The coming and going of Batesian mimicry in a Holarctic butterfly clade

Fiedler BMC Biology 2010, 8:122 http://www.biomedcentral.com/1741-7007/8/122 CO M M E N TA R Y Open Access The coming and going of Batesian mimicry in a Holarctic butterfly clade Konrad Fiedler* See research article http://www.biomedcentral.com/1471-2148/10/239 Abstract A study using phylogenetic hypothesis testing, published in BMC Evolutionary Biology, suggests that non-mimetic forms of the North American white admiral butterfly evolved from a mimetic ancestor. This case might provide one of the first examples in which mimicry was gained and then lost again, emphasizing the evolutionary lability of Batesian mimicry. Comparative studies of butterfly wing patterns by naturalists such as Henry Bates, Roland Trimen and Fritz Müller marked the birth of the scientific concept of mimicry right at the onset of the Darwinian age. Even today, the factors governing the origin, maintenance and dynamics of mimicry systems remain a challenge for evolutionary biologists, and butterfly wing patterns continue to provide prime models for developing and testing new ideas concerning the mechanisms governing how mimetic phenotypes arise and what factors may regulate their maintenance [1]. The emergence and maintenance of Batesian mimicry, in which palatable mimics share conspicuous warning color patterns with unpalatable models that are protected from predation by their aposematic pattern, is particularly intriguing in evolutionary terms. Here, fitness benefits will accrue to the non-toxic mimic only as long as the toxic model remains present, and in large enough numbers, to ensure that predators are familiar with it and are thus warned off by its characteristic appearance. Otherwise, selection should favor the disappearance of palatable mimics, which suffer from higher predation risk than incon spicuous phenotypes. [2]. White admirals, that is, the Holarctic butterfly genus Limenitis (Nymphalidae), have been the target of *Correspondence: Department of Animal Biodiversity, University of Vienna, Rennweg 14, 1030 Vienna, Austria research into the function and evolution of mimicry for more than 40 years. The genus comprises about 25 species in Asia, Europe and North America. Most of them show disruptive wing coloration [3]: dark brown with white bands stretching across fore and hind wings, and undersides similar. Within the four North American species, sometimes referred to as subgenus Basilarchia, two radically different phenotypes occur that exemplify two different mimicry syndromes. On the one hand, Limenitis archippus, the viceroy, is orange colored and forms a Müllerian mimicry ring with toxic Danaus plexippus (the monarch) and D. gilippus [4]. In a Müllerian mimicry ring, all species share a common warning color pattern, and since they are all unpalatable to predators, they collectively benefit from this common signaling. On the other hand, L. arthemis (the white admiral) comprises an experimentally proven example of Batesian mimicry [5]. Its northern two subspecies, arthemis and rubrofasciata, show the disruptive colora tion usual for the genus and are non-mimetic. However, the southwestern (arizonensis) and southeastern (astyanax) subspecies are bluish without white bands, and with conspicuous red dots ventrally. They are mimics of the toxic pipevine swallowtail Battus philenor. The four forms of Limenitis arthemis freely interbreed in nature as well as in captivity and thus belong to the same species under the biological species concept. The past 5 years have seen an interesting controversy as to whether mimetic forms in the L. arthemis complex have evolved once (monophyletic mimicry hypothesis (MMH); Figure 1b), or whether the non-mimetic arthemis phenotype might constitute an example of the reversion to an ancestral phenotype from a mimetic one (reversion hypothesis (RH); Figure 1a). A first sequencebased phylogenetic analysis [6] did not support the MMH, but soon after, Savage and Mullen [7] concluded the MMH to be more appropriate on the grounds of amplified fragment length polymorphism (AFLP) data, disputing support for the RH obtained from mitochon drial sequence data [8]. This controversy might appear of © 2010 Fiedler; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Fiedler BMC Biology 2010, 8:122 http://www.biomedcentral.com/1741-7007/8/122 (b) (a) L. a. arthemis Page 2 of 3 L. a. astyanax L. a. arizonensis Reversion hypothesis L. a. arthemis L. a. astyanax L. a. arizonensis Monophyletic mimic hypothesis Figure 1. Two contrasting hypotheses of mimicry evolution in the Limenitis arthemis species complex. (a) According to the reversion hypothesis, mimetic L. a. astyanax is sister to non-mimetic L. a. arthemis. Under this hypothesis, the mimetic phenotype arose in the common ancestor to all L. arthemis and was subsequently lost in the L. a. arthemis lineage. (b) In contrast, the monophyletic mimic hypothesis predicts that the mimetic lineages L. a. astyanax and L. a. arizonensis are most closely related to each other and Batesian mimicry evolved only in the stem group of these two subspecies. Recent phylogenetic hypothesis testing [9] provided evidence in favor of the reversion hypothesis. Figure modified from [9]. little general significance, were it not for the fact that evidence for a reversion from a mimetic to an ancestral phenotype is extremely rare so far [2]. Phylogenetic hypothesis testing allows for new insights In a study recently published in BMC Evolutionary Biology, Oliver and Prudic [9] now revisit the case. They use sequence information from eight nuclear loci com bined with coalescent simulation of gene trees to evaluate a range of models of population structure and evolu tionary history. By using multiple loci they compensate, at least partially, for the problem of gene-tree/speciestree discrepancies. The main difference from earlier approaches, however, is that Oliver and Prudic use sophisticated statistical models to measure how well their simulations fit the empirical data. Parameters for these 15 models were estimated divergence times (for species evolution) and migration rates (for population structure), taken from earlier studies of the same species. The advantage of this approach is that explicit models for contrasting evolutionary scenarios are compared with each other. Hence, inference is based on rigorous statistical tests of explicitly formulated alternatives. Oliver and Prudic found that the MMH had to be rejected: the only model that fitted the data in all aspects was a scenario that assumes moderate migration rate of the butterflies plus divergence times of about 655,000 years for the split of arthemis from astyanax, and 1,075,000 years for the split of the western ari (...truncated)