Spider mimics bird dropping

June 5, 2014 • 5:12 am

Readers know by now that I love mimicry. This is for many reasons, but I suppose foremost among them is that it shows the power of natural selection to “mold” an animal to closely resemble something else. (That’s a metaphor, of course, for natural selection is not something “outside” that “molds” an animal or plant, but simply a process of the accumulation of genes that, in this case, help keep their carrier from being killed.)

A common “something else” is bird droppings, which of course are unpalatable to predators like birds and wasps. Ergo, many insects and spiders, as well as vertebrates like frogs, have evolved to resemble bird droppings, hiding themselves from predators. This would of course be favored by selection, for any resemblance to a dropping reduces your chance of being nommed, and increases the chances of passing on your genes. Over time, genes would accumulate that would make you look, within developmental and ecological constraints, as close to a bird dropping as possible.

According to a paper in Nature’s Scientific Reports by Min-Hui Liu et al. (reference below; free access), this has happened in the spider Cyclosa ginnaga from East Asia.

It’s long been known that this spider weaves decorations into its web that look like droppings, but some spiders of this genus have also evolved to look like droppings. When a spider sits in its decoration, the combination looks remarkably like a bird dropping, thereby avoiding attention from predators. Here’s a photo from the paper showing the spider sitting in the decoration it’s woven in its web (a), along with a normal bird dropping (b). The scale bar is 5 mm. (about 1/5 inch). Note that the spider’s abdomen, and perhaps the rest of it, resembles bird droppings as well.


Here’s another photo, from a Smithsonian article about the paper, showing the similarity between bird droppings alone (first and third rows) and spiders on their webs (second and fourth rows). The spiders are oriented on a vertical web, just as many bird droppings are on vertical surfaces:

Photos by Min-Hui Liu

The authors wanted to test the hypothesis that the mimicry was not only something that deceived the eyes of wasp predators, but also reduced predation on the spiders. To do this, they first did a spectral analysis of the spiders on their decorations and compared their “colors” to a computer model of the sensitivity of the hymenopteran eye (wasps, along with bees, are hymenopterans).  They found that the spiders were indistinguishable in this spectrum from both their web decorations and from real bird droppings—again using the model of what a wasp eye could see.

The authors then changed the colors of spiders and their webs using a black marking pen to darken the spiders’ bodies, and black carbon powder to darken the web decorations.  There were three treatments: darkened spider on darkened decoration; unmarked, light spider on darkened decoration; darkened spider on unchanged light decoration, and the control: unmarked spider on normal, undarkened decoration. They then used video cameras to record incidents of wasp attacks.

Only one of the treatments significantly increased the predation rate on spiders: the normal, light spider sitting on a decoration that had been blackened with carbon powder. None of the other treatments affected predation. Here’s a graph from the paper showing the elevated predation in the normal spider/blackened decoration treatment (fourth bar to the right) compared to the control and the other two treatments. Predation was increased more than fourfold, a significant selective force:

Screen shot 2014-06-05 at 6.50.45 AM

It’s not hard to understand why a black spider on a blackened decoration (first bar to the left on graph above) isn’t spotted by predators so easily, because it’s camouflaged. But it’s more puzzling why the blacked spider on an unblackened decoration (third bar from the left) didn’t experience higher predation. Nor do the authors (or the Smithsonian piece) discuss this anomaly. Perhaps the spider is sufficiently hidden by being on a larger spot that looks like a bird dropping, or perhaps the blackness of the spider simply looks like a bird dropping, too, for droppings contain black bits. But that doesn’t explain why the spider itself would evolve body markings resembling bird droppings. Why would that happen if its body color was unimportant compared to the color and design of the web decoration it makes?

There are possible answers, including the presence of other predators (like birds) that weren’t detected. And the effect of hiding yourself from your prey wasn’t considered.  What the paper does show is that the color of the spider’s decoration helps hide it. What it doesn’t show is that the spider itself, sitting in that decoration, is part of the mimicry. The authors’ conclusion that “C. ginnaga‘s decoration and body coloration forms a bird dropping masquerade”, then, seems a bit dubious to me.

Perhaps readers have other explanations.

h/t: Jim

Reference: Liu, M.-H., S. J. Blamires, C.-P. Liao, and I.-M. Tso. 2014. Evidence of bird dropping masquerading by a spider to avoid predators. Scientific Reports 4:10.1038/srep05058.


25 thoughts on “Spider mimics bird dropping

  1. I don’t pretend to know the answer. I do know that the question regarding the function of web decorations still isn’t fully settled. My co-author, Catherine L. Craig, was one of the first researchers to address this question experimentally, and she had results that strongly suggested that many decorations attract prey. Wasps get more spiders than birds do, but both get spiders. Spiders’ prey are, like wasps, arthropods. Arthropod and bird eyes function differently. My guess is that there’s some balancing act going on here between attracting prey and repelling different kinds of predators. There’s no end of experiments that could be carried out on spider webs. My other guess is that, given the vast diversity of spider species, what’s happening with one species may not necessarily predict what’s happening with another.

    1. I have a related question. Spiders in the Argiope genus (commonly known as garden spiders) will sit in the center of their orb webs, and the center is decorated with thick, zig-zaggy silk. I know there has been long debate about the purpose of the decoration. Is there any settling of the issue?

      1. Cay Craig’s experiments were on Argiope. A number of papers since have supported the hypothesis that the decorations do attract prey. I don’t think it’s settled that that is their only function, but, at least for the Argiope species involved in the various studies, prey attraction looks to be a selective force.(We have a chapter on this in our book and you can look for Cay’s papers and her OUP monograph if you want something more technical. There’s also an interesting 2012 paper in Behavioral Ecology and Sociobiology by Kim, Kim, and Choe, which has references to papers by others.)

  2. But it’s more puzzling why the blacked spider on an unblackened decoration (third bar from the left) didn’t experience higher predation

    While its hard to say why the web matters more than the spider, the experimental findings are at least consistent with the more general observation you made above, that people have long known about the web adaptations but are just discovering the spider adaptations. All other things being equal, we might reasonably expect the web adaptation to be better mimicry if it has a stronger impact on fitness.

    1. So, possibly various forms of selection bias?

      Then the spider color is not part of the mimicry (as per the predation figures). And the spectral match a coincidence or even a bad observation.

    2. It appears that when the spider is on the decoration, the spider’s color variation doesn’t matter because of natural variations in the color of mostly white bird poop. However, the spider is vulnerable when away from the decoration. This may be when not looking like white poop becomes a disadvantage.

      1. They did change the color of some spiders and it didn’t change the predator attack probability.

        Ah, but you made me realize I had sloppily read that as the predation/death rate. Maybe there were more selection bias at work and they only looked at spiders when they were on the spots?

  3. Amazing mimicry! I posted to G+ without using the provided share button so I could choose a better photo than what was automatically shown, that is, of the chart.

    Maybe the ducklings in the campus pond (from an earlier post) should try disguising themselves in this way?

    1. =D

      Else, I wish it was so simple!

      The Fermi question has too many possible false negatives to be useful as far as I can see.

      And a first order economical analysis says, unless I am mistaken, that there isn’t much of an economical incentive to be trying to communicate (too expensive) or explore (too expensive compared to what we can do from here) or travel (too expensive) or trade (too expensive). Even colonization is cheaper to Oort cloud bodies than to planets, and maybe there is where putative colonists ‘hide’; all over the place, between stars – but again communication in between rapidly diverging biospheres is … (too expensive, and ultimately futile).

      It isn’t much of a “paradox” given known constraints.

      But yes, if one wants to look at 100+ possible other answers, it is alluring. Reminds me of looking at traits in species. 😉

      1. When you work through the economics, it turns out that, in order to even be able to afford interstellar travel, you have to have access to a significant fraction of a star’s energy. That leads to a corollary: that the only reason for interstellar colonization is that you’re exhausting the energy available from your own star. That then puts you on an exponential growth curve such that, even if you posit that it takes as long for a new colony to grow to its parent civilization’s resource usage than it has for humans to evolve from the first population of Hominina (~2.5 MYA), if that species got its start about the same time as the dinosaurs went extinct (~93 MYA) then the galaxy would be overrun already. (You’d need log2(300,000,000) = 28 doublings; 93 / 2.5 = 36.8, or enough for 500 times the number of stars in the galaxy. Exponential growth is a real bitch.)

        As such, we can be reasonably confident that there are practical limits to growth, whatever they may be.



        1. Where’d your 93 come from? K-Pg boundary is about 66 mya, so (if I follow your calculation correctly) 66/2.5 = 26.4 and the galaxy should not necessarily be overrun yet. Whew!
          But I think the solution to the Fermi Paradox (cf. Drake Equation) is that technologically advanced species arise on very few of the myriad planets with life. It’s not necessary to assume those few quickly kill themselves off completely (or get killed off by cosmic sharks/spiders), but returning to a basic, sustainable level of tech after one or two overpopulation spikes would look much the same from where we sit.

          1. Except that UN’s figures now point to that we will have no such spike. The human population numbers will (as they are projected right now, obviously) level out smoothly and avoid an even faster and perhaps riskier drop than the rise.

            So we can’t automatically assume there will be spikes, and the current evidence speaks against.

            It seems to be the biology consensus that language capable intelligence is rare. And yes, perhaps that is enough.

            On the other hand, colonization is an investment over generations with uncertain outcome, and if the source is fairly well off and with balanced population (which we seem to converge on by the UN’s et cetera recent data), there isn’t much incentive. It will happen anyway of course, people climb mountains and go to the Moon “because it sits there”.

            I think colonization over interstellar distances is a lot more silent and haphazard than people account for.

            Any of those mechanisms, or both together, makes for negatives. My point with the colonization example is that the constraints of the question potentially makes for false negatives as the natural outcome. Meaning that the Fermi question is neither a paradox nor much useful. We can for example from the absence of other intelligences not chose between “intelligences are rare” and/or “colonization is silent”.

            1. You’re looking too narrowly at population figures. Yes, they’re showing signs of leveling off…long past the time when petroleum reserves will have been essentially depleted. And it’s petroleum that’s feeding those people today — petroleum in the form of fertilizer for the crops, petroleum in the form of diesel fuel to run the farm machinery and the trucks to get the produce to market, petroleum in the form of fuel to power other huge segments of the civilization’s economy necessary to glue it all together.

              And those figures are also forecasted past the time when pollution from all the fossil fuels we’ve been burning and will continue to burn will have dramatically reshaped the agricultural landscape such that many areas now fertile will be useless to growing crops.

              All in all, I’d say there’s no chance of avoiding a population crash at this point. Maybe we can cushion the impact a bit, but we’re not even showing any interest in doing so. Jesus is coming, donchano? And it would hurt the economy….


              1. Population figures is the main driver.

                Re oil, “Peak Oil” is another meaningless term. Of course there is a maximum production rate if you drain fossil fuel faster than they renew (over geological time scales). But the original theory was describing _single oil fields_ and were shown to be incomplete already then. (I.e. didn’t predict the correct peak nor tail.)

                The alternative is to use economists market models, in which case the ‘peak’ is a flat plateau. (And definitely no wars.) FWIW, I think the IMF come up with the first realistic model last year. It combines oil field behavior (akin to peak oil) to get at price shocks, with market models to get at price evolution.

                The prediction is that oil will become very expensive but there is no shortage to make war over. Other energy sources will be able to take over nicely. E.g. gas & solar.

                Conversely, I don’t think there is any evidence for problems avoiding a population crash. Our population figures have been robust (albeit increasing) since the last Out Of Africa Event. The earlier instability had to do with insufficient technology.

              2. I’m not familiar with the UN figures you mention.

                But the basic, undisputed facts are that we’ve used roughly half of the world’s petroleum reserves; that the remaining reserves are low-quality and expensive-to-reach deposits such as tar sands and deep sea fields; that demand is already outstripping production with the gap expected to widen as demand grows and production falls; and that there’s no alternative available right now to take the place of petroleum.

                That’s a recipe for global financial catastrophe that will make the Great Depression look like an Hawaiian vacation. If you thought the price shocks of the ’70s oil crises were bad…well, that’s nothing compared to the perfect storm that’s looming on the near horizon.

                Sure, market meltdowns will curb the demand for petroleum and thus stretch out remaining reserves…but most of the demand for petroleum is inelastic: it’s food (fertilizers, farm equipment fuel) and transportation (without which there is no global economy). It’s one thing to shave a few fractions of a percent off growth rates by encouraging public transit and carpools for commuters; it’s another to have food riots because of skyrocketing food prices because farmers are spending all their profits on fertilizer and fuel, because yields are still down because they’re using less fertilizer, and because the truckers are passing on their fuel costs to the growers and wholesalers.

                And, ohdidImention? Croplands are already stressed due to pollution / climate change. California, the breadbasket of the nation, is in a serious crisis due to a prolonged drought with a distinct possibility that it could last for centuries, not just the decades it’s already lasted. Commercial coffee may soon go extinct because there won’t be any more overlap between favorable climates and favorable soils.

                …and how, exactly, is industry supposed to innovate a way to new energy prosperity when runaway inflation means they can’t afford do keep their own lights on?

                Our entire economic system is predicated on the presumption of perpetual ~2.5% annual growth. Take that away — which is exactly what rapidly-declinging oil reserves are about to do — and the whole house of cards comes crumbling down.



              3. I would add that the latest UN’s projection, and the IMF models, were complete surprises for me. Like Pinker’s next to latest book, about dwindling violence. (Man, he writes fast!)

                Earlier I though we were in trouble, srsly. Now, not so much.

        2. I’m not sure “afford” is the correct measure. You want “return on investment” to have a useful economy. And no matter the energy resources the distances make trade expensive vs local markets. (And analogously.)

          Besides, “exhausting the energy available from your own star” is a futurist meaningless extrapolation.* We don’t know how much energy a civilization can capture in an affordable way. And as I argued above, expansion would take people away from the star to live off of fusion of volatiles (hydrogen mostly).

          *Capture irradiation close to the star is perhaps doable, at least it seems economical on already present bodies. (Earth, Mars rovers, et cetera – out to Jupiter at the present.)

          But then they go off into insanity and speak of having access to a galaxy radiation output worth of energy. Well, perhaps if you colonize that much, but never in one place. The energy density maxes out for every star and its energy-matter content, you can’t trade it elsewhere (by the previous analysis).

          I don’t know if I’m annoyed or amused by the futurists. “It’s hard to predict, especially about the future.” Indeed.

          1. “We don’t know how much energy a civilization can capture in an affordable way.”

            By my own analysis, I meant “with ROI”. 🙂

          2. I’m taking this from rough back-of-the-envelope figures.

            A ten-year mission to send a Shuttle-sized spacecraft on a one-way trip to the Centauri system would take about as much energy as the entire civilization currently consumes in an entire year. In terms of resource availability, there’s a bigger difference between us (meaning the modern global civilization) and that sort of expedition than there is between Alexander’s Greece and us.

            It’s only when you start imaginatively constructing massive solar power plants inside Mercury’s orbit and using the power to create antimatter as a storage fuel that you can wind up with the types of proportional reserves historically available to simple exploration — and that’s still far short of colonization.

            It should be obvious even to futurists at that point that even this level of extrapolation is at least as absurd as Alexander asking his generals to draw up contingency plans for a moon invasion.


    2. Reminds me of the classical Berserker sci fi series by Saberhagan. This answered the question ‘where is everybody’? with huge, ancient interstellar death machines that kill all life on planets once they start to advertise their presence in radio.

  4. Two C. ginnagas walk into a bar. First one says, “Gee, Bob, you look like shit today.” Second one says, “Why, thank you!”


Leave a Reply