“How extremely stupid not to have thought of that.”

The quote above is, of course, from Thomas Henry Huxley; it was his reaction after learning about Darwin’s simple but powerful idea of natural selection.  But I had exactly the same reaction after reading a new paper in The Proceedings of the National Academy of Sciences by Shevtsova et al. (reference below; free online access). The paper shows that insect wings, viewed under certain conditions of light, display stable patterns of iridescence that might be of profound behavioral (and evolutionary) significance.

I’ve spent 42 years looking at flies—ever since I was a sophomore in college—and I must have seen millions of them under the microscope.  Usually we look at them against a white background, which we use for inspecting the beasts for mutations or sorting them into males and females for mating.  Very occasionally, as when I dissected out female ovaries to count the number of ovarioles, I’d put them on a black background, which makes the white eggs more visible.

On those occasions, I could see an iridescent pattern in the wings, like what you see on the surface of a soap bubble or an oil slick on wet pavement.  I never paid that pattern any attention, and that was my mistake.

Unlike every other fly person who’s seen the patterns, the authors of the PNAS paper—Ekaterina Shevstova, two Swedish colleagues (C. Hansson and J. Kjaerandsen), and Dan Janzen, the well-known American naturalist and entomologist—didn’t ignore the colorful patterns. They found that these patterns (see examples below) are produced by interference between light reflected off the two surfaces of the chitinous wing membrane, that they are stable within an individual but often variable among sexes and among species, and that they’re found in most Hymenoptera (wasps and bees) and Diptera (flies).  They call the patterns “wing interference patterns” (WIPs), and discovered that they’re best seen against dark backgrounds, and when the light strikes at certain angles.

The paper is well explained by Ed Yong at Not Exactly Rocket Science, so I won’t go over the findings in extenso (plus, the paper is pretty easy to read).  First, look at a few of the WIPs.  Here’s Figure 1 of their paper (click to enlarge):

G, at the top right, is a hymenopteran (Closterocerus coffeellae), showing how the difference in background can suddenly reveal a hidden iridescent pattern.  H and I, below that, show the common fruit fly, Drosophila melanogaster (the one I’ve seen millions of), showing the same background-dependent pattern.  J and K are the wings of another species of Drosophila (D. guttifera), which has patches of pigment on the wings under “normal” illuniation but bright iridescence before a black background. The authors suggest that maybe the pigment spots are there not to be seen on their own, but to control the patterns of iridescence, which is what the other flies are really looking for.

A and B show two different but related species of hymenopterans, demonstrating that the WIPs can be highly visible when displayed in the right circumstances, and invisible otherwise.  C shows the wasp Neorileya: like many insects, it has a dark abdomen, and so the WIPs are visible when the wings are folded against the body.  D shows a sepsid fly displaying a bright pattern of color during a wing display (this may be a mating display; in many flies the males vibrate and waggle their wings when courting females).

Finally, E and F show the WIPs appearing in two fly species against different backgrounds, including the green of a leaf.

What are these patterns for?  The authors suggest that they may be of profound significance in insect communication. (See Ed Yong’s article for a precis).  As I suggested above, for example, they could be used in mating displays, with females preferring the patterns of males from their own species, since related species may differ in pattern.   The paper provides intriguing evidence that within a species of flies and wasps, the WIPs of males differ from those of females, making these patterns (like the colors or plumage of many birds) sexually dimorphic traits.

Whenever males and females differ in an ornamental trait without obvious adaptive significance, one hypothesis is that the trait evolved by sexual selection (see WEIT for an explanation).  This is particularly true if, in a group of related species, the males differ in pattern but the females don’t.  This invariance of a female trait coupled with strong variation in a male trait that could be related to mating is a strong signal of sexual selection.  And that’s the pattern we seem to see, at least in part of Fig. 3 from their paper:

In this figure, each column represents one of three related species of parasitoid wasp in the genus Achrysocharoides.  Two “replicate” males of the species are above the line, and the female is below the line.  So, for example, B and C are males of one species, and D is the female of that species.  F and G are males of a related species, H the female, and so on.

What you see is that while the WIPs of males vary among the species (but don’t vary much within a species), the females of different species (D, H, and L) are all pretty much the same.  To me, this implies sexual selection: the males diverge by sexual selection driven (perhaps) by female choice, while there’s no selection on the females to diverge.

This is precisely the pattern seen in insect genitalia: very often closely related species show big differences in male genital shape, while the females of those species are all the same.  As William Eberhart has pointed out often, this suggests sexual selection involving female choice. (See his wonderful and underappreciated book, Sexual Selection and Animal Genitalia.)

WIPs have other evolutionary significance.  They might be used not to recognize individuals of the same species for mating, but to discriminate against individuals of different species that have different WIPs.  That is, they could constitute a reproductive isolating barrier that contributes to speciation.

The authors (probably Dan Janzen, who studies fig wasps) point out another possible “use” of WIPs.  Female fig wasps pollinate the figs (and lay their eggs) by squeezing through the tiny hole at the bottom of the fig (the ostiole).  To get through, they break off their wings and leave them on the outside of the fig, at the same time secreting a fluid from their abdomen that glues the wings in a “protruding and visible position” on the fig.  This might serve as a signal to other females to leave that fig alone, since it’s already occupied with the reproductive output of a previous female.

Here’s a female about to enter the ostiole of a fig.  After she enters, and lays her eggs, she dies inside. Every time you eat a fig, you’re eating at least one dead wasp.

The WIPs are useful taxonomically, too: the authors have used them to diagnose “sibling species” of wasps: species that are almost morphologically identical.  It turns out that despite their similarity, the species have diagnostic wing patterns.  Once you separate species based on those patterns, you can begin to see subtler differences in morphology that you might have missed.

This paper doesn’t cure cancer or anything, but it’s a really nice presentation of a ubiquitous pattern in nature that may have important evolutionary explanations.  And how stupid of all of us drosophilists (and other entomologists) not to have paid attention to it!


Shevtsova, E., C. Hansson, D. H. Janzen, and J. Kjaerandsen. 2010. Stable structural color patterns displayed on transparent insect wings.  Proc. Nat. Acad. Sci. USA Early edition. doi/10.1073/pnas.1017393108.


  1. Tulse
    Posted January 7, 2011 at 9:20 am | Permalink

    It is amazing to me that something so simple to observe had not been noticed before. It’s delightful that such basic science is still available to be done.

  2. Dominic
    Posted January 7, 2011 at 9:21 am | Permalink

    Absolutely fascinating! Prof – your link was to the University of Chicago log in –
    – this should work I hope.

  3. Doc Bill
    Posted January 7, 2011 at 9:30 am | Permalink

    Put me right off figs, I can tell you that!

    • Dominic
      Posted January 7, 2011 at 9:41 am | Permalink

      We probably all consume a large amount of insect protein without knowing it – in flour for example. New word joy – ostiole!

      • Posted January 7, 2011 at 10:19 am | Permalink

        That doesn’t have to be bad. Why don’t we eat insects? A nice compact form of protein… ;-))

        (agree–ostiole…cool word! Remember…remember…)

        • Posted January 7, 2011 at 10:59 am | Permalink

          We do eat insects, usually without knowing it. This probably prevents some trace nutrient deficiencies for many people, so it’s not a bad thing.

        • Torbjörn Larsson, OM
          Posted January 7, 2011 at 11:13 am | Permalink

          Anybody who has visited the swedish mountains north in the summer knows that insects are a vital ingredient in outdoor food. It is virtually impossible to prepare and eat a meal without having several seen and unseen mosquitoes drown in it… (O.o)

          • Torbjörn Larsson, OM
            Posted January 7, 2011 at 11:16 am | Permalink

            Adding to Ray’s commentary, this ‘practice’ is of course a means for recapturing some of the iron the mosquitoes drain. 😀

        • Hempenstein
          Posted January 7, 2011 at 12:02 pm | Permalink

          In college a vegetarian friend used to share his fried grasshoppers with me. They came in a relatively large tin and tasted about like the brown fringe of a well-fried egg.

          • Tulse
            Posted January 7, 2011 at 12:18 pm | Permalink

            Since when are grasshoppers plants?

            • Hempenstein
              Posted January 7, 2011 at 4:57 pm | Permalink

              There are apparently all manner of vegetarians. Some will eat anything that can’t run away from you, hence snails are OK. I think my old pal’s OK with anything invertebrate.

              • Diane G.
                Posted January 7, 2011 at 5:36 pm | Permalink

                Wow. Then they’re either lousy biologists or lazy etymologists.

          • Dominic
            Posted January 10, 2011 at 3:23 am | Permalink

            I love that – “tasted about like the brown fringe of a well-fried egg”! And Diane’s comment!

    • daveau
      Posted January 7, 2011 at 10:33 am | Permalink

      Not a fan of Bear Grylls, eh?

    • Thanny
      Posted January 7, 2011 at 1:58 pm | Permalink

      Fig wasps are absorbed by the fig flowers (which produce enzymes for the purpose) after they die. If you wait long enough and put bags over the flowers to control how many wasps have access (as commercial fig growers do), you get a wasp-corpse-free product.

      • Doc Bill
        Posted January 7, 2011 at 6:20 pm | Permalink

        Actually, I once worked for a food company and we had an allowable amount of “parts” in the product. By “parts” I mean wings, legs, stingers, compound eyes and all the other crunchy stuff.

        From Monty Python: You mean this has frog bones in it? Well, we couldn’t call it Crunchy Frog now could we?

        At the fish counter last night I was looking at snow crab legs and thought, hmmmm, could be grasshopper …

        All a matter of what you’re used to. Tell you one thing, though, Ray Comfort put me right off bananas.

  4. daveau
    Posted January 7, 2011 at 9:48 am | Permalink

    Sexual selection on chitinous membrane interference patterns. I wonder what else we don’t notice?

    • daveau
      Posted January 7, 2011 at 9:51 am | Permalink

      It also occurs to me that this is a good example of why IDers think science is wrong. Because we don’t know everything, therefore we don’t know anything.

      • Posted January 7, 2011 at 11:36 am | Permalink

        It is an example of why free marketeers think socialists are wrong. The Plan will not accommodate happy accidents.

  5. Frank
    Posted January 7, 2011 at 10:16 am | Permalink

    Since these are interference patterns, one wonders if an insect can alter them by changing the degree of hydration of the wings – some shiny beetles do this with their wing covers. Also, I wonder if these patterns would look different under the relevant insect’s different domains of vision (compared to humans) – even if they are not pigment-based colors.

  6. Posted January 7, 2011 at 10:18 am | Permalink

    Thank you! Didn’t Darwin say something about being “observant” was what helped him be so successful?

  7. chlorion
    Posted January 7, 2011 at 10:55 am | Permalink

    It’s that damned box.
    Reminds me of Robert Frost’s “Neither Out Far Nor In Deep.

    “The people along the sand
    All turn and look one way.
    They turn their back on the land.
    They look at the sea all day.

    As long as it takes to pass
    A ship keeps raising its hull;
    The wetter ground like glass
    Reflects a standing gull.

    The land may vary more;
    But wherever the truth may be–
    The water comes ashore,
    And the people look at the sea.

    They cannot look out far.
    They cannot look in deep.
    But when was that ever a bar
    To any watch they keep?”

    • TreeRooster
      Posted January 7, 2011 at 11:33 am | Permalink

      Thanks! Yes, I know that box well. I hadn’t seen this poem though. Frost was certainly the observant one–at least one person at the beach that day thought to look in a different direction.

      No need to feel stupid, though, Jerry! Not that I don’t know how that feels, and I do assume you mean it as a complement to the authors (not backhanded; since when we use the phrase about ourselves we expect that others know we don’t really feel that way!) However, it is just the nature of research that sometimes solutions or undiscovered territory sit right in front of us all, waiting for the lucky/inspired one who tries the box at that particular corner.

  8. Posted January 7, 2011 at 1:43 pm | Permalink

    So I have two questions…

    How were these photos taken — with normal light? Through polarizing lenses? Have the colors been exaggerated in these photos? [I admit I haven’t read the paper — perhaps it’s all set forth.]

    And, related to Frank’s comment above, what does this suggest about insects’ vision? For one thing, these patterns are extremely tiny, so can compound eyes resolve this level of detail? For another, it would seem that the colors play an important part — can insect eyes perceive these colors? And are there yet more colors in the ultraviolet and infrared in these wings, that the insects are seeing but we haven’t noticed yet?

  9. Diane G.
    Posted January 7, 2011 at 1:51 pm | Permalink

    Wow, wow, wow, wow, wow! This is just SO COOL! (I get all erudite when excited…)

    Dan Janzen just keeps bringin’ it. Talk about 42 years ago–it was only slightly less than that that I first ‘discovered’ his work, and it seemed to be the epitome of all that drew me to biology.

    RE “How extremely stupid…”–we can be so grateful that remark stems from the era it does. “Doh!” just wouldn’t resonate the same way…

  10. אביגיל
    Posted January 7, 2011 at 2:39 pm | Permalink

    Fig Newton’s not a cookie! It’s fruit and cake and wasp.

  11. Alex SL
    Posted January 7, 2011 at 2:43 pm | Permalink

    Intriguing. I just wonder why insects would actually need that, as it seemed that they would rely a lot on pheromones for these purposes (and the eyes of many do not seem complex enough to appreciate the detail of these patterns as much as we can anyway…).

  12. Prodigal Sun
    Posted January 7, 2011 at 4:24 pm | Permalink

    I wonder how we taste to insects.Like humans?

  13. Xenithrys
    Posted January 7, 2011 at 5:56 pm | Permalink

    Orchids that are pollinated by pseudocopulation (tricking a bee or wasp into attempting to mate with an insect-like flower) have pairs of iridescent smooth stripes on the labellum that mimic insect wings. That aspect of the mimicry was well-known, but I guess no botanist wondered about how iridescence was part of the insect mating display. It’d be interesting now to see how closely they mimic their pollinators, and if this works towards pollinator specificity.
    Given Bateman’s principle, I’d predict males would try to mate with anything that looks vaguely like a female, unless precise patterns were associated with species recognition.

    • Ipswitch
      Posted June 27, 2012 at 1:12 pm | Permalink

      Hey, you better do some more research on Bateman’s principle:

      “Of overarching concern is how a 64-year-old classic that influenced
      “legions of graduate students” (and uncounted legions of undergrads,
      too) was fatally flawed yet went unchallenged all those years. “

  14. abadidea
    Posted January 8, 2011 at 10:56 pm | Permalink

    Thank you so much for this article. This is exactly the sort of science gem I love to find.

    But I am now totally grossed out about figs.

    • Dominic
      Posted January 10, 2011 at 3:28 am | Permalink

      But they are tiny tiny tiny wasps!

One Trackback/Pingback

  1. […] Jerry Coyne has a very interesting post about the light patterns reflected by fly’s wings. Really, the post is […]

%d bloggers like this: