When Flat-Head Met Dome-y

One of the most interesting aspects of ornithischian phylogeny has been (alongside whether heterodontosaurs are ornithopods or whatever) is the nature of many pachycephalosaur taxa. Recently, Schott et al. (2003) have done some great detective work in nailing down what they expect to be a firm argument for expected ontogeny in pachycephalosaurs.

A long-standing debate in the literature has been what, precisely, to do with AMNH FR 5450. This specimen, originally noted by Galton (1971) while discussing the (likely not a pachycephalosaur) Yaverlandia bitholus, is a peculiar “flat-headed” frontoparietal fragment of the skull, with a division between the parietal and the frontal forming two, slightly domed prominences. Galton, however, referred the dome as a probably female form of Stegoceras validum, arguing that relative doming was a sexual dimorphism, and present in the material. It helped Galton’s argument that the specimen he described and the type (lectotype CMN 515) and some of the best known specimens (e.g., complete skull UALVP 2) of Stegoceras validum were each from the Dinosaur Park Formation (Campanian) of Alberta, Canada. But the morphology is peculiar, and rather than be satisfied at a potentially younger or female Stegoceras validum specimen, Wall and Galton (1979) described AMNH FR 5450 as Stegoceras browni (in honor of Barnum Brown, who collected the specimen from Alberta’s Badlands for the American Museum of Natural History).

AMNH FR 5450, holotype of Ornatotholus browni (Wall and Galton, 1979), referred to Stegoceras validum (Galton, 1971, following Schott et al., 2011). Frontoparietal in A) dorsal, B) left lateral, and C) posterior views (left side) and in coronal, sagittal and transverse sections (right side) using high-resolution computed tomography (HRCT).

Still not satisfied, Galton and Sues (1983) then separated browni from a complex of apparent Stegoceras-like species as Ornatotholus browni, entrenching the idea then that the material was neither juvenile-enough or female-enough to prohibit taxonomy to its utmost degree (though I suspect had they chosen to separate it further with “Ornatotholidae,” that might be pushing it). However, going back to Galton (1971) proved attractive as more and more material indicated many of the supposed species of Stegoceras were, in fact, Stegoceras validum. While some of this material was known from elsewhere (further south into Montana), and across time (from the older Oldman Formation), it implied a largely homogenous group of taxa that could very well be one species, and that would be the oldest affirmed one, Stegoceras validum. Goodwin et al. (1998) felt strongly enough about this argument, using further “flat-headed” specimens, that they argued that AMNH FR 5450 should be referred to Stegoceras validum (where it started, no less).

Williamson and Carr (2002) followed Goodwin et al. (1998) in affirming the ontogeny of Stegoceras validum and, using a relatively broad sample of material from the Belly River Group (=Judith River Group in America, including the Foremost, Oldman and Dinosaur Park Formations in Alberta and Saskatchewan in Canada and the Judith River Formation in Montana, USA, a range of time representing around 6ma), referred AMNH FR 5450 to Stegoceras validum, thus sinking browni as a synonym of validum (we’ll get to this part in a bit, and my particular interest here). Sullivan (2003) followed suit. Dissatisfied with the methodologies used, as they were tenuous and based on general principles of ontogeny but no affirmative data. These issues are, rather, prone to the complex into which you measure your material against. Schott et al. (2011) instead chose to use detailed high-resolution computed tomography (or HRCT, a non-invasive technique imaging inside the domes of pachycephalosaurs, supplementing — or supplanting — the invasive and destructive sectioning technique used previously).

Skulls of specimens referred to Stegoceras validum, arranged by relative size (specimen numbers within the figure). In dorsal (above) and right lateral (below) views. Specimens are all to scale.

This material thus becomes a strong baseline from which to discuss relative ontogeny in pachycephalosaurids, a subject that has come up recently, especially as some of the arguments made in the paper would seek to reference this. For example, in discussing the relative synonymy of Dracorex hogwartsi and Stygimoloch spinifer with Pachycephalosaurus wyomingensis, Goodwin and Horner (2009) used the relative size and vascularity of the domes of these three Hell Creek Formation pachycephalosaurids to argue that they were, in fact, a single taxon (the third), with the “juvenile” form being Dracorex hogwarsti, argued previously by the authors to be adult due to neurocentral fusion and its cranial ornamentation (Bakker et al., 2006). This was despite contradiction by Williamson and Carr (2002), who affirmed Goodwin et al. (1998) in distinguishing these taxa on the basis of relative horn size and dome shape: In Stygimoloch spinifer, the corner nodes are elongated into horns and are very large, and the dome is very narrow, despite bearing open sutures between left and right halves, while in Pachycephalosaurus wyomingensis the corner nodes are very short and sometimes rounded with the skull at larger size, and the dome is broad and rounded instead of bearing steepled sides, and the midline suture is obliterated even when domes of smaller size are present.

I discuss this here, where I distinguished the growth series of Horner and Goodwin from that proposed for Stegoceras validum (using Williamson and Carr, 2002 and Sullivan, 2003). One would think then that the baseline would be useful, but instead assertions were made to invoke extreme transformation using a process known as metaplasia whose effects were nonetheless not fully known. Were metaplasia involved, the argument would go, resorption can result in a complete reduction in form of the peripheral nodes from long spikes to small rounded knobs, and thus result in tall distinct domes transitioning to broad, low domes. But this trend is not apparent in Stegoceras validum, and Schott et al. (2011) do not touch on this inconsistency in growth changes. Perhaps only Pachycephalosaurus wyomingensis suffered this issue, but this is not precisely clear.

Quo vadis, Stegoceras?

But there is a further issue, and that issue is actually Stegoceras validum. Or rather, what exactly Stegoceras validum is.

Gilmore (1924)'s illustration of the partial skeleton of Stegoceras validum, with the skull of UALVP 2. Stolen from Scott Hartman's EXCELLENT discussion of the history of the "outline" form of skeletal reconstructions.

I might be called for “reaching” for this argument, but I think I’m on fairly secure ground (but of course, I would think so as I am biased). Implicitly, despite cladistic analysis on several fronts, “stegoceratine” systematics is touchy. Initially, there were multiple species: Stegoceras validum, lambei, sternbergi, edmontonense (=edmontonensis), breve (=brevis)[n1]. Sullivan (2003) separated lambei and sternbergi into new “genera,” Colepiocephale and Hanssuesia, respectively; breve and edmontonense were sent over to Prenocephale as a representatives of that otherwise Mongolian taxon in North America, forming Prenocephale brevis and edmontonensis, respectively; and that left poor validum to itself. The primary problem here is that most of these taxa are separated in time: edmontonense is known only from the Horseshoe Canyon Formation of Alberta (early Maastrichtian), while lambei is known only from the Foremost Formation of Alberta (middle Campanian); validum, edmontonense, sternbergi and breve were all known from the Oldman and the Dinosaur Park Formations, while some referred specimens were noted from other formations roughly equivalent to these levels. Yet other taxa in the Dinosaur Park and Oldman formations show temporal replacement regimes, indicating successive replacement of species.

This is evident in taxa from the Belly River Group such as the chasmosaurines where Chasmosaurus russelli is succeeded by Chasmosaurus belli, which is in turn succeeded by Vagaceratops irvinensis (formerly Chasmosaurus irvinensis) and Mojoceratops perifania (potentially synonymous with Chasmosaurus russelli); lambeosaurines, where Corythosaurus casuarius is replaced by Lambeosaurus lambei, which is replaced by Hypacrosaurus sternbergi, which is in turn replaced by Lambeosaurus magnicristatus; and centrosaurines, where Centrosaurus brinkmani (it of the “popcorn” frill) is followed by Centrosaurus apertus, in turn replaced by Styracosaurus albertensis. These indicate relative faunal segregation as well, where the appearance of Lambeosaurus and Styracosaurus calls to an end of Corythosaurus and Centrosaurus (sensu lato, in the broad sense) (Arbour et al., 2009). while this succession is evident in many larger taxa, the sampling of smaller taxa is more difficult, in part owing to the difficulty in cataloguing many of the earlier collected material (including the types of most of the smaller theropods in the formations) but also the tendency to ignore potential faunal succession within the formations when assessing the location of many of the smaller fragments. As such, the stratigraphic range of the collected Stegoceras domes are known, but their precise placement in regard to one another hasn’t been touched on. This might be indicative of broadly sampling for taxonomic inclusion, or overextending the similarity of potentially related but distinct taxa. As such, there may be many more taxa present in the sample referred to Stegoceras validum, which spans material collected from the Foremost through Dinosaur Park Formations like few other dinosaurs are known from.

If Stegoceras validum may be over-sampled, exactly what does this say of the integrity of the ontogenetic series? This is the same argument I made when dealing with Triceratops as a supposed container to Torosaurus. Virtually all frontoparietal domes from the Belly River Group have been referred to Stegoceras, either as Stegoceras sp., validum, or to one of the other species that co-occur with it (such as breve and sternbergi). This may even imply that many of the specimens are intermixed among the taxa, that the taxa are over-split, or that the taxa are under-split (which itself implies that the features used to support the taxa by Sullivan [2003] are sampling broader taxa, or are suggestive that more features of the skeleton may be useful for sampling). Reliance of frontoparietal domes as exemplary of diagnostic features is a product itself of biased sampling: As there are very few other remains, an over-reliance on this material as indicative is potentially or actually problematic, an aspect that also plague tooth-based taxonomy (TBT).

Not Always a Juvenile…

MPC-D 100/1201 (formerly, GI SPS 100/51), holotype of Homalocephale calathocercos Maryańska and Osmólska (1974).

Occasionally, one of these papers takes a stab at Homalocephale calathocercos (Maryańska and Osmólska, 1974), which is based on a partial skull missing the snout (shown above, with a portion of the broken snout left out due to incompleteness and lack of superficial preservation) and skeleton. As summarized in Evans et al. (2011), authors have generally agreed that Homalocephale calathocercos is unique, although in how they treat it among other pachycephalosaurs differs. The recent argument (e.g., Bakker et al., 2006; Butler and Sullivan, 2009; Sullivan, 2006, and implied in two abstracts which portend forthcoming work on the subject of ontogeny; Longrich et al., 2010, in naming a new pachycephalosaur, Texacephale langstoni) is that it represents a juvenile pachycephalosaur, in which case it may be synonymous with another taxon. This is supported not only by open neurocentral sutures, but also the lack of a dome (as in the series shown above), open frontal-parietal suture, and only minimal thickening of the frontals and parietals.

Longrich et al. (2010) argue that Homalocephale calathocercos may be a likely juvenile of Prenocephale prenes:

Homalocephale and Goyocephale also exhibit a laterally deflected and paddle-shaped preacetabular process of the ilium, and a ‘crown’ formed by the lateral expansion of the squamosal and postorbital; these features are shared with Prenocephale (Maryańska and Osmólska, 1974) but are absent from Stegoceras (Gilmore, 1924). Given this, and that Homalocephale and Prenocephale are remarkably similar in the construction of the palate, occiput, and pelvis, it seems almost certain that Homalocephale is a juvenile or a female of Prenocephale, which occurs in the same formation.

[Longrich et al., 2010:pg.281. All citations in quotes included below.]

This is unusual as Longrich et al. (2010) included a cladistic analysis whose results differ substantially from that of previous authors, including Sereno (2000), Williamson & Carr (2002), Sullivan (2003) and Maryańska et al. (2004) in placing Homalocephale calathocercos, Goyocephale lattimorei and Wannanosaurus yansiensis into a small nested group close to the “crown” of Pachycephalosauridae, which along with Prenocephale prenes and Tylocephale gilmorei formed the sister taxon to “Pachycephalosaurinae” (Pachycephalosaurus wyomingensis, Dracorex hogwartsi, and Stygimoloch spinifer + Alaskacephale gangloffi + Sphaerotholus sensu lato). These results are surprising because typically, and as argued by Schott et al. (2011), juveniles tend to group towards the base of the tree, rather than the crown, and the same pattern has been observed for other taxa, especially tyrannosaurids (see Carr, 1999). So recovering these “juvenile” taxa towards the crown may actually represent an odd reversal of the argument, without refuting it. For example, Goodwin and Horner (2009) argued that more adult-seeming features are present in Pachycephalosaurus generally, lost as it ages and forms a complete dome. Of course, it may also represent a problem in the selection of supposed juvenile features as equivalent to adult, or a rejection of the argument that these taxa are actually juveniles.

Based on a restudy of the skull of Homalocephale calathocercos, Evans et al. (2011) specifically treated the varying and odd features and agreed that yes, it was a juvenile. But they differed from Longrich et al. by concluding:

The degree of suture closure in the skull and vertebral neurocentra suggests an immature status for the holotype specimen of Homalocephale calathocercos. Therefore, the degree of cranial thickening or doming at maturity remains unknown in Homalocephale, and the hypothesis that Homalocephale is pedomorphic in the development of the skull roof cannot be either supported or refuted at this time. The possible juvenile nature of MPC-D 100/1201 makes assessment of its phylogenetic position more difficult. Although intraspecifically variable, the parietosquamosal ornamentation has an influential role in defining the structure of pachycephalosaurid interrelationships and taxonomy. Several aspects of cranial ornamentation described here for Homalocephale, including the enlarged medial nodes and the presence of two prominent lateroventral corner nodes, are likely important for assessing its systematic position. These characters can now be more easily compared to other pachycephalosaurs and incorporated into future phylogenetic analyses of the clade. The unique pattern of parietosquamosal ornamentation together with differences in the maxilla and dentition between H. calathocercos and P. prenes suggest that synonymy of these taxa may be premature, and we tentatively reaffirm the presence of at least two pachycephalosaur species in the Nemegt Formation[.]

[Evans et al., 2011:pg.91.]

So despite it’s seeming juvenility, it may still be basal, or may be basal and undomed, or (in a strange coincidence) affirm Bakker et al. in arguing a derived clade of domeless taxa existed. The data is nonetheless relatively inconclusive.

Synthesis

The bulk of this post has been to cover what people are talking about when developing a means of assessing whether various named taxa are “real,” or whether they are synonyms of some other forms. Unfortunately, it has gotten into some people that arguing a taxon is a juvenile means it is not valid, regardless of whether the specimen at hand has a unique, hithertofore unseen feature that can be diagnostic. It also seems problematic that some authors disregard the potential that ontogenetic features in and other themselves can be diagnostic even when lost in later stages. There is, for example, no other pachycephalosaur quite like Stygimoloch spinifer. I will go into more depth on this argument later.

Nonetheless, the recognition of a general series of Stegoceras validum through ontogeny progresses from a Homalocephale-like stage (UCMZ(VP) 2008.001, in the series above) towards a decidedly Pachycephalosaurus-stage (ROM 53555) has problems, in that it is largely artificial rather than recognized. These specimens derive from from isolated pieces of skull, without the detail of potential variation in closely related taxa that might be revealed should they undergo explicit and detailed examination with respect to their stratigraphy. If faunal replacement occurs in the Belly River Group, and pachycephalosaurs are segregated among them, or split ecologically, how precisely will cranial doming distinguish these taxa? These questions should be answered before the taxa become lumped together or split apart. This is a conservative position, but one I think we should allow ourselves to be beholden to.

[n1] Due to a unique rule in Linnaean Systematics, as overseen by the ICZN, specific and generic names when formed together need to agree in gender. When the specific name is created or is transferred from one taxon to another, the author is required to change the name to agree in gender with the genus, if necessary. While this rule was discarded in the 4th edition, it was in place for the earlier editions during which the taxa named were transferred about (formerly in the now-theropodan “genus” Troodon, validum was originally named validus by Gilmore (1924): transfer to a new “genus” Stegoceras required the otherwise masculine validus to become validum, agreeing with Stegoceras‘ feminine name. Personally, while this is correct Latin (and Greek), I think it’s a silly joke when each change is typically recorded in a synonymy list and looks like multiple taxa named validus or validum were coined. This is one reason, I think, the rule was dropped, although I am not in the particulars. As such, validus becomes validum, edmontonense became edmontonensis, brevis became breve, then back again. My “originalist” credo is fairly simple: If we don’t need to change the name for a functional reason (like a second species ending up close to a different genus) then we shouldn’t, and that includes changing the gender of the name in question.

Arbour, V. M., Burns, M. E. & Sissons, R. L. 2009. A redescription of the ankylosaurid dinosaur Dyoplosaurus acutosquameus Parks, 1924 (Ornithischia: Ankylosauria) and a revision of the genus. Journal of Vertebrate Paleontology 29(4):1117-1135.
Bakker, R. T., Sullivan, R. M., Porter, V., Larson, P. & Saulsbury, S. J. 2006. Dracorex hogwartsia, n. gen., n. sp., a spiked, flatheaded pachycephalosaurid dinosaur from the Upper Cretaceous Hell Creek Formation of South Dakota. New Mexico Museum of Natural History and Science Bulletin 35:331–345.
Butler, R. J. & Sullivan, R. M. 2009. The phylogenetic position of the ornithischian dinosaur Stenopelix valdensis from the Lower Cretaceous of Germany and the early fossil record of Pachycephalosauria. Acta Palaeontologica Polonica 54:21–34.
Carr, T. D. 1999. Craniofacial ontogeny in Tyrannosauridae (Dinosauria, Theropoda). Journal of Vertebrate Paleontology 19(3):497–520.
Evans, D. C., Brown, C. M., Ryan, M. J. & Tsogtbaatar K. 2011. Cranial ornamentation and ontogenetic status of Homalocephale calathocercos (Ornithischia: Pachycephalosauria) from the Nemegt Formation, Mongolia. Journal of Vertebrate Paleontology 31(1):84-92.
Galton, P. M. 1971. A primitive dome-headed dinosaur (Ornithischia: Pachycephalosauridae) from the Lower Cretaceous of England and the function of the dome of pachycephalosauris. Journal of Paleontology 45(1):40-47.
Galton, P. M. & Sues, H.-D. 1983. New data on pachycephalosaurid dinosaurs (Reptilia: Ornithischia) from North America. Canadian Journal of Earth Sciences — Revue canadienne de sciences de la Terre 20(5):462-472.
Gilmore, C. W. 1924. On Troodon validus, an orthopodous dinosaur from the Belly River Cretaceous of Alberta. Alberta University Bulletin 1:1–43.
Goodwin, M. B., Buchholtz, E. A. & Johnson, R. E. 1998. Cranial anatomy and diagnosis of Stygimoloch spinifer (Ornithischia: Pachycephalosauria) with comments on cranial display structures in agonistic behavior. Journal of Vertebrate Paleontology 18(2):363-375.
Horner, J. R. & Goodwin, M. B. 2009. Extreme cranial ontogeny in the Upper Cretaceous dinosaur Pachycephalosaurus. PLoS One 4(10):e7626.
Longrich, N. R., Sankey, J. & Tanke, D. 2010. Texacephale langstoni, a new genus of pachycephalosaurid (Dinosauria: Ornithischia) from the upper Campanian Aguja Formation, southern Texas, USA. Cretaceous Research 31:274–284.
Maryańska, T., Chapman, R. E. & Weishampel, D. B. 2004. Pachycephalosauria. pp.464–477 in Weishampel, Dodson & Osmólska (eds.) The Dinosauria, second edition. University of California Press (Berkeley).
Maryańska, T. & Osmólska, H. 1974. Pachycephalosauria, a new suborder of ornithischian dinosaurs. Paleontologica Polonica 26:133–182.
Sampson, S. D., Loewen, M. A., Farke, A. A., Roberts, E. M., Forster, C. A., Smith, J. A. & Titus, A. L. 2010. New horned dinosaurs from Utah provide evidence for intracontinental dinosaur endemism. PLoS ONE 5(9):e12292.
Schott, R. K., Evans, D. c., Goodwin, M. B., Horner, J. R., Brown, M. & Longrich, N. R. 2011. Cranial ontogeny in Stegoceras validum (Dinosauria: Pachycephalosauria): A quantitative model of pachycephalosaur dome growth and variation. PLoS One 6(6):e21092.
Sereno, P. C. 2000. The fossil record, systematics, and evolution of pachycephalosaurs and ceratopsians from Asia. pp.480-516 in Benton, Shishkin, Unwin & Kurochkin (eds.) The Age of Dinosaurs in Russia and Mongolia. Cambridge University Press (Cambridge).
Sullivan, R. M. 2003. Revision of the dinosaur Stegoceras Lambe (Ornithischia, Pachycephalosauridae). Journal of Vertebrate Paleontology 23(1):181-207.
Sullivan, R. M. 2006. A taxonomic review of the Pachycephalosauridae (Dinosauria: Ornithischia). New Mexico Museum of Natural History and Science Bulletin 35:347–365.
Wall, W. P. & Galton, P. M. 1979. Notes on pachycephalosaurid dinosaurs (Reptilia: Ornithischia) from North America, with comments on their status as ornithopods. Canadian Journal of Earth Sciences — Revue canadienne de sciences de la Terre 16(11):1176-1186.
Williamson, T. E. & Carr, T. D. 2002. A new genus of derived pachycephalosaurian from western North America. Journal of Vertebrate Paleontology 22(4):779-801.