Silurian arthropod dragged its offspring around tethered to its body like kites

The Irish paleontologist and Yale professor Derek Briggs—no relationship to the other famous Irish paleontologist Sir Arthur “Artie” O’Dactyl—is famous for his work on the Burgess Shale fauna. He’s actually speaking today on that fauna at Chicago’s Field Museum, but I’ll be unable to attend. But we can all still marvel at some new work on younger specimens just published by Briggs and his colleages, reported in the early online edition of the Proceedings of the National Academy of Sciences (reference below, not sure if there’s free download for non-subscribers). There’s also a description of the work for nonscientists at the BBC’s site.

Briggs et al. describe a Silurian fossil (about 430 million years old) from a formation in the UK, a fossil that appears to have a unique method of brood care. It was a tiny fossil, only 1 cm long, and finding out what it really was took careful preparation: grinding it away  bit by bit (and of course destroying the specimen), and imaging it at each stage to produce a three-dimensional reconstruction. The animal proved (see below) to be an early arthropod.

What Briggs et al. found in the reconstruction was remarkable. Tethered to the “tergites” of the specimen (the post-cranial segments of the beast) were ten capsules, each attached by a long filament. And each of those roughly triangular, kite-shaped capsules (ranging 0.5 to 2.0 mm in size) consists of an outer shell containing a mass of tissue, some with limbs visible. The capsules are tethered to the parent specimen with long filamentous threads. Here’s what it looks like in reconstruction:


Screen Shot 2016-04-06 at 8.01.46 AM

What were these weird attachments? The most likely explanation is that they’re offspring of the specimen, being carried around—perhaps for protection of the developing embryos.

Although weird, this is not completely unknown in animals. As the authors point out, the developing embryos of the freshwater crayfish Astacida are atttached to the mother by smaller stalks, and I’ve managed to find a photo of that in a paper from 2004:

Screen Shot 2016-04-06 at 8.08.25 AM

Crayfish embryos attached to mother. From Figure 1 of Vogt and Tolley (2004).

However, these aren’t the long filaments (or tough embryo-containing capsules) described in the Briggs et al. paper. In that respect, what they found in this specimen, named Aquilonifer spinosus, is unique among animals. By the way, the source of the name is described by the authors:

The name of the new taxon refers to the fancied resemblance between the tethered individuals and kites, and echoes the title of the 2003 novel The Kite Runner by Khaled Hosseini (aquila, eagle or kite; –fer, suffix meaning carry; thus aquilonifer, kite bearer; spinosus, spiny, referring to the long lateral spines on the tergites).

I can’t think of any other animal named after a novel, but I’m sure there must be at least one.

The authors suggest, and reject, two other possibilities for these tethered kite-like structures: they could be parasites, or they could be epizoans (nonparasitic organisms that colonize others). They rule out parasites because there doesn’t appear to be any advantage for a parasite to absorb nutrients from a host through such long threads, and because the places where the threads attach to the “host”—on its spines—aren’t a great place to suck nutrients from.

They also argue that epizoans are unlikely because none are known that attach in this way, because ten epizoans probably would have killed the specimen (which was apparently alive and healthy when preserved), and because A. spinosus could have cleaned off such epizoans with its long head appendages. I agree with the authors that these capsules, particularly because some contain tissue with legs, are likely to represent a heretofore unknown form of brood care.

Finally, where does this new species fit? As I noted above, it’s an arthropod, at least based on the cladistic analysis conducted by the authors. The cladogram based on many morphological characters puts it with the arthropods (node 1), but in particular with the Mandibulata (node 4), the subgroup that includes centipedes, millipedes, crustaceans, and “hexapods” (insects and three other and much smaller groups):

Screen Shot 2016-04-06 at 7.54.40 AM

Fig. 2. Cladogram showing the phylogenetic position of A. spinosus gen. et sp. nov. Shown is a strict consensus of the 12 most parsimonious trees of 142.16612 steps (consistency index = 0.513; retention index = 0.870), produced using New Technology search options in TNT (tree analysis using new technology) and using implied character weighting with a concavity constant of three. Numbers above nodes are GC support values. 1, Euarthropoda (crown- group); 2, total-group Chelicerata; 3, Artiopoda; 4, total-group Mandibulata; 5, Mandibulata (crown-group).

The upshot: the paper doesn’t really produce new generalizations about life, but rather the discovery of a particular way of life that was completely surprising. There’s nothing wrong with such an anecdotal observations, for that’s the kind of thing—the multifarious and unexpected variety of life—that keeps our wonder alive.

h/t: Barrie

Addendum, by Greg Mayer: Jerry did not get a chance to go to hear Derek Briggs at the Field Museum yesterday but I did, and Jerry asked for a report.

Briggs talked mostly about his work on the “kite runner” (which, he noted, he named after Khaled Hosseini’s 2003 novel), so I won’t restate what Jerry covers admirably above. Briggs mentioned that the reviewers were less certain than he was that the ‘kites’ were juveniles, rather than parasites or something else, and that he did see the reviewers’ point, but still thinks they are juveniles. He showed a number of neat 3-dimensional rotating videos of their fossil reconstructions. For a museum audience, it was a bit wince-inducing, but understandable, to know that the method of preparation destroyed the specimen. Briggs is a also a museum guy, and is working to develop non-destructive forms of imaging, and was consulting on this trip with physicists at Argonne National Laboratory. Such imaging would also be enormously time saving, as the specimens come in nodules, and they don’t know what fossil is in a nodule till it’s ground through a considerable ways. He also quipped that PNAS (where his paper was published) stands for “Probably Not Acceptable in Science“, which is a “nerdy science joke“. (BTW, I think Jerry’s Artie O’Dactyl also eminently qualifies as a “nerdy science joke“!)

He made two other interesting points. First, the apparent extinction of many of the unusual soft-bodied forms at the end of the Cambrian seems to be a preservational artifact. There is a period from the late Cambrian into the Ordovician from which no lagerstatten are known. (Lagerstatten are deposits with unusual preservation in which soft parts are fossilized, such as the Burgess Shale of British Columbia.) Cambrian “weirdos” are now turning up in these later lagerstatten. For example, anomalocarids (well known in the Burgess Shale), are also now known from the Fezouata, Morocco, lagerstatte, which is Ordovician. There are a lot of taxa represented in the Ordovician, which is the peak of diversity origination, referred to as “GOBE” (the Great Ordovician Biodiversification Event).

Second, he talked a fair amount about limb evolution in arthropods, and noted that an early horseshoe crab, Dibasterium, had an extra row of legs compared to modern Limulus. In Limulus, it turns out that important “leg genes” are also activated in the embryo in a row of small spots parallel and lateral to the actual legs– in just the places where Dibasterium‘s extra legs are! (The developmental work was done by someone else.) This reminded me of the phenomenon in vertebrates with reduced numbers of toes in which toe primordia develop a bit, and then regress.


Briggs, D. E. G., D. J. Siveter, D. J. Siveter (one is Derek, the other David!), M. D. Sutton, and D. Legg. 2016.  Tiny individuals attached to a new Silurian arthropod suggest a unique mode of brood care. Proc. Nat. Acad. Sci. USA: Published online before print, April 4, 2016, doi:10.1073/pnas.1600489113


  1. rickflick
    Posted April 6, 2016 at 12:31 pm | Permalink

    Surprising. It is even further evidence for Steven J. Gould’s suggestion that rewinding the tape of life would show the results are unpredictable.

    I’ve seen mothers “walking” their young children on long leashes in the park. Surely an example of parallel evolution.

    • Posted April 6, 2016 at 12:52 pm | Permalink

      Well, if you’re a strong determinist like me, this might reappear if the tape of life were replayed so long as mutations aren’t quantum-mechanical phenomena! See my discussion in FvF on whether the appearance of humans was inevitable.

      • rickflick
        Posted April 6, 2016 at 1:25 pm | Permalink

        As a strong determinist myself, I agree. But whenever I think of Gould’s hypothetical I always assume, for the sake of argument, that he really means start all over with ‘approximately’ the same conditions. Then I think he makes more sense.

        • Posted April 6, 2016 at 1:36 pm | Permalink

          I’m not so sure–that’s not what I got from Gould. His notion of “contingency” is really one that was unclear and needed further discussion.

    • Mark Sturtevant
      Posted April 6, 2016 at 1:03 pm | Permalink

      I have not understood the strong opposition to this Gouldian idea, but I have not also understood why one would claim that it would turn out completely different.
      A replayed life history would produce different groups that occupy this or that ecological niche. But the species that would emerge would still show some resemblance to those seen in our life history by virtue of convergent evolution.

      • eric
        Posted April 6, 2016 at 1:27 pm | Permalink

        That’s not necessarily a good bet. Kangaroos, the giant tortoise, Abrictosaurus, and ostrich all occupy or occupied the ‘100 kg land herbivore’ niche along with the more ubiquitous deer. Do they resemble each other?

        Moreover the further back you go, the wider the possible result. Kangaroos and deer both occupy that niche now (and abrictosaurus doesn’t) because mammals became widespread. But go back in time to where there was only one or a few mammal progenitors competing with species in other clades for niches, and if the mammals lose instead of win, the critters in the “100 kg land herbivore” niche could be very, very different.

        Always assuming that QM indeterminacy would have an impact on macro-scale events. I’m not saying it necessarily did, just that if it did, I would not expect a rewind to produce species that resembled current species.

        • Mark Sturtevant
          Posted April 6, 2016 at 2:23 pm | Permalink

          Giant tortoise versus glytodonts.
          Ostriches versus ornithomimids.
          Flying avians versus pterosaurs.

          If you rewind waay back, then maybe the replacements would not be vertebrates, I admit.

      • rickflick
        Posted April 6, 2016 at 1:29 pm | Permalink

        Ignoring the subatomic level for a moment, I think rewinding would produce many gross similarities, such as bilateral symmetry, and some big surprises like kites (or birds with propellers?).

    • eric
      Posted April 6, 2016 at 1:17 pm | Permalink

      Yes, I was going to point out that 430 million years later, humans developed the same adaptation.

      Re: JAC’s comments about whether mutations are quantum mechanically influenced or not; I would bet heavily that some are. The rate of radioactive decay in the human body is about 5 kBq, or about 5000 decays per second. Mostly beta decay from K-40 and the gammas associated with that. Some of those are going to hit DNA bases and break bonds etc. for sure.

      However I doubt it’s a significant factor in evolution. The vast majority of such decay-induced mutations would occur in adult cells or nonreproductive cells, simply because that’s what the vast majority of our bodies are. Once you start talking about decays/individual sperm/month or decays/egg/cycle or decays/embryo/pregnancy, the number of mutations is probably some small fraction per developing individual. IOW, for most of us the answer is “zero.”

      • lkr
        Posted April 6, 2016 at 1:55 pm | Permalink

        Eric: My impression is that germ line mutations per generation [eg, new in my zygote, not present in zygote of either parent] are well above zero. I recall claims that the number of “new” mutations were in the range of 15-30; the current Wiki article cites estimates around 60 per generation in the germline, based on typical mutation rates across eucaryota. If your point is that not all are quantum-radiation induced, but in some way deterministic, that may be another matter.

        • eric
          Posted April 6, 2016 at 8:00 pm | Permalink

          I think the latter is the case but I don’t have a citation for you. I expect most mutations are due to chemical and biological causes, not radiation (i.e., that subset of chemistry consisting of high energy photon photochemistry). And in the radiation category, there are probably more hits from cosmic rays than there are from internal nuclear decays. So my qualitative SWAG is that they’re only going to account for a small fraction of germ line mutations per generation.

  2. Mark Sturtevant
    Posted April 6, 2016 at 12:47 pm | Permalink

    Marvelous. I am not sure I am convinced of legs in the embryos, but that these are attached offspring seems a reasonable explanation.
    In the decapod crustacea (and perhaps other crustacea, I don’t know), females produce fertilized eggs from distinct openings on specific basal leg segments. It is perhaps a stretch, but it would be worthwhile and simple to look for similar openings at the basal leg segments in this specimen. A positive finding would strengthen the case. A negative result would do no harm since other arthropods produce eggs from different locations.

  3. Paul S
    Posted April 6, 2016 at 2:29 pm | Permalink

    Cool stuff. Too bad I won’t be able to get to the Field Museum for a while. I wonder if the Museum will add a display.

  4. gravelinspector-Aidan
    Posted April 7, 2016 at 1:59 pm | Permalink

    Damn. Missed this post while I was away. And I just spent an hour writing up this very topic and mailing it to PCC(E).

  5. Igor
    Posted April 7, 2016 at 7:38 pm | Permalink

    So cool! I sure hope we get more posts like this one! 🙂

  6. TJR
    Posted April 8, 2016 at 3:58 am | Permalink

    Going back to comment on it this time, its *very* interesting to hear that the apparent extinctions may well be an artifact of preservation.

    I’ve often wondered if that was the case, and pretty much every course I teach on I warn the students to think about sampling bias.

    In particular, anomalocarids in the Ordovician, I had no idea.

    Love all this very-early-life stuff, don’t miss the Attenborough programmes First Life if you’ve not already seen them.

  7. Posted April 8, 2016 at 8:15 am | Permalink

    Wow! That really is amazing!

  8. Torbjörn Larsson
    Posted April 10, 2016 at 2:55 pm | Permalink

    It is a wonderful world. Thanks for writing this up!

    [Catching up on the web after a prolonged hospital stay. Came in dying, left after intensive treatment ‘likely not, let’s see how it develops’ – so the world are at times less wonderful too.]

    • Posted April 10, 2016 at 8:15 pm | Permalink

      Please don’t die! You would be missed, here! Wishing you a full and fast recovery.

    • rickflick
      Posted April 10, 2016 at 8:28 pm | Permalink

      Yes, hang in there Torbjörn. You’d be a terrible loss.

%d bloggers like this: