Phylogenetics of mimicry
Mimicry is well understood and heavily studied within specific mimicry complexes. However the evolutionary and phylogenetic relationships between mimic-model or co-mimicry pairs are less apparent. The difficulty many researchers face in trying to build phylogenies for mimicry complexes is to discriminate between analogous traits, shared characteristics developed through convergent evolution, and homologous traits, shared characteristics that are due to a shared common ancestor.
To build phylogenies for these groups of mimics, scientists first need to understand which species is the mimic and which is the model, then afterwards determine how evolution will have proceeded to increase the instances of the shared characteristics.
In Müllerian mimicry, first described by Müller in 1862, defended species have evolved similar appearances as a means to share the cost of predator learning. The classic example of Müllerian mimicry is the Heliconius of butterfly. There are 54 species of this unpalatable genus, with over 700 names applied to its various forms (phenotypes). It functions as a perfect Müllerian mimic because all species of the Heliconius are inedible and form symbiotic relationships. Extensive research on Heliconius butterflies has shown not just phenotypic similarities, but behavioral commonalities within an overlapping territory.
It is commonly accepted that the driving force for mimicry is predator behavior, as can be shown in various computer and mathematical models  and comparative evidence. However, what is lacking is experimental evidence that shows how predatory pressures select for a common phenotype. Mimicry rings, the groups of mimetic species that evolve as a result of an ‘initial condition' phenomenon, also often overlap and the maintenance of their heterogeneity is also not well understood. The co-existence of multiple mimicry rings may contain individuals who do not completely overlap, which may result in lack of pressure to converge. Therefore, if different mimicry rings contain forms distinct enough, intermediate forms would be disadvantaged and forced to converge and match the phenotypes within the mimicry ring.
In Batesian mimicry, one defended organism is known as the model and the non-defended organism is the mimic (Bates, 1862). Batesian mimicry is often perceived as a parasitic relationship because the mimic benefits from the presence of the model and predatory learning as a result of experience with the model, however the model gains no benefit from the mimic. Interestingly, monarchs within their own species also form instances of Batesian mimicry as each butterfly varies in toxicity depending on its diet as a caterpillar, so those with a weaker defense depend on individuals with a stronger one. Polymorphisms are also more prevalent in Batesian mimicry than in Müllerian mimicry, with several important studies focusing on the female Papilio dardanus which mimics several species of defended danaid butterfly.
What is not well understood is how these species evolved through an intermediate stage and what phenotypes they might have displayed prior to successfully mimicking their models. After breeding Papilio dardanus individuals with differing phenotypes, results showed that the cross-breeding yielded variable, non-mimetic offspring. Experiments by Nijout (1991) and Turner (2000) have also suggested the possibility that the main pattern gene is contained within a supergene, a locus containing several tightly linked genes that control wing pattern phenotypes. Their experiments also showed the expression of the genes controlling for wing pattern depend on yet other genes, called modifiers, which co-evolved within a very specific population Kunte et al. (2014) found that the Batesian mimicry exhibited by the butterfly species Papilio polytes is controlled by the doublesex gene.
Understanding the function of the modifiers and the possible role of supergenes, however, does not make drawing up the mimicry ring's phylogeny any easier, nor do scientists yet know how mimicry is achieved.
Mimicry is often celebrated as one of the most straightforward examples of evolution by natural selection, however, several cases of imperfect mimicry have been documented. The phenomenon may help biologists understand the efficiency of and need for selection within a population. It is unclear why selection does not further improve the imperfect mimics and why it allows imperfect mimicry to persist within a population.
One hypothesis suggests that a limitation in a predator's cognitive ability permits imperfect mimicry. Predators may not be able to use all aspects of the prey's phenotype to distinguish an edible from an inedible species. For example, an edible snake species, L. elapsoides, imperfectly mimics the deadly species, M. fulvius. Although the mimic differs from the model by the order of colorful rings surrounding their body, their phenotypes match in other respects, keeping predators at bay nonetheless. Imperfect mimics evolve only the signals necessary to deceive predators.  Furthermore, if the mimicked traits are equal in their level of warning and one of the more salient traits seen within the model, deception will occur. As long as the imperfect mimic is deceiving its predators, the imperfectly mimicked traits will persist into the next generation and natural selection will not occur.
Another hypothesis suggests that mimics using multiple models may evolve imperfect mimicry as an intermediate form, rather than strongly resembling the several models they mimic. A study conducted with the octopus species, T. mimicus, revealed that despite its ability to impersonate a handful of different sea creatures imperfectly, enough confusion within the predators will occur, allowing the mimic to escape an unfavorable situation. 
Finally, a study conducted by Vesley et al revealed that the existence of predators and prey within the natural world condition the predators to avoid a certain type of prey and driving the predators to choose from a variety of different prey sources. Previous experience with the unpalatable prey increases the protection of the imperfect mimic, strongly affecting their selective advantage. Also, the model must outnumber its mimic within the natural world because the predator must experience the unpalatable prey more frequently in order to correlate unpleasantness with specific phenotypic traits. 
The prevalence of imperfect mimicry shows that natural selection does not always occur without fault. However, as long as the course of evolution increases the frequency of genotypes that produce phenotypes with higher fitness, the imperfections within mimics do not need to be selected against.
- Ruxton, G. D., T. N. Sherrat, and M. P. Steed. 2004. Avoiding attack: the evolutionary ecology of crypsis, warning signals, & mimicry. Oxford Biology.
- Müller, 1862
- Brower, 1996
- Mallet, J. and L. E. Gilbert, Jr. 1995. "Why are there so many mimicry rings? Correlations between habitat, behaviour and mimicry in Heliconius butterflies". Biol. J. Linn. Soc. 55: 159-180.
- Franks, D. W. and J. Noble. "Batesian mimics influence mimicry ring evolution". P Roy Soc Lond B Bio. 271: 191-196,
- Turner, J. R. G., E. P. Kearney, L. S. Exton. "Mimicry and the Monte-Carlo predator – the palatability spectrum and the origins of mimicry". Biol J Linn Soc. 23: 247-268
- Symula, R., R. Schulte, and K. Summers. 2001. " Molecular phylogenetic evidence for a mimetic radiation in Peruvian poison frogs supports a Mullerian mimicry hypothesis." P. Roy. Soc. Lond. B. Bio. 268: 2415-2421.
- G. D. Ruxton and M. P. Speed. "How can automimicry persist when predators can preferentially consume undefended mimics?" P. Roy. Soc. Lond. B. Bio. 273: 373-378
- Clarke, C.A. and P.M. Sheppard. "Further studies on genetics of mimetic butterfly papilio-memnon l". Philos. T. Roy. Soc. B. 263: 35-70.
- Nijhout, H. F. "Developmental perspectives on evolution of butterfly mimicry". Bioscience. 44: 148-157.
- Kunte, K., Zhang, W., Tenger-Trolander, A., Palmer, D. H., Martin, A., Reed, R. D., ... & Kronforst, M. R. (2014). "Doublesex is a mimicry supergene". Nature, 507(7491), 229-232.
- Wilson, J., Jahner, J., Williams, K., & Forister, M. 2013. "Ecological and Evolutionary Processes Drive the Origin and Maintenance of Imperfect Mimicry". PLOSONE 8(4):1.
- Kikuchi, D., & Pfenning, D. 2010. "Predator Cognition Permits Imperfect Coral Snake Mimicry". The American Naturalist 176(6): 830-834.
- Howse, P. E., & Allen, J. A. 1994. "Satyric Mimicry: The Evolution of Apparent Imperfection". The Royal Society 257(1349): 111-114.
- Huffard, C., Saarman, N., Hamilton, H., & Simison, B. 2010. "The Evolution of Conspicuous Facultative Mimicry in Octopuses: An example of Secondary Adaptation?" Biology Journal of the Linnean Society 101: 68-77.
- Vesely, P., Luhanova, D., Praskova, M., & Fuchs, R. 2013. "Generalization of Mimics Imperfect in Colour Patterns: The Point of View of Wild Avian Predators". The International Journal of Behavioral Biology 119: 138-145.