A novel but clever behavior: ants form daisy chain to haul prey back to nest (and fly lagniappe)

I believe it was a reader who sent me a link to the first video below, but I can’t remember that reader’s name, so my apologies. But when I saw the video of ants pulling prey back to the nest by forming a long daisy chain, linking their bodies by biting each other to increase pulling power, I couldn’t believe it. I had neither heard of nor seen any insect behavior like this. Watch these ants hauling a millipede back to the nest:

I wrote to Dr. Phil Ward, one of our three Official Website Entomologists™, asking him if these were army ants. He responded this way, “Pretty cool, eh? These are ants in the genus Leptogenys, forming daisy-chains to more effectively pull their prey back to the nest. Alex Wild posted about this last year.” But they are not army ants; as Phil said, “No, they are group-raiding ponerine ants, with some army-ant like behavior: army ant wannabes.”

He then gave me a link to a post of Alex Wild, another Official Website Entomologist™ (Piotr Naskrecki is the third), who had posted about this behavior on his website Myrmecos:

The quality isn’t great, but the clip appears to show an Asian Leptogenys daisy-chaining their bodies in parallel lines to haul away a large millipede. I have spent the morning searching the technical literature for mention of this unusual behavior, and am coming up empty. Some Leptogenys species, including L. diminuta, L. nitida, and L. processionalis, are known to forage in groups and transport prey “cooperatively” (source, source). What is meant by “cooperative” is often vague. (For more, see this excellent recent review of cooperative transport by Helen McCreery).  Yet I didn’t find any explicit description of workers linking up, mandible to abdomen, to pull together.

Yes, they’re forming these chains by biting each other’s butts! You can see that more clearly in the video below. Alex continues:

Is ponerine daisy-chaining an unknown behavior? Possibly. It is also possible my search skills aren’t up to the task. If you know of a description of it, please drop a note in the comments.

Alex found another video and posted it later:

Alex also posted a photo of similar behavior, also in the same genus, taken by Steve Shattuck in Borneo. In this case the ants appear to make the chain by biting each other’s legs:

Leptogenys1

In his post, Wild quotes ant expert Christian Peeters who saw the behavior in Borneo:

The behaviour was very stereotyped: mandibles grab preceding ant’s gaster (between first and second segment).

Seiki Yamane identified it as Leptogenys sp. 47, closely related to L. chalybaea described from Borneo by Emery (but stronger sculpture especially on gastral tergites).

The millipedes were 130mm long, identified as order Spirostreptida (Diplopoda). Ant is 16mm long.

Back then I reviewed the literature and found no other record of chain behaviour in Ponerinae. No record of millipede predation in Leptogenys.  Specialized hunting on millipedes is restricted to Thaumatomyrmex,Probolomyrmex and Gnamptogenys, but these are solitary hunters on a very different kind of millipedes (polyxenids).

Finally, Alex notes that because the first video appeared on LiveLeak, without attribution or information, we know nothing about who filmed it nor about where or when it was filmed, making it hard to do further study. The behavior is fascinating but so far has not been published in the scientific literature. I’m also curious about whether the leg-biting vs. abdomen-biting method of forming the chains are behaviors specific to particular species (I expect this is the case).

When you think about how this evolved, remember that the ants doing the dragging are sterile workers (they’re eusocial, as we’ve been discussing), so they themselves gain no reproductive advantage by behaving this way. Any evolutionary scenario based on natural seletion has to show an advantage to these ants that can contribute genes for the behavior to the next generation. I’ll leave it to readers to solve this apparent puzzle, which holds for all evolved behaviors of sterile workers.

By the way, head over to Piotr’s wonderful site, The Smaller Majority, and see his latest post about a red-headed fly from Mozambique. There is a lot of swell biology in the post, but I’ll just show the flies, which will make Matthew Cobb wet his pants. (Captions fare from Piotr’s post.)

bromophila

Red-headed flies (Bromophila caffra) are striking and common animals in East and southern Africa, but little is known about their biology.

bromophila5

Red-headed flies, which in Mozambique emerge at the end of the rainy season, like to hang in clusters on leaves.

bromophila3

One peculiar morphological characteristic of the Red-headed flies is the absence of the ocelli, which are typically found on the head of other flies.

25 Comments

  1. Posted May 9, 2015 at 9:49 am | Permalink

    Any evolutionary scenario based on natural seletion has to show an advantage to these ants that can contribute genes for the behavior to the next generation. I’ll leave it to readers to solve this apparent puzzle, which holds for all evolved behaviors of sterile workers.

    Doesn’t the colony effectively operate as a single breeding unit? That is, if the queen produces effective workers, she’s more likely to create daughter colonies that will, in turn, be successful.

    b&

    • Posted May 9, 2015 at 10:54 am | Permalink

      Yes, but where does the mutation occur? If it occurs in the eggs or sperm that produced the queen (or in a single male who inseminated her), then half of her workers will have the new cooperative behavior and the colon will show the behavior. But what if the first mutation occurred in, say, a sperm that produced a worker, so there was only one cooperative worker in the colony. What good would that do? He’d want to hook legs up with others, but nobody would get on board, so there’s no cooperative dragging of prey to the nest.

      • Posted May 9, 2015 at 11:25 am | Permalink

        I might be missing something…but isn’t that, from the gene’s perspective, not unlike the problem caused by recessiveness? The odds are similarly stacked against the spread of a novel recessive gene, but that just means that…well, the odds are stacked against it, not that it can’t happen.

        It also seems like it’d make more sense to note a distinction between germlines that can get passed to new colonies and ones that terminate in sterile workers. The latter would, evolutionarily, I should think, be indistinguishable from non-germline stem cells.

        The genes in your muscle cells are an evolutionary dead end unless they provide a benefit to the identically-encoded genes in your germ cells. Similarly, only the genes in the queen and her mates that produce new queens really count, with the identically-encoded genes that get expressed in the sterile workers playing only a supporting role.

        …or am I missing something…?

        b&

        • Posted May 9, 2015 at 11:45 am | Permalink

          This gene seems to have no adaptive advantage unless more than one ant has it. That’s the problem I’m posing.

          • Posted May 9, 2015 at 12:03 pm | Permalink

            I’m still not seeing why that should be a problem.

            Once a queen has the gene, either all or some Mendelian fraction of her drones expresses it, no?

            So the mutation either has to happen in the queen’s germline or one of her ancestor’s germlines…but how’s that different from other organisms?

            You could have a muscle cell develop a mutation that makes it stronger (or whatever), but that mutation is irrelevant because one muscle cell out of the billions in your body isn’t going to make a difference. But if that same mutation were to happen in one of your germ cells and said cell become half of the genetic material of a child, now the mutation has a chance to develop — not in you, of course, but in your child.

            Same with the ants. A sterile drone that developed the mutation would be useless, but a queen that did would potentially outcompete the other queens.

            b&

          • Gregory Kusnick
            Posted May 9, 2015 at 1:23 pm | Permalink

            I can imagine a scenario in which a mutation in a single worker ant can produce this sort of behavior.

            Suppose there’s already an existing behavior in which worker ants form living bridges across gaps in the trail by linking up fore-to-aft. We know that some species of ants do this (although I have no idea if this species is one of them).

            Bees have similar linkup behaviors in forming swarm balls to protect the queen. So there’s reason to think this is not uncommon among hymenoptera.

            Perhaps all that’s needed then is for a single ant to trigger the existing linkup behavior in a novel situation (e.g. load-dragging). If a bunch of other ants then fall in and join the linkup as they’re already primed to do, the new behavior might become adaptively established, even if only a tiny minority of workers are prone to initiate it.

            • winewithcats
              Posted May 9, 2015 at 7:51 pm | Permalink

              Does there even need to be an existing linking behavior to work from? It seems to me that the only requirement is that the other ant in the pair doesn’t resist against the link, but instead continues going about its business as though it were working solo. Even a single worker with linking behavior could then provide a special benefit to the overall success of the colony, when compared against an otherwise identical colony lacking any linker ants.

        • Andrikzen
          Posted May 9, 2015 at 11:50 am | Permalink

          The male ant is haploid so 100% of his genes are passed on to his daughters (workers), assuming monogamous mating with queen. The daughters are more closely related, so perhaps, the trait has a better chance to express itself in the colony and fix in subsequent populations.

          I wonder if the behavior is a simple “bite and pull” algorithm in response to some type of chemical “I’m pulling too much weight” signal?

          I notice not all of the ants participate in the chain, evidence perhaps of only a subset of the colony having the trait.

          • Diane G.
            Posted May 9, 2015 at 9:46 pm | Permalink

            There’re always a lot of ants cheering on the pullers, though! 😀

            Yeah, I was wondering about the milling-arounders. Some seemed to be helping by attaching themselves to the body of the millipede. Could the others possibly relieve the pullers now and then, if necessary? Or maybe they’re there to guard the prey and/or protect the pullers, who are otherwise occupied.

  2. John
    Posted May 9, 2015 at 9:49 am | Permalink

    Absolutely amazing stuff. It sounds as if we’re witnessing fairly new science in real time. Thanks PCC!

  3. notagod
    Posted May 9, 2015 at 11:33 am | Permalink

    That’s how humans learned to move all the huge rocks around that christians are always puzzling, their gods or UFOs had to do it? Maybe though, when the humans wanted a huge rock moved they would place an ant colony where they wanted the rock to end up, then, glue millipedes to the rock at its origin site. Sit back and have a beer.

    Maybe less important to christians but possibly more interesting, the flies in the first picture appear to have spoilers(?) at the wing tips, is that due to handling by humans or is it characteristic of the fly?

    • Diane G.
      Posted May 9, 2015 at 9:48 pm | Permalink

      The first thing I though of when I saw the vids was, “the pyramid builders had nothing on these guys!”

  4. Sciencefictionfan
    Posted May 9, 2015 at 11:36 am | Permalink

    If men will becomeextinct some time in the future, ants will rule the world. Such an “intelligent” behavior (but not an intelligent design 😉). Or think about the leafcutter ant, the gardener of teh tropics.

  5. dodger
    Posted May 9, 2015 at 12:15 pm | Permalink

    How do the ants grab the prey, are they walking backwards?

    • Diane G.
      Posted May 9, 2015 at 9:51 pm | Permalink

      I think they are! If you watch the first vid, somewhere around 25 seconds or so, you can see an ant peel off of the lowest chain; then it appears that it’s abdomen is facing in the direction the millipede is heading. And the ant then quickly turns around to move forward normally.

  6. Gregory Kusnick
    Posted May 9, 2015 at 1:02 pm | Permalink

    “army ant wannabes”

    So militia ants then.

  7. Diana MacPherson
    Posted May 9, 2015 at 2:19 pm | Permalink

    Those red-headed flies look huge! And look at their feet; I know all flies probably have feet like this, but it is like they are wearing little tap shoes.

    • Mark Sturtevant
      Posted May 9, 2015 at 4:22 pm | Permalink

      I thought at first that these were fake, plastic insects and that this was some kind of April fools joke. Damn, they look weird.

      • Diana MacPherson
        Posted May 9, 2015 at 6:00 pm | Permalink

        Like weird, tap-dancing flies from the Cretaceous.

  8. Jim Knight
    Posted May 9, 2015 at 6:19 pm | Permalink

    The ant video is spectacular! That is an amazing behavior…

    How big are those red-headed flies? They appear to be fairly large?

  9. Diane G.
    Posted May 9, 2015 at 9:55 pm | Permalink

    In addition to the facial coloring, those pure black wings are very cool looking as well. Looks like the flies have been dipped in black paint.

    • ladyatheist
      Posted May 9, 2015 at 9:59 pm | Permalink

      It looks like they’re wearing motorcycle leather

      • Diane G.
        Posted May 10, 2015 at 1:54 am | Permalink

        Yes! 😀

  10. ladyatheist
    Posted May 9, 2015 at 9:58 pm | Permalink

    I shouldn’t have read this post right before bed. I’m going to have nightmares and feel creepy crawlies all over

  11. infiniteimprobabilit
    Posted May 10, 2015 at 3:37 am | Permalink

    Ack! My sympathies are entirely with the millipede.

    Urrgh, nature is ‘orrible. Specially ants.

    I do seem to remember some nonsense in the Bible about ants being somehow good, which vague impression somehow survived for many decades my complete deconversion.


%d bloggers like this: