Principal Investigator University of Pittsburgh Pittsburgh, Pennsylvania, United States
Background/Question/Methods
Classic parasite-mediated sexual selection hypotheses—the Hamilton-Zuk and Contagion-Indicator hypotheses—predict that a sexually selected trait should be negatively correlated with parasite load within populations and positively between, but evidence remains ambiguous in many systems. Ambiguous evidence is often attributed to failure to measure the parasite actually linked to the sexually selected trait, but other reasons could explain why predictions are not met: variation in the strength of preference and the feedback of preference on the relationship between sexually selected traits and parasitism. Stronger preference in a population may reduce parasite prevalence while greater variation in the strength of preference within a population weakens the correlation between a sexually selected trait and parasite load. For directly transmitted parasites, like Gyrodactylus spp. parasitizing guppies, Poecilia reticulata, preference may dictate transmission opportunities. The sexiest mates may thus be most exposed to parasites, either weakening the correlation between the sexually selected trait and parasite load, or making the correlation non-linear—even if the trait advertises parasite resistance. Trinidadian guppies—a classic sexual selection system in which females prefer to mate with more colorful males—and their gyrodactylid parasites allow me to test predictions unraveling the complex relationship between preference and parasitism.
Results/Conclusions
We combine data from the literature with dynamic models to update the predictions for the relationship between preference, sexually selected traits, and parasitism. A wealth of literature already exists on Trinidadian guppies. However, no single study quantifies both female preference and the covariance between male coloration and parasite load or prevalence. By matching literate estimates across studies, we found that preference could potentially predict prevalence but is confounded by predation regime. The relationship between male coloration and parasite load within populations that differ in the strength of female preference suggests that greater preference may switch the sign of the correlation between male coloration and infection. However, binary measures of preference in the field (preference vs. no preference) likely obscure the true relationship. To complement these data, we created a modified SIR-model to allow preference to affect transmission-relevant contact rates. Simulation results showed a nonmonotonic relationship between a trait advertising resistance and prevalence: preference may be strong enough that the increased exposure may overwhelm or obscure the inherent resistance of sexy males while still reducing overall population-level prevalence. Planned transmission trial and mate choice experiments will further confirm that preference may modify classic parasite-mediated sexual selection predictions.
