Left: Xenoposeidon proneneukos holotype NHMUK PV R2095 in dorsal view (anterior to top), from Taylor (2018: figure 1A). Right: FIFA World Cup 2018 logo.

You can’t tell me that’s a coincidence.


Supplementary information

It’s coming home.



I’m delighted to announce the publication today of my new paperXenoposeidon is the earliest known rebbachisaurid sauropod dinosaur”. This is the peer-reviewed version, in my favourite journal PeerJ, of the manuscript that became available as a preprint eight months ago — which was in turn a formalisation of a blog-post from 2015.

Taylor (2018: Figure 3). Autapomorphies of Xenoposeidon proneneukos NHMUK PV R2095, mid-posterior dorsal vertebra, highlighted in red. A. anterior view. B. left lateral view. Numbers pertain to the numbering of autapomorphies in the text. 1a, neural arch covers whole of centrum, and 1b is contiguous with posterior articular facet. 2, neural arch is inclined forward by 30–35 degrees relative to the vertical. 3a, inclined ridge-like lamina marks ventral margin of 3b broad featureless area of bone. 4, large teardrop-shaped anterior fossa. 5a, vaulted laminae bound this fossa, but are not the medial CPRLs (5b, drawn in finer lines), which continue up to the presumed location of the prezygapophyses.

In a sense, then, this paper is old news. It doesn’t contain any startling new insights that readers of this blog wouldn’t already have been aware of. But it’s become more rigorous, better argued and justified, better illustrated (the image above is one of two new figures), and generally toughened in the forge of peer-review. It’s also now, of course, officially part of the scientific record.

I’m delighted about this paper for several reasons. First, of course, because Xenoposeidon is a beautiful specimen and now turns out to be rather more important than I’d previously realised. Second, because I hope this paper’s inclusion of the high-resolution full-colour 3D model as a supplementary file will help to establish this as common practice. But also third, because it’s my first paper in ages.

In fact, if you were being harsh, you could say it’s my first real paper since the annus mirabilis of 2013 when Matt and I had four good, solid papers come out in a single year. My CV lists five papers between then and now, but a case can be made that none of them really count:

  • Taylor 2014 is essentially an addendum to my and Matt’s PLOS ONE paper the year before.
  • Upchurch et al. 2105 is a significant and substantial piece of work, but almost all the credit on that one is due to Paul and Phil.
  • Taylor 2016 is more of an advocacy piece than a scholarly paper.
  • Ansolabehere et al. 2016 is merely a report summarising a multi-day discussion, and I am in any case only one of nine(!) co-authors.
  • Taylor 2017 is just a short comment on someone else’s ICZN petition. (In fact that one is so feeble I should just remove it from my CV.)

Putting it all together, it’s been the best part of five years since I made a significant contribution to the scientific record, and to be honest I was starting to wonder whether I could still do it. (My deep thanks go to Paul Upchurch and Phil Mannion for keeping my publication record on life-support with that Haestasaurus paper!)

The challenge for me now is, having got back on the horse, to ride it hard. In particular:

That’s not even mentioning other long-in-the-works projects like the descriptions of Apatosaurusminimus and “Biconcavoposeidon”. Sheesh. I’m so lazy. Nearly as bad as Darren.



Peter Falkingham and Nick Gardner independently put me onto Sketchfab: a website that provides a way to view and navigate 3D models without needing to download any software beyond the browser that you’re already running.

So get yourself over to the live Xenoposeidon model! Verify for yourself that the laminae are as I described them, that the posterior margin of the neural arch really does grade into the posterior articular surface of the centum, etc. Really, this is worth ten times whatever set of illustrations I might have provided.

Truly, we are living in the future!

UPDATE, 23 November 2017: see also this beautiful 3d model of the skull of Triceratops horridus, photogrammetrised from images taken at the Museum National d’Histoire Naturelle, Paris, France, by Benoît Rogez; and the same creator’s Nanotyrannus lancensis model, also from MNHN photos. And, most astonishingly, his model of the whole MNHN palaeontology gallery!

In writing the recent preprint “Xenoposeidon is the earliest known rebbachisaurid sauropod dinosaur” (Taylor 2017), it was invaluable to have a 3D model of the Xenoposeidon vertebra available. Here’s a short clip of viewing the model in the free MeshLab program. (It’s well worth full-screening to get the full impact.)

As I pan around, I look first at the upper margin of the posterior articular facet of the centrum, showing how the posterior margin of the neural arch shades into it — something that is not really apparent from photos, but needs the shifting perspectives that 3D offers to eliminate the interpretation that this contiguous border is due to damage.

Then I zoom in on the complex of laminae at the top of the left side of the neural arch, and explore the shapes of the intersections (ACPL with lateral CPRL, and PCDL with CPOL).

Finally I look at the distinctive sets of laminae on the anterior face of the vertebra which enclose the big, teardrop shaped centroparapophyseal fossa: lateral CPOL coming in from the lateral face of the arch, medial CPOL emerging from the pedicels, and the additional arched laminae that bound the space.

It’s just great to be able to do this. Time and again as I was preparing that manuscript, I went back to the model to check some detail — much as, twenty years earlier, Matt kept driving into the OMNH late at night to look at the Sauroposeidon holotype, to check out some idea he’d had as he worked on the description. The difference is, I didn’t need to drive into Norman, Oklahoma — or even London, England. The idea now of going back to trying to understand fossils from photos seems ridiculous.

A few years back, Matt wrote:

The idea of superseding photographs with 3D photogrammetric models is not original. I got religion last week while I was having beers with Martin Sander and he was showing me some of the models he’s made. He said that going forward, he was going to forbid his students to illustrate their specimens only with photographs; as far as he was concerned, now that 3D models could be cheaply and easily produced by just about everyone, they should be the new standard.

I’m totally on board with that, and said as much in the concluding paragraph of the new preprint.

The last thing I want to say here is to acknowledge the enormous amount of help I’ve had from Heinrich Mallison, digitizer extraordinaire at the Museum für Naturkunde Berlin. He’s invested many, many hours building models for me from my photos, pointing me to programs that I can use to view them, and helping me get started on making my own models. The greatest regret of my palaeontological life is that, when I happened to be in Berlin on 19th November 2008 and Heinrich invited me to come and watch the Germany-England friendly at his place, I couldn’t do it, and missed out on a pretty unique chance to see England beat Germany, in Germany, with a German. I doubt that chance will come up again any time soon.

I leave you with EmperorDinobot‘s life restoration of Xenoposeidon, which I stumbled across a few days ago. Obviously it’s wildly speculative, but I’m cool with that.


  • Taylor, Michael P. 2017. Xenoposeidon is the earliest known rebbachisaurid sauropod dinosaur. PeerJ PrePrints 5:e3415. doi: 10.7287/peerj.preprints.3415 [PDF] [PeerJ page]


There’s just time before midnight strikes to wish Xenoposeidon a very happy tenth birthday. It came along just a month and a half after SV-POW! itself — in fact, I can’t even remember now, a decade on, whether part of the reason we started SV-POW! in the first place was so we’d have somewhere to talk about it when the paper (Taylor and Naish 2007) came out.

Taylor 2017: Figure 4. NHMUK R2095, the holotype and only vertebra of Xenoposeidon proneneukos, in left lateral view, interpreted as a rebbachisaurid. This interpretation is modelled primarily on MNHN MRS 1958, a posterior dorsal vertebra from the holotype specimen of Rebbachisaurus garasbae. The CPOL passes through a sheetlike PCDL, as in Rebbachisaurus; but the lateral CPRL forms a cross-shaped junction with the ACPL, each of these laminae equally interrupting the trajectory of the other. Abbreviations as used in the text. Scale bar = 200 mm.

For the last few days, I have been working away like a trojan, trying to ready a new manuscript for launching on this day. I’ve taken two days off from my day-job to get it done before this arbitrary deadline, and here I am writing about it with just 15 minutes to go!

The title of this new manuscript (Taylor 2017) is “Xenoposeidon is the earliest known rebbachisaurid sauropod dinosaur”, and it explains in detail the argument that I made informally sixteen months ago when I first saw the rotating video of the Rebbachisaurus garasbae that Jeff Wilson and co put out with their then-new redescription of that species. I got it submitted (to PeerJ, natch) a bit more than an hour ago, and at the same time I clicked the “Make this available as a preprint” button. So as I write this, I am periodically checking back in the other window to see whether it’s made it through the basic editorial checks yet.

The thing is, I really love Xenoposeidon. I admit that a surprising number of people (my wife, Matt, his wife, Heinrich Mallison, his wife) seem to think it looks like a turd. But I honestly think it’s the most beautiful single bone I’ve ever seen. It’s a privilege to work on it.

Taylor 2017: Figure 5. NHMUK R2095, the holotype and only vertebra of Xenoposeidon proneneukos, in left anteroventrolateral view, highlighting the three sets of laminae related to the prezygapophyses. The trajectories of the medial CPRLs (which emerge from the neural arch pedicels) and the lateral CPRLs (which intersect with the APCLs) indicate the approximate position of the prezygapophyses. The additional arched laminae form the margins of the large, teardrop-shaped CPRF, but meet at a position some way below and posterior to the presumed location of the prezygapophyseal facets. Breakage of both medial CPRLs and the left ACPL and PCDL is indicated by cross-hatching. Note that, from this perspective, the lateral CPRL appears to turn a corner where it intersects with the ACPL, such that the posteroventral portion of the lateral CPRL appears contiguous with the dorsal portion of the ACPL. This is an illusion brought about by the eminence at the point of intersection. As always, this is much easier to see in three dimensions. Abbreviations as used in the text.

Anyway, I’ll link to the preprint as soon as it’s up. In the mean time, I’m just going to bask in the beauty that is Xenoposeidon.

Immediate update

Four minutes after posting this, and just three minutes before midnight, I got the notification from PeerJ that the preprint is up! So you are welcome to leave comments about the science on that page if you wish: I will take them into account when I revise the manuscript in light of the formal peer-reviews that will be coming along in due time.



Now that the new Wilson and Allain (2015) paper has redescribed Rebbachisaurus, we can use it to start thinking about some other specimens. Particularly helpful is this beautiful rotating animation of the best dorsal vertebra (here captured at the point of the rotation where we’ve viewing it in right anterolateral):


As I briefly discussed on Twitter, seeing this made me think of my baby, Xenoposeidon. Now that specimen, beautiful though it is, preserves only the lower one third of the vertebra. But there are some clear commonalities, and they’re clearer if you look at the animation.


Most obviously, there are laminae running up and down the anterior and posterior margins of the lateral face of Xeno’s neural arch, and those same laminae seem to exist in Rebbachisaurus. We didn’t name these laminae in the Xeno paper, but if they’re the same thing as in Rebbachisaurus, then they’re ACPLs and PCDL — anterior centroparapophyseal and posterior centodiapophyseal laminae.

If that’s right, then we misinterpreted the site of the parapophysis in Xenoposeidon. We (Taylor and Naish 2007) thought it was at the cross-shaped junction of laminae near the anterodorsalmost preserved part of the vertebra. In Rebbachisaurus, this cross exists, but it’s merely where the CPRL (centroprezygapophyseal lamina) intersects the ACPL.

But there’s more. In Xenoposeidon, the base of the CPRL (if that’s what it is) forms a “V” shape with an accessory lamina that proceeds posterodorsally from the same origin. (This is one of the features that’s apparent on the more damaged right side of Xeno as well as the nicer left side.) That lamina also seems to exist in Rebbachisaurus — but with the whole vertebra to consider, we can see that it’s not an accessory lamina, but a perfectly well-behaved CPOL (centropostzygapophseal lamina).

So if Xeno is indeed a rebbachisaurid, then the two branches of the “V” go to support the pre- and postzygs, and the laminae running up the anterior and posterior margins of the centrum support the parapophyses and diapophyses respectively. There are actually two crosses on each side of the neural arch: one at the intersection of CPRL and ACPL, the other at the intersection of CPOL and PCDL; but in the Xeno specimen, the posterior cross is lost, having been just above where the break occurs at the top of the neural arch.

Here’s what I mean:


In case it’s not clear, the grey lines are an (extremely crude) reconstruction, the blue lines label the important laminae, and the red circles highlight the two crosses.

Hmm. The more I look at this, the more convincing I find it.

But there’s more! The anterior aspect of the Rebbachisaurus vertebra also bears a notable resemblance to what we see in Xeno, with a pair or arched laminae forming a vaulted roof to the neural canal.



Jeff Wilson spotted the same thing in a sequence of comments on my tweets, saying:

That’s not a bad call—the infrazygapophyseal region of that vert is elongate, and there is a nice CPRF and those closely positioned TPRLs could mean that prz’s are close to one another or even conjoined. It’s tantalizing, but not much to go on. Would be nice to prep out CPRF & work out laminae on lat sfc.

Jeff is right that more preparation would help to figure this out.

Not that everything about the Rebbachisaurus dorsal is Xeno-ish. Most notably, the lateral foramen is nothing like that of Xeno, being an uninspiringly dull and simple oval rather than the much more elegant foramen-within-a-fossa arrangement that we see in Xeno. But there are other points of commonality, too, such as the flat stretch of bone above the fossa and the way the posterior margin of the neural arch reaches the posterior margin of the centrum.

All in all — while there is plenty of work yet to do — I am increasingly inclined to think that the evidence we currently have suggests Xenoposeidon is a rebbachisaurid. If that’s right, it would be quite an exciting result. It would be the earliest known rebbachisaur, and the only named one from the UK. (Mannion 2009 described, but did not name, a rebbachisaurid scapula from Wessex formation of the Wealden). Could Mannion’s scapula be Xenoposeidon? Unlikely, as it’s 10 million years more recent. But it could be a close relative.

Exciting times!



Get your red/cyan anaglyph glasses on, and feast your eyes:


Click through for stupidly high resolution.

Those of you who are still too cheap to have sprung 99¢ for a pair of glasses, you can make do with this grossly inferior wigglegram: