reflectance values of ‘dull’ versus ‘red’ categorized females used in simultaneous choice trials. Reflectance values were assessed in a tetrahedral color space model to obtain a single value for female throat color, a measure of chroma. Females were then assigned to a ‘red’ or ‘dull’ throat color category using this value, the highest value from each pair receiving the ‘red’ classification. Average subject male throat reflectance values included for comparison. Error bars represent ±1 S.E.
Fig 4.
Intruder male throat color affects subject male behavioral response.
Regression analysis of conspecific male trials shows that males display a significant, negative response to intruder male throat color (R2 = 0.188, df = 29, P = 0.016).
Fig 5.
Male response differs between female and male presentations.
Paired t-test revealed that in sequential choice trials, males (A) zigzagged significantly more toward females than males, but showed no difference in the number of (B) bites. Error bars represent ±1 S.E