An Investigation into the Ecology and Evolution of Caterpillar Eyespots

Eyespots are conspicuous circular patterns on the body of an animal which superficially resemble vertebrate eyes. These odd markings have long captured the interest of biologists and naturalists alike, many of whom have suggested that eyespots mimic the real eyes of dangerous animals and thereby protect prey species from their attackers. Eyespots are particularly widespread and diverse in lepidopteran caterpillars, and these caterpillars with eyespots are typically assumed to be snake-mimics. Yet a dearth of empirical investigation has left us without evidence that eyespots can protect caterpillars, or the ability to substantiate our subjective belief that these caterpillars are mimicking snakes. Using a combination of field and lab experiments my research provides robust empirical evidence demonstrating that eyespots are an adaptation in caterpillars, but reveals that other factors, including body colour and body size, influence their efficacy as a deterrent. Moreover, my work reveals that the defensive posture, which many caterpillars with eyespots adopt when harassed, also provides protection independent from the possession of eyespots. Using an objective measure of mimetic fidelity I showed that this defensive posture is likely a form of behavioural mimicry which increases the caterpillar’s resemblance to viperine snakes when viewed dorsally. Caterpillars from three distinct families showed analogous behavioural mimicry, suggesting strong selection for convergent phenotypes across lepidopteran larvae. Finally I provide an explanation for why some caterpillars have eyespots, and others do not. I show that body size critically influences the evolution of eyespots using a phylogenetically-controlled analysis of Macroglossinae (Sphingidae) caterpillar traits. Using field and lab experiments I reveal that the protective value of eyespots to caterpillars is in fact size-dependent: eyespots are only intimidating to attackers when on large-bodied prey that are already easy to detect, yet eyespots hinder small prey by making them easy to detect and failing to intimidate attackers. I argue that caterpillar eyespots have evolved to serve a mimetic function and were generated and are currently maintained by the innate fear of predator eyes harboured to varying degrees by numerous species of insect-eating birds.