A very early fish

August 30, 2014 • 8:47 am

Many readers sent me a note about this paper, but, given my schedule, I simply hadn’t gotten around to reading it. Fortunately, Greg did, and gives us a nice summary of what it means.

by Greg Mayer

In the latest issue of Nature, Simon Conway Morris and Jean-Bernard Caron provide a detailed description of Metaspriggina walcotti, a poorly known and enigmatic fossil from the famous Burgess Shale of British Columbia. Originally described by Alberto Simonetta and Emilio Insom in 1993 from a single specimen that had been collected by Charles Walcott around 1910, in 2008 Conway Morris referred a second Walcott specimen to the species, and redescribed the species based on these two specimens. It was Conway Morris who first referred Metaspriggina to the chordates (the group that includes vertebrates, lancelets and tunicates). He and Caron have now redescribed Metaspriggina again, this time on the basis of 100 new specimens from British Columbia, and also referred specimens from a few other North American localities to the genus. The new material is very well preserved, and allows a much more detailed reconstruction.

Metaspriggina reconstruction by M. Collins.
Metaspriggina reconstruction by M. Collins.

The results of their studies are very interesting. Metaspriggina has the basic chordate features of a notochord, postanal tail, and gills. In addition it has segmented muscles, and the “vertebratey” features of eyes, nasal sacs, and perhaps cranial cartilages and arcualia. (The latter are arrayed along the notochord, and would be ancestral vertebrae.) Most interesting to me is that there are seven sets of paired branchial bars (gill arches), all but the most anterior supporting laterally directed gill filaments. The direction of the latter is significant. In jawed fishes and their descendants, the gills extend laterally from the skeletal arches, while in lampreys and hagfish (the cyclostomes) the gills extend medially from the arches. The lateral gills in Metaspriggina suggest that this is the primitive condition, retained by jawed fishes, and that the medial placement in cyclostomes is derived. The seven pairs of arches also suggest that the many arches of cyclostomes is another derived feature, and that therefore Metaspriggina more closely resembles the jawed fish ancestor. The anteriormost arch, perhaps the homologue of the jaw, differs from the other arches in being more robust.

An, anus; Brv, branchial bars (ventral element); Brd, branchial bars (dorsal element); Brp, branchial bar processes; Es, oesophagus; Ey, eyes; Gu, gut; He?, possible heart; Li, liver; Mo?, possible position of mouth; My, myomere; Na, nasal sacs; No, notochord; Ph, pharyngeal area . (From Fig. 2)
An, anus; Brv, branchial bars (ventral element); Brd, branchial bars (dorsal element); Brp, branchial bar processes; Es, oesophagus; Ey, eyes; Gu, gut; He?, possible heart; Li, liver; Mo?, possible position of mouth; My, myomere; Na, nasal sacs; No, notochord; Ph, pharyngeal area . (From Fig. 2)

In their phylogenetic analysis, Metaspriggina is close to Haikouichthys and Myllokunmingia, two other very early Cambrian vertebrates from the Chengjiang of China. Metaspriggina (ca. 500-515 mya) is slightly younger than the Chinese forms (ca. 520 mya). An interesting side result of their phylogenetic analysis is that Pikaia, formerly the only known Burgess Shale chordate, and usually considered a cephalochordate (i.e. a relative of lancelets, which is what they look like to me) comes out crownward of the cephalochordates, in fact as the sister-group (i.e. closest relative) of the vertebrates.

This paper is a real advance in our knowledge of vertebrate evolution. It is becoming clear that there is an at least modestly diverse Cambrian fauna of jawless fishes, and that these fossils will help us understand the origin of vertebrates and jawed vertebrates in a way that the extant jawless fishes (the cyclostomes) cannot, due to the latter being collateral relatives with their own long separate evolutionary history and corresponding suite of derived characters. These Cambrian fossils seem to provide a better model for the ancestral vertebrates than do the modern lampreys and hagfishes.

I would also add that these early Cambrian vertebrates look very much like what ancestral vertebrates were hypothesized to look like prior to their discovery.

Conway Morris, S. 2008. A redescription of a rare chordate, Metaspriggina walcotti Simonetta and Insom, from the Burgess Shale (Middle Cambrian), British Columbia, Canada. Journal of Paleontology 82:424–430.

Conway Morris, S. And J.-B. Caron. 2014. A primitive fish from the Cambrian of North America. Nature 512:419-422. abstract only

Simonetta, A. M. and E. Insom. 1993. New animals from the Burgess Shale (Middle Cambrian) and their possible significance for the understanding of the Bilateria. Bolletino di Zoologia 60:97–107. pdf

A note for taxonomy geeks: In his 2008 paper, and again in this latest one, Conway Morris refers to the second specimen collected by Walcott as the “lectotype”. This is completely wrong. The specimen upon which a new species is based is the holotype; additional specimens upon which the original description is based are called paratypes. In the old days, before the current rules were codified, authors would frequently base a new species description on several specimens, without specifying a particular specimen to be the name-bearer or holotype.  When no holotype was designated, all the specimens had equal status, and were called syntypes. This is bad in case it should turn out that the original set of specimens actually comprised more than one species (this happened a lot). A later author, in order to insure nomenclatural stability, is permitted to choose from among the syntypes a single specimen to be the primary type of the species. This selected syntype then becomes the lectotype, the other syntypes becoming paralectotypes. A lectotype has the same status as a holotype, in that it fixes application of the name should the type series prove to be composite.

So what’s wrong with the second specimen being a lectotype? Well first off, the second specimen was not part of the original type series, and so cannot possibly be a lectotype. (Simonetta and Insom mentioned this specimen, but thought it taxonomically distinct from Metaspriggina, so it was not part of the basis of the description of Metaspriggina— only the first specimen was.) If a holotype is lost or destroyed, or for some other very good (and rare) nomenclatural reason, it is possible to designate a new primary type, which would be called a neotype. But the holotype of Metaspriggina (the single original specimen) is still in existence and readily observable, so there is absolutely no need to try to designate a new primary type. This is inside baseball and perhaps small beer, but it’s really puzzling how Conway Morris seems not to understand the rules of nomenclature, and how this has not been caught by reviewers or editors.

23 thoughts on “A very early fish

  1. Really bizarre looking creature. Lends credence to Haldane’s remark that “nature is queerer than we can suppose.” If this is a common ancestor, what will Ken Ham make of this? More confirmation of evolution!

    1. Really bizarre looking creature.

      That probably depends on your meaning of the word “bizarre”. I’ve spent decades familiar with the description of Amphioxus, was reading the Haikouichthys and Myllokunmingia papers (they were published back to back, if I remember correctly) over a beer in the late 90s, and spent days puzzling over the cranial and spinal anatomy of lampreys and hagfish. The origin of vertebrates is an interesting, if somewhat parochial, field (us being vertebrates, after all). Arguably, it deserves no more study than the murky roots of the myriads of classes, orders and taxa of the arthropod radiation, but it does have the advantage that most people can relate the anatomy under discussion to the anatomy they are intimately familiar with. Which makes it undeniably popular.
      In that background – of Mid-Late Cambrian fishy blobs – this looks very familiar. Familiar enough that I’m downloading the paper this very moment, to turn into dead tree form.

  2. Looks like you have one too many ms in Myllokumnmingia, which is a good thing, because otherwise I had no idea how to pronounce it.

  3. A very impressive finding indeed. Imagine how much more impressive this science could be if Simon Conway Morris and Jean-Bernard Caron gave up their dead-end reductionist paradigm, chomped some peyote, and collectively linked their minds, letting their findings flow naturally from the universal consciousness. Time is running out. We must act now.

  4. Beautiful animal.

    The perfect example of the power of metamerism and evolution. I speculate that multiple repeating subunits can be modified and selected with less risk to individual organisms.

  5. Of course, over at the DiscoTute’s mendaciously named site “evolution news” they are using this to try to say, “Haha! Evolutionismists worng again!” Last line is “for Darwinists, it should hardly be more surprising to find than a Precambrian rabbit.”

    1. Whelp, that’s it then. The Burgess Shale is obviously just another part of God’s perfect plan to fool us with fossils.

  6. “[I]t’s really puzzling how Conway Morris seems not to understand the rules of nomenclature, and how this has not been caught by reviewers or editors.”

    Puzzling, maybe, but unfortunately not surprising. You don’t get grants or Nature papers by understanding taxonomy or biological nomenclature.

  7. Now I aren’t saying I’m old, but we used to catch these things down by the old primordial pond. If I remember correctly, the best bait was Hallucigenia on a number three hook. I once caught one > THIS < big.

  8. Interesting eye placement on the little beastie, presumably it was expecting predation from above. I wonder what it ate with that funny little mouth.

  9. Haven’t read the paper in detail yet, but an issue that might be relevant in phylogenetic analysis of early chordates is that post-mortem decay can mimic primitive character states and push a taxon further down the tree than it should be, as described in this open-access paper. Given the large sample of specimens rapidly buried in very fine sediment, I think it’s unlikely to be a problem here.

    Another issue for zoological taxonomy geeks that came up recently was in a couple of papers from the ‘pathology group’ of critics of the Pleistocene ‘Hobbit’ Homo floresiensis. They published in PNAS using the egregious ‘contributed by Academy member’ route that gives them (essentially) a free pass on peer review. Both papers presume that proving abnormality in the type specimen (skull and partial skeleton LB1) would invalidate any name based on it, citing section 1.3.2 of the Code, which is certainly an incorrect reading and should not have been allowed by editors or even the tamest referees.

  10. Fascinating post, I love the stories of these strange and long-extinct creatures. I wish there were more documentaries about them. Attenborough’s “First Life” was excellent. I wish it was a whole series.

Leave a Reply