How does predator perception affect the evolution of Batesian mimicry?
In 1862 Henry Walter Bates published a paper hypothesizing that the mimetic resemblances between tropical butterflies were caused by natural selection upon some harmless species to resemble unpalatable ones in order to avoid being eaten by predators. Charles Darwin was so excited by this idea that he wrote to Bates and exclaimed
“You have most clearly stated and solved a most wonderful problem. Your paper is too good to be largely appreciated by the mob of naturalists without souls; but rely on it that it will have lasting value.”
Predators are the selective agents that drive mimicry, and so to understand mimicry, we must understand the predators. Both ultimate and proximate aspects of predator behavior affect the evolution of mimicry and the maintenance of mimetic phenotypes. Projects that address these themes are described below.
I study the deadly coral snake (Micrurus fulvius) and its nonvenomous mimic, the scarlet kingsnake (Lampropeltis elapsoides). These snakes are native to the southeastern United States.
Predator generalization can allow mimicry to evolve
For decades, biologists have debated how mimicry evolves. A species evolving mimicry must sacrifice cryptic coloration because model species are conspicuous. Therefore, intermediate stages of evolution between crypsis and mimicry might not be protected against predators. Some believe that this would forestall the evolution of mimicry, and as a result, have proposed that mimicry can't evolve unless it starts with a major mutation so that the organism looks somewhat like the model. On the other hand, if intermediate stages of evolution don't sacrifice too much crypsis or if they receive protection from mimicry even when they resemble models very little, there is no need for evolution to start with a major mutation.
Thinking about things from a predator’s perspective lets us understand the conditions under which predators would attack intermediates. Predators should want to maximize their payoff from attacking prey of different appearances. If the likely cost of attacking a model is high compared to the benefit of attacking a mimic, predators should avoid intermediates, but not if costs of attacking a model are low. One factor that predators would have to consider in calculating costs & benefits is model abundance. I explored the effects of model abundance on the adaptive landscape of mimicry by making replicas of cryptic, intermediate, and mimetic snakes. I placed them in areas of high and low model abundance and then measured predator attack rates on each phenotype to approximate fitness.
In areas where coral snakes were highly abundant, intermediate phenotypes were not less fit than cryptic and mimetic phenotypes. However, where coral snakes were rare, intermediates were disfavored, as has classically been predicted. In sum, some conditions such as high model abundance may permit the evolution of Batesian mimicry through a plurality of mechanisms.
Predator cognition can permit imperfect mimics to persist
Batesian mimicry is often imprecise. Although generalization might explain some imperfect mimicry, it is difficult to invoke when models and mimics can be distinguished with 100% reliability because there is no overlap in the phenotype of models and mimics. An alternative explanation for imperfect mimicry is that predators might not be able to perceive all dimensions of prey phenotype and thus permit imperfect mimicry to persist. I conducted a field experiment to test whether or not predators can distinguish coral snakes from scarlet kingsnakes. The order of colored rings that encircle the bodies of each species is different. Despite this imprecise mimicry, we found that scarlet kingsnakes that match coral snakes in other respects are not under selection to match the ring order of their model. Scarlet kingsnakes appear to have evolved only those signals necessary to deceive predators. Generally, imperfect mimicry might suffice if it exploits limitations in predator cognitive abilities.
Can predator sensory bias drive the evolution of mimetic coloration?
Mimicry should be tailored to deceive not only the cognitive systems of predators but also their sensory systems. Much mimicry is based on prey coloration. Predators’ perception of color is subjective because their photoreceptors are most sensitive at specific wavelengths of light. Therefore, mimicry ought to be best at the wavelengths at which predators are most sensitive, just as the sexual signals of cichlids have adapted to the wavelengths to which they are most sensitive in their own habitats (Seehausen et al. 2008). I will quantify the coloration of models and mimics using objective readings taken from a spectrophotometer, and use those data to test whether mimetic coloration has evolved to match models significantly more closely at the wavelengths to which predators are most sensitive.
Scarlet kingsnake on Macbeth Color Checker
The same snake in UV light
Top - a scarlet kingsnake from North Carolina; bottom - a scarlet kingsnake from Florida
Replica of mimetic phenotypein situ
Cryptic phenotype replica after attack by large mammal
