The Saga Continues…
Today marks the publication of a new paper describing spinosaurine material from the Kem Kem of Morocco, specifically a set of quadrate bones, the upper portion of the original tetrapod jaw joint. It’s freely available in PLoS ONE, and you can follow that link to read it now if you wish. As normal, I have something to say about this in the general context of things, which is largely parroting what others have said. So this is really not going to be much of a unique post.
This one’s been in the works for a while, this time by a team including Christophe Hendrickx, Octávio Mateus, and Eric Buffetaut. This stems from Christophe’s work describing the anatomy (this is a preprint, pre-publication inviting commentary, it is not formally published yet), and terminology, of non-avian dinosaur quadrates (both of those papers are freely available for perusal; send PeerJ useful feedback!). And this paper remarks on a subset of them.
The story of Spinosaurus has been torrid, torn apart by war, the ignominious flow of time, and by the attempts of systematists to make sense of a slew of bones from across northern Africa. From the mid-1990s, it seemed clear to some that there were probably more than one morphology of spinosaur in the north of Africa, specifically in Morocco and surrounding nations. Others were far more skeptical, referring to variation as a symptom of our lack of knowledge; that the differences weren’t systematic, or taxonomically viable, but merely either regional, ontogenetic, or less significant than thought. Spinosaurs inflated from the eastern Spinosaurus aegyptiacus to the that eastern form but also the western forms Spinosaurus maroccanus and Sigilmassasaurus brevicollis. And while some works treated these variously, it would not be for another decade that they received new treatment, where it was deemed by Brad McFeeters amongst others that the last of these was distinct; then later, that the last of these included the second of these. That perhaps there were two distinct spinosaurs in the north of Africa. Speculation abounded to bones in Egypt as well, whether there were two taxa. Discussion of a skeleton uncovered in Morocco and acquired through various means, then described by Nizar Ibrahim and colleagues, followed by Serjoscha Evers’s discussion on the topic, raised the problems that became clear, asking ultimately the question:
How Many Spinosaurs?
All of this is covered here, and at posts linked to in those first paragraphs.
Two works now underscore something that Evers’ work argued. In the Moroccan Kem Kem, there were probably two different, distinct, spinosaurine spinosaurids. One of them might very well be Spinosaurus aegyptiacus, albeit with regional differences occluded by the fact that the skeleton has been claimed to be chimerical. The other is Sigilmassasaurus brevicollis. A number of other remains from the north of Africa include partial articular bones, snout fragments, vertebrae, and scores upon scores of teeth. Hendrickx et al. concerned themselves with quadrates, that portion of the jaw joint articulating with areas of the skull, and determined there were two distinct morphotypes. Through the use of taking geometric landmarks and a best-fit algorithm, they found that these quadrates separate into two distinct types, and referred one to Spinosaurus aegyptiacus (morphotype I, including five specimens) and Sigilmassasaurus brevicollis (morphotype II, including two specimens). The explicit reasons for which specimens make it to which taxa aren’t underscored; reasonably, one might not choose which taxa, leaving this to future discoveries. However, accepting Evers et al.’s claims on taxic identity, Hendrickx et al. opt to use the two taxa as the valid options, and thus sort the material into each of these containers.
[Edit: Hendrickx et al. also chose to accept that Ibrahim et al. have correctly identified FSAC-KK 11888 as the neotype, and a specimen otherwise, of Spinosaurus aegyptiacus; this, despite concerns raised by Evers et al. on the topic. Needless to say, two quadrate morphologies in the Kem Kem support two distinct spinosaurine taxa. It remains my personal opinion that the holotype of Spinosaurus aegyptiacus, formerly intact and not dust, is substantiated by its description and plates, and a specimen from 3000 miles away, from a different apparent stratigraphic layer, and without resort to collection in Egypt, is insufficient for erection of a neotype. Nevertheless, whatever FSAC-KK 11888 is – and there is reason to believe it is very similar to the Egyptian taxon – it is probably representative of the taxon that owned the morphotype 1 quadrates. This does not in any way substantiate Spinosaurus aegyptiacus in the Kem Kem.]
Their study is interesting in that it demonstrates these quadrates all occupy a distinct area of morphospace from other theropod dinosaurs in which the orientation of the distal condyles, which articulate to the lower jaw, are more eccentric, and angled further outward than in other theropods.
From Hendrickx et al., 2016. doi:10.1371/journal.pone.0144695.g007 Morphospace along two coordinate axes reflecting relative distal width across condyles to condyle length and to angle of the condyles to the wide axis across them. Specimens near the base of both axes, where spinosaurids are, have broader condyles and higher angles, versus those at the opposite end which have narrow condyles, and lower angles, such as the supposed fishing abelisaur Masiakasaurus, oviraptorids, etc. Not many sampled taxa occupy the extreme space, due to the overall wideness of the distal quadrate in most theropods. Tyrannosaurs shown have angled condyles, but the condyles have a low aspect ratio, being narrow and long.
This meant many things, covered in their paper, such as how the jaw behaves when it is opened, increasing its width at the joint, and thus increasing the angle between the left and right mandibles. In some animals, this permits extreme depression (opening of the jaw) by allowing the attachment point of the mandibular depressor muscle at the retroarticular process to move lateral to the quadrate itself. This isn’t possible in spinosaurids due to articular shape, but it still permits extreme gape.
I was given the opportunity to illustrate this:
From Hendrickx et al., 2016, doi:10.1371/journal.pone.0144695.g015 The jaw joint in lateral and rostral views, Ibrahim et al.’s reconstruction based on the conflated “Spinosaurus,” illustrating cranial anatomy and mandibular depression in Spinosaurus, and Jason Poole’s work at bottom. Only A-D, and F-G are my own work.
Before you ask, the skull is based on a generalized reconstruction, rather than a specific set of bones, so the shapes aren’t meant to be “perfect.” They capture what they are meant to show … artistically.
A mandibular symphysis without the extremely rugose texturing indicating a thick cartilaginous symphyseal tissue suggests a looser tissue would permit more mobility. The intramandibular joint, between dentary/splenial and angular/surangular, is poorly known, as no complete jaw of a spinosaurid is available for analysis.
Little further attention has been made of a further analysis, one by Ute Richter and colleagues, assessing teeth collected from the Kem Kem — one of the “low value” fossils Moroccan authorities permit locals to collect, sell, and export — in which they note that “spinosaurine” teeth found can be further clarified into three, rather than two, distinct morphotypes, dependent on fluting of the surface, which side of the crown if more heavily fluted, and wrinkling between the flutes. Work assessing dental variation in a single jaw is also further complicated by the absence of good, intact material, save for material referred to baryonychines. And so we continue to be left wondering.
As the story goes, more evidence piles up that there are not one north African, Cenomanian spinosaurid, but many, diverse and peculiar from one another. We move from knowing next to nothing and having nothing, to knowing much more, and having more, yet still end up with more questions. We don’t know the answer to the question. But we have better tool to answer it.
Hendrickx, C., Mateus, O. & Buffetaut, E. 2016. Morphofunctional analysis of the quadrate of Spinosauridae (Dinosauria: Theropoda) and the presence of Spinosaurus and a second spinosaurine taxon in the Cenomanian of north Africa. PLoS ONE 11 (1): e0144695.
Richter, U., Mudroch, A. & Buckley, L. G. 2012. Isolated theropod teeth from the Kem Kem Beds (early Cenomanian) near Taouz, Morocco. Paläontologische Zeitschrift 87 (2): 291-309.
Jaime A. Headden
The Bite Stuff 
Hendricks referred one quadrate morphotype to Spinosaurus since it is the same as those of the two quadrates of the Ibrahim et al specimen. They follow Ibrahim et al. taxonomy to refer the new specimen to Spinosaurus aegyptiacus. Even if we may disagree with such taxonomy, their analysis confirm that even the few cranial bones of the specimen belong to a Spinosaurinae.
With respect to how the paper by Hendricks et al. fits in the context of the paper by Evers et al., Hendricks et al. are careful not to act like splitters by assigning the non-Spinosaurus quadrates to Sigilmassasaurus, since they know that no cranial material is known for Sigilmassasaurus (Ibrahim et al. 2014 were close, but not quite correct about Sigilmassasaurus, even though Evers et al. and Hendricks et al. agree with Ibrahim that Sigilmassasaurus is a spinosaurid).
Interested to what degree the lower mandible could bow out to swallow prey. I have seen some photos of pelicans take on some unbelievably large fish. Not in agreement that this paper or the Evers really “casts doubt” on the general skeletal morphology of the Ibrahim version. The bits and pieces now ascribed to Sigillmassasaurus which Ibrahim et al had sunk into Spinosaurus were not pivotal in their skeletal proportions.
I don’t see how the work by Evers at the first doesn’t cast doubt on Ibrahim et al.’s skeletal reconstruction. They both, along with Richter, indicate the presence of more than one spinosaurid morph, and moreover Evers et al. and McFeeters et al. suppose the morphology of a neck rather unlike that of Ibrahim et al.s assumption, forcing it as a chimera. Ibrahim et al. assumed that all the material is consistent with a single taxon, without substantive information to support them, assumting one large taxon period in the incoherent time of the Kem Kem.
There are at least two neck morphologies, and they aren’t particularly consistent with one another. One of them includes anterior dorsals without tall spines, cervicals with virtual no spines up into the anterior neck. The other has tall spines throughout. I’m not sure how we can assume the Ibrahim et al. reconstruction is consistent with this data.
I guess, to clarify, what sense of chimera I speak of that; 1) FSAC-KK-11888 is a chimera which I don’t agree with or 2) that the skeletal composite that Ibrahim et al came up with is a chimera, which does appear to be the case at this point.
However given that the neotype – or Spino “C” whatever you choose – and Stromer’s Spino B both still converge on a similar morphology of short legs and lengthened vertebrae suggests that Ibrahim’s composite skeleton has some element of truth in general proportion but is due for a revision taking into account the neck morphology. Is that the gist of it all?
I do recall from some illustrations in the Ever’s paper that they drew up the neck as relatively long and mobile for Sigillmassasaurus although I don’t recall them explicitly stating that in the paper or how they deduced that from the isolated pieces – maybe I have to reread it. So, I guess to clarify Spinosaurus had more of a “straight” neck but Sigillmassasaurus a more typically S-curved theropod neck?
I think I get it now. From Evers paper they argue pretty convincingly that Spino B and FSAC-KK-11888 differ in several manners and even if they are both respectively not chimeras they are not likely the same taxa. Additionally I found the part, buried deep in the paper, where they argue for a sigmoidal neck in Sigillmassasaurus. I am starting to wonder if FSAC-KK-11888 would have been better off if treated as its own beast – a new species.
“It remains my personal opinion that the holotype of Spinosaurus aegyptiacus, formerly intact and not dust, is substantiated by its description and plates, and a specimen from 3000 miles away, from a different apparent stratigraphic layer, and without resort to collection in Egypt, is insufficient for erection of a neotype.”
Isn’t this pretty much exactly what happened to Iguanodon? In that case the type specimen was clearly dubious, having been based on teeth alone, but something like the Maidstone specimen would have been a much better choice of neotype than the Bernissart specimens from a great distance away and possibly a different stratigraphic age. Not that it’s right, just nothing new.
The type of Iguanodon anglicus was crap, though, and has become a reasonable trend, we’ve moved away from the use of teeth, isolated or part of blocks of jaw bone, as holotypes. And for good reason. There were two choices, the first being anglicus needed a new type specimen to substantiate it, and whether anglicus then was substantial enough for Iguanodon. The latter, it was deemed, was that it didn’t work; I disagree, but whatever. It is difficult to tell what the type teeth belonged to, which robust and gracile iguanodonts available. Moreover, the proposed neotype was from the same general area and stratigraphic layer. So, it’s a little [actually, a lot] different.
” the proposed neotype was from the same general area and stratigraphic layer.”
I was under the impression that the original type teeth of I. anglicus were Valanginian, while the neotype I. bernissartensis is Aptian. Is that not correct? That’s a least a 10 million year temporal separation. The holotype and proposed neotype Spinosaurus are both Cenomanian aren’t they?
What implications does this have for cranial reconstructions of spinosaurines? I remember talking to Ocatvio Mateus on Google+ about this and he pointed out that the new data we have on spinosaurine quadrates suggests a relatively shorter and boxier preorbital region than previously though (or at least, that’s what I thought he said). I also think he said that even a rostrum as long as MSNMv-4047 would have just come from an animal with a total skull length of not much more than 150cm.
It shouldn’t mean much. The shape of the rear of the mandible is suggested by the front, and the reconstruction provided in the paper, as produced by me, is a modification of the merging of Irritator’s skull with the snout of MSNM V4047, which is assumed to be a large spinosaur of the type we see in either Morocco or Egypt. Thus, their skulls would see similar. Sergey Krasovksy’s reconstructions imply such similarity. The major differences involved appear to refer to how wide the distal quadrate was, so the narrower morphology would have slightly less widening of the mandibles during gape.