Tracking the evolution of warning signals

Abstract

EVOLUTIONARYstudies are hampered by a lack of experimental ways in which to test past events such as the origination of aposematism1–7, whereby unpalatable or poisonous prey signal their unprofitability, often by being warningly coloured. Inexperienced predators do learn to avoid unpalatable prey as a result of such signals8–10, but in addition there may be an inherited cautiousness about attacking when common or conspicuous warning signals are evident11–16. As current predators are not naive in the evolutionary sense, it is still not resolved3–7,17,18 whether aposematism originated only in aggregations of prey19,20 or among solitary prey as well21–23. Here we explore this controversy in evolutionarily naive predators by creating a novel world with warning signals not found in the environment. Initially, the aggregation of prey favoured the warning signals supporting Fisher's view24 of kin aggregations as the evolutionary starting point of aposematism. However, once predators had experienced warning signals, pre-existing avoidance seemed to facilitate evolution of Mullerian mimicry complexes25 with similar types of signals even among solitary prey.