A flower hints at the appearance of an extinct bee

I’ve written before about “bee-mimicking orchids,” whose petals have evolved into a shape roughly resembling that of a female bee or wasp.  Randy male hymenopterans, lured by the shape (they can’t see very well), and, I believe, by the flower’s fragrance—which in some cases has also evolved to resemble insect pheromones—land on the flower and try to copulate with it.  They fail, of course, but during their fornicatory struggles the flower’s pollinia (sacs of pollen) detach from the flower and affix themselves to the insect’s head.

The frustrated insect flies away, but not long thereafter is fooled again by another flower of the same species. When it lands and tries to copulate again, the pollinia from the previous flower are transferred to the new orchid and pollinate it.  In this way flowers have evolved to use the insect as a sort of flying penis, a way to effect pollination through trickery.  Here’s a short video in which David Attenborough shows how it works:

Today’s xkcd describes such an orchid, Ophrys apifera, which is found throughout Europe.  The kicker here is that, according to the comic (but see below), the pollinator has gone extinct, though its appearance can still be discerned through the appearance of the flower. Since there’s no insect around to pollinate it, the flower has evolved self-pollination (that is, those individuals that were able to self-pollinate were the only ones to pass their genes to the next generation).


It’s a nice story, and the part about the extinct pollinator and the evolution of self-pollination is probably true, but I’m not sure one can discern the appearance of the extinct pollinator from the flower’s appearance. The Orchids Wiki notes this:

The flowers are almost exclusively self-pollinating in the northern ranges of the plant’s distribution, while pollination by the solitary bee Eucera occurs in the Mediterranean area. The sepals are marginal and spread out, coloured mauve to pink, often with a greenish central line. The flower lip is furry to the touch and is quite variable in the pattern of coloration, but is usually brownish-red with yellow markings.

The pollinia are produced on the inner face of a greenish column overhanging the lip, ready to deposit the pollen on visiting bees. Eucera bees in the past have influenced the evolution of bee orchids. Male bees, over many generations of cumulative orchid evolution, have favoured plants with the most female-bee-shaped lip through trying to copulate with flowers, and hence carrying pollen.

The flowers have a generalized shape, and I’m not sure whether individuals of O. apifera in southern Europe, where they mimic an extant insect, differ in appearance from those in the northern part of the range, where they’re self-pollinated and supposedly resemble a “ghost insect.”  If they don’t differ, then we can’t say, as the cartoon does, that the shape of flowers in, say, England tells us something about the appearance of the extinct pollinator. There must be work on this, but I don’t know of it.

It would also be worth studying whether the northern flower still retains a hymenopteran-like fragrance and, if so, whether it differs from the fragrance of southern flowers that attract Eucera bees.

Regardless, today’s xkcd is poignant, and does impart a nice lesson about biology and evolution.

h/t: David, infiniteimprobability


  1. Jesper Both Pedersen
    Posted September 2, 2013 at 5:00 am | Permalink

    I like that cartoon. It made me aww a bit.

  2. Posted September 2, 2013 at 5:10 am | Permalink

    Would the most parsimonious explanation not be that the orchid evolved to be pollinated by the Mediterranean bee, then a mutant occurred that could self-pollinate and was thus independent of pollinators so that it could spread further north where the bee does not occur?

    Similar things happen all the time. Here in Australia are various daisies that have outcrossing and self-pollinating morphs, and the latter are usually much more widespread. One can easily imagine how that happens: if you are outcrossing and only one of your seeds gets dispersed to a new place, it is toast (especially considering that those daisies are generally annuals). If you are a selfer, one seed is enough to establish a new population. Of course, in the long run it is probably a dead end, lack of recombination and all that…

    • Posted September 2, 2013 at 5:27 am | Permalink

      That is a good hypothesis, and it would be testable by looking at the age of the separation dates of the populations.

    • Whocares?
      Posted September 2, 2013 at 10:52 am | Permalink

      Almost all daisies have crossed pollination simply because you can’t have a thousand flowers (cause each daisy is, in fact, a bunch of flowers resembling a single one) without them pollinating one another. Anyway most high flowers have self-polliating emergency sistems, so if they are not pollinated they self-pollinate.

      My point is, the self-pollinating daisies are not bound to a dead end because they can still do crossed pollination so in the end they have more success because they can expand with and without pollinators.

      • Gregory Kusnick
        Posted September 2, 2013 at 11:13 am | Permalink

        Sorry, I’m not following you.

        Each daisy plant is a single genetic individual, no? So all of those thousand florets are genetically identical. So I can’t see how it counts as cross-pollination if one floret pollinates another. The plant is still fertilizing itself.

      • Posted September 2, 2013 at 3:01 pm | Permalink

        You may be confusing a breakdown of self-incompatibility with an adaptation to self-pollination here. Yes, they could, theoretically, cross-pollinate if a pollen grain got to the right place. But these Asteraceae usually evolve towards a reduction in capitulum size, less wide opening of the capitula and florets, and at least an order of magnitude reduction in pollen amount. Compare, just for one example genus, how that looks in Podotheca:


        In other words, they evolve to be less attractive to pollinators and to have so little pollen that it becomes less likely to fertilize somebody else even if there is accidental transport. It becomes increasingly hard to evolve out of that situation again.

  3. Posted September 2, 2013 at 5:30 am | Permalink

    If the bee went extinct long ago, then there is no longer selective pressure on the flower to mimic the bee, so we would expect increased variability in flower color and some loss of accuracy. Still, the flower should give us some clues about the female bee’s appearance (as seen through the eyes of the male bee, as noted in the cartoon).
    Quite a cartoon!

  4. marksolock
    Posted September 2, 2013 at 5:58 am | Permalink

    Reblogged this on Mark Solock Blog.

  5. jesse
    Posted September 2, 2013 at 6:32 am | Permalink

    There is an excellent book that somebody reading this post might like to know about. It’s called Orchid Fever, by Eric Hansen.

  6. teacupoftheapocalypse
    Posted September 2, 2013 at 6:33 am | Permalink


    • Merilee
      Posted September 2, 2013 at 8:30 am | Permalink


      • Sagra
        Posted September 2, 2013 at 8:42 am | Permalink

        Sub means “I don’t have anything to contribute at this time but I want to subscribe to follow-up comments via email, which I can’t do without posting a comment.”

  7. hyperdeath
    Posted September 2, 2013 at 6:34 am | Permalink

    Determining the appearance of the bee is difficult, because the flower’s appearance is just as much a product of the bee’s visual system as of the appearance of the bee. It’s not necessary for the flower to be a realistic depiction, it just has to be able to trigger recognition in a primitive visual system.

    In fact, realism may be a suboptimal solution. A flower with shapes and colours which exaggerate certain visual cues may be more recognisable.

  8. geckofeet
    Posted September 2, 2013 at 6:42 am | Permalink

    I’m sorry, could someone explain this to me? If the pollinating bee is extinct, where do the males bees come from? If the males are a different species, why should the flower mimic the extinct bee?

    • Whocares?
      Posted September 2, 2013 at 10:55 am | Permalink

      The male and female bees are both exctint. When they were still alive the flower used the males to spread their pollen. Now that they have gone extinct the only way for the orchid to survive is to self-pollinate because no bee is atracted to it anymore.

      • geckofeet
        Posted September 2, 2013 at 1:53 pm | Permalink

        Ah, got it. Thanks!

  9. Diana MacPherson
    Posted September 2, 2013 at 6:58 am | Permalink

    I wonder if other insects check it out anyway or even hummingbirds. I see hummingbirds check out my gigantic Canna lilies. Or, is the lily just doomed to self pollination all the time?

    • Posted September 2, 2013 at 7:37 am | Permalink

      The canna is bred from ancestors that are hummingbird-pollinated. The new hybrids still have most of the cues of the hummingbird syndrome.

  10. Kurt Helf
    Posted September 2, 2013 at 7:56 am | Permalink

    It is a good strip but, while the cartoonist was introducing a neat natural history, couldn’t they have gotten the binomial nomenclature right?

    • Gregory Kusnick
      Posted September 2, 2013 at 11:07 am | Permalink

      Looks right to me, other than capitalization. But then all text in such comics is capitalized by convention, so the caps aren’t wrong in this context.

    • John Scanlon, FCD
      Posted September 5, 2013 at 10:17 am | Permalink

      If you look carefully, you’ll see that the O in Ophrys is a bit taller than other O’s in the same panel. That’s how Randall indicates capitalization of names, subtly but consistently; there is nothing wrong about it.

  11. ridelo
    Posted September 2, 2013 at 8:17 am | Permalink

    For those bees an Alpine meadow must look like a Koranic paradise.

  12. Gregory Kusnick
    Posted September 2, 2013 at 11:09 am | Permalink

    For some reason I read your second sentence as “…lured…by the flower’s flagrance…”

    I have to say I like my version better.

  13. Posted September 2, 2013 at 3:51 pm | Permalink

    Well, it *is* only a memory… Evolution can’t look forward but it sure likes to get nostalgic.

    It would do the world good to have more interpretations of science in this kind of humanistic or mythological context – People who become gut bacteria metabolizing meaning for those unable to digest raw facts on their own.

    • Shanehar Allmighty
      Posted September 3, 2013 at 3:06 pm | Permalink

      I just quoted you on facebook because that was elegant and full of truth I identify with.

  14. I of Horus
    Posted September 2, 2013 at 4:02 pm | Permalink

    If the shape of the flower is indicative of preferential selection by an EXTINCT bee, why hasn’t the flower lost its distinctive shape? After all, as there is no evolutionary pressure to keep looking like this, I would expect changes to arise fairly rapidly (rapidly in biological terms that is).
    This lead me to assume that the extinction is recent. Is it known which bee used to pollinate that orchid? And if not, can the extinction be dated by changes in the phenotype?
    Or do the bees in the Southern end of its range exert that evolutionary pressure? If so, how does that pressure extend all the way to the UK?

  15. Ken Kukec
    Posted September 2, 2013 at 5:58 pm | Permalink

    So The Designer created one entity whose continued existence depends upon its prick-teasing another of His designed creations? (Predator/prey and parasite/host wasn’t enough inherent cruelty?) That’s some all-loving deity there!

  16. Posted September 2, 2013 at 6:52 pm | Permalink

    Fascinating stuff. I recently read about this very thing in The Greatest Show on Earth, by Prof. Richard Dawkins.

  17. Diane G.
    Posted September 3, 2013 at 12:23 am | Permalink


  18. Joseph Allen
    Posted September 8, 2013 at 5:09 pm | Permalink

    Is everyone here confusing bees with wasps?

    Aren’t all worker bees who pollinate flowers really infertile females with no male suitors in sight?

    Is there some species of bee in which workers mate male-to-female?

    Am I living in an entomological ghetto of gross bee generalizations?

    Or has no one else noticed this basic error in an otherwise amusing cartoon?

    • Gregory Kusnick
      Posted September 8, 2013 at 11:53 pm | Permalink

      All your questions can be answered by clicking through to the Ophrys apifera article Jerry linked to in his post.

      • Joseph Allen
        Posted September 9, 2013 at 3:02 am | Permalink

        Well, I’ll be damned. Eucera bees. Solitary bees.

        I’ve seen a billion carpenter bees and bumble bees, but for some reason, I never thought of them as bees even though they’re called bees. But they’re bees.

        Thanks, Gregory. Now I’m out of the ghetto and into the know.

  19. Posted September 11, 2013 at 11:54 am | Permalink

    The TV extract, incidentally, also shows beings who lack revulsion for homosexual sexual contact, albeit accidental contact.

  20. Saša Iskrić Smrekar
    Posted September 17, 2013 at 7:35 am | Permalink

    A humble attempt at reconstructing the unfortunate bee that exists now only as a painting on a flower …

5 Trackbacks/Pingbacks

  1. […] the risk of introducing science into an otherwise charming strip, one commenter at the Why Evolution is True blog points out […]

  2. […] usual, the science of this orchid story is a little more complicated than made out (check out this post and the great comments to learn why). It’s just as likely that self-pollinating orchids […]

  3. […] Vill du läsa mer så gå till bloggen Why Evolution is True. […]

  4. […] usual, the science of this orchid story is a little more complicated than made out (check out this post and the great comments to learn why). It’s just as likely that self-pollinating orchids […]

  5. […] Then discussion on: https://whyevolutionistrue.wordpress.com/2013/09/02/a-flower-hints-at-the-appearance-of-an-extinct-be… […]

%d bloggers like this: