- Some time back (about a year ago, in fact), I wrote about the hypothesis where Heterodontosaurus tucki, a small ornithischian from the Early Jurassic of southern Africa, might actually have been a basal marginocephalian, joining the ranks of blatant look-a-likes like Goyocephale lattimorei and Yinlong downsi. This hypothesis is covered in some level there, so I won’t go into it that much here. But as the title of this post might suggest, there was a rebuttal.Heterodontosaurids received several new pieces of publication technology, expanding the level of data with which we might assess them.
First, there was the description of Manidens condorensis (Pol et al., 2011), from the Cañadon Asfalto Formation of Chubut Province, Argentina, which is Middle to Late Jurassic in age. This was based on a partial skeleton with skull from South America, relatively close in its original time to southern Africa, although much later than those heterodontosaurids from the Stormberg Series. Then, Porro et al. (2011) described new specimens from said Stormberg Series of South Africa which included an extremely large (and possibly the largest) heterodontosaurid specimen known (NM QR 1788). This, in context with the juvenile skull described by Butler et al. (2008), SAM-PK-K10487, increases our knowledge of the breadth of heterodontosaurs in time and space. A newer paper redescribes the skull from the holotype (SAM-PK-K337), the “good” specimen SAM-PK-K1332, and additional specimens, in further detail than any other analysis. This is then supplemented with a cladistic analysis and in depth review of the characteristics linking heterodontosaurs to various clades.
The question of which group of ornithischians heterodontosaurs belonged to has occupied few workers interested in actually figuring out the answer, while other authors considered a few linking attributes. Even I offered a suggestion, although it was not novel. This suggestion was featured in the last post on sabre-toothed ornithischians, and the figure above represents part of that argument.
As I wrote before, there are varying theories on the evolution of heterodontosaurids. Despite having very divergent and unqiue crania, there are several fundamental cranial and postcranial aspects that force heterodontosaurs as basal ornithischians, including the presence of an abbreviated prepubic process of the pubis, a fully pentadactyl manus (although the fifth digit lacks an ungual) and a large antorbital fossa with distinct and large antorbital fenestra. However, these are plesiomorphies: we would expect them in basal taxa of more derived lineages. And therein lies the problem, as heterodontosaurs have a unique dental form and arrangement that link them to some marginocephalians (pachycephalosaurs, ceratopsians), some pelvic and limb features that link them to ornithopods (iguanodontoids, “hypsilophodonts” and hadrosauroids), and even some pelvic features that link them to thyreophorans (ankylosaurs, stegosaurs, “scutellosaurs”). These have led to short list of various positions, some of them based on phylogenetic conclusions, as to where they belong.
I threw my hat in this ring as part of a long-standing attempt to get research going on this topic along with Pete Buchholz, who eventually submitted a short paper and talk to the Symposium in Paleontology and Geology held at the Tate Geological Museum at Casper College, in Casper, Wyoming (USA). This was eventually published, and then cited, but that is another story and not really mine to tell. Nonetheless, one of the conclusions of this analysis placed heterodontosaurs at the base of the marginocephalian radiation, which results in a sequential timing of marginocephalian features from what are essentially ornithopodan features. But it also stressed many of the cranial features I alluded to in this post.
But that was then.
In a newly published study, Norman et al. (2011) builds on the earlier work of Butler et al. (2008) and describes the skull of Heterodontosaurus tucki on the basis of all available skull material, including the well-known SAMP-PK-K1332, the less-well known holotype, and a large variety of other specimens.
The authors spend the first part of this paper redescribing the skull, using as their baseline that of SAM-PK-K337, and with this do two things. First, they reassessed all the other heterodontosaur material, including that which has been described from the Lower Jurassic Elliot and Clarens Formations of southern Africa (South Africa and Lesotho) as Geranosaurus atavus (Broom, 1911), Lycorhinus angustidens (Haughton, 1924), Abrictosaurus consors (Thulborn, 1974; originally Lycorhinus consors, where Hopson (1975) provided a generic distinction) and also Lanasaurus scalpridens (Gow, 1975), as well as several other bits that do not seem to belong to these taxa based on explicit comparisons and minimum representation of specimens to types and original referred material. This was supplemented later by Pol et al. (2011) in describing a new taxon (Manidens condorensis) from Patagonia’s Middle Jurassic Cañadón Asfalto Formation, who expanded on Butler et al.’s cladistic analysis.
In both Butler et al. (2008) and the adaptive analyses of Norman et al. (2011) and Pol et al. (2011), heterodontosaurids have come out on bottom, as it were. They are the most basal ornithischians known, with the exception of Pisanosaurus mertii and Echinodon becklessi. And even then, that’s questionable. Below, I provide a simplistic cladogram of the phylogeny of Butler et al. (2008) following Pol et al. (2011), adding in Manidens condorensis:
The phylogeny thus shows that the stem leading to Marginocephalia, here shown as “stem leading to Cerapoda” is interspersed with many ordinarily-known “ornithopods” such as Agilisaurus louderbacki and Lesothosaurus diagnosticus. This phylogeny is particularly at odds with typical analyses (such as the early analysis of Cooper (1985) or the much more recent ones of You et al. (2003) and Xu et al. (2006) — which used the same base dataset) where Heterodontosauridae is placed as the sister taxon of Marginocephalia. This is supported in Xu et al. (2006) by nine features, leading them to coin the name Heterodontosauriformes for “a node-based group that includes the most recent common ancestor of Heterodontosaurus and Triceratops and all their descendents.” Norman et al. (2011), however, appear to have issues with this, so a little backstory is in order.
Cooper (1985) coined the name Neornithischia to refer to the clade containing Ornithopoda and Marginocephalia, but not Thyreophora. Unaware, Sereno (1986) coined the name Cerapoda for the same clade. Although Sereno (1998) dropped use of Neornithischia in favor of Cerapoda, the latter has come under greater useage, despite useage where the two generally contain the same taxa. Sereno (1998) defined Neornithischia as “All genasaurs closer to Triceratops than to Ankylosaurus,” but his usage now corresponds to a more complex definition, “The most inclusive clade containing Triceratops horridus Marsh 1889 but not Ankylosaurus magniventris Brown 1908 and including Parasaurolophus walkeri Parks 1922″ [Sereno, Taxon Search]. However, Norman et al. (2011) note that Heterodontosauriformes is a “junior synonym” of Neornithischia, I presume on the basis of application of the definition to various phylogenies. However, Cooper’s 1985 phylogeny is fairly different:
Thus Norman et al. (2011) conflate the definitions provided of some fairly defunct taxa. Neornithischia is not used currently, while the clade containing strictly Heterodontosaurus tucki and any member of Marginocephalia would be called either Neornithopoda or Heterodontosauriformes. While the former is unlikely to be resurrected, the latter may come into more common usage if the clade is ever further supported. It is also possible, if we apply all taxon definitions as “viable,” despite them currently overlapping, “Heterodontosauriformes” would correspond to Ornithischia, not Neornithischia (and therefore Norman et al.’s usage of Cerapoda).
What Lies Between
Nonetheless, 9 characters is a big number to overcome! Norman et al. handle this by assessing the eighteen features that have been used by various authors to link Heterodontosaurus tucki exclusively with Marginocephalia, finding potential support for only three of them (I’d actually say the number is higher, as they conflate a few features, but…): Reduced premaxillary dentition; enlarged first dentary canine with associated depression on the maxilla in the diastema between premaxillary and maxillary tooth rows; fewer presacral vertebrae. Some of these are unknown in pachycephalosaurs (the last) or are absent in ceratopsians (the third). Arguably, a lateral depression on the surface of the maxilla in IVPP V12738 (holotype of Liaoceratops yanzigouensis) and IG-2004-VD-001 (holotype of Auroraceratops rugosus) may represent remnants of this diastema depression.
When dealing with the remainder of Xu et al. (2006)’s characters, Norman et al. (2011) are careful to point out the absences and presences:
On a contact between the premaxilla and lachrymal: As shown above, this is present in virtually all ceratopsians, absent in Goyocephale lattimorei (it is absent in Stegoceras validum and Prenocephale prenes, just to note), but present also in Dryosaurus altus, Camptosaurus dispar, Theiophytalia kerri, and Jeholosaurus shangyuanensis. With the distribution of taxa between Dryosaurus altus and Heterodontosaurus tucki is vast, including a large number of “hypsilophodonts,” I can reasonably assume that even were Heterodontosaurus tucki a basal member of Ornithopoda, this would be considered convergent, and therefore useless to dismiss the viability of this feature. On the other hand, Jeholosaurus shangyuanensis represents a particularly basal position in Ornithopoda in Pol et al. (2011) and Butler et al. (2008), and separated from Heterodontosaurus tucki by several taxa without the feature. Thus, I’d argue this is convergent, and does not invalidate the use of this character, although it implies there may be various other taxa with the feature. However, if it represents a unique feature to connect heterodontosaurs with marginocephalians, then this may be viable in any case, just more ambiguously.
On the jugal “boss” or “flange:” Norman et al. merely draw a strong line between the morphologies of the broad triangular thickening of the jugal in ceratopsians and the boss in heterodontosaurids. The problem here is semantic, and not one I think is very useful. Essentially, the authors distinguish the form of the thickening of the jugal in basal ceratopsians as a ridge or bowing out of the jugal as precipitous to the pointed, triangular structure seen in psittacosaurids and neoceratopsians. Distinctly, Yinlong downsi (Xu et al., 2006) differs from virtually all basal ceratopsians by the presence of a vertical crest that divides the jugal into anterior and posterior faces, while in other basal forms this is a dorsal and ventral face. A thickening, formed of a small tuberculate circular region, is present in some taxa such as Jeholosaurus shangyuanensis, Changchunsaurus parvus, Orodromeus makelai and Zephyrosaurus schaffi.
However, based on the typical distribution of a jugal with a circular growth region among basal ornithopods, this character is a little touchy. However, the flange, a developed feature that divides the jugal into dorsal and ventral faces, matches most basal ceratopsians, and is quite unlike that of any other ornithischian. This simply tells me the feature is being incorrectly defined in analyses, where a circular tuberculate surface, a horizontal ridge, or a developed process are more correct elements, and they may correspond to both a step-wise transformation phylogenetically, where one precedes the other, or are explicitly unique to one another. SAM-PK-K337 (holotype of Heterodontosaurus tucki) actually lacks this flange, while SAM-PK-K1332 (above) possesses it; in its place, SAM-PK-K337 bears a small circular region raised into a dome, but this feature is smooth unlike that seen in the other taxa said to possess the “boss.” In this case, the feature may not be related to age.
On the relative length of the supratemporal fenestrae to basal skull length: Norman et al. note that while this feature rings true, it is absent in pachycephalosaurs (to a degree that adult taxa shrink or even obliterate these fenestra; Schott et al. (2011) — which I discussed here), and is only just slightly above the ratio in Lesothosaurus diagnosticus and Hypsilophodon foxii, which bracket it in some phylogenies (e.g., Sereno, 1999) but are both basal ornithopods in others (e.g., Butler et al., 2008). So like the premaxilla-lachrymal contact, this one is a shaky rejection, and I think can be just as viable a character as the three tacitly accepted.
On the length of the squamosal process of the postorbital: The comparative distribution of this feature, where the process of the jugal extends above that of the squamosal and forms virtually the entire dorsal margin of the interfenestral bar in lateral view, is also otherwise known in basal ornithopods by Butler et al. (2008)’s analysis. Otherwise, this feature is bracketed with some basal taxa (e.g., Lesothosaurus diagnosticus; Sereno, 1999). And so once again I’m going to have to call this one an ambiguous optimization that favors a strict synapomorphy.
On the relative size of the infratemporal fenestra to the orbit: I don’t really get Norman et al. on this one. They actually support the distribution in heterodontosaurids, exclusive of basal ornithopods, and present at least in basal ceratopsians (Yinlong downsi and psittacosaurids), but because this feature is absent in pachycephalosaurids, it is “relatively ambiguous.” Perhaps a clue can be found in a feature they accept as useful for defining a heterodontosaur + marginocephalian clade, that of relative reduction in premaxillary dentition. There, the authors note that while the number of premaxillary dentition varies in other taxa, it’s stable in Heterodontosauridae, Pachycephalosauridae and Ceratopsia. But that’s not necessarily fair: If Camptosaurus dispar, which lacks teeth, can be used to reject premaxilla-lachrymal contact, why can’t its toothless premaxilla be used to reject this character? Maybe it has to do with presence in pachycephalosaurs for the feature, but that doesn’t make sense as pachycephalosaurs are not known for presacral counts, but the authors are supportive of this feature to imply the “Heterodontosauriformes” relationship. I’ll have to toss this one in the “maybe” pile with the rest of them.
On the formation of a “dental battery,” reducing space between teeth: Teeth in ornithischians come in three flavors: Spaced as in rostral teeth in the maxilla, premaxilla and dentary so that the teeth do not appear to overlap at all in side view; or en echelon, where the teeth are closely set and slightly rotated so that the distal margin of the more anterior tooth overlaps the mesial margin of the more posterior tooth; or so closely set that no space is readily discernible. In heterodontosaurids, Heterodontosaurus tucki, Geranosaurus atavus and Lycorhinus angustidens are the only named taxon to possess this feature, other taxa showing spacing and en echelon conditions; this is also true in pachycephalosaurs, and most basal ceratopsians, so this does not optimize for the supposed link very well, and even the very small SAM-PK-K10487 shows tight spacing and “battery formation,” making even age in this case problematic for the character. So this one is a good rejection.
On the relative depth of the preacetabular ala of the ilium to the postacetabular ala: Norman et al. seem to give this one to “Heterodontosauriformes,” but as it appears to be actually true, but give a caveat in that it’s not enough (much like the length of the supratemporal fenestra). So like above, I’m not convinced this is a good rejection.
On the eversion of the ilium’s dorsal margin: Norman et al. appear puzzled by this feature, noting only that it appears the ilium is merely “thickened” laterally, thus mediolaterally expanded relative to the portion of the ilium above the acetabulum and pubic and ischiadic peduncles. In lateral view, there is a “ridge” that progresses from the preacetabular ala along virtually the entire dorsal margin of the ilium, and is expanded laterally from the surface of the ilium so that it forms a slight “shelf;” above the ischiadic peduncle, this “ridge” moves along the lateral surface of the ilium until it reaches the posterior end, where a thickened “boss” is formed with the posterior margin of the postacetabular ala. In a sense, then, the dorsal margin of the ilium moves from an actual dorsal margin, slightly overlying the lateral surface, and in the postacetabular ala turns (or “everts”) onto the lateral surface; this divideds the lateral aspect of the ilium (rather than lateral “side”) into a dorsal and a ventral facet, with the lower continuous with the rest of the lateral surface of the ilium. Norman et al. (2011) dismiss this character, however, because Yinlong downsi appears to lack it, where the thickened dorsal margin of the ilium never turns (or “everts”) below 1/3 of the height of the postacetabular ala. In Goyocephale lattimorei (Perle et al., 1982) and Homalocephale calathocercos (Maryańska and Osmólska, 1974) the dorsal margin is thickened in this sense and is continuous to the distal end of the ilium, rather than forming a “boss” as in SAM-PK-K1332. If this prevents the dorsal margin of the ilium from being “everted,” I have no idea how to argue against the position of Norman et al. (2011). The simplest explanation is that the character refers to the thickened border and the formation of a dorsal “face” of the postacetabular ala. If so, then it is present in all taxa noted.
Thus, in 9 of the 18 characters that Norman et al. (2011) reject, they actually appear to support the data on two, but dismiss them anyways; one of them look to be legitimately rejected; and the remaining five appear to be supportive either ambiguously (present in a distribution of bracketing taxa or a second lineage) or strongly and the reasoning given rejecting just a little stronger than that used to reject the two actual favorable features. This means that of the three features they support, we can add two to the “sure” pile and five to the “maybe” pile. The remaining six features are somewhat easier to deal with:
On the relative preorbital length to the basal skull length: Norman et al. note that while true, the preorbital skull length differs from some other basal ornithopods by as little as 3-5%, and the preorbital skull length is greater in some pachycephalosaurids, so that the extremely short facial region in marginocephalians is not consistent with heterodontosaurids. However, this is about drawing a large, thick line between features, and as much with the squamosal features noted above, goes into the “maybe” pile.
On the contact between left and right palatal process of the maxilla: This feature is proposed as a “similar” feature by Butler et al. (2008) to the condition in Marginocephalia, where the maxillae exclude the premaxillary palatal processes from contacting the margin of the internal bony naris. Because this is an “approximate” feature, and not directly analogous, with a lot of “iffy”ness involved in exactly what’s preserved in Heterodontosaurus tucki (Norman et al., 2011), this is certainly a point in the “not” pile. It either is, or it isn’t.
On the presence of planar wear facets in “subcylindrical” crowns, while the denticulation is restricted to the distal third: Large wear facets cover up to a third of the labial surface of the dentary crowns and the lingual surface of the maxillary crowns in virtually all heterodontosaurids save Fruitadens haagarorum, causing the teeth to bear an essentially “chisel-shaped” (or scalpriform) appearance, abetted by the cylindrical form and restriction of denticulation (when apparent) to the very apical third of the crowns. Norman et al. (2011) doubt the utility of these essentially three features as useful because several basal ornithopods (Agilisaurus louderbacki and Hexinlusaurus multidens — you know, my mascot) possess them as well, providing doubt as to their consistency. Consistency or not, as I noted above, they are synapomorphic if they are mutually present. The robustness of this synapomorphy is a different matter. Absence in a basal heterodontosaur, however, is more problematic. Fruitadens haagarorum (Butler et al., 2010) does not actually lack wear facets; in the adult holotype (LACM 115747), the wear are nearly transverse to the crown long axis, rather than oblique to it, and instead of indicating a shearing action where the upper dentition move as a unit across the out, labial surface of the lower dentition, instead appear to indicate a more “grinding” action aided by the jaw moving forward and back against the upper dentition with imbricating apices, rather than shearing action. This form is more similar to some ornithopods. I’m gonna give these ones to Norman et al..
On the supposedly “akinetic” skull and the “asymmetrical” manus: While covered separately by Norman et al., I will contain them here, as they are largely answered in the same way. An akinetic skull is so prevalent among ornithopods and maybe represent the plesiomorphic condition as it is present in thyreophorans, that it should be taken as a given. The same is true for the manus, in which manual digits 1 through 3 (md1, 2, 3) increase in side externally, but then md4 and md5 are extremely small, shorter than md1 each. This leads to the “asymmetry” argument, but is true generally for basal ornithischians and indeed for basal dinosaurs in general. There is a strong sense here that instead of a potential convergent feature, these two represent unreversed plesiomorphic trends. While this still serves to connect them, it makes the utility of these features extremely weak. So I’ll give these to Normal et al.
On “fangs:” I was fairly certain I would have a lot to say about the upper caniniform premaxillary teeth, because they interest me so much. Primary among this is that the number and size of the premaxillary teeth are used fairly regularly to assess the value of “fangs.” in the upper jaw. Norman et al. assess strictly the “peg-like” shape that has been used to connect these teeth to those of basal ceratopsians, almost certainly due to their cylindrical form and the short, bluntness of premaxillary crowns 1-2, while crown 3 is almost always a very large “fang.” As seen in the first image above, not all basal ceratopsians have these “fangs,” Abrictosaurus consors has only two small sets, Tianyulong confuciusi has just the one big pair of “fangs” preserved, and the third crowns in Prenocephale prenes and Stegoceras validum are small and not very “fang”-like. These teeth may be somewhat leaf-shaped (phylloform), with a constricted base above the root, or an expanded base of the crown. So what actually connects the teeth of these taxa? I think this is where Norman et al. spend much of their time looking at the diversity among the dentitions, rather than on their distinctive similarities. In all of Ornithischia, only Agilisaurus louderbacki has large teeth of the form of the premaxilla seen in any heterodontosaurs, ceratopsian, or pachycephalosaur, relative to the size of its premaxilla. These teeth are long, with the crowns recurved and the aspect low, indicated a low mesiodistal basal length relative to basoapical crown height. The aspect of the crowns deviate from the phylloform shape, instead of having large denticulations along their edges they are smooth-edged or minutely serrated, and are cylindrical with a distinct distal carina and tear-drop aspect, lacking a central ridge or logitudinal “thickening.” These features characterize all heterodontosaurids with the exception of Fruitadens haagarorum, and in all pachycephalosaurids and most ceratopsians with the exception of Auroraceratops rugosus and psittacosaurids, where the teeth are simply round-sectioned, rounded over the apex, and relatively low-crowned (in comparison). And that’s just in general. In short, this is a complex of characters, and it does these taxa no service treating these as a single feature to be assessed in a single discussion, thus dismissing the suite because the complex is not by itself a cut-and-paste carry over across all taxa. Very large rostral dentition with larger posterior-most premaxillary teeth and large mesial dentary dentition characterize heterodontosaurids, but also basal pachycephalosaurids, and some basal ceratopsians possess a very short fang-like first dentary crown and large, recurved, denticulated premaxillary crowns. It is not possible to lump “fangs” into a “yes/no/maybe” distinction, but I can only assess this as Norman et al., have, and put this into the “maybe” pile.
In total, then, Norman et al. (2011) have turned down six, leaving five “sure” and six “maybe” features. This is a lot of balance, then, on what they initially posited as “weak” (n=4), “complex” or “convergent” (n=2), “ambiguous” (n=2), “little” or an outright “no” or a flat “fails” (n=7), or a single “may provide” verdict. I may be very wrong on my assessments, and I certainly haven’t spent as much time as David Norman and Richard Butler on the subject of ornithischians. However, the sense above that a feature may have to be autapomorphic to be useful systematically appears odd; instead, the authors dismiss almost every single synapomorphic characteristic if it was also present in some basal ornithopods, or was unknown or not present in one of the included groups, even though it was present in the other. This indicates the authors are plying the line between “yes” and “no” very thinly, and erring against the side of Heterodontosauriformes (=Neornithopoda) because their systematic analysis fails to recover it. They may then have their greatest strength in that analysis, whereas their list does not optimize well when enforcing Heterodontosauriformes. I am unaware of an attempt by the authors to constrain their topology to alternative positions, and one of its greatest failures is that despite including 221 characters, they included only 14 taxa, of which several were suprageneric categories. The condensing of taxa into “groups” obscures transformations and therefore polarities among basal forms, especially in weeding out especially interesting transformations among taxa. The omission of several basal dinosaurian taxa, especially sauropodomorphans, “guaibasaurs,” and other interesting outliers such as “silesaurids” impairs the analysis from a basal polarity viewpoint. The analysis of Pol et al. (2011) chose 51 taxa and increased the matrix to 230 characters, including splitting the pachycephalosaurs and basal ceratopsians; in this, I think they achieved closer to their goals of determining the finer relationships among taxa, although they continued to omit several useful basal taxa and did not split their higher-level taxa into individual species, and I suspect as the most complete phylogenetic analysis done to date on the basal relationships among ornithischians it will serve as a benchmark.
Norman et al. (2009) have done a substantive job assessing the phylogenetic position of Heterodontosauridae, and I cannot fault their analysis on this, despite my previous assumptions for alternate positions. They’ve tested that position by positing these supposed sister group positions in their analysis, although they did not do this using the phylogenetic analysis. I am concerned over the effort to dismiss over half of the implied connecting characters that place heterodontosaurs with marginocephalians, but without a competing and equally- or more-comprehensive analysis to back up any of my arguments, my concerns fail against the test. Sabre-toothed dinosaurs are spread out among the origin of ornithischians in this analysis, and despite some constraints (including a lack of them in the basal Ornithopoda) they appear to be part of the radiation of the Genasauria (which would include by definition both Neornithopoda sensu Cooper (1985) as well as Heterodontosauriformes Xu et al. (2006) and Genasauria sensu Sereno (1999) as junior heterodefinitional synonyms).
Butler, R. J., Upchurch, P. & Norman, D. B. 2008. The phylogeny of the ornithischian dinosaurs. Journal of Systematic Palaeontology 6:1-40.
Butler, R. J., Galton, P. M., Porro, L. B., Chiappe, L. M., Henderson, D. M. & Erickson, G. M. 2010. Lower limits of ornithischian dinosaur body size inferred from a new Upper Jurassic heterodontosaurid from North America. Proceedings of the Royal Society, B 277:375-381 [Published online October 21, 2009.]
Cooper, M. R. 1985. A revision of the ornithischian dinosaur Kangnasaurus coetzeei Haughton, with a classification of the Ornithischia. Annals of the South African Museum 95:281–317.
Crompton, A. W. & Charig, A. J. 1962. A new ornithischian from the Upper Triassic of South Africa. Nature 196:1074-1077.
Gow, C. E. 1975. A new heterodontosaurid from the Red Beds of South Africa showing clear evidence of tooth replacement. Zoological Journal of the Linnean Society 57:335–339.
Haughton SH. 1924. The fauna and stratigraphy of the Stormberg Series. Annals of the South African Museum 12:323–497.
Hopson, J. A. 1975. On the generic separation of the ornithischian dinosaurs Lycorhinus and Heterodontosaurus from the Stormberg Series (Upper Triassic) of South Africa. South African Journal of Science 71:302–305.
Maryańska, T. & Osmólska, H. 1974. Pachycephalosauria, a new suborder of ornithischian dinosaurs. Palaeontologia Polonica 30:45-102.
Norman, D. B., Crompton, A. W., Butler, R. J., Porro, L. B. & Charig, A. J. 2011. The Lower Jurassic ornithischian dinosaur Heterodontosaurus tucki Crompton & Charig, 1962: Cranial anatomy, functional morphology, taxonomy, and relationships. Zoological Journal of the Linnaean Society 163(1):182-276.
Perle A., Maryańska, T. & Osmólska, H. 1983. Goyocephale lattimorei gen. et sp. n., a new flat-headed pachycephalosaur (Ornithischia, dinosauria) from the Upper Cretaceous of Mongolia. Acta Palaeontologica Polonica 27(1-4):115-127.
Pol, D., Rauhut, O. W. M., Becerra, M. 2011. A Middle Jurassic heterodontosaurid dinosaur from Patagonia and the evolution of heterodontosaurids. Naturwissenschaften 98:369-379.
Porro, L. B, Butler, R. J., Barrett, P. M., Moore-Fay. S. & Abel, R. L. 2011. New heterodontosaurid specimens from the Lower Jurassic of southern Africa and the early ornithischian dinosaur radiation. Earth and Environmental Science Transactions of the Royal Society of Edinburgh 101:351-366.
Santa Luca, A. P. 1980. The postcranial skeleton of Heterodontosaurus tucki (Reptilia: Ornithischia) from the Stormberg of South Africa. Annals of the South African Museum 79:159–211.
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. 1999. The evolution of dinosaurs. Science 284:2137-2147.
Thulborn, R. A. 1974. A new heterodontosaurid dinosaur (Reptilia: Ornithischia) from the Upper Triassic Red Beds of Lesotho. Zoological Journal of the Linnean Society 55:151–175.
Xu X., Forster, C. A., Clark, J. M. & Mo J. 2006. A basal ceratopsian with transitional freatures from the Late Jurassic of northwestern China. Proceedings of the Royal Society of London, B 273:2135–2140.
You H.-l., Xu X. & Wang X.-l. 2003. A new genus of Psittacosauridae (Dinosauria: Ornithopoda) and the origin and early evolution of marginocephalian dinosaurs. Acta Geologica Sinica 77:15–20.
Zheng X.-t., You H.-l., Xu X. & Dong Z.-m. 2009. An Early Cretaceous heterodontosaurid dinosaur with filamentous integumentary structures. Nature 458:333-336.