Male preference for conspecific females depends on male size in the splendid darter, Etheostoma barrenense
Description
We sought to expand existing knowledge of mate preference in the darter fish species Etheostoma barrenense by investigating whether variation in preference for conspecifics among individual males correlates with body size and coloring. We observed that only large males preferred conspecific females and no effect due to coloration. This result may indicate that preferences are expressed only by older males, or males with a competitive advantage, as both may be correlated with larger body size. If larger males are indeed older, our finding suggests a possibility of learned mate choice in darters. Standard body lengths of all individuals were measured with a digital caliper, and males were designated as ‘small’ (< 45 mm) or ‘large’ (> 45 mm). Dichotomous mate choice trials were conducted presenting two female stimuli, one E. barrenense and one Etheostoma zonale, to a male E. barrenense. Time spent in the conspecific and heterospecific association zone was recorded and used to calculate proportion of time spent associating with the conspecific and heterospecific female and strength of preference (SOP). SOP was calculated as SOP = (TC¬ – TH) / (TC¬ + TH), where TC is the time spent in the conspecific association zone and TH is time spent in the heterospecific association zone. SOP ranges from +1 to -1, representing complete preference for the conspecific or heterospecific female, respectively. Following trials, fish were euthanized and photographed. Males’ standard lengths were measured digitally, and coloration was quantified using the software ImageJ. We quantified proportion of red body coloration and a ‘red score,’ calculated by measuring the average red(R), green(G) and blue (B) pixel values for red body area and dividing R by the sum of the averaged RGB pixel values. Larger red scores indicate more saturated red coloration, with a score of 1 indicating pure red. Data is recorded in the file “raw_data.csv.” We used R to perform appropriate hypothesis tests, treating male size as a categorical variable, and to generate a linear model testing the effect of male size and colour, as continuous variables, on SOP. Lastly, we performed post hoc analyses to determine any effect of female size on our findings, including a comparison of female size within trials and a linear model comparing the standard length of the male and his preferred female. All analyses are available in the file “all code.R.” Trials were videorecorded from ‘top’ and ‘side’ viewpoints. No significant difference was found for trials with recordings from both. One view was randomly chosen for analysis for these trials. These data are recorded in the file “view.csv.” Three photographed males could not be matched to their behavioural trial and one photographed male was not used in behavioural trials. Therefore, behavioural trials were performed on N=17 males, colour analysis was performed on N=18 males, and linear modelling was performed on N=14 males.