Xenoturbella and Xenacoelomorpha

Xenoturbella, Acoela, and Nemertodermatida are worm-like simple bilateral animals, which might give insights into the earliest steps in the evolution of Bilateria. Phylogenomic analyses (Philippe et al. 2011; Cannon et al. 2016) and some ultrastructural characters (Franzén & Afzelius 1987; Lundin 1998) group them into a monophyletic group, which has been named Xenacoelomorpha (Philippe et al. 2011; Tyler & Schilling 2011).

Xenoturbella has been kind of a puzzle for zoologists since it was described in 1949. First considered to be a primitive turbellarian flatworm, occasionally affiliated with Hemichordata, suggested to be a sister group to other bilaterians, for a moment mistaken for a mollusc, then suggested to be a deuterostome, then again together with Acoelomorpha (Acoela and Nemertodermatida) as a sister group to Bilateria, and then back to deuterostome affinity, as a sister group to Ambulacraria (Echinodermata and Hemichordata) (Telford 2008; Hejnol et al. 2009; Philippe et al. 2011). It is exciting that four new species of Xenoturbella have now been found in the Pacific ocean (Rouse et al. 2016). Before, only one small (up to 4 cm) species (X. bocki) was known from the west coast of Sweden (the second, X. westbladi, from the same place is probably the same species). Three of the new species are much bigger than X. bocki, some of them 20 cm or longer, but all are otherwise anatomically and in their habits (feeding on bivalves on the muddy ocean floor) quite similar to each other, which is also consistent with the genetic data. I heard about a new large Xenoturbella species found in the Pacific in 2010 while at EDIT taxonomy course in Kristineberg, Sweden. Sounded pretty unbelievable. I was hoping that this amazing new discovery will help to solve the phylogenetic position of Xenoturbella and Xenacoelomorpha more reliably, but it seems that the debate will continue for years to come.

Rouse et al. (2016) gathered also phylogenomic data for one of the new species and sequenced complete mitochondrial genomes of all the new species. Their phylogenetic analyses were incongruent to each other. Phylogenomic analyses based on nearly 1200 nuclear protein coding genes found Xenacoelomorpha (composed in this dataset of one Acoela and two Xenoturbella species) as sister group to other Bilateria (Nephrozoa) (using site homogeneous models) or as a sister group to Protostomia (using site heterogeneous CAT model). Analyses based on mitochondrial proteins grouped Xenacoelomorpha (4 acoels and 4 xenoturbellid species) with deuterostomes. In the same issue of Nature where Rouse et al's paper was published, Cannon et al. (2016) investigated the placement of Xenacoelomorpha more thoroughly, although using only one Xenoturbella species, but with many more Acoela species (7) and including Nemertodermatida (4 species). Cannon et al. (2016) added impressive amount of new transcriptome data (including for Xenoturbella), which enabled them to have pretty complete datasets for phylogenetic analyses (phylogenomic datasets tend to be gappy if not based on fully sequenced genomes). Cannon et al. (2016) results were much more consistent compared to Rouse et al. (2016), finding support only for sister-group relationship between Xenacoelomorpha and Nephrozoa (all other Bilateria, hypothetically having ancestrally nephridia in contrast to Xenacoelomorpha, which lack them). However, when fast-evolving acoelomorphs where excluded (retaining only clearly slower evolving Xenoturbella), statistical support for Nephrozoa and Deuterostomia decreased (from 99–100% to 81% in both cases). Unfortunately, Cannon et al. (2016) did not analyse this dataset with more complex CAT model and I suspect if they had, support for Nephrozoa and Deuterostomia would have disappeared entirely and Xenoturbella might have jumped next to Ambulacraria (as in Philippe et al. 2011). Cannon et al. (2016) did analyze their main dataset (which included Acoelomorpha) also with CAT model fully supporting Nephrozoa and Deuterostomia, but the internal branches leading to Nephrozoa and Deuterostomia of the resulting phylogenetic tree were clearly shorter (barely visible in the published figure) than in the analyses using simpler phylogenetic models. Simpler site homogeneous models underestimate the amount of sequence evolution compared to site heterogeneous models (like CAT) and can therefore produce artefactual groupings (grouping of fast-evolving, long-branch taxa with each other or with a distant out-group, regardless of their actual phylogenetic affinities). It could be that the fast-evolving Acoelomorpha (which is clearly evident in the long branches they display in phylogenetic analyses, the shortest ones perhaps still about twice as long as the Xenoturbella branch) caused the whole Xenacoelomorpha to shift artefactually at the base of Bilateria, away from more slowly evolving Deuterostomia and this way creating the Nephrozoa. CAT model was able to shorten the branches supporting monophyly of Nephrozoa and Deuterostomia, but not entirely to overcome possible non-phylogenetic signal present in the fast-evolving Acoelomorpha.

In conclusion, I think it is too early to exclude deuterostome affinity of Xenacoelomorpha. Curiously, while looking at the pictures of the new large species of Xenoturbella, I thought that they looked a bit like enteropneusts (Hemichordata). Could it be that the ring furrow of Xenoturbella, which divides the body into the head and tail region, is homologous to the collar (or to part of it) of Hemichordata (Deuterostomia: Ambulacraria)?


Cannon JT, Vellutini BC, Smith III J, Ronquist F, Jondelius U, Hejnol A (2016) Xenacoelomorpha is the sister group to Nephrozoa. Nature 530: 89–93. doi: 10.1038/nature16520
Franzen A, Afzelius BA (1987) The ciliated epidermis of Xenoturbella bocki (Platyhelminthes, Xenoturbellida) with some phylogenetic considerations. Zoologica Scripta 16: 9–17. doi: 10.1111/j.1463-6409.1987.tb00046.x
Hejnol A, Obst M, Stamatakis A, Ott M, Rouse GW, Edgecombe GD, Martinez P, Baguñà J, Bailly X, Jondelius U, Wiens M, Müller WEG, Seaver E, Wheeler WC, Martindale MQ, Giribet G, Dunn CW (2009) Assessing the root of bilaterian animals with scalable phylogenomic methods. Proceedings. Biological sciences / The Royal Society 276: 4261–4270. doi: 10.1098/rspb.2009.0896
Lundin K (1998) The epidermal ciliary rootlets of Xenoturbella bocki (Xenoturbellida) revisited: new support for a possible kinship with the Acoelomorpha (Platyhelminthes). Zoologica Scripta 27: 263–270. doi: 10.1111/j.1463-6409.1998.tb00440.x
Philippe H, Brinkmann H, Copley RR, Moroz LL, Nakano H, Poustka AJ, Wallberg A, Peterson KJ, Telford MJ (2011) Acoelomorph flatworms are deuterostomes related to Xenoturbella. Nature 470: 255–258. doi: 10.1038/nature09676
Rouse GW, Wilson NG, Carvajal JI, Vrijenhoek RC (2016) New deep-sea species of Xenoturbella and the position of Xenacoelomorpha. Nature 530: 94–97. doi: 10.1038/nature16545
Telford MJ (2008) Xenoturbellida: the fourth deuterostome phylum and the diet of worms. Genesis 46: 580–586. doi: 10.1002/dvg.20414
Tyler S, Schilling S (2011) Phylum Xenacoelomorpha Philippe, et al., 2011. In: Zhang, Z.-Q. (Ed.) Animal biodiversity: An outline of higher-level classification and survey of taxonomic richness. Zootaxa 3148: 24–25.


  1. I find this post extremely fascinating.

    Being no expert in the field, I am having difficulty reconciling what you say with Casey Dunn's earlier rebuttal.


    Any clarification would be greatly appreciated.

    While I am here, could anybody please explain to me why Ctenophora and Cnidaria are not commonly regarded as Bilateral?

    Thanks in advance to one and all for any help for any help proffered.

  2. Of course, I meant to say this post as well as the post immediately before.

  3. Thanks for your comments and the link to Casey Dunn's blog. I have to read it more carefully and hopefully can comment on this.

    Cnidaria and Ctenophora are not regarded as bilateral, because they do not display bilateral symmetry, that is you cannot divide their body into left and right nor to ventral (belly) and dorsal (back) side. Still, things are a bit more complicated, as they usually are, because anthozoan Cnidaria can display bilateral symmetry internaly and because of that Cnidaria has sometimes been considered to be bilaterian. Because the phylogenetic branch Bilateria is anatomically and genetically well defined, clearly separate from Cnidaria, Ctenophora, Porifera, and Placozoa and this clade needs a name, we might as well stick with the well known name Bilateria.

  4. Hello Marko

    Thank you for helping me out... this is definitely outside my comfort zone and very distant from my level of expertise.

    Some taxonomists argue that Cnidarians are descendants of ancient bilateral coelomates and not the other way around. Biologists have known since the 1920s that Cnideria had a directive axis which gave them right and left-hand sides. Volker Schmidt goes on to argue that non-radially organized hydrozoan larvae have an anterior concentration of sensory and ganglionic nerve elements, suggesting that a fundamental genetic toolkit for the establishment of bilateral and polarized anatomies was already present before the Cnidaria-Bilateria divergence. He goes so far as to suggest that diploblastic status of adult Cniderians is derived and that true mesoderm can be even be detected during Cniderian embryogenesis. I admit according to my limited reading that last argument is particularly contentious.

    Here is my problem - when I stare at Cnidaria - I witness two vertical axes of symmetry perpendicular to each other. The point is moot as you point out about internal structure:

    Regarding Ctenophore anatomy. The oral-aboral axis is the axis of symmetry around which is arrayed AGAIN the two perpendicular axes of biradial symmetry.

    I guess that would mean Ctenophores may have a "left and a right" or at least some version of "two sides" but do not have an obvious dorsal vs. ventral side.

    I was wondering out loud whether this interpretation was correct. I was wondering about Jordi Garcia-Fernàndez and his great review where he claims

    "...in fact, the simplest bilaterian metazoans, having a single or only a few Hox-like genes and several NK genes. This would imply that no basal non-bilaterian animals currently exist. An intriguing exception might be the placozoans..."


    I am very perplexed... any help you could provide would be greatly appreciated!

  5. Tom, seems that you know about these things more than I do :)

    I haven't followed the literature about various hypotheses on the origin of Bilateria, I have been mostly interested in phylogenetic relationships.

    My understanding, based mostly on phylogenetic literature, is that some genetic toolkit was present before Cnidaria-Bilateria or even Porifera-Bilateria divergence, which enabled animals to become more complex, but it didn't necessarily happen (examples are Porifera and Placozoa). To me it starts to look increasingly more likely that complex anatomies (mesoderm, different symmetries, nerves and muscles) evolved in parallel among the three main groups of eumetazoa (Ctenophora, Cnidaria, Bilateria) rather than from a common ancestor sharing all these characters. I'm relying here mostly on papers (mentioned in a previous post), that found parallel expansions in gene families specifying nervous systems and muscles:

    Similar parallel origins of complex characters (coeloms, segmentation, large brains etc) appear to be norm among Bilateria. Andreas Hejnol has written with colleagues in that direction, e.g:
    http://dx.doi.org/10.1016/j.devcel.2009.07.024 - especially illuminating

    To me, theories postulating complex ancestor for Bilateria, or for Cnidaria-Bilateria ancestor, seem less likely.

    But if you want me to go into details, then I have to admit my ignorance. So unfortunately I cannot give you any good answer and solve the perplexities. I guess many of the answers do not even exist yet, and still need to be researched.

  6. Hello Marko,

    I am no expert - I am less than a rank amateur. In fact, I am an aging Canadian High School Biology teacher desperately attempting to stay current.

    I am delighted at your citations of Hejnol & Martindale! I have long been fans of theirs, especially after reading their chapter in this book.


    I confess I remain somewhat bewildered and need to revisit your citations - especially the last. I remain unconvinced about parallel expansions in gene families specifying nervous systems & muscles, etc ... In my limited experience, convergent evolution is often as not a cop-out. My limited understanding of EvoDevo suggests that molecular toolkits were present in ancestors for a reason - and that what appears at first glance to convergent evolution really isn't. I am thinking of the various lineages of Mollusc Eye evolution as a case in point.

    Mind you rereading what I just wrote makes me uneasy and we may be talking at cross-purposes and really be essentially agreeing with each other.

    Up to now, the acoelomate worms were viewed as the crucial link between simple animals like sponges and jellyfish and more complex organisms. It has now emerged that these animals did not always have as simple a structure as they do today.

    ITMT, I am convinced the primitive UrBilateran was probably pretty complex but rapidly split into different lineages in response to what Hejnol & Martindale deem the shift of blastopore formation from the animal pole to the vegetal pole creating developmental problems that required quick resolution. I resorting to pure speculation here, but perhaps the diffuse nerve net of the UrTriploidBilateran ancestor needed to concentrate ventrally or dorsally due to developmental constraints imposed by the blastopore axis shift.

    I still remain confused regarding the definition of Bilateral. Aren't the Planulae of Cnidarians bilateral? So why are not Cnidaria considered bilateral no differently than Echinoderms?

    Meanwhile - I remain particularly perplexed by Xenoturbella. Does Xenoturbella embryology indeed recapitulate Lophotrocophora ontogeny? If so - wow!


    I have over-stayed my welcome already. I am still curious to your reaction to the Dunn paper I cited at the beginning.

    best regards

  7. I commented Dunn's paper

    Thanks for the link to the book.

    I'm no expert either. Still, seems that I have a different opinion regarding the nature of Urbilateria. Genetic toolkit specifying some complex morphologies in distant lineages (e.g. metamerism in arthropods, annelids, and chordates) doesn't automatically mean that their ancestor had these morphologies (I think this is what Hejnol argues; one another paper about nervous systems: http://dx.doi.org/10.1098/rstb.2015.0045, although I have only read the abstract). Certainly, the genes which are now part of some genetic toolkit did something in the ancestor, but effects of those genes didn't have to manifest in complex morphologies. The genes might have been adopted independently to solve similar problems. Maybe only few changes have to made in gene regulation to turn simple animal into a complex one, without inventing many new genes.

    I think Xenoturbella has a good chance, regardless of its phylogenetic position (among deuterostomes or sister to Nephrozoa), to give some indication how first bilaterians looked like. Perhaps it has simplified in some respects, but maybe not so much. Parasites are usually the ones that can simplify extraordinarily, free living organism not so much, I think (although it appears that several annelids living between sand grains have gone through some degree of simplification: http://dx.doi.org/10.1016/j.cub.2015.06.007).

    Analogues situation seems to be with multicellularity in eukaryotes. I think no one argues seriously that ancestor of extant eukaryotes was multicellular. The more genomes of protists become available the more difficult it becomes to specify which genes are the true signatures of multicellularity (one recent example from Amoebozoa http://gbe.oxfordjournals.org/content/8/1/109.abstract). Many genes that were thought to be essential for multicellularity keep turning up in single celled protists. It seems that the eukaryote ancestor had quite a gene rich genome which made evolution of multicellularity possible. But it didn't have to happen.

    It might well be that Cnidaria should be regarded as bilaterian on anatomical grounds, but the name "Bilateria" is almost unversally used to refer a particular monohyletic group of animals that doesn't include Cnidaria. I think it would be better to keep this name, even if it might be somewhat misleading.

  8. Hi Marko

    I cited you here:

    best regards

  9. Hi Marko,

    You state that
    " Many genes that were thought to be essential for multicellularity keep turning up in single celled protists."

    A simple explanation therefor could be that those particular single celled exemplars had multicellular ancestors. A case in point would be Yeast.

    I respectfully disagree with you on conserving old convention with taxonomic nomenclature. For example, I suggest it prudent to retire tired and useless terms such as "Protists" and "Bacteria". When maps no longer correspond to the landscape, it's time to draw a new map, IMHO.

    best & grateful regards


  10. Hi again Marko... thinking over your response, I would like to elaborate on my earlier raising of eyebrows at when you said,

    Parasites are usually the ones that can simplify extraordinarily, free living organism not so much...

    I don't think that is in fact the case...

    Some coelomate Locotrophozoan ancestor gave rise to pseudocoelomates (today’s Rotifers). Some coelomate Ecdysozoan ancestor also gave rise to peudocoelomates (today’s Nematodes). Some coelomate Locotrophozoan ancestor also gave rise to acoelomates (today’s Flatworms).

    You yourself contend that some coelomate Deuterostome ancestor also gave rise to acoelomates (Xenoturbellida and Acoelomorpha)…

    In other words, all modern acoelomates and pseudocoleomates are probably derivative. Meanwhile, it is becomingmore and more likely that the original Ur-Triplobalstic-Bilateran may be ancestral to Cnidaria and Porifera.

  11. Hi Tom,

    Ah yes, yeast is a nice example. I remembered also tunicates, which quite likely evolved from lancelet-like (Cephalochordata) ancestor. Still, to me seems that parasites go to bigger extremes (e.g. Microsporidia that are practically prokaryotes).

    It is not at all certain that rotifers, nematodes, flatworms and many other pseudocoelomates or acoelomates evolved from colemates (see http://dx.doi.org/10.1016/j.cub.2015.06.068). Quite likely coelemates evolved many times independently from acoelomates or pseudocoelomates (I prefer this scenario). Sure, some pseudocoelomates quite clearly have evolved from coelomates (e.g. different interstitial annelid lineages).

    I now even think that among deuterostomes, regardless of its internal phylogeny or if it is monophyletic or not, coelomes might have evolved independently in Ambulacraria and Chordata...

    I really have a different opinion. I don't believe that Porifera is simplified either, to me there appear to be no serious evidence for that. Considering that monophyly of Porifera is not even certain, or if it is, then the two or three lineages of sponges (Demospongiae+Hexactinellida, Homoscleromorpha, Calcarea) diverged from each other in such a short time that they are effectively independent lineages. This shows that body plan that sponges have, has been the same practically since the root of animal tree of life (phylogenetic analyses show branching of these sponge lineages very close to the root, clearly before Bilateria started branching). It would be unparsimonious to postulate that these two or three sponge lineages independently simplified to become so similar (it's no accident that Porifera is usually classified as one phylum).

    But I must leave it like that, just an opinion. It's not my field of research (I study taxonomy and phylogeny of sawflies), but have been interested about phylum level animal phylogenetics for a long time and reading the literature has given me such an impression.

    I see you're having quite a discussion at theskepticalzone.com