Velvet look-alikes: a most astonishing mimicry complex

18 December 2013 by Christopher Buddle, posted in Evolution, Wasps

At the Entomological Society of America's annual meeting last fall, I heard a talk by Justin Schmidt about sand ecosystems in the SW USA, and their fascinating and diverse arthropod fauna.  Justin showed some photographs of a series of velvet ant species in the genus Dasymutilla. Velvet ants, by the way, are not ants: these critters are parasitioids of bees and wasps, and are known to deliver a heck of a powerful sting (e.g., the 'cow killer').  The wingless females are quite conspicuous, 'velvet-like' in appearance, and are active during the day. Dasymutilla is a diverse and lovely genus of velvet ants, as depicted with photos of three species found in Justin's study area (reproduced here, with permission from Justin):

 

Dasymutilla klugii (top), D. magnifier (middle) and D. nogalensis (bottom)

Credit: J. Schmidt. Dasymutilla klugii (top), D. magnifier (middle) and D. nogalensis (bottom)

I was amazed and curious when viewing these, I hope you are equally interested - why would these different species all look so similar? Thankfully, Joe Wilson and colleagues published an excellent paper on the topic. They argue effectively that the Dasymutilla complex in North and Central America is comprised of a species-groups which are Müllerian mimics of each other. These kind of mimics look alike because they mutually benefit from this strategy - and in the case of velvet ants, the species-groups share a suite of secondary defences against predators.  Many of these velvet ants exude defensive chemicals, use stridulation to make warning sounds to predators, have a hard cuticle, and have a painful sting. So, their predators (mostly lizards, it is argued) learn to avoid trying to eat velvet ants, so that over evolutionary time, the species benefit from looking similar to each other. Wilson et al. (2012) investigated 65 species of Dasymutilla in North America and determined that the Müllerian colour patterns in Dasymutilla are the result of independent evolutionary events rather than a result of species groups looking more similar to each other because they are more  closely related (i.e., a shared phylogenetic history). Their main figure, reproduced below (with permission) tells the story:

 

Figure 1, from Wilson et al. Here, 65 Dasymutilla species are depicted: the mimicry 'rings' are defined by a shared colour, pattern and region. The small numbers under each represents a species position in a phylogenetic tree (note: they are all over the place, supporting independent evolutionary events).

Figure 1, from Wilson et al. Here, 65 Dasymutilla species are depicted: the mimicry 'rings' are defined by a shared colour, pattern and region. The small numbers under each represents a species position in a phylogenetic tree (note: they are all over the place, supporting independent evolutionary events).

These velvet ants are truly stunning, from the fuzzy, white species in the desert, to the tropical yellow and black species, to Justin's study species. To quote Wilson et al. "To our knowledge, the 65 Dasymutilla species treated here represent one of the largest Müllerian mimicry complexes yet to be identified, particularly in temperate regions". Indeed! Even the well-known butterflies mimicry complexes in the genus Heliconius only contains forty species (and they are featured in the Wikipedia entry about Müllerian_mimicry).  Without a doubt, arthropods continue to surprise and delight, and the marvellous mimicry in Dasymutilla is no exception.

Dasymutilla gloriosa

Credit: A. Wild. Dasymutilla gloriosa (fuzz-ball)

To finish, I'll add two quick facts.

1) One important piece of information from Justin is that despite the similar appearances among his Dasymutilla, the species are far from similar. For example, Justin tells me that of the four nearly identical species in his study area, D. klugii and D. magnifica are both about 250 mg in mass whereas D. zelaya and D. nogalensis are ~80 mg in mass. So: size doesn't seem to matter within this mimicry complex - similar appearances will do the trick.

2) You may have recognized Justin Schmidt's name from a RadioLab episode which highlights his fascinating pain scale for arthropod bites and stings, described here. A '2' with his scale is 'average pain', equated to a honey bee. The scale is logarithmic, so the '3' that he attributes to the sting of the larger Dasymutilla species described above, is, well, about 10 times more painful than a honey bee sting. Ouch. (oh, and the scale only goes to 4)

(note: a HUGE thanks to Justin for the photos and discussions about Dasymutilla)

Reference

Wilson JS, Williams KA, Forister ML, von Dohlen CD, & Pitts JP (2012). Repeated evolution in overlapping mimicry rings among North American velvet ants. Nature communications, 3 PMID: 23232402


3 Responses to “Velvet look-alikes: a most astonishing mimicry complex”

  1. Roy Niles Reply | Permalink

    Similar appearances for similar purposes (rather than having similar sizes as well) indicates an intelligently chosen process rather than the mimicry typical of an accidental selection process.

    • Christopher Buddle Reply | Permalink

      Thanks, Roy, for the comment. I'm not entirely sure what you mean, but natural selection is certainly a process that can explain the convergence of similar appearances in velvet ants, in response to shared predation pressure.

Leave a Reply


8 × = forty eight