I have written about mimicry before, describing why most mimics are imperfect and how some mimics imitate not only the appearance of other animals but also their sounds and behaviour. Now, I need your help with an ambitious experiment to test theories about the evolution of mimicry. Most people know that there are harmless animals that have yellow and black stripes to look like stinging bees and wasps. But did you know that there are many thousands of such species, all with different degrees of “bee-ness” or “waspiness”? The new experiment is designed to compare 56 harmless hoverflies with 42 wasps and bees to measure how similar they are. That’s 2,352 unique comparisons! This information will allow us to test exciting new ideas about the evolution of mimicry. There’s only one catch…
This particular experiment will use the human brain as a processing tool and the power of the crowd to generate data. It’s a bit like “Strictly Come Mimicking” (or “Mimicking with the Stars“, if you’re in the US): you just need to rate how similar you think the two insects appear out of 10. I’d appreciate it greatly if you could take some time to run through the experiment below. Don’t do it thinking that there is an end, though – there are 2,352 combinations, remember, and the images are randomly paired on each screen! You can access the experiment here:
My goal is to reach 10 ratings of each pair of insects. That means a total of 23,520 ratings. I know this is a long shot, but that’s the aim, people! Please do share it far and wide! I’ll share regular updates on the blog as the ratings come in (however many or few there are!).
I just had my first article published at the Conversation – an excellent online collaboration between journalists and academics. As part of their publishing model, anybody can share any articles. So here’s mine!
To bee or not to bee – why some insects pretend to be dangerous
In the summer of 2011, panic gripped a small community in Gatineau, Quebec. Hundreds of small, striped insects were buzzing around a children’s playground. The playground was evacuated and entomologists were called in to establish whether or not the animals were dangerous. The answer was no, but it is easy to see why local residents were concerned. The animals that had taken up residence in the playground were hoverflies, a family of harmless fly species that have built up quite an arsenal of tricks to convince would-be predators that they are dangerous.
The panic that a swarm of hoverflies can generate belies the fact that they are immensely beneficial insects. Many of their larvae (the baby hoverflies that look like maggots) crawl around on plants feeding on the aphids that would otherwise eat our flowers and crops. Meanwhile the adults –- the stripy, flying insects that instil such terror –- spend their days pollinating flowers as they feed on nectar and pollen. But flying around in the open leaves hoverflies vulnerable to predators, a problem they have solved by evolving to resemble the stinging, pollinating insects such as bees and wasps with which they share the flowers.
Yet the story is not quite so simple. For every hoverfly that presents an exquisite example of mimicry (like the wonderful Spilomyia longicornis pictured above) there are several that really do not seem to be trying at all. Given that mimicry can obviously benefit hoverflies, why don’t they all evolve such excellent abilities?
Researchers found a potential solution to this Darwinian puzzle in 2012, when they looked into the characteristics of mimicking and non-mimicking hoverflies. You might expect that birds would prefer to eat larger species of hoverfly, since those hoverflies represent a bigger, more rewarding meal. Those larger species would therefore have more to gain from mimicry because they are under greater pressure from predators. Sure enough, it turns out that the colour patterns of the largest hoverflies (which are effectively flying buffets for birds) bear a close resemblance to the yellow, black, and white stripes of wasps and bees. The smallest species (which are barely worth chasing) do not show such similarity.
However, hoverflies have more than just wasp-like costumes. Some species also have considerable acting talents. It has been known for decades that certain hoverflies will pretend to sting when attacked, or hold their dark front legs in front of their heads to make it appear as though their antennae are long like those of wasps.
A recent extensive field survey showed that the species that behaved like wasps and bees were comparatively rare (just like the species that look like wasps and bees). This behavioural mimicry also tended to occur only in those species that already showed a strong visual resemblance to wasps and bees. In other words, those species that had the costumes also had the acting skills.
Insect sound bites
One of the most fascinating aspects of hoverfly mimicry has recently been dissected in great detail. As well as looking like wasps and bees, and acting like wasps and bees, some species also sound like wasps and bees. As part of our most recent project, my colleagues and I caught 172 insects from 13 species of hoverflies and nine species of wasps and bees, and brought them into a soundproofed recording studio. There, they recorded the sounds the insects made during regular flight and when the animal was attacked (simulated using a sharp poke with a pair of tweezers).
When they ran a statistical analysis on these sounds, the researchers found that some species of hoverfly make sounds when they are attacked that are indistinguishable from the high-pitched alarm buzz of bumble bees. The high-pitched buzz that bumble bees make seems to be produced by the bee unhooking its wings from the muscles that drive them, resulting in a completely different sound. This is a bit like what happens when you take your car out of gear and rev the engine – a lot of noise and you don’t go anywhere. It seems that hoverflies are capable of the same behaviour.
But just because statistical analysis can’t tell the difference, that doesn’t mean natural predators can’t. To test for the benefits of this sound imitation in the wild, researchers presented pastry models of insects to wild birds with the different sounds. Pastry has approximately the same nutritional content as the insects that the birds forage on naturally, being part fat and part carbohydrate. The pastry can also be painstakingly painted to resemble insects as well, as in the photo to the right. To the surprise of the researchers, the birds only avoided the bee sounds. This was despite the fact that the hoverflies sounded identical to the computer-based analysis.
So we are left with a situation where an animal brain outperforms human researchers and their technical wizardry, which is not altogether surprising. Birds have evolved alongside a host of potential prey, developing the ability to find safe prey while avoiding animals that sting. While the hoverflies have a complex and fascinating suite of acting skills to dissuade would-be predators, they are still part of an evolutionary “arms race” where predators either keep up or starve.
The best part of this particular story is that it is possible to watch it unfold in your back garden. Next time you see or hear an animal that makes you reach instinctively for the rolled up newspaper, take a minute to check that it isn’t one of nature’s great actors.
I blogged some time ago about a Cafe Scientifique talk I gave on the topic of “Avoiding Attack” (broadly mimicry and camouflage in animals). I stole the title of the talk wholesale from the excellent book of the same name written by former colleagues Mike Speed and Tom Sherratt along with Graeme Ruxton). After giving that talk, I was asked to contribute to the Leeds Festival of Science – a great initiative where University of Leeds staff engage local people (particularly schools) with their research through on-campus and external events. As part of that event this year I took part in the “schools roadshow” where researchers go out into schools to teach about their work. I thought I would post the resources that I used here with some notes so that teachers can make use of the materials that I produced. Everything here is released on a Creative Commons license (CC-BY 4.0).
Background: Animals and plants can benefit by resembling other species. For example, some plants have spots that look like ants to deter herbivores, cuckoos look like hawks to frighten smaller birds from their nests, and harmless snakes have striped bodies that resemble highly venomous species. However, there are other modes of resemblance: animals and plants can smell, sound or act like another species in addition to (or instead of) having visual resemblance. However, we don’t know much about how different types of mimicry interact in the wild.
What we did: Heather Penney, a MSc student at Carleton University, collected individuals from 57 species of hoverfly. Hoverflies are famous for having some examples of very close visual mimicry of stinging wasps and bees, but in some species this mimicry is “imperfect”. It is also known that hoverflies can exhibit behaviours that are characteristic of wasps and bees, and so Heather tried to elicit these responses from each of the species that she caught. She found that only 6 out of 57 species exhibited behavioural mimicry, and that these species belonged to only two genera (i.e. they were all closely related). Furthermore, there was some evidence that only animals that looked a lot like wasps also had behavioural mimicry.
Importance: While behavioural mimicry has been described a number of times in the wild, it is rarely surveyed using such a comprehensive approach – Heather tested every species in a community so that we know that there are a range of species that do not exhibit these behaviours. Also, we show that the behaviours are constrained to relatively few high quality visual mimics which suggests that behavioural mimicry acts to enhance morphological mimicry where that morphological mimicry already exists.
This is part of a series of short lay summaries that describe the technical publications I have authored. This paper, entitled “The relationship between morphological and behavioral mimicry in hover flies (Diptera: Syrphidae)”, was published in the journal American Naturalist in 2014. You can find this paper on the publisher’s website or for free at Figshare.
Image credit: Photos by Brent Lamborn, used with permission.
There are lots of ways to fool an observer, and I mentioned quite a few in my post on the Cafe Scientifique talk that I gave in September. However, one aspect that I didn’t mention there was “behavioural mimicry” – where an animal acts like another animal in order to fool a potential predator or prey. This sort of behaviour has been reported plenty of times in the field, but has never been studied in a systematic way. My collaborators over at Carleton (led by Tom Sherratt and Heather Penney, who collected the data as part of her MSc thesis work) and I have just published a paper (press release here) which provides just such an overview, and tests a few key evolutionary hypotheses along the way.Read More »
In September I gave a Cafe Scientifique talk at the Leeds City Museum on the evolution of mimicry and camouflage. For those of you who aren’t familiar with the concept, Cafe Scientifique offers an opportunity for scientists to give short (or long, depending on how it is run) talks on their research to a general audience and then take questions in an informal setting. I have always been a fan of this kind of outreach, and when Clare Brown, the curator of Natural History at Leeds Museum asked if I wanted to give a talk I jumped at the opportunity. I spent a bit of time pulling resources together for the talk and I thought I would post them here in case anybody else could find a use for them. I have outlined the talk I gave below:Read More »
A few colleagues and I recently had a paper published in Nature on “A comparative analysis of the evolutionary of imperfect mimicry”. Those of you fortunate to have a Nature subscription can read the paper here. Alternatively, you can email me and I’ll send you a copy. Unfortunately, I can’t make the paper available due to issues with copyright from Nature (see elsewhere for details of scientists’ love-hate relationship with publishers…) but I can summarise the paper here. Read More »