Same old story: Naish plans to blog on long-promised subjects, Naish gets distracted by cool new stuff, Naish ends up writing about cool new stuff and delaying long-promised subjects for even longer. Here, inspired by a paper I recently published with University of Bristol's Barbara Sánchez-Hernández and Mike Benton (Sánchez-Hernández et al. 2007), I've made a concerted effort to finish writing about the Mesozoic tetrapods of the Galve region of Teruel Province, NE Spain. In the previous post I covered crocodyliforms and pterosaurs. This time we get to the dinosaurs [adjacent image shows some heterodontosaurids: painting by Greg Paul]. As nice as dinosaurs are, I have to say that something far bigger and more newsworthy is on the horizon - it's coming soon here on the blog, and I hope that it will generate a lot of interest. Stay tuned.
By the way, today I saw two Roe deer Capreolus capreolus. Both were feeding in the middle of a field: apparently a 'new' behaviour that they indulge in regularly now. And in a shallow pool adjacent to the M27 near Portsmouth, I counted eight Little egrets Egretta garzetta: this bird was nationally very rare prior to the 1980s but now occurs everywhere in Britain; it's been breeding here since the late 1990s. Anyway...
Among the dinosaurs of Galve, we report new material of theropods, sauropods and ornithischians (Sánchez-Hernández et al. 2007). I'll keep it brief - as who wants to hear more about theropods and sauropods - but one particularly interesting new discovery we report is the presence of (drumroll)... spinosaurines. That's right, not just spinosaurids (the group that includes both baryonychines and spinosaurines), but spinosaurines. The only well known spinosaurines are Spinosaurus from northern Africa and Irritator from Brazil, and this is the first record of the group from Europe (to my knowledge, and disregarding old, incorrect suggestions that Becklespinax and other such taxa might be close kin of Spinosaurus). This is significant, as - assuming that we are correct in our identification - it now means that spinosaurines have a fossil record that extends as far back as that of their sister-taxon, Baryonychinae. It also means that the oldest spinosaurines are European. Furthermore, the teeth aren't particularly big, with crown heights that are mostly less than 20 mm. Are the teeth from juveniles, or do they represent a small-bodied taxon? This is an awful lot to stack up on a bunch of isolated teeth, sure, but I strongly disagree with the idea that we should just sweep this sort of evidence under the carpet and pretend it doesn't exist [adjacent image shows the mounted Spinosaurus holotype, photographed in Munich prior to 1944].
Teeth belonging to baryonychines, allosauroids, dromaeosaurids and indetermine coelurosaurs have also been recovered at some of the Galve sites. As is the case with several other Spanish baryonychine teeth, the Galve specimens differ in detail from those of Baryonyx walkeri, and might belong to another species (for more on this idea seen Ruiz-Omeñaca et al. 1997, Naish 2002 and Naish & Martill 2007) [adjacent image shows multiple hypothetical Baryonyx species]. Some weird coelurosaur teeth from the Galve outcrops of the El Castellar Formation look a bit like Upper Cretaceous teeth that Sankey et al. (2002) identified as avian, but that's as far as we could go.
Sauropods at Galve are represented by Aragosaurus ischiaticus, Galveosaurus herreroi and the immense Turiasaurus riodevensis [I blogged on Turiasaurus at ver 1 here. Though well behind the behemoth that is Amphicoelias fragillimus (go here for more on that taxon), it is one of the biggest known sauropods]. Aragosaurus has been identified as a camarasaurid, a titanosaur and (most recently) as a eusauropod of uncertain affinities, Galveosaurus was described as a cetiosaurid but later argued to be a basal macronarian (Barco et al. 2006), and Turiasaurus was argued by its describers to represent a new clade close to neosauropod ancestry, Turiasauria (Royo-Torres et al. 2006). The last study also concluded that Galveosaurus was a turiasaur, and not a cetiosaurid or macronarian.
Galveosaurus has taken centre stage in a debate on ethics and publishing rights. Barbara published a 2005 paper on this taxon, naming it Galveosaurus herreroi, in the journal Zootaxa (Sánchez-Hernández 2005). In the same year, a team led by José Luis Barco published a less formal article in the magazine Naturaleza Aragonesa (Barco et al. 2005), and here they named the same taxon Galvesaurus herreroi (viz, with a difference of one letter). There is a lot that could be said about this case and both sides have presented different arguments. I was totally unaware of this situation when I began working with Barbara and am keen to avoid getting involved in what appears to be an internecine debate. Working on dinosaurs is not as fun as it might be: there are apparently not enough specimens to go round [adjacent image shows Barbara's diagrammatic interpretation of the Galveosaurus remains, though I would argue that she shouldn't have used Carol Abraczinskas' silhouette of Shunosaurus as a template].
Anyway, besides these named taxa, isolated remains referred to camarasaurids, brachiosaurids and diplodocids have also been reported from the Galve area. We re-evaluated this and argued that, while some of the isolated sauropod material from Galve can't definitely be identified beyond Eusauropoda, there is material from the locality that appears to belong to a Camarasaurus-like taxon, and to a titanosaur (Sánchez-Hernández et al. 2007). The Camarasaurus-like taxon is represented only by teeth from the El Castellar Formation: they are more like the teeth of Camarasaurus than the teeth of any other sauropod, but their Hauterivian-Barremian age makes them substantially younger than Camarasaurus, and hence likely to belong to a separate taxon. Here we come again to that 'cryptic diversity' theme that I keep coming back to: the idea that the fossil record includes tons of material that surely belongs to new species, but species based on material that is generally too poor for us to name. Finally on sauropods, our article was submitted long before Rafael Royo-Torres and colleagues published Turiasaurus (Royo-Torres et al. 2006). Could it be then that some of our remains belong to turiasaurs? I have to say that it doesn't look like it.
Finally, we come to ornithischians, a dinosaur group that has pretty much been a no-show on the blog so far, but not because I don't like them. Come on, how could anyone not be interested in all those elaborately crested hadrosaurs, the stegosaurs, the ceratopsians... it's just that I've obviously been too busy with lizards, mice, passerines and sheep :) Of the Galve material that we describe, arguably the most interesting is that belonging to heterodontosaurids. These are a mostly Lower Jurassic clade, best known from southern Africa, but with an undoubted presence in the form of little Echinodon becklesii from (probably) the Berriasian Lulworth Formation of Dorset.
Conventionally argued to be basal ornithopods (predominantly because of tradition, and more recently because of two not-particularly-convincing cranial characters), several authors (including Naish & Martill 2001) have interpreted heterodontosaurids as close relatives of marginocephalians, and the recently named Yinlong downsi led Xu et al. (2006) to formally recognise a heterodontosaurid + marginocephalian clade that they named Heterodontosauriformes (skull of Yinlong shown in adjacent photo). A soon-to-be-published analysis (by ornithischian worker Richard Butler) argues instead that heterodontosaurids are right down at the base of Ornithischia, and while we're here it's worth noting that several major studies on this fascinating group are due to be published within the next few years.
Again, the Galve material isn't great: we're talking about isolated teeth. As Richard reminds me on occasion, saying that isolated ornithischian teeth are of heterodontosaurid identity is dicey given that several ornithischian clades have teeth of this sort. Even so, the Galve teeth look more like those of heterodontosaurids than those of anything else, and they're also different from the teeth of Echinodon. The Galve taxon must have been tiny, with teeth that are less than 6 mm in crown height. As we discuss in the paper, various other ornithischian teeth from Lower Cretaceous Spain also seem referable to heterodontosaurids.
So there we have it. While the material we describe is certainly not the best in the world, it is potentially significant in suggesting some new range extensions, and in hinting at some previously undocumented taxa. As is so often the case, we wait for better discoveries to confirm our identifications.
Refs - -
Barco, J. L. , Canudo, J. I. & Cuenca-Bescós, G. 2006. Descripción de las vértebras cervicales de Galvesaurus herreroi Barco, Canudo, Cuenca-Bescos & Ruiz-Omeñaca, 2005 (Dinosauria, Sauropoda) del tránsito Jurásico-Cretácico en Galve (Teruel, Aragón, España). Revista Española de Paleontología 21, 189-205.
Naish, D. 2002. Thecocoelurians, calamosaurs and Europe's largest sauropod: the latest on the Isle of Wight's dinosaurs. Dino Press 7, 85-95.
- . & Martill, D. M. 2001. Ornithopod dinosaurs. In Martill, D. M. & Naish, D. (eds) Dinosaurs of the Isle of Wight. The Palaeontological Association (London), pp. 60-132.
- . & Martill, D. M. 2007. Dinosaurs of Great Britain and the role of the Geological Society of London in their discovery: basal Dinosauria and Saurischia. Journal of the Geological Society, London 164, 493-510.
Royo-Torres, R., Cobos, A. & Alcalá, L. 2006. A giant European dinosaur and a new sauropod clade. Science 314, 1925-1927.
Ruiz-Omeñaca, J. I., Canudo, J. I. & Cuenca-Bescós, G. 1997. First evidence of baryonychid dinosaurs (Saurischia: Theropoda) in the upper Barremian (Lower Cretaceous) of Vallipon (Castellote, Teruel, Spain). Beca del Museo de Mas de las Matas 17, 201-223.
Sánchez-Hernández, B. 2005. Galveosaurus herreroi, a new sauropod dinosaur from Villar del Arzobispo Formation (Tithonian-Berriasian) of Spain. Zootaxa 1034, 1-20.
- ., Benton, M. J. & Naish, D. 2007. Dinosaurs and other fossil vertebrates from the Late Jurassic and Early Cretaceous of the Galve area, NE Spain. Palaeogeography, Palaeoclimatology, Palaeoecology 249, 180-215.
Xu, X., Forster, C. A., Clark, J. M. & Mo, J. 2006. A basal ceratopsian with transitional features from the Late Jurassic of northwestern China. Proceedings of the Royal Society of London B 273, 2135-2140.
- Log in to post comments
More like this
Er. Isn't this true by definition?
Hi Darren,
Interesting post, interesting paper. For what it's worth, I think your identification of the teeth as heterodontosaurid looks very plausible (although I have been dubious about the previously described Spanish "heterodontosaurid" material). However, you only chose to compare them with the teeth of basal ornithischians and ornithopods. From what I can make out of the photos, the teeth also compare reasonably well with basal neoceratopsians, and I would have like to see the comparison made, even if such an identity is ruled out. If they are heterodontosaurid, then they are certainly a different taxon to Echinodon. Very small heterodontosaurids aren't unusual: Echinodon and the undescribed Morrison and Kayenta taxa are all tiny as well.
I will be publishing a phylogenetic analysis positioning heterodontosaurids at the base of Ornithischia in the next two months, as part of another paper. A more detailed analysis with similar results will be published in the Journal of Systematic Palaeontology in spring 2008. While I don't think the heterodontosaurid-Marginocephalia link is well-supported, it is certainly plausible. I'm going to be examining Yinlong in a few weeks time; after that I'll be attempting to integrate my dataset with that of Xu Xing and colleagues.
A final point: the captions to figures 15A and B are mixed up - 15B is Abrictosaurus not 15A.
Rich
Bigger? Than Amphicoelias fragillimus? I hope not!
BTW, you forgot to put Royo-Torres et al. (2006) into the references section.
This post brings up a few questions. First, Martill et al. (1996) consider Suchomimus to be synonymous with Baryonyx (as does Hutt & Newberry 2004) but that the former should keep its specific name due to features of the vertebrae and locality differences. I'm all in favor of this reclassification, actually, but was wondering what all you folks think of it.
Also, the status of the Marginocephalia seems to be ever-changing. When Yinlong was found, it not only strengthened the monophyly of that group but also linked them firmly to heterodontosaurs. But then Dracorex was discovered and Bob Bakker claimed that it actually weakened the Marginocephalia. And just recently Robert Sullivan published a review of the Pachycephalosauridae in which he doubts the connection to ceratopsids.
And NOW you're telling me that heterodontosaurs might be booted down to the very base of the Ornithischia? I didn't realize that ornithischian phylogeny was so volatile.
In your review of English dinosaurs you state that Becklespinax might be an allosauroid. What are its diagnostic features indicative of this proposed affinity as against the old spinosaurid claim? The inflection in the length of the preserved neural spines also appears rather drastic -- but is there really a sign of injury as you show in your interpretive figure?
All in all that Brit dinosaurs paper was very useful. As a kid I grew up in the US and was captivated by the rich dinofauna of North America. But I relocated to spend the next several years of youth in a place where most dinosaur books were from England, and learned of these new names like Zanclodon and Metriacanthosaurus. More recently I have been wanting to catch up with the fate of these tantalizing fragments condemned to nomina dubia. This paper gave a one stop answer.
Hey, Darren. I'm not new here, just new at commenting. I have to say that you're doing an excellent job of providing some information that would be hard to find otherwise. do you think you find the time in your busy schedule to post on borophagine canids? it would really be a good thing, as really good information on things like their anatomy is hard to come by. Thanks. Tayo
What I notice about "Carol Abraczinskas'[s] silhouette" is front toes. Is it that, or something else wrong with using it?
Thanks to all for their interesting, informative and helpful comments. Here are brief responses to everything above that needs responding to. Thanks Richard for the update on your work: we all look forward to seeing it published. Thanks for the correction WRT to the heterodontosaur figure (dammit, another typo that was missed in the proofs).
Zach: whether Suchomimus and Baryonyx warrant synonimization is of course subjective, as (to my knowledge) everyone agrees that they are separate species. Following data presented by Angela Milner (but still unpublished by her), I and Dave Martill (in Dinosaurs of the Isle of Wight, 2001) advocated synonimization (Martill et al. and Hutt & Newberry should have cited this; I can't recall if they did though). Most of the differences purported to differentiate the two are erroneous or minor, and they are certainly 'alike enough' to be within the same 'genus'.
As for Marginocephalia and the issues you raise regarding pachycephalosaurs, the monophyly of marginocephalians is not really in doubt and this is a well supported clade (I wouldn't worry about Bakker and Sullivan's claims: they did not take account of most of the character data we now have). Maybe the phylogenetic position of heterodontosaurs will change, but I suspect this is mostly because of insufficient study, not because the relationships among Ornithischia as a whole are volatile. Cue Butler...
Rajita: glad you like the British dinosaurs review, part II (on ornithischians) is due to appear soon. I could talk for ages about Becklespinax and will leave that for another time I'm afraid - I am about to submit a manuscript on that taxon and will discuss it on the blog in the future.
Tayo: I collected a ton of stuff on borophagines from Xiaoming Wang a few years back and am interested in the diversity they evolved. Little known is that many (most?) of them were fox-like generalists, and they even included procyonid-like omnivores. I would like to blog on them one day.
Nathan: Shunosaurus was a poor choice, as there is no indication that this sauropod (it's a basal eusauropod) is particularly close to Galveosaurus (note that the tail club had to be removed). Furthermore, Abraczinskas' silhouette shows the animal as semi-plantigrade; correct me if I'm wrong, but this isn't favoured for any sauropodomorph any more (I think that Jeff Wilson favoured the idea of plantigrady in basal sauropods when that restoration was produced).
Boreophagines - now that is something to look forward to. Dinosaurs are wonderful I admit (I have never gotten over my childhood obsession with Triceratops - I actually prayed for God to bring them back from extinction, and I treasured a picture I had in a book of a Tyranosaurus meeting a grisly end on the horns of my beloved Triceratops), but to me the Mesozoic was a dry run for what was to come. I think that the Miocene was the 'high summer' of life on earth, apparently unequalled in terms of diversity of flora and fauna. I am particularly fascinated by the groups that didn't make to the present - Amphicyonids, boreophagine & hesperocyonine canids, nimravids, creodonts, hemicyonines etc, and thats just some of the placental carnivorous forms - I'll stop now before I really get carried away (possibly drooling in a straight jacket!). But if you were to blog a little more on the cenozoic it would go down very well with me! :)
Zach,
For the most part, ornithischian phylogeny has proved relatively stable at the global level, despite a recent renewal of interest and new work by people such as Xu Xing, Cathy Forster, and myself. For example, although a few (notably Bob Bakker and Robert Sullivan) disagree, there has been no serious published threat to the monophyly of Marginocephalia (i.e. no-one has identified a credible alternative using a numerical analysis), and most workers accept its validity. However, there are a few major problematic areas, and the position of heterodontosaurids is the most controversial. Few (with the possible exception of Paul Sereno) favour the traditional position as basal ornithopods - the character data supporting this idea is weak. Most popular is the idea that heterodontosaurids and Marginocephalia are sister groups, and this has been supported by Xu Xing, Cathy Forster, Hailu You and Darren, amongst others. This idea has a lot going for it, but needs testing with more data. Finally, my analyses suggest that heterodontosaurids may be non-genasaurians and amongst the most basal known ornithischians - this fits with their early stratigraphic position (records from the Triassic) and their retention of numerous very plesiomorphic features. The reason for such disagreement is two-fold: heterodontosaurids have a very weird and difficult to interpret combination of features, and heterodontosaurids simply haven't been studied enough to date. We must hope that future large-scale phylogenetic analyses incorporating more character and taxon data will help with the first problem; as for the second, there are a large number of people working on heterodontosaurids right now (including Peter Galton, Paul Sereno, Laura Porro, David Norman, Paul Barrett and myself), so hopefully our knowledge of the group should improve greatly in the next few years.
Interesting about roe-deer feeding in open fields being a new behaviour in Britain, since in Sweden it has been common for decades. As a matter of fact I once counted 69 roe-deer in sight simultaneously on the intensively cultivated plain north of Lake Tåkern in Östergötland. This was at a time when the fox population was at an all-time low due to disease, they are not nearly so common now. Incidentally Fallow Deer also feed in open fields here, but not Red Deer as far as I know.
I've been taught that roes are not really forest animals, only retreating to forests when there are too many people around, and indeed I've seen more outside than inside forests. (Never far from a forest, though.)
Hi Darren,
Interesting to know you may have other Lower Cretaceous heterodontosaurids in Europe: a relief that Echinodon is not the only one! (have always been a bit worried about making that call, but it would ease my mind greatly if some Jurassic ones showed up to fill in the gap). The idea of marginocephalian relationships for heterodontosaurids is not a new one - it was mentioned in some of the very early cladistic studies of ornithischian phylogeny by Cooper, Maryanksa & Osmolska and possibly others.
Sorry, but a bit of a moan. Why keep calling robust spatulate sauropod teeth 'Camarasaurus-like'? I know you're not alone in this - lots of people continue to do it - but it's essentially meaningless. You might as well call them 'Omeisaurus'-like, 'Jobaria'-like or pick any other of a suite of non-diplodocoid non-titanosauriform eusauropods. However, once you even mention that such teeth are 'Camarasaurus-like' this implies a relationship with Camarasaurus, which other less discerning people then go on to build into all sorts of castles-in-the-air regarding lineage ranges and palaeobiogeographic scenarios. Similarly, no taxon (or bit of a taxon, whether it's a tooth or anything else) should be called 'Camarasauridae indet.'. No recent phylogenetic analysis of sauropods recovers a monophyletic clade of taxa more closely related to Camarasaurus than to any other taxon (Camarasauridae is essentially a monogeneric family at the moment, quite different from the classifications of even 15 years ago). Of course, I realise that this could change with the discovery of new material and further analyses: but for the moment use of the term is not useful.
Apologies for such a pedantic post, but this is a bug-bear of mine: it's frustrating trying to stop people continuing with this practice, which has its roots in pre-cladistic studies of sauropod relationships.
Cheers, Paul
Thanks for your comment Paul, much appreciated. It might not be obvious to many dinosaur aficionados that the idea of heterodontosaurids being allied to marginocephalians is almost as old as the better known/more standard idea that they are basal ornithopods, which is frustrating given how weak the hypothesized ornithopod affinity has always been.
On the 'camarasaurid' teeth, I certainly agree that classifying teeth on the basis of what is apparently a plesiomorphic morphology (viz, robust and spatulate) should be avoided. However, the Galve teeth were not identified as 'Camarasaurus-like' on this basis. Here is the relevant section of text, from Sánchez-Hernández et al. (2007), pp. 198-199...
... though having said that I'd be interested to know if you agree!
And yeah, no phylogenetic analysis of Sauropoda has yet recovered a clade that includes Camarasaurus and any other taxon more closely related to it than to anything else. From the standpoint of current phylogenetic nomenclature, Camarasauridae doesn't exist because it hasn't been defined. But whether that means anything depends entirely on your point of view, and given that there is (excluding the Galve material) at least one taxon that has been identified as being closer to Camarasaurus than to any other sauropod (the unnamed Khadir Island taxon described by Moser et al. 2003), the term is - arguably - useful.
Ref - -
Moser, M., Mathur, U. B., Fürsich, F. T., Pandey, D. K. & Mathur, N. 2006. Oldest camarasauromorph sauropod (Dinosauria) disovered in the Middle Jurasic (Bajocian) of the Khadir Island, Kachchh, western India. Paläontologische Zeitschrift 80, 34-51.
Hi Darren,
I haven't had a chance to take a look at the new paper yet, but congratulations ahead of time. I saw the comments on silhouettes and choice of foot posture and wanted to add my thoughts. The silhouette that Sánchez-Hernández used in her 2005 Galveosaurus paper is from Wilson & Sereno (1998:foldout). She didn't acknowledge that she used our silhouette, although it is obviously recognizable if you were able to pin it down to Shunosaurus and to Carol Abraczinskas. It may seem minor, but figures (even parts of them) should be credited in publications and in public lectures - not only to recognize the authors but also to inform the reader where the information came from.
Returning to the Shunosaurus reconstruction....although Carol Abraczinskas did a wonderful job on that silhouette, as she did on the others from that paper and everything else she touches, the credit is not hers for the image. Paul Sereno and I designed and drafted all the silhouette reconstructions and the skull reconstructions...credit for getting things wrong or right should be ours.
If I were to do a reconstruction of Shunosaurus today, I would elevate the manus into a more digitigrade pose, although not as elevated or as tightly recurved proximally and distally as in later sauropods. I would leave the pes more or less as it is. It looks flat-footed (i.e., plantigrade) because footprints tell us that the pes was supported by a heel pad, which we decided to include in the silhouette. You will notice in the original skeletal reconstruction that the metatarsus is not arranged in a plantigrade posture; it is elevated into a 'semi-digitigrade', 'sub-unguligrade' or 'semi-plantigrade' pose - depending on one's interpretation of terms and posture (I discussed this a bit in my 2005 Paleobiology paper). In my view, this pedal posture was retained by all later sauropods.
JAW
Many thanks for that Jeff, your thoughts are appreciated. On the way the hands and feet of your Shunosaurus are depicted, I made a mistake: as you state, semi-plantigrady is universally regarded as the correct pedal morphology for eusauropods. Sorry for confusing things (and, yes, I do feel silly for making such an amazing mistake [reference]).
Hi Darren, thanks for the clarification on your identification of the sauropod teeth. However, all of the features that you mention as close to Camarasaurus are also present in other sauropods, including Omeisaurus, for example. None of the features, or the combination of features proposed, is synapomorphic of a restricted Camarasauridae. As for use of Camarasauridae, when a phylogeny appears that identifies such a clade, fine, I'll use it. Until such time, however, the term should be avoided to prevent misleading people over the affinities of material that is effectively indeterminate on the basis of current knowledge of character distributions.
Cheers, Paul
Ok, fair enough. Thanks for commenting.
Regarding putative "camarasaurids": as well as the various species of Camarasaurus itself and Moser et al.'s putative Indian camarasaur, Peng et al. (2005) considered both Datousaurus and their new genus Dashanpusaurus to be camarasaurs. But Moser et al.'s (2006, not 2003) referral of their specimen is based only on a partial fibula, and since Peng et al's paper is mostly in Chinese, I can't tell what the basis of their referrals is. So it seems that there are a fair few candidate camarasaurs out there, but nothing very solid.
I'd like to see what would happen if each Camarasaurus species was coded as a separate OTU. Does anyone know if someone's working on that?
PENG, G., YE, Y., GAO, Y., SHU, C. and JIANG, S. 2005. Jurassic dinosaur faunas in Zigong. Sichuan People's Publishing House, Zigong, 236 pp.
Bakker and Sullivan's views regarding _Dracorex_ and the monophyly of Marginocephalia are subject to serious question. Their suggested relationships of pachycephalosaur taxa have not held up so far when tested in the matrices of Williamson and Carr (see this).
Nor that of Sullivan's own (see this).
Perhaps it's possible that outgroup selection in both cases has an effect on the resulting ingroup topology. It would be good to see an adequately extensive and inclusive work that examined the position of pachycephalosaurs (and _Stenopelix_ as well) with respect to other ornithischians. TMK, there has never been a more viable alternative put forth, in spite of Sullivan's murmuring towards Coombs' proposed ankylosaur-pachycephalosaur relationship in the late 70s.
I must say I am more than pleased though to read of the possibility of more studies being put forth into Ornithischia though. :)
An ankylosaur-pachycephalosaur group? You've gotta be kidding me!
Coombs did not regard _Scelidosaurus_ as a basal ankylosaur, or even as a relative of ankylosaurs at all (Coombs, 1978; 1979), but instead believed it was a basal ornithopod Thulborn's assessment of it (Thulborn, 1977) since he felt it lacked the derived characters present in ankylosaur pelvis. It is for this same reason that Coombs rejects a relationship between ankylosaurs and stegosaurs as well, by ignoring any shared characters, and focusing solely on the lack of ankylosaur-like characters in their pelves. Coombs advocated the ankylosaur-pachycephalosaur clade on the basis of a few pelvic characters (pubis at least partially excluded from participation in the acetabulum; preacetabular process of ilia horizontally oriented and overlying the posterior dorsal ribs; an apparent lack of an ilial "antitrochanter"; no ischial obturator process) and some cranial ones (armored skull roof; a "tendency to" closed supratemporal fenestra; closure of interorbital space). But most of these are either plesiomorphic (such as the lack of the ilial "antitrochanter"), are not present in basal pachycephalosaurs or basal ankylosaurs (such as the closure of the supratemporal fenestra), or are false (pachycephalosaur skulls are thickened, but are not covered in armor plating as in ankylosaurs). Sullivan, however, does not address most of these, and only states that he finds this "more compelling" (p. 349 of Sullivan, 2006). In Sullivan's defense, he does argue that _Stenopelix_ is not a basal pachycephalosaur and that "flat-headed" pachycephalosaurs are actually derived, and therefore their open supratemporal fenestrae are a reversal from what he believes is the plesiomorphic condition. The former's position has not been tested, and it's possible outgroup selection in existing pachycephalosaur analyses has had an effect on the position of "flat-headed" pachycephalosaurs, so frankly, this could be easily resolved with a large scale analysis of the Ornithischia. I am doubtful however that the relationships supported by Sullivan or Coombs will hold up in light of one. :)
Coombs, W.P. 1978. The families of the ornithischian dinosaur order Ankylosauria. Journal of Paleontology, 21:143-170.
Coombs, W.P. 1979. Osteology and Myology of the Hindlimb in the Ankylosauria (Reptilia, Ornithischia). Journal of Paleontology 53 (3): 666-684.
Sullivan, R.M. 2006. A taxonomic review of the Pachycephalosauridae (Dinosauria: Ornithischia). New Mexico Museum of Natural History and Science Bulletin 35:347-365. (available for free as a PDF from http://www.robertmsullivanphd.com)
Thulborn, R.A. 1977. Relationships of the Lower Jurassic dinosaur Scelidosaurus harrisonii. Journal of Paleontology 51(4):725-739
Nick,
My coauthors and I do have a large-scale (about 50 taxa, 220 characters) ornithischian analysis in press at Journal of Systematic Palaeontology which I hope makes a reasonable stab at testing the position of pachycephalosaurs (as well as other ornithischian groups) - we find good support for Marginocephalia, and Stenopelix as a basal pachycephalosaur, although the support for this latter result is weak (and, having seen the specimen first-hand I am far from convinced of its pachycephalosaur affinities - more to come at a later date). Although I understand Bob Sullivan's concerns, there is no credible alternative to Marginocephalia at the moment (the character evidence for an ankylosaur-pachycephalosaur clade is extremely weak), and I for one find the support for the clade reasonably compelling.