Insects are crustaceans!

The phylogeny of arthropods has always been messy.  One reason is that studies trying to discern their evolutionary relationships often use too few taxa (this is, after all, the most species-rich of all animal groups), and, especially, too few genes.  Conclusions have been based, for example, on only 18S and 28S rRNA and mtDNA (the latter is, of course, effectively one gene).  And this has led to conflicting conclusions, some of which contravene morphologically-based systematics.

For example, morphology seems to define a group called the “mandibulata”: all those arthropods that have mandibles.  But some molecular work has lumped the myriapods (centipedes and millipedes), which have mandibles, together with the chelicerates, which don’t have mandibles but a nonhomologous biting structure called chelicerae. (Chelicerates include spiders, horseshoe crabs, pycnogonids, and the like; see Fig. 1 below).

Now, a new paper in Nature by Regier et al. has come up with a near-definitive family tree of arthropods that resolves many of the questions that arose from studies using lesser resolution.  In their work, Regier et al. used many arthropod species (75, to be exact), and 62 single-copy genes that were orthologous in species from flies to humans.  This is a huge genetic sample, allowing for a good, well-supported tree based on 41 kilobases of DNA sequence.

We needn’t go into the messy details, but there are three quite important findings.

1.  Insects (“Hexapoda”) are not a sister group of crustaceans, as was indicated by some molecular studies. Nor are they the sister group of myriapods, the traditional arrangement supported by morphology.  Instead, insects are nested within crustaceans (see Figure 1).  In the same sense that birds are dinosaurs, then, insects are crustaceans.

2.  The sister group of insects within crustaceans comprises two rather obscure taxa that were long thought to be primitive: the cephalocarids and the remipedes.  Regier et al., however, find these two groups (see below) belong to the monophyletic clade called Xenocarida.  Their “primitiveness” is thus deceptive.  Now the fact that some crustaceans, like the Xenocarida, are more closely related to insects than to other crustaceans means that the group “Crustacea” is paraphyletic, since, by not including insects, it doesn’t include all descendants of the common ancestor.  If we want to be punctilious taxonomists, we’d have to dump the name “Crustacea”, or else reclassify insects as crustaceans.

3.  The Mandibulata, mandible-carrying arthropods, is now confirmed as a real monophyletic group, since the myriapods now join all the other mandible-bearing beasts, as they should.

Maybe this seems arcane to those who aren’t interested in arthropods, but it really does seem to settle long-standing questions about where the insects came from.  And it sets the standard for the numbers of species and genes that should be included in a good phylogenetic analysis.

Figure 1 (from Regier et al.). The phylogeny of arthropods.

Fig. 2.  Speleonectes tulumensis, a marine crustacean, a remipede, and a member of Xenocarida, the sister group of insects

Fig. 3.  A cephalocarid, also a member of the Xenocarida

h/t: Phil Ward, Cliff Cunningham


Regier, J. C.  J. W. Shultz, A. Zwick, A. Hussey, B. Ball, R. Wetzer, J. W. Martin and C. W. Cunningham. 2010. Arthropod relationships revealed by phylogenomic analysis of nuclear protein-coding sequences. Nature 463:1079-1083.


  1. NewEnglandBob
    Posted March 4, 2010 at 1:29 pm | Permalink

    I have nothing to contribute. I just want to subscribe to comments.

  2. Posted March 4, 2010 at 1:47 pm | Permalink

    Ive been arguing for this position since I was four.

    Mom: “Why dont you want to eat crab/lobster/crawdads/insert ‘crustacean’ here”


    Picky eater for the phylogeny win! hehehe!

    • KP
      Posted March 4, 2010 at 2:24 pm | Permalink

      EXACTLY!!! Ever since my brief undergraduate and side project grad student days working in the rocky intertidal, I have always referred to them as “cockroaches of the sea!” I never found them appealing as food for that reason.

      Happy to see that I have phylogenetic data to support my discrimination…

      • Revereche
        Posted June 5, 2011 at 1:04 pm | Permalink

        And to think, I just thought they made cockroaches look delicious . . .

    • Posted March 4, 2010 at 4:32 pm | Permalink

      I don’t know… Since insects are crustaceans maybe it’s an argument *for* eating insects?

      “Honey, the grasshopper is delicious!”

      • MadScientist
        Posted March 4, 2010 at 4:41 pm | Permalink

        Grasshoppers are OK if you prepare and roast them properly, but having eaten them I think it would take nothing short of the desperation of a war to put them on my menu.

        As for other insects, I’m not trying Blatta Americana as a cheap substitute for shrimp.

        • Otto
          Posted March 9, 2010 at 3:23 pm | Permalink

          And what about the house fly ?

      • KP
        Posted March 4, 2010 at 5:35 pm | Permalink

        Hey speaking of grasshoppers, shouldn’t they have been used as an outgroup since they don’t really belong in the Hexapoda? (Leviticus 11:20-23)???

  3. Posted March 4, 2010 at 2:01 pm | Permalink

    I’m a fan of the arthropic principle: The universe was fine-tuned for ticks, roaches and scorpions.

    • Drunk Dude
      Posted March 5, 2010 at 1:12 am | Permalink

      Wouldn’t that be the arthropic principal?

      • Drunk Dude
        Posted March 5, 2010 at 1:13 am | Permalink

        D’oh, meant arthropodic

  4. Posted March 4, 2010 at 2:18 pm | Permalink

    ERV: See I use that as a good reason why I don’t mind dropping live ones head first into boiling water. They’re bugs, tasty bugs, but still bugs.

    • MadScientist
      Posted March 4, 2010 at 4:46 pm | Permalink

      Shhh … don’t let the pro-animal terrorists hear about how you cook ’em.

    • Posted April 9, 2014 at 12:57 pm | Permalink

      I don’t think that is a good analogy. Crustaceans are a very vast group, and saying that lobsters and insects are the same might be like saying lancelets (invertebrate worm-like ancestors of vertebrates) are the same as humans.

  5. Neil
    Posted March 4, 2010 at 3:13 pm | Permalink

    Mmmmm. Scorpions with melted butter. My favorite.

  6. Barry
    Posted March 4, 2010 at 3:20 pm | Permalink

    Insects are crustaceans! According to a new cladistic study, which may be obsolete tomorrow, insects for the time being may be interpreted as crustaceans.

    • blue
      Posted March 4, 2010 at 3:46 pm | Permalink

      Really? You read WEIT and to you science is some random walking, vacillating string of opinions that never settles on any one consensus? You don’t see it as a lens, that keeps getting polished to a better degree, bringing reality into increasing focus? Huh. Ok, then…

      • Sven DiMilo
        Posted March 4, 2010 at 6:03 pm | Permalink

        Barry’s right, in a way. All phylogenies are hypotheses, vulnerable to more and better data. We can never know “for sure” that we have reconstructed the correct phylogeny.

        Of course, this one seems to have a lot of good data…

        • Posted March 6, 2010 at 3:13 pm | Permalink

          All scientific statements of whatever order are hypothesis, and can be revised or overthrown by the next good solid falsification some bright grad student can demonstrate to the satisfaction of his peers. That’s why they call it science.

    • whyevolutionistrue
      Posted March 4, 2010 at 6:08 pm | Permalink

      Yeah, exactly in the sense that the idea of evolution itself may be obsolete tomorrow. All science is tentative. However, the support for putting insects in crustacea is VERY VERY strong, and is based on multiple taxa and genes. I’d bet a good sum that it’s not going to be “obsolete tomorrow.”

      • Barry
        Posted March 4, 2010 at 7:00 pm | Permalink

        Whoever would have thought that such a harmless little quip about the stability of cladistic classifications would generate any comments?

        • Insightful Ape
          Posted March 5, 2010 at 11:22 am | Permalink

          It is called trolling, Barry.

        • David Marjanović
          Posted September 15, 2010 at 6:13 am | Permalink

          If you call it “cladistic classification”, that shows you haven’t understood cladistics. It’s not classification, it’s phylogenetics. The science of phylogenetics, as opposed to the art it used to be.

  7. Neil
    Posted March 4, 2010 at 3:26 pm | Permalink

    Okay. Okay. Cockroaches with melted butter then.

    • Otto
      Posted March 9, 2010 at 3:27 pm | Permalink

      You don’t like them just roasted over an open fire ?

  8. blue
    Posted March 4, 2010 at 3:41 pm | Permalink

    Is this really novel? I recall long hearing that insecta are nested within crustacea… [runs off to wikipedia]. Nope, guess not, the crustacea entry only has this relationship as a recent edit. Well then, like other commenters said, I guess I just always KNEW this; shrimp = cockroach! 🙂

    • Sven DiMilo
      Posted March 4, 2010 at 6:07 pm | Permalink

      No, this is definitely not the first study that has had crustaceans paraphyletic with respect to hexapods.

  9. Hempenstein
    Posted March 4, 2010 at 5:55 pm | Permalink

    Good that this is getting more widespread recognition. Maybe 6yrs ago a friend generating expressed sequence tags (EST’s) from lobster told me that most often the closest hit he would get was to Drosophila, since that was usually the closest organism in the database at the time.

  10. Torbjörn Larsson, OM
    Posted March 4, 2010 at 7:53 pm | Permalink

    Not that the question has bugged me personally, but I find the answer fascinating.

    Funny, my gut reaction was the same: should I abstain from eating arthropods or include the insects?

    As for the later, I believe that the fatty larvae is considered the most delicious among the food cultures that make use of them. But spiders may be different; I’m reminded of my dad recounting an episode from a biology book, where some arachnid specialist who used to sample species remarked that some have ‘a nutty taste’. It may even be true!

    [The story was that he, absent-minded, picked such an individual that lowered itself by thread in a bus and used it as a convenient snack. Thereupon a lady witness fainted soundly next to him. :-D]

    it sets the standard for the numbers of species and genes that should be included in a good phylogenetic analysis.

    So, any predictions how many species and genes that needs to be sampled to tease out the bacteria radiation vs bacteria/archaeabacteria sister (and archaeabacteria/eukaryote sister, for that matter) phylogenetic hypotheses with enough “goodness” to satisfy?

    Or in other words, will it happen any year now or do we have to wait decades?

    • Torbjörn Larsson, OM
      Posted March 4, 2010 at 7:58 pm | Permalink

      Um, of course I’m not trying to put spiders within insects – the food aspect suggested itself. Sorry for the confusion!

  11. Jason
    Posted March 4, 2010 at 8:24 pm | Permalink

    It’s settled! The uniramia do not exist!!

    • Gerdien
      Posted March 5, 2010 at 12:42 am | Permalink

      Did anyone else than Steven Jay Gould defend Uniramia?

      • Posted March 5, 2010 at 8:40 am | Permalink

        While Gould’s opinions about arthropod classification may have been informed by wide reading in the area, he was not an arthropod systematist. The concept of Uniramia was championed by the late S.M. Manton, the renowned systematist who advocated a polyphyletic origin of the Arthropoda. Her ideas were controversial at the time, and the paper discussed in this post by Jerry provides fairly strong evidence against (some of) them.

  12. Josh Slocum
    Posted March 4, 2010 at 8:28 pm | Permalink

    I freakin’ knew . So glad all of you above recognize what I’ve always known. . . lobsters are simply Sea Cockroaches. Everyone laughed at me, but who’s laughing now?

    Odd, though. . . I still eat them without a hint of squeamishness. Show me a roach, though, and I’m into a panicked fit of sterilizing, boric acid powder, and Xanax. No, it’s not rational. I know. You don’t have to tell me.

  13. Josh Slocum
    Posted March 4, 2010 at 8:28 pm | Permalink

    Whoops, sorry about the html fail above!

  14. Miranda
    Posted March 5, 2010 at 5:14 am | Permalink

    “1. Insects (“Hexapoda”) are not a sister group of crustaceans, as was indicated by some molecular studies. Nor are they the sister group of myriapods, the traditional arrangement supported by morphology. Instead, insects are nested within crustaceans (see Figure 1). In the same sense that birds are dinosaurs, then, insects are crustaceans.”

    I’m not sure I’d write “in the same sense.”

    PhysOrg writes:

    “The weight of the evidence is now suggesting that not only did birds not descend from dinosaurs, Ruben said, but that some species now believed to be dinosaurs may have descended from birds.”

    • llewelly
      Posted March 5, 2010 at 6:21 am | Permalink

      Finding that a few non-flying theropods previously not thought to be avian are in fact avian is no more evidence against birds being descended from dinosaurs than ostriches. Furthermore – I don’t see anything in the article you link that is not refuted here.

    • llewelly
      Posted March 5, 2010 at 6:39 am | Permalink

      Oh, and see also here.

      • Miranda
        Posted March 5, 2010 at 12:00 pm | Permalink

        Hmmm, what does this say about peer review? On the one hand, your site seems pretty solid, but on the other hand, the study I referred to was peer-reviewed.

        • Insightful Ape
          Posted March 5, 2010 at 2:58 pm | Permalink

          Sorry. You are wrong.
          Evidence that birds are living dinos comes from many lines of evidence-fossils, anatomy, embryology, biochemistry, and most recently, respiratory anatomy and physiolgy.

          When you have the opinion of one individual against consensus built on numerous lines of evidence, that is called denialism.

          • Miranda
            Posted March 5, 2010 at 3:47 pm | Permalink

            So the PNAS was wrong?

            • Sven DiMilo
              Posted March 8, 2010 at 8:23 am | Permalink

              That particular article in PNAS is very likely to turn out to be wrong, yes. Ruben is riding his hobbyhorse.

              Peer review at PNAS is…variable.

    • blue
      Posted March 5, 2010 at 9:49 am | Permalink

      John Ruben, the Maniraptoran Denialist par excellence.

  15. Posted March 5, 2010 at 10:00 am | Permalink

    I find it fantastic that new evolutionary finds are accessible for me. Thank you, Dr. (Prof.?) Coyne.

  16. Barney
    Posted March 5, 2010 at 12:27 pm | Permalink

    So where does that leave trilobites in the arthropod family tree? On their own at one side?

    • Sven DiMilo
      Posted March 8, 2010 at 8:24 am | Permalink

      an earlier branch than any extant arthropods

      • David Marjanović
        Posted April 2, 2010 at 1:56 pm | Permalink

        Well, no. Nobody knows, because there’s no DNA left of any trilobite, and because nobody has done a sufficiently big morphology-based analysis as far as I know.

        A sister-group relationship between Trilobita (or a slightly larger clade?) and Chelicerata has occasionally been suggested, and found by small morphology-based analyses.

  17. TheBrummell
    Posted March 5, 2010 at 12:44 pm | Permalink

    This study is obviously deeply flawed. Figure 1. shows Onychophora used to root the tree, when EVERYONE KNOWS there’s a reticulation point there since they mated with the other ancestors of holometabolous insects to produce the far-too-complex-for-evolution complex life histories (caterpillars, in other words – gahd, it’s sooooo obvious!)

    /sarcastic idiocy

    Sorry, I just had to bring that up. Seriously, that study does set the bar for high-level phylogenetic analysis.

    • Sven DiMilo
      Posted March 8, 2010 at 8:24 am | Permalink

      like I just typed:
      Peer review at PNAS is…variable.

  18. Mike from Ottawa
    Posted March 5, 2010 at 12:52 pm | Permalink

    Hmmm. So the efforts to persuade folk in Australia to eat locusts by calling them ‘sky prawns’ isn’t so silly afterall!

    In one of Jack Vance’s ‘Demon Princes’ books, the hero is on another planet where he is served a meal that included a course of ‘sea insects on leaves’, which seemed odd until I thought that wasn’t far off a salad with shrimp and thus not so disgusting. I imagine Vance would be pleased to find out about this study.

  19. Posted April 5, 2010 at 9:40 am | Permalink

    I’ve long felt it was obvious that it all boils down to “lives in water = let’s eat it. Lives on land = YUCK!”

  20. Mary
    Posted August 5, 2011 at 5:20 am | Permalink

    This show the cultural bias we have when thinking of food. I dont undersand why people react with such a repulsion when proposing eating insects. You see a lobster and it remembers a lot an insect, and considering a cricket is herbivore and a crab or lobster is carnivore-detritivore, well it could be considered more disgusting eating a crab than a cricket.

  21. Chanel
    Posted December 12, 2011 at 3:54 pm | Permalink

    It was just last night that I learned that lobsters are sometimes called “insects of the sea”. I LMBO at the time because it is so rediculous. It is interesting that crustacians may be paraphylic. The lobster nickname may not as far-fetched. It makes sense when reversed. Insects may be called “crustacians of the land”. BTW In other cultures, humans eat insects. Eating insects may seem yucky to us, but I belive it is still possible. However cockroaches are cool but disgusting. I wouldn’t want to eat them because they are scavengers. Yet they may not be any worse than lobsters or crabs.

  22. Irving
    Posted August 25, 2012 at 2:49 pm | Permalink

    This study can be supported by neurophylogeny; in 2009 Straulsfeld linked the origin of insects with the group malacostraca in base of the comparation between some important neuropils of the brain, the fact of the genetic relationship of this groups implies that the evolution of their respectives nervous systems is not a simple evolutive convergence. What is your opinion?

2 Trackbacks/Pingbacks

  1. […] […]

  2. […] various arthropod groups should be grouped together. The current wave of molecular data shows that insects and crustaceans are more closely related to each other than they are to the other arthropods (the “pan-crustacean” hypothesis), while insects + […]

%d bloggers like this: