The paradox of inaccurate mimicry

Many species gain a selective benefit through mimicry by converging on the phenotypic attributes of another unrelated species which has resulted in many examples of mimics bearing a striking resemblance to their model. It is assumed that natural selection should drive the evolution of accurate mimicry while eliminating inaccurate mimics from the population. Paradoxically, there are many instances of mimics that inaccurately resemble their model. The many hypotheses proposed to explain the occurrence of inaccurate mimics are reviewed, with no one hypothesis providing a complete explanation. This thesis explored a mostly overlooked hypothesis, the perfecting hypothesis that predicts that inaccurate mimics are in an intermediate stage in a transition toward accurate mimicry, using ant-mimicking spiders as a model. To investigate this concept, three methods of quantifying mimic accuracy in ant-mimicking spiders were evaluated for their efficacy. Following this, the phylogenies for two subfamilies of ant-mimicking spiders were reconstructed to map the distribution of mimic accuracy and the traits involved in ant mimicry and estimate ancestral states. The results indicate that mimic accuracy, and traits such as the constriction of the body, evolve via an incremental process, supporting the prediction of the perfecting hypothesis and further elucidating potential evolutionary processes in mimicry systems.