Plant seeds evolve to mimic antelope droppings, and dupe dung beetles roll and bury the seeds

December 3, 2020 • 9:30 am

The first paper below is five years old, but I just read it yesterday because it’s a remarkable example of mimicry. In this case, seeds of a plant in South Africa have apparently evolved a size, shape, appearance AND smell that makes them resemble antelope droppings. Dung beetles, thinking that the seeds are fecal matter, roll them to a safe place and bury them, ensuring that the seeds are protected, dispersed a bit, and get planted. The paper, from Nature Plants, is below (click on screenshot), the pdf is here, and the reference is at the bottom. (If the article is paywalled, a judicious inquiry will yield it.)

This is one of the very few examples in which plant seeds have evolved to deceive animals, either physically or chemically, and in which the plant benefits but the animal loses. This is, in fact, the evolution of a plant that parasitizes an animal.

Here’s a later paper (2016) from the South African Journal of Science with a free pdf (click on screenshot):

The plant that’s evolved mimicry is Ceratocaryum argenteum, a shrubby plant that’s endemic to the Cape Province of South Africa:

Unlike seeds from other plants in the family Restionaceae—which are normally pretty flat, with a smooth, dark seed coat as well as elaiosomes (fleshy bits that are edible to ants, who carry the seeds to their nest, feed the elaiosomes to their larvae, and then discard the rest of the seed, which thereby gets dispersed)—C. argenteum has a “rough tuberculate and brown outer seed coat” which, to the authors’ noses, “has a pungent scent similar to herbivore faeces”.

Below: what the seed looks like (a-c) in contrast to other seeds in the area (h-j). (g) shows the dung of an antelope (a Bontebok). Note that the C. argenteum seeds are about the size and shape of the Bontebok dropping, and are round to facilitate rolling. Dung beetles roll balls of dung to a nearby location, bury them, and lay an egg with the dung so its larvae can feast on the feces. The beetle observed burying seeds was Epirinus flagellatus.

ac, Vertical (a) and side (b) views of a C. argenteum seed as well as one that has been cracked open (c) showing the endosperm and thick woody inner seed-coat layer and the outer tuberculate layer which together form the husk. d,e, Scanning electron microscopy (SEM) of the outer, tuberculate layer and inner seed-coat, with white silicon granules at the boundary between the two layers. fE. flagellatusg, Bontebok faeces. h,i, Vertical (h) and side (i) views of an L. sessile seed. jCannomois grandis seed with white elaiosome.

And the dung-maker, the small antelope most common in the area (80-100 cm or 31-39 inches at shoulder): the bontebok, Damaliscus pygargus pygargus. 

The authors hypothesized that the size and smell of the C. argenteum seeds would facilitate them being buried, and so they put out seeds along with some camera traps.  They observed four-striped grass mice (Rhabdomys pumilio) eating husked seeds but never burying them.  In contrast, of 195 seeds put out after a rain (when dung beetles are active), at least 55 were buried (they used fluorescent threads to mark the seed paths).

In no case did the buried seeds have a dung beetle egg on them, so the beetles were first fooled, and then realized that something was wrong—but only after they had rolled away the seeds and buried them.

As I noted above, resembling dung to fool beetles is a good way to perpetuate your genes, as you get dispersed, protected by the soil from mice, and buried (planted). Further, C. argenteum plants can’t re-sprout after a fire, and thus the persistence of plant genes depends on a way to escape fire—by getting its seeds buried! For many reasons, then, selection might favor the seeds resembling dung, and because beetles detect dung by its odor, you’d want to smell like dung, too. The dung beetles are simply dupes, doing a lot of work and not getting anything out of it.

The authors also did gas chromatography and mass spectrometry to measure the amount of volatile compounds on seeds and dung, and found that the seeds had a significantly larger amount of volatiles than other seeds in the area, even when old and when corrected for surface area, and resembled the amount of volatiles in dung. Further, compounds in the seed volatiles were also identical to compounds in the volatiles of bontebok and eland dung (another antelope), with “various acids, the benzenoid compounds acetophenone, phenol, p-cresol and 4-ethyl-phenol, as well as the sulphur compound dimethyl sulphone.”

Here’s a two-dimensional plot showing the resemblance of the C. argenteum seed volatiles to dung volatiles; note that other seeds (green triangles) don’t have dung-like profile of volatiles:

(from the paper): Similarity in the composition of volatile blends of seeds and animal droppings is based on non-metric multidimensional scaling. Symbols for other Restionaceae (Methods, Supplementary Table 3) that overlap are slightly offset for clarity. The composition of scent sampled from Ceratocaryum seeds is very similar (R = 0.75, P = 0.33) to that of dung of local herbivores (eland and bontebok), but differs markedly (R = 1.0, P = 0.028) from that of seeds of other Restionaceae (nested ANOSIM permutation test).

In the second paper, the authors observed another dung beetle, Scarabaeus spretus, burying the seeds, flying rather than crawling to the piles of seeds put out. (It’s clear that odor rather than appearance is a major attractant, and one S. spretus flew directly into a paper bag of seeds!) This species moved seeds only about a quarter of a meter, while E. flagellatus could move them up to 2 meters away from the pile. (As you see, the dispersal is quite limited!) Here’s a figure showing beetles of both species rolling away the seeds and burying them:

(a) Epirinus flagellatus rolling a Ceratocaryum argenteum seed; (b) Scarabaeus spretus rolling a seed (the arrow indicates a Sphaerocerid Lesser Dung Fly); (c) the large hole made by Scarabaeus spretus for burying several seeds (the arrow indicates the location of the Dung Beetle); and (d) a female Sarcophagid Fly on a seed. Midgley & White (2016).

Further, the bontebok eats different kids of grasses from the eland (Taurotragus oryx), a larger species shown below, and the different species of grass have different ratios of nitrogen and carbon isotopes. By looking at the isotope ratios in the beetles (whose juvenile stages eat the dung), and in the antelope dung itself, the authors found that the ratios of the dung beetles (green diamonds and purple triangles) resemble the dung of the eland (light blue triangles) more closely than the dung of bontebok (red circles), as shown in the diagram below.

Conclusion: the dung used by both species of beetles is likely to be from eland rather than from bontebok. But as the authors showed above, the volatiles of both antelope dung are pretty similar, and still resemble the volatiles of the seeds.

The one puzzle is that the size of C. argenteum seeds are more similar to that of bontebok droppings than to eland droppings. Being much bigger, elands have larger scat—about twice as big. But since dung beetles can form smaller balls out of larger droppings, and because it may be too onerous for the plant to produce a seed twice as large as it does, this may not be a problem.

An eland:

So we have mimicry here that deceives the beetle, who comes to its senses only after it rolls away and buries a seed. In this case it doesn’t adhere to the Who’s dictum, “Won’t get fooled again.” It would be interesting, though, to do lab experiments with dung and seeds to determine if beetles eventually learn to avoid rolling and burying these mimetic seeds. It’s a lot of effort for nothing, and the beetle “knows” it since it doesn’t lay an egg on the seed.

h/t: Jean


J. J. Midgley, J. D. M. White, S. D. Johnson and G. N. Bronner. 2015. Faecal mimicry by seeds ensures dispersal by dung beetlesNature Plants 1, 15141,


34 thoughts on “Plant seeds evolve to mimic antelope droppings, and dupe dung beetles roll and bury the seeds

  1. The beetles must have two separate programs they run.

    1) “If it is round, about x size, and smells like this, roll and bury it somewhere”
    2) “if you have a poop-like-thing buried in the soil that feels soft (or something?) then lay an egg”

    I’m guessing they don’t have the “brains” to suddenly realize that they have wasted a bunch of energy burying something so why bother leaving an egg.

    1. This seems very likely. Much like the burrow-locating programs studied in wasps. They will just keep running the program over and over if you mess with the stimuli.

    2. That’s pretty much what I was going to say. The seed provides the perceptual cues for the rolling, digging, and burying behaviors, but not for the egg-laying behavior, and the beetle walks away none the wiser.

      1. If it cost the beetles too much effort to the point of being detrimental to them, the grass would presumably suffer. It would be interesting to try excluding beetles from an area to see how the grass coped.

        Another thing -are there any seed-eating birds that would wipe out the grass? Can seeds survive passing through the guts of a bird (I suppose not)? How frequent are fires? Can seeds successfully germinate only if buried?

  2. So …for the plant to reproduce it fools the beetle into thinking it is doing something of critical importance for itself, all the while the plant is robbing the beetle of a significant part of its own chances of survival.

    Change “plant” with “the rich” and “beetle” with “most everyone else” and it is something of a parable for our times.

    1. Everything can be seen more clearly through Darwinian spectacles. But the beetles can clearly afford this extra effort. Also, it is in the interests of antelope to have grass regenerating, & without grasses there would be no antelope & thus the beetles would suffer.

      All of evolution, & society it seems to me, is about avoiding cheats & freebooters. Any group can tolerate a small number, but if it gets too big it cannot be sustained.

      1. I’d dispute the “All of evolution…” statement. Much evolved behavior includes cheating because it is advantageous to cheaters, up to a point. It’s all about cheating as much as you can and preventing others from cheating as much as you can. One man’s cheater is another man’s success story, so to speak.

  3. It’s fun to laugh at the dumb beetles, especially when there is some poop involved.
    It’s also easy for us to laugh at something, resolve it, & then move on to something else that engages our need for discovery.
    Entomologists are experts at this — we are perhaps even addicted to it!
    [Several of my patronyms are dung beetles.]
    But there are deep layers of adaptation here between the Ceratocaryum shrubs that produce the seeds.
    Clearly, the plants are benefitting from their co-evolution with the beetles.
    Should we delve a bit deeper to see what might be in it for the beetles?
    Like, maybe …
    1] it’s adaptive for the beetles to help this plant in the environment?
    2] this shrub feeds the antelopes that drop the poop that the beetles need?
    3] the buried seeds are good decoys to deflect parasites away from beetle larvae?
    4] it’s good practice for newly-eclosed adult beetles?
    5] the beetles are having fun?
    Beware” the marshmallow test”!
    To wit — Delaying gratification is productive of discovery!

    1. While all of those possible, none are required to explain how this behavior came to be. In fact, the last two are (I assume) just offered in jest. I don’t know of any evidence of insects engaging in play.

      1. Thanks for your perceptive attention.
        Yes — some bits of the comment were offered jocularly.
        I have engaged beetles in play many times — I am not sure of their perceptions.
        But sometimes, they won.

    2. The key test is not whether planting seeds would somehow be good for beetles in general; it’s whether it’s good for the particular beetle doing the planting. It’s hard to see how some of your proposals meet that criterion.

      1. No, I would say it cannot be doing it harm or the behaviour would not exist, for beetles that suffered would not reproduce as much. What I would like to know is – how many balls of dung does a beetle bury in a year, so is it laying eggs every day? Every week? Every month? At a particular time of year?

        It cannot harm a beetle or it would not happen. It must happen to all the beetles living near these grasses, so one would not carry any more cost than its neighbour.

        Do seeds form all year or only in one season? If so, how does that match any seasonality in beetle reproduction if any? Presumably these things are known, maybe covered in the article (I’d struggle to read on this iPhone).

        1. Parasitism that harms the host happens all the time in nature. What we see here is not different in principle from the behavior of cuckoos who lay their eggs in the nests of other birds, hijacking the host’s reproductive instincts (and thereby causing them harm) to further the reproductive success of the cuckoo.

  4. “Devil’s dung” or Asafoetida is a root used in Indian cooking. It has a terrible odor. I’m intrigued if the chemical odorant got trapped in an evolutionary well, or something…

    1. I love the smell of Asafoetida (and the flavor). Somewhat like onions and it is used in place of onions by certain Indian religious sects where onions are taboo. To each their own. 🙂

  5. Nothing to add to the comments; just a word of thanks to PCC(E) for yet another fascinating science post.

  6. If I may just brag a bit as a Cape Town resident. The Cape Floral Kingdom is one of six in the world. It is the smallest and richest per unit area. Pretty cool then to learn about the mimicry of the plant Ceratocaryum argenteum, We breed them shrewd and tough in the Cape 🙂

  7. If plants can do this, perhaps we should stop eating plants! They may be sentient. 😱

    Jokes aside, this was a fascinating and illuminating post. The feats of evolution never cease to amaze.

Leave a Reply