A PHYLOGENETIC ANALYSIS OF THE BASAL ORNITHISCHIA (REPTILIA, DINOSAURIA)
Marc Richard Spencer
A Thesis
Submitted to the Graduate College of Bowling Green State University in partial fulfillment of the requirements of the degree of
MASTER OF SCIENCE
December 2007
Committee:
Margaret M. Yacobucci, Advisor
Don C. Steinker
Daniel M. Pavuk
© 2007
Marc Richard Spencer
All Rights Reserved iii
ABSTRACT
Margaret M. Yacobucci, Advisor
The placement of Lesothosaurus diagnosticus and the Heterodontosauridae within the
Ornithischia has been problematic. Historically, Lesothosaurus has been regarded as a basal ornithischian dinosaur, the sister taxon to the Genasauria. Recent phylogenetic analyses, however, have placed Lesothosaurus as a more derived ornithischian within the Genasauria. The
Fabrosauridae, of which Lesothosaurus was considered a member, has never been phylogenetically corroborated and has been considered a paraphyletic assemblage. Prior to recent phylogenetic analyses, the problematic Heterodontosauridae was placed within the
Ornithopoda as the sister taxon to the Euornithopoda. The heterodontosaurids have also been considered as the basal member of the Cerapoda (Ornithopoda + Marginocephalia), the sister taxon to the Marginocephalia, and as the sister taxon to the Genasauria. To reevaluate the placement of these taxa, along with other basal ornithischians and more derived subclades, a phylogenetic analysis of 19 taxonomic units, including two outgroup taxa, was performed.
Analysis of 97 characters and their associated character states culled, modified, and/or rescored from published literature based on published descriptions, produced four most parsimonious trees. Consistency and retention indices were calculated and a bootstrap analysis was performed to determine the relative support for the resultant phylogeny. The Ornithischia was recovered with Pisanosaurus as its basalmost member. A monophyletic Genasauria was recovered with two major clades: Eocursor + Cerapoda and Fabrosauridae + Thyreophora. Lesothosaurus was recovered within the Fabrosauridae, along with a clade consisting of Stormbergia + Agilisaurus. iv
The Heterodontosauridae was recovered as the basalmost taxon of the Cerapoda. In the bootstrap analysis, however, several taxa, including the Fabrosauridae, collapse down to form a polytomy at the base of the Genasauria consisting of Eocursor, Lesothosaurus, Stormbergia +
Agilisaurus, Cerapoda, and Thyreophora. This analysis is the first to recover a weakly-supported
Fabrosauridae clade, and lends further support to the placement of the Heterodontosauridae within the Cerapoda, but outside of the Ornithopoda. Additionally, the implication of the phylogenetic relationship of Eocursor and the Cerapoda indicates, paleobiogeographically, that they evolved in the southern supercontinent of Gondwana in present-day southern Africa. v
This thesis is dedicated to my loving mother, Pamela Yerry, and to the memory of my loving grandfather, the late Richard Murgatroyd. Your love and guidance have made me the man that I am today and you have taught me that strength is far more than merely a physical attribute; it is a measure of character and will, how one stands up after being knocked down, and how the love of
one can affect many. I am forever indebted to you both. Thank you. vi
ACKNOWLEDGEMENTS
A project such as this can never be completed by just one person without the help of
others. For this, I am deeply grateful for the guidance and encouragement of my advisor, Dr.
Peg Yacobucci. You were always willing, regardless of the time of day and how much work you
had piled up, to entertain the most trivial questions, which would inevitably turn into at least
half-hour discussions and, more likely, half-hour digressions. I am grateful to my committee
members Dr. Don Steinker and Dr. Dan Pavuk for their patience throughout a long and arduous
thesis project that began with a thesis proposal that was ultimately, as you so correctly pointed
out, more along the lines of several doctoral dissertations. Dr. Richard Butler of the Natural
History Museum (London) generously provided me with a copy of his 2005 data matrix
regarding basal ornithischian dinosaurs. I would like to express my appreciation to the BGSU
Geology Department faculty and staff for their support. To my fellow graduate students and
friends in the BGSU Geology Department and, in particular, Chris Pepple, Chris Wright, and
Chris Klug – you helped me through this process, perhaps unknowingly, with encouragement and, most importantly, with your company (as drinking buddies). To my family – my mother,
Pam Yerry, my sister, Tracy Liuzzi, my brother-in-law, Frank Liuzzi, my brother, Scott Spencer, and my sister, Amanda Yerry – your overwhelming love and support cannot be quantified, but it can be felt and for that I am forever thankful to you all. To my friends – your support, indirectly, through your sarcastic jokes and endless mocking banter, as well as your genuine encouragement, has helped me through this project and kept me grounded. Collectively, my family and friends have all helped to make this possible, and I sincerely thank you. vii
TABLE OF CONTENTS
Page
1. INTRODUCTION ...... 1
1.1. Purpose of Study...... 3
2. PREVIOUS WORK ...... 6
2.1. Phylogenetic Systematics...... 6
2.2. Dinosauria ...... 8
3. REVIEW OF TAXONOMIC UNITS ...... 18
3.1. Operational Taxonomic Units ...... 18
3.1.1. Outgroup OTUs ...... 19
Marasuchus lilloensis ...... 19
Saurischia ...... 21
3.1.2. Ingroup OTUs ...... 24
Pisanosaurus mertii ...... 26
Eocursor parvus ...... 31
Lesothosaurus diagnosticus ...... 33
Stormbergia dangershoeki ...... 41
Heterodontosaurus tucki ...... 43
Abrictosaurus consors ...... 47
Agilisaurus louderbacki ...... 50
Hypsilophodon foxii ...... 53
Thescelosaurus neglectus ...... 56
Iguanodontia ...... 59 viii
Pachycephalosauria ...... 62
Ceratopsia ...... 65
Scutellosaurus lawleri ...... 68
Emausaurus ernsti ...... 70
Scelidosaurus harrisonii ...... 72
Stegosauria ...... 75
Ankylosauria ...... 80
4. METHODS ...... 87
4.1. Characters and Character States...... 87
4.2. Taxon-Character Matrix ...... 88
4.3. Phylogenetic Analysis ...... 90
5. RESULTS ...... 94
5.1. Node 1 – Ornithischia ...... 101
5.2. Node 2 – Genasauria ...... 102
5.3. Node 3 – Unnamed Clade (Stormbergia + Agilisaurus) ...... 103
5.4. Node 4 – Cerapoda ...... 104
5.5. Node 5 – Heterodontosauridae ...... 105
5.6. Node 6 – Unnamed Clade (Hypsilophodontidae + Iguanodontia +
Marginocephalia) ...... 106
5.7. Node 7 – Hypsilophodontidae ...... 106
5.8. Node 8 – Marginocephalia ...... 107
5.9. Node 9 – Thyreophora ...... 108
5.10. Node 10 – Thyreophoroidea ...... 108 ix
5.11. Node 11 – Eurypoda ...... 109
6. DISCUSSION ...... 110
7. SUMMARY AND CONCLUSIONS ...... 115
REFERENCES ...... 118
APPENDIX 1. – CHARACTER-CHARACTER STATE DESCRIPTIONS ...... 130
APPENDIX 2. – TAXON-CHARACTER MATRIX ...... 138 x
LIST OF FIGURES
Figure Page
1. Map of Pangaea during the Triassic ...... 3
2. Gauthier’s (1986) phylogeny of the Archosauria ...... 9
3. Benton’s (1999) phylogeny of the Archosauria ...... 10
4. Benton’s (2004) phylogeny of the Archosauria ...... 11
5. Pelvic articulation of saurischian and ornithischian dinosaurs ...... 12
6. Traditional view of ornithischian phylogeny ...... 16
7. Reconstruction of Marasuchus lilloensis ...... 20
8. Reconstructions of basal saurischian dinosaurs ...... 22
9. The manus of various archosaurs ...... 24
10. Right mandible of Pisanosaurus mertii ...... 26
11. Cranial and mandibular material of Pisanosaurus mertii ...... 27
12. Pelvic impression of Pisanosaurus mertii ...... 28
13. Right crus of Pisanosaurus mertii ...... 29
14. Reconstruction of Eocursor parvus ...... 32
15. Reconstructed left mandible of Lesothosaurus diagnosticus ...... 36
16. Reconstructed cranium of Lesothosaurus diagnosticus ...... 38
17. Reconstructed pelvis of Lesothosaurus diagnosticus ...... 39
18. Hindlimb material of Lesothosaurus diagnosticus ...... 40
19. Reconstruction of Lesothosaurus diagnosticus ...... 41
20. Left ilium of Stormbergia dangershoeki ...... 42
21. Skull of Heterodontosaurus tucki ...... 45 xi
22. Pelvis of Heterodontosaurus tucki ...... 46
23. Reconstruction of Heterodontosaurus tucki ...... 46
24. Skull of Abrictosaurus consors ...... 48
25. Dentary tooth of Abrictosaurus consors ...... 49
26. Skull of Agilisaurus louderbacki ...... 51
27. Left ilium and femur of Agilisaurus louderbacki ...... 52
28. Skull of Hypsilophodon foxii ...... 55
29. Teeth of Hypsilophodon foxii ...... 55
30. Pelvis of Hypsilophodon foxii ...... 56
31. Reconstruction of Hypsilophodon foxii ...... 56
32. Reconstruction of the skull of Thescelosaurus neglectus ...... 58
33. Reconstructions of the skulls of Tenontosaurus tilletti, Zalmoxes robustus, and
Camptosaurus dispar ...... 60
34. Manus of Tenontosaurus tilletti ...... 60
35. Pelvis of Tenontosaurus tilletti and Dryosaurus altus...... 61
36. Reconstructions of the skulls of Prenocephale prenes and Stegoceras validum ...... 64
37. Sacropelvic region of Homalocephale calathocercos ...... 64
38. Reconstruction of Psittacosaurus mongoliensis and Pentaceratops sternbergii ...... 66
39. Skull of Psittacosaurus mongoliensis ...... 68
40. Reconstruction of Scutellosaurus lawleri ...... 69
41. Skull of Emausaurus ernsti ...... 71
42. Hindlimbs of Scutellosaurus lawleri and Scelidosaurus harrisonii ...... 73
43. Pelvis of Scutellosaurus lawleri and Scelidosaurus harrisonii ...... 74 xii
44. Reconstruction of Scelidosaurus harrisonii ...... 75
45. Reconstructions of various stegosaurs ...... 76
46. Skull of Huayangosaurus taibaii ...... 78
47. Postcrania of Stegosaurus armatus ...... 79
48. Left humerus of Stegosaurus armatus ...... 79
49. Pelvis of Stegosaurus sp...... 80
50. Skull of Euoplocephalus tutus ...... 82
51. Skulls of various ankylosaurs ...... 83
52. Synsacrum of nodosaurid and ankylosaurid ankylosaurians ...... 84
53. Reconstructions of ankylosaurid and nodosaurid ankylosaurians ...... 85
54. Four MPTs generated by PAUP* in the phylogenetic analysis ...... 98
55. Strict consensus of the four MPTs ...... 99
56. Bootstrap majority rule consensus tree ...... 100
57. Ornithischia cladogram ...... 115 xiii
LIST OF TABLES
Table Page
1. Location and geologic age of the 12 species-only ingroup OTUs ...... 25
2. Parsimony-uninformative characters ...... 91
3. Character indices ...... 95
1
1. INTRODUCTION
The Dinosauria, first described by Richard Owen (1842), was formally divided into two groups by Harry Govier Seeley (1887), the Saurischia and the Ornithischia. Dinosaurs first appear in the fossil record during the Carnian (Late Triassic, approximately 228 million years ago) having already diversified into these two groups (Benton, 2004). Three taxa of similar age,
Herrerasaurus ischigualastensis, Eoraptor lunensis, and Pisanosaurus mertii, were uncovered in the Ischigualasto Formation, Argentina (Carnian). Two other taxa, Saturnalia tupiniquium and
Staurikosaurus pricei, were uncovered in the Santa Maria Formation, Brazil (late Ladinian or early Carnian). Eoraptor, Staurikosaurus, and Herrerasaurus are considered to be basal saurischians (Langer, 2004; Langer and Benton, 2006); however, their placement as basal theropods has also been evaluated (Sereno, 1997, 1999; Rauhut, 2003) as well as the potential for a clade consisting of Herrerasaurus + Staurikosaurus to be placed as the immediate outgroup to
Dinosauria (Langer et al., 1999; Fraser et al., 2002). Saturnalia is considered to be either a basal saurischian or a basal sauropodomorph (Langer et al., 1999; Langer, 2004; Galton and Upchurch,
2004a) and Pisanosaurus is known to be a basal ornithischian (Casamiquela, 1967; Bonaparte,
1976; Weishampel and Witmer, 1990b; Sereno, 1991; Norman et al., 2004a). The Elliot
Formation of the Stormberg Series, South Africa, has yielded several species of archosauriforms, including basal ornithischian dinosaurs such as Heterodontosaurus tucki, Lesothosaurus diagnosticus, and Stormbergia dangershoeki (Butler, 2005). These dinosaurs were uncovered in the Upper Elliot Formation (Lower Jurassic). However, a newly described ornithischian dinosaur, Eocursor parvus, was discovered in the Lower Elliot Formation (Upper Triassic 2
[Norian Stage]; Butler et al., 2007). Aside from Pisanosaurus, Eocursor is the first definitive
Triassic ornithischian uncovered to date (Irmis et al., 2007).
Based on the age of the rock strata in which these taxa were discovered and the already
derived nature of the fossils (the most basic distinctions – ornithischian, saurischian – had been
established), Sereno (1997, 1999) opines that dinosaurs evolved sometime in the Middle Triassic
and diversified into the respective sister taxa (Saurischia and Ornithischia) well before the close
of the Triassic period. Additionally, based on the discovery of the aforementioned taxa, it
appears evident that dinosaurs – in particular ornithischian dinosaurs (Irmis et al., 2007) –
evolved in the southern portion of the supercontinent Pangaea (Fig. 1; broken up into northern
Laurasia and southern Gondwana during the Late Triassic and Early Jurassic; Olsen, 1997;
Weishampel et al., 2004) and radiated outward to establish a global terrestrial diversity prior to
the close of the Triassic (Forster, 1999; Taylor, 2006).
3
Figure 1. Map of Pangaea during the Triassic. Modern political divisions and landmarks have been overlain for orientation purposes. The red square approximates the broad geographic area in which Eocursor, Eoraptor, Herrerasaurus, Staurikosaurus, Saturnalia, and Pisanosaurus were uncovered. As stated in the text, these six taxa are considered to be among the oldest and earliest known dinosaurs. (Modified from Scotese, 2001.)
1.1. Purpose of Study
Norman (1984) and Sereno (1984, 1986) independently published phylogenetic analyses of the clade Ornithischia, but did not use computer-based phylogenetic techniques that have been used on the Saurischia (e.g., Gauthier, 1986; Rauhut, 2003; Benton, 2004). Despite the extensive literature on the taxonomic members of the Ornithischia, no known numerical phylogenetic analysis has been conducted and published on the whole of the group (Weishampel, 2004). It is not known why such an analysis has yet to be published on the Ornithischia; however, one can speculate that more interest is invested in the saurischian dinosaurs due to their relevance in the 4
popular culture as well as their link to an extant crown group (i.e., the relationship of modern
birds and other theropod dinosaurs). Another potential reason for the lack of phylogenetic
studies may be due to the very sparse Triassic record for ornithischian dinosaurs, which at present consists of only two known taxa, Eocursor parvus and Pisanosaurus mertii (Parker et al.,
2005; Butler et al., 2007; Irmis et al., 2007; Nesbitt et al., 2007). Recently, however, several workers have published phylogenetic analyses including basal members of the Ornithischia
(Butler, 2005; Langer and Benton, 2006; Butler et al., 2007). Butler (2005) presented a phylogenetic analysis of several basal ornithischians, including the first published description of
Stormbergia, which challenged long-standing assumptions of the placement of Lesothosaurus.
Langer and Benton (2006) analyzed the relationships of basal dinosaurs and dinosauriforms using Pisanosaurus and a composite-coded Ornithischia, rather than all known referred basal taxa of the Ornithischia. Butler et al. (2007) presented a phylogeny of several basal ornithischians, including a thorough description of Eocursor, but excluded the ornithopod dinosaurs (hypsilophodontids and iguanodontids). These three analyses will be discussed further in Section 2.2. This thesis reevaluates the relationships among the basal ornithischian dinosaurs and their placement within the Ornithischia with respect to its other well-established clades (i.e.,
Iguanodontia, Ceratopsia, Pachycephalosauria, Ankylosauria, and Stegosauria). One of the intriguing results from the phylogeny of Butler (2005) was the placement of Lesothosaurus. In
his analysis, Lesothosaurus was recovered as the basalmost member of the Cerapoda, not as the
sister taxon to all other ornithischians (the Genasauria) as most other analyses have presumed.
Butler et al. (2007) corroborated this position for Lesothosaurus. The placement of the
Heterodontosauridae has also proved problematic in myriad analyses. Some workers have
placed it as the basalmost member of the Ornithopoda (e.g., Sereno, 1986, 1999), others have 5 placed it as the basalmost member of the Cerapoda (e.g., Butler, 2005), some have placed it as the sister taxon to the Marginocephalia (e.g., Xu et al., 2006), and some have placed it, along with Pisanosaurus, as the sister taxon to all other ornithischians (e.g., Butler et al., 2007). The affinities of the problematic Heterodontosauridae and Lesothosaurus will be tested here.
6
2. PREVIOUS WORK
2.1. Phylogenetic Systematics
A phylogeny is an evolutionary history of a species or group of species and systematics is
a methodology for classification. Thus, phylogenetic systematics is a method of reconstructing
evolutionary relationships among organisms. While phylogenetic systematics, or cladistics, has
been used in neontology since the translation (1966) of Hennig’s original publication (1950),
cladistic analyses are still relatively new in the world of paleontology. The issue of how and
why to utilize this methodology is still eliciting papers that call for a universal alternative that
can work concurrently with the classic Linnaean binomial nomenclature, but remove all higher
rank-order taxonomic designations (e.g., Kingdom, Phylum, and Order). For instance, under the
classical Linnaean taxonomy, the class Aves is now known to have been derived from a group of
theropod dinosaurs, which are nested within the order Saurischia, which in turn is within the
class Reptilia. By this classification, the class Mammalia is also a group nested within the class
Reptilia, because mammals evolved from a group of Permian reptiles known as therapsids.
Bakker and Galton (1974) and Bonaparte (1976) independently came to the conclusion
that Ornithischia is a monophyletic clade. A monophyletic clade is a natural group of species
that is derived from a more recent common ancestral species than any other that is classified outside of the group (Hennig, 1966). Bakker and Galton (1974) also were the first to indicate that the whole of the Dinosauria is also a monophyletic clade; that is, the Saurischia and
Ornithischia share a common ancestor. Sereno (1998) indicates that the Dinosauria is a node-
based clade defined as the most recent common ancestor of Triceratops, Neornithes (extant
birds), and all its descendants. Sereno (1998) also indicates that both the Saurischia and the 7
Ornithischia are stem-based clades. A node-based clade, as defined by Sereno (1998), is a taxon defined to include a common ancestor and all of its descendants, whereas a stem-based clade is a taxon defined as all descendants more closely related to a reference taxon than to any other.
The positive and negative aspects of cladistic taxonomic methodology have been the subject of many papers and have even been the focus for convening conferences (e.g., Sereno,
1998; Doolittle, 1999; Benton, 2000; Brochu, 2001; Brochu and Sumrall, 2001; Dyke, 2002;
Cantino and de Queiroz, 2004; Taylor, 2007). Phylogenetic systematics, however, is not entirely infallible. It is well understood, perhaps rather frustratingly, that established phylogenetic relationships are always subject to change (Sereno, 2005a). For example, the stem-based clade
Herrerasauridae, as defined by Sereno (1998), is based on the specifier Herrerasaurus ischigualastensis, whereas Langer (2004) treats the Herrerasauridae as a node-based clade within the larger stem-based Herrerasauria. Sereno (2005b) suggests that the lack of specimens does not warrant a higher-level taxon to further define this clade, considering that it is already positioned outside of the clade Theropoda, and thus Langer’s (2004) approach seems redundant.
Sereno (2005b) suggests that Langer’s (2004) definition is thus a first-order revision of the stem- based clade Herrerasauridae. A first-order revision, as defined by Sereno (2005a), is simply a modification or edit of the original definition that does not alter a clade’s membership.
Conversely, a second-order revision is one in which the definition is changed (e.g., stem-based to node-based) or the specifiers are changed and/or substituted.
Challenges from opponents of cladistic methodology as well as the nuanced quarrels among its proponents have had and likely will continue arguments at conferences and in the published literature. However, regardless of these disagreements, this thesis will employ cladistic methodology to reconstruct evolutionary relationships based on: (1) the limited 8
information that the dinosaur fossil record yields – cladistics makes no a priori assumptions
about the completeness of the fossil record or that a geologically older taxon is necessarily
ancestral to the younger taxa, whereas the stratophenetic method assumes that the ancestral
species is that which is found lower in the rock record than those superficially similar taxa from
geologically younger strata, regardless of plesiomorphies or apomorphies; and (2) because it is
standard practice among the peer-reviewed journals and within the vertebrate paleontological
community.
2.2. Dinosauria
Gauthier (1986) was the first to apply cladistic methodology to the Archosauria. Through this analysis, he was able to determine that Archosauria is a monophyletic clade. It also followed that the Dinosauria represented a monophyly, as mentioned above, an idea that was put forth a decade earlier (Bakker and Galton, 1974) but uncorroborated until Gauthier’s (1986) analysis.
Figures 2, 3, and 4 are phylogenies from Gauthier (1986), Benton (1999), and Benton (2004) that illustrate archosaur phylogenies, respectively. These phylogenies, although some taxa differ, similarly illustrate the interrelationships of the clade Archosauria. In the case of Benton’s (1999,
2004) phylogenies with the application of bootstrap values, the Dinosauria was very well supported. The phylogenies maintained extremely robust bootstrap values of 100% and 98%, respectively. 9
Figure 2. Gauthier’s (1986) phylogeny of the Archosauria. (Redrawn after Gauthier, 1986.) 10
Figure 3. Benton’s (1999) phylogeny of the Archosauria. Several of the nodes have been labeled; those nodes with numbers indicate supporting bootstrap values. Note that not all nodes have the associated bootstrap values; the polytomy indicates that those clades collapsed in Benton’s bootstrap analysis (had a bootstrap value <50%). (Redrawn after Benton, 1999.) 11
Figure 4. Benton’s (2004) phylogeny of the Archosauria. Several of the nodes are labeled; the numbers beneath the nodes indicate bootstrap values. Polytomies indicate collapsed clades (bootstrap values <50%). Note that the position of Herrerasaurus is ambiguous here as opposed to Benton’s earlier phylogeny presented in Figure 5. (Redrawn after Benton, 2004.) 12
As mentioned earlier, Dinosauria is subdivided into two clades, the Saurischia and
Ornithischia (Seeley, 1887). To a first-order approximation, the simplest distinction between
saurischian and ornithischian (literally, “lizard-hipped” and “bird-hipped,” respectively)
dinosaurs is the articulation of the pelvis (Fig. 5). However, noting that the ornithischian
dinosaurs share a similar pelvic arrangement with more derived saurischians (i.e.,
dromaeosaurids, troodontids, and modern birds – a group of derived theropods [Aves] as first
defined by Gauthier [1986] and later amended by Padian et al. [1999]) as well as concerns with
using one apomorphic character to diagnose an entire clade of vertebrate animals, it is clear that
utilizing pelvic articulation solely as a defining character is problematic.
The stem-based clade Saurischia is defined as all dinosaurs more closely related to
Tyrannosaurus rex than to Triceratops horridus (Holtz and Osmolska, 2004). This clade will be
discussed in further detail below. It is important to note that Saurischia is divided into two major
clades, the Theropoda and Sauropodomorpha, and numerous researchers consider that these two
clades, along with the Ornithischia, are the three major dinosaurian groups.
Figure 5. Pelvic articulation of saurischian and ornithischian dinosaurs. Note the posteroventral rotation of the pubis in the ornithischian pelvis. (From Lucas, 2000.) 13
Weishampel (2004) indicates that the Ornithischia is a stem-based clade defined as all
dinosaurs more closely related to Triceratops than to Tyrannosaurus. Norman (1984) and
Sereno (1984, 1986), independently, were the first to compile an exhaustive list of synapomorphies that diagnose the clade. Dental synapomorphies that Sereno (1984, 1986) listed are: loss of the recurvature in the cheek teeth; low, triangular-shaped crowns on the cheek teeth; distinct neck that separated the root from the crown; overlapping adjacent crowns in the cheek
teeth; and the central or posterocentral cheek teeth achieve maximum size. Several of the cranial
characters listed are: toothless (at least the width of one tooth position) tip of the snout, which
may indicate the presence of a rhamphotheca (a horny beak); a premaxillary palate that is
horizontal to broadly arched; robust posterolateral premaxillary process; ventral margin of the
antorbital fossa is parallel to the maxillary tooth row; antorbital fenestra is relatively small;
elevated coronoid process formed by the posterior portion of the dentary; an ossified palpebral;
and a neomorphic predentary bone. Several of the postcranial characters listed are: five sacral vertebrae; ossified epaxial tendons at least along the sacral vertebrae; elongate preacetabular
process of the ilium; posteroventrally deflected pubis; pubic, ischial, and puboischial symphyses
are restricted to the distal portions; pubis contains an obturator notch and the obturator foramen
is formed between the pubis and the ischium; and a pendant fourth trochanter. It should be noted
that the dental characters listed above and used by numerous authors (e.g., Chatterjee, 1984;
Hunt and Lucas, 1994; Heckert, 2002) to solely identify putative basal ornithischian dinosaurs
(e.g., Technosaurus smalli, Revueltosaurus callenderi, Tecovasaurus murryi) have been recently
shown to exist convergently in other contemporaneous Triassic archosaurs (Parker et al., 2005;
Irmis et al., 2007; Nesbitt et al., 2007). However, within Dinosauria these dental characters, used 14
in conjunction with other cranial and postcranial characters, do serve as apomorphic features of
ornithischian, not saurischian, dinosaurs.
Historically, Ornithischia has been subdivided into the basal ornithischians, consisting of
Lesothosaurus diagnosticus and Pisanosaurus mertii, and Genasauria (Sereno, 1997, 1999;
Weishampel, 2004; Norman et al., 2004a). The Genasauria (the “cheeked” dinosaurs, based on the buccal emargination of the maxillary tooth row) is a node-based clade defined as the most recent ancestor of Ankylosaurus and Triceratops and all of its descendants (Sereno, 1998).
Thyreophora and Cerapoda (=Neornithischia; Cooper, 1985; Sereno, 1997, 1999) are the two
subdivisions that comprise Genasauria (Sereno, 1986, 1997, 1999; Weishampel, 2004). The
thyreophorans are the “armored” dinosaurs that include the Stegosauria (Huayangosaurus taibaii
+ Stegosauridae) and the Ankylosauria (Nodosauridae + Ankylosauridae) as well as the basal
sister taxa, Scutellosaurus lawleri, Scelidosaurus harrisonii, and Emausaurus ernsti (Norman et
al., 2004b). Thyreophora is a stem-based clade that is defined as all genasaurs more closely
related to Ankylosaurus than to Triceratops (Sereno, 1999; Weishampel, 2004). Cerapoda is a
more diverse clade than its sister taxon Thyreophora and is a stem-based clade defined as all
genasaurs more closely related to Triceratops than to Ankylosaurus (Sereno, 1999; Weishampel,
2004). Cerapoda can be divided into Ornithopoda and Marginocephalia (Sereno, 1997, 1999;
Weishampel, 2004). The Ornithopoda, according to Sereno (1998), is a node-based clade
defined as the most recent common ancestor of Heterodontosaurus and Parasaurolophus and all
of its descendants, whereas Norman et al. (2004c) indicate that it is a stem-based clade defined as
all cerapodans more closely related to Edmontosaurus than to Triceratops. Regardless of the
disagreement over the phylogenetic definition, Ornithopoda traditionally is further
subcategorized into the Heterodontosauridae and the Euornithopoda (Weishampel, 1990; 15
Norman et al., 2004c). The Heterodontosauridae is a stem-based clade defined as all ornithopods
more closely related to Heterodontosaurus than to Parasaurolophus (Sereno, 1998). The
euornithopods are a stem-based clade defined as all ornithopods more closely related to
Parasaurolophus than to Heterodontosaurus (Sereno, 1998) and consist of the hypsilophodontids and iguanodontians; however, there is some suspicion about the monophyly of the hypsilophodontids (Weishampel et al., 2003; Norman et al., 2004c). The sister taxon to
Ornithopoda, Marginocephalia, is a node-based clade defined as the most recent common ancestor of Pachycephalosaurus and Triceratops and all its descendants (Weishampel, 2004).
The marginocephalians are divided into the Pachycephalosauria and the Ceratopsia (Sereno,
1986, 1997, 1999; Weishampel, 2004). Figure 6 reflects the traditional view of ornithischian phylogeny as described above.
Figure 6. Traditional view of ornithischian phylogeny. Note that here the Ornithopoda consists of Heterodontosauridae + (Hypsilophodontidae + Iguanodontia). (From Weishampel, 2004.) 16
Recent cladistic analyses have shown that there may be some suspicion about heterodontosaurid affinities (Butler, 2005; Xu et al., 2006; Butler et al., 2007). The
marginocephalians and ornithopods may share a more recent common ancestor to the exclusion
of the heterodontosaurids and place the Heterodontosauridae as the basalmost member of the
Cerapoda (Butler, 2005). Heterodontosaurids may be more closely related to the
marginocephalians, thus creating a new clade, the Heterodontosauriformes (Xu et al., 2006).
Other research has suggested that the heterodontosaurids and Pisanosaurus may be the sister
taxa to all other ornithischians (Butler et al., 2007). Sereno (2005b) suggests that if analyses that
place the heterodontosaurids more closely related to another clade than to the euornithopods are
corroborated, then his node-based definition (Sereno, 1998) would be inapplicable. In all recent
phylogenetic analyses (e.g., Butler, 2005; Xu et al., 2006; Butler et al., 2007), the
heterodontosaurids have not been recovered as the sister taxon to the euornithopods. In these analyses, the euornithopods have been referred to as the Ornithopoda (i.e.,
Hypsilophodontosauridae + Iguanodontia).
Numerous cladistic analyses have been performed on more exclusive clades of
ornithischian dinosaurs (e.g., Forster, 1990; Weishampel and Heinrich, 1992; Coria and Salgado,
1996; Weishampel et al., 2003; Dodson et al., 2004; Galton and Upchurch, 2004b; Horner et al.,
2004; Maryańska et al., 2004; Norman, 2004; Norman et al., 2004b, c; Vickaryous et al., 2004;
You and Dodson, 2004; Maidment et al., 2006; Xu et al., 2006). There are, however, only three
known analyses that consider basal ornithischians (Butler, 2005; Langer and Benton, 2006;
Butler et al., 2007). Langer and Benton (2006) conducted their analysis on basal dinosaurs,
including basal saurischians and dinosauriforms, to evaluate dinosaur origins. The analysis did
not compare the interrelationships among the basal ornithischians as it just considered 17
Pisanosaurus mertii and a composite coded Ornithischia versus all other dinosaurs and
Silesaurus opolensis, a putative dinosauriform from southern Poland. Butler’s (2005) analysis evaluated the basal ornithischian affinities with the oldest and most primitive known taxa, which include a newly described ornithischian, Stormbergia dangershoeki. His analysis challenged long-standing phylogenies of basal ornithischian dinosaurs and included Lesothosaurus as the basalmost member of the Cerapoda. Traditionally, Lesothosaurus has been considered a priori as the outgroup to all other ornithischians (Genasauria) likely based on the lack of buccal emargination and the presence of an external mandibular fenestra that Lesothosaurus possesses.
Aside from Butler (2005) and Butler et al. (2007), there are no known analyses that place
Lesothosaurus as an ingroup taxon to be tested with other basal ornithischians. Butler et al.
(2007) considered basal ornithischian affinities and included a complete description of a new taxon, Eocursor parvus, which the previous study of Butler (2005) did include, but referred to as the then undescribed specimen, SAM-PK-K8025. This most recent analysis suggests that the heterodontosaurids + Pisanosaurus are basal to all other ornithischians and is also in agreement with Butler (2005) regarding the placement of Lesothosaurus as the basalmost member of the
Cerapoda.
18
3. REVIEW OF TAXONOMIC UNITS
3.1. Operational Taxonomic Units
An operational taxonomic unit (OTU) is a designation for the groups used in any
phylogenetic analysis. This study will employ both individual species (e.g., Pisanosaurus mertii,
Lesothosaurus diagnosticus) and larger composite taxa (e.g., Saurischia, Iguanodontia). The
monophyly of these composite OTUs (discussed below) has been well supported in the literature
and does not need to be reevaluated here, as it is beyond the intended scope of this thesis and
these clades are sufficiently derived to not have a systematic impact on the resulting phylogeny.
However, several basal species of well-established monophyletic clades (i.e.,
Heterodontosauridae, Ornithopoda, Euornithopoda, and Thyreophora) will be utilized for this
analysis. Although the monophyly of these clades has not been seriously challenged, the basal
members will be included to evaluate the potential affinities for other putative basal
ornithischians. For example, Butler et al. (2007) recovered Pisanosaurus and the
Heterodontosauridae as sister taxa to the remainder of all ornithischians; therefore, it is appropriate to evaluate the status of Pisanosaurus with other heterodontosaurids.
The institutional abbreviations used below are as follows: BMNH, Natural History
Museum, London, UK; MNHN, Museum National d’Histoire Naturelle, Paris, France; PVL,
Fundación Miguel Lillo, Universidad Nacional de Tucumán, Argentina; SAM, South African
Museum, Cape Town, South Africa; UCMP, Museum of Paleontology, University of California,
Berkeley; UPLR, Museo de Paleontología, Universidad Provincial de La Rioja, Argentina;
ZDM, Zigong Dinosaur Museum, Zigong, Sichuan Province, China; ZPAL, Instytut
Paleobiologii PAN, Warsaw, Poland. 19
3.1.1. Outgroup OTUs
Two outgroup taxa have been chosen for comparison with the basal members of
Ornithischia. These outgroup taxa are the closest known successive sister taxa to the
Ornithischia (Benton, 2004; Langer, 2004; Sereno, 2007) and thus obviate the question of potential basal dinosaur or dinosauriform affinity for some of the ingroup OTUs presented below.
Marasuchus lilloensis
Since its description (Sereno and Arcucci, 1994), Marasuchus (Fig. 7) has been
considered the sister taxon to all other dinosaurs. Previously, Lagosuchus talampayensis was
considered to be the immediate sister taxon to Dinosauria (Romer, 1971); however, Sereno and
Arcucci (1994) note that the holotype of Lagosuchus (UPLR 09) does not exhibit any
autapomorphies that could distinguish it from other dinosauromorphs; therefore, they declared
Lagosuchus talampayensis a nomen dubium. However, another species of “Lagosuchus” was described by Romer (1972; PVL 3871) and given the name Lagosuchus lilloensis. This specimen, given the new generic name Marasuchus by Sereno and Arcucci (1994), does possess several autapomorphies and shares several synapomorphies with Dinosauria that exclude other dinosauromorphs such as Lagerpeton and pterosaurs, uniting them as Dinosauriformes (Novas,
1992). Those autapomorphies as identified by Sereno and Arcucci (1994) include: anterodorsally inclined neural spines on its posterior cervical vertebrae; a depression ventral to the transverse process on the posterior cervical vertebrae; subtriangular neural spine on the middle and posterior dorsal vertebrae; centrum length of the caudal vertebrae increases distally; the length of the anterior chevron is three times the length of the first caudal vertebra centrum; 20
width of the scapular blade is more than 25% of the blade length; pubic blade is deflected
posterolaterally; and a lateral margin on the fibular facet of the calcaneum. The synapomorphic
features that Sereno and Arcucci (1994) identified which unite it with Dinosauria forming a
monophyletic Dinosauriformes include: parallelogram-shaped centrum in the presacral vertebrae;
a forelimb/hindlimb ratio of 0.5 or less; an open iliac-pubic acetabular wall; an antitrochanter on
the posterior wall of the acetabulum; the pubis at least three times longer than the diameter of the
acetabulum; the articular surface of the femoral head extends ventrally; an anterior (lesser)
trochanter; a femoral trochanteric shelf; and a posteromedial flange on the distal tibia.
Figure 7. Reconstruction of Marasuchus lilloensis. Marasuchus is a dinosauriform, the group considered to be the sister taxon to all dinosaurs. Scale bar = 5 cm. (From Sereno and Arcucci, 1994.)
Dinosauriformes also includes Pseudolagosuchus major and Lewisuchus admixtus
(Arcucci, 1997). However, based on incomplete morphological data from these two taxa and from Marasuchus, the affinities to Dinosauria are unclear, thus creating a polytomy (Arcucci,
1997; Langer and Benton, 2006). In their analysis of basal dinosaurs, Langer and Benton (2006) also included Silesaurus opolensis (Langer and Benton actually included Silesaurus within the ingroup), a dinosaur-like animal from the Keuper Claystone (Carnian) of southern Poland (Dzik, 21
2003) but did not include other dinosauriforms such as Marasuchus. Although Silesaurus may
represent the immediate sister taxon to Dinosauria, the morphological evidence seems suspect at
this time (Irmis et al., 2007). The holotype (ZPAL Ab III/361) was culled from more than four
hundred bones among several different semi- and disarticulated skeletons (Dzik, 2003).
Additionally, Langer and Benton’s analysis produced a phylogeny that recovered Silesaurus as
the sister taxon to all dinosaurs (2006:fig. 15), and without the inclusion of other dinosauriforms,
did not resolve the relationships of the members of Dinosauriformes to the basal Dinosauria.
Based on the lack of uncorroborated morphological data substantiating the unambiguous nature
of the holotype and due to its recovery as the sister taxon to Dinosauria similar to that of
Marasuchus, Silesaurus does not appear to be more closely related to Dinosauria than does
Marasuchus; therefore, this study will not use Silesaurus as an outgroup OTU.
Saurischia
Saurischia, first defined by Seeley (1887), is comprised of two major clades of dinosaurs,
the Theropoda and the Sauropodomorpha. The Theropoda are characteristically considered to be
bipedal carnivorous dinosaurs. However, it is now presumed that at least two groups of theropod dinosaurs exhibit at least partial herbivory: Avialae (Gauthier, 1986; Padian et al., 1999; Padian,
2004) and Therizinosauroidea (formerly Segnosauria; Barsbold and Maryańska, 1990; Russell,
1997; Clark et al. 2004). Regardless of the dietary habits of these derived theropods, all known
carnivorous dinosaurs belong to the Theropoda. The Sauropodomorpha is divided into the
Prosauropoda and the Sauropoda. The sauropodomorphs are generally considered to be the long-
necked titans that constitute the largest and heaviest land animals in history. They were
herbivorous and generally either facultatively bipedal or obligatorily quadrupedal. The basal 22
members such as Thecodontosaurus (and possibly Saturnalia), however, are more similar to
basal theropods/saurischians such as Herrerasaurus, Eoraptor, and Staurikosaurus in that they
were likely small bipedal cursors (Fig. 8; Langer, 2004; Galton and Upchurch, 2004a).
Figure 8. Reconstructions of basal saurischian dinosaurs. A, Herrerasaurus ischigualastensis; B, Eoraptor lunensis; C, Staurikosaurus pricei. Scale bar = 25 cm. (From Langer, 2004.)
Saurischia is here coded as a composite OTU based on numerous cladistic studies that have solidified its status as a monophyletic clade (e.g., Gauthier, 1986; Novas, 1992; Holtz, 23
1994; Sereno, 1997, 1999; Rauhut, 2003; Holtz and Osmolska, 2004; Langer, 2004). Based on the majority of studies (e.g., Novas, 1993, 1997; Sereno, 1993, 1997, 1999; Sereno and Novas,
1993; Rauhut, 2003; Langer, 2004; Langer and Benton, 2006; but see Novas, 1992, Langer et al.,
1999; Fraser et al., 2002), Herrerasaurus (as a characteristic member of the Herrerasauridae) will be considered within the composite saurischian OTU.
Several synapomorphic characters that unite the Saurischia to the exclusion of the
Ornithischia based on Gauthier’s (1986) analysis include, but are not limited to, the following: little or no contact between the maxillary process of the premaxilla and nasal; temporal musculature extends onto the frontals rather than limited to just the dorsolateral surface of the parietals; posterior cervical vertebrae are elongate; axial postzygapophyses are positioned lateral to the prezygapophyses; epipophyses present on anterior cervical postzygapophyses; manus is at least 45% of the length of the upper and lower arm (humerus + radius); asymmetrical manus – digit II is the longest, whereas digit III is the longest in the ornithischian dinosaurs and other archosaurs; and pollex (digit I, the “thumb”) is robust and offset from the remainder of the manus, unlike any other archosaur manus (Fig. 9). 24
Figure 9. The manus of various archosaurs. Left manus in dorsal view of A, Crocodylus sp.; B, Heterodontosaurus tucki; C, Thecodontosaurus antiquus/Efraasia diagnostica; D, Massospondylus carinatus; E, Syntarsus rhodesiensis; F, Allosaurus fragilis; G, Struthiomimus altus; H, Deinonychus antirrhopus; I, Archaeopteryx lithographica. Note that C-I (all saurischians) have digit II as the longest as opposed to A and B (crocodylomorph and ornithischian, respectively) which have digit III as the longest. (From Gauthier, 1986.)
3.1.2. Ingroup OTUs
Cope (1866) considered the group of dinosaurs now known as ornithischians to have been descended from a single common ancestor. Since its first formal designation by Seeley
(1887), the Ornithischia has been generally accepted as a monophyletic clade (e.g., Thulborn, 25
1971a; Galton, 1972; Bakker and Galton, 1974; Bonaparte, 1976; Gauthier, 1986; Weishampel and Witmer, 1990b). The taxa described below are considered a priori unambiguous ornithischian dinosaurs based on several characters that are – coupled with a combination of other synapomorphies – derived features of ornithischians not present in other archosaurs. Table
1 lists the 12 species-only ingroup OTUs used in this analysis with their occurrence and geologic age.
Table 1. Location and geologic age of the 12 species-only ingroup OTUs. The references in parentheses following each taxon refer to the works that identified the locations of each OTU. The geologic ages presented here have been updated to reflect the most current understanding of the relative geologic time scale applied to the rock units listed.
Stratigraphic Rock Unit and Ingroup OTU Geologic Age Occurrence
Pisanosaurus mertii Ischigualasto Formation, Carnian (Casamiquela, 1967) Argentina
Eocursor parvus (Butler et. Lower Elliot Formation, South Norian al., 2007) Africa
Lesothosaurus diagnosticus Upper Elliot Formation, Lesotho Hettangian-Sinemurian (Ginsburg, 1964)
Stormbergia dangershoeki Upper Elliot Formation, South Hettangian-Sinemurian (Butler, 2005) Africa and Lesotho
Heterodontosaurus tucki Upper Elliot Formation, South Hettangian-Sinemurian (Crompton and Charig, 1962) Africa
Abrictosaurus consors Upper Elliot Formation, South Hettangian-Sinemurian (Thulborn, 1974) Africa and Lesotho
Agilisaurus louderbacki Lower Shaximiao Formation, Bathonian-Callovian (Peng, 1992) People’s Republic of China
Hypsilophodon foxii (Galton, Wessex Formation, Isle of Barremian-Aptian 1974b) Wight, UK
Thescelosaurus neglectus Lance Formation, Wyoming, Campanian-Maastrichtian (Galton, 1974a) USA 26
Scutellosaurus lawleri Kayenta Formation, Arizona, Hettangian-Sinemurian (Colbert, 1981) USA
Emausaurus ernsti (Haubold, Unknown Unit, Germany Early Toarcian 1990)
Scelidosaurus harrisonii Lower Lias, England, UK Late Sinemurian (Thulborn, 1977)
Pisanosaurus mertii
The type and only known specimen of Pisanosaurus was discovered in the Ischigualasto
Formation (Carnian) of Argentina and described by Casamiquela (1967; PVL 2577). The only known materials recovered are partially articulated skeletal pieces including: fragmentary dentary and maxilla (Figs. 10, 11); central portion of the right pelvis, including the proximal ischium and pubis as well as the femoral head (Fig. 12); the right crus (Fig. 13); and fragments of the pes and several dorsal (and possibly cervical) vertebrae. Bonaparte has produced the only published reconstruction of Pisanosaurus (1976:fig. 8).
Figure 10. Right mandible of Pisanosaurus mertii. The mandible (PVL 2577) in A, lateral view and B, medial view. Abbreviations: “emf,” putative external mandibular fenestra as reported by Sereno (1991); mfo, medial mandibular fossa. Scale bar = 1 cm. (From Irmis et al., 2007.) 27
Figure 11. Cranial and mandibular material of Pisanosaurus mertii. Fragmentary right maxilla and teeth in A, lateral view; B, medial view; C, cross-sectional view with medial surface to the left. Right mandible in D, lateral view; E, medial view; F, cross-sectional views with lateral surface to the left. Abbreviations: a, angular; ar, articular; co, coronoid; d, dentary; pd, reconstructed predentary; pra, prearticular; q, fragment of quadrate; sa, surangular; spl, splenial. Scale bar = 1 cm. (From Bonapaprte, 1976.) 28
Figure 12. Pelvic impression of Pisanosaurus mertii. A, Bonaparte’s (1976) illustration indicating a posteroventrally deflected pubis (dashed lines); B, Sereno’s (1991) illustration indicating an anteroventrally deflected pubis. Abbreviations: f, femoral head; il, ilium; is, ischium; p, pubis. Scale bar = 1 cm. (A, modified from Bonaparte, 1976; B, modified from Sereno, 1991.) 29
Figure 13. Right crus of Pisanosaurus mertii. A, tibia and astragalus in lateral view; B, distal tibia and astragalus in posterior view; C, distal tibia and astragalus in anterior view; D, fibula and calcaneum in lateral view; E, distal fibula and calcaneum in anterior view; F, astragalus and calcaneum articulated with the tibia and fibula in ventral view. Scale bar = 1 cm. (From Bonaparte, 1976.) 30
Casamiquela (1967) indicated that Pisanosaurus is the earliest known ornithischian dinosaur and placed it within the Ornithopoda as a basal member within a new clade termed
Pisanosauridae. Galton (1972) indicated that, although it exhibits several plesiomorphic ornithischian features, Pisanosaurus belonged to the hypsilophodontids whereas the basal ornithischians Fabrosaurus australis and Echinodon becklesii were members of a new family, the Fabrosauridae. Confusingly, Galton placed Pisanosaurus as the sister taxon to the hypsilophodontids (1974b), then reverted back to his original assertion (1972) and placed
Pisanosaurus as a basal member of the Hypsilophodontidae (1974a), concluding that the hypsilophodontids were derived from cursorial fabrosaurids. Bonaparte (1976) concluded that
Pisanosaurus represented a basal member of the Heterodontosauridae, despite the lack of heterodonty as exhibited in Heterodontosaurus and Lycorhinus. Weishampel and Witmer
(1990b) contend that although Pisanosaurus may share several features with heterodontosaurids such as occlusal tooth wear, until further discoveries are made these characters appear to be the product of convergent evolution. Sereno (1991), Langer (2004), Butler (2005), and Langer and
Benton (2006) considered Pisanosaurus to be the sister taxon to all known ornithischians.
The presumed presence of the predentary and buccal emargination are characters that make Pisanosaurus an ornithischian dinosaur. Casamiquela (1967), Bonaparte (1976), and Irmis et al. (2007) indicate a derived ornithischian feature, the absence of an external mandibular fenestra (contra Sereno, 1991 and Butler, 2005). The presence of an opening in the posterior mandible in Figure 10, as indicated by Irmis et al.’s (2007) personal observations of the irregular broken margins, seems to be an unnatural break (also indicated by Sereno’s [1991:174] assertion that the medial mandibular fossa maintains openings at the anterior and ventral margins – clearly 31
this putative external mandibular fenestra has fragmented further between Sereno’s observations
and Irmis et al.’s [2007] observations). Another derived ornithischian feature of Pisanosaurus is the extensive wear facets on the maxillary and dentary teeth (Bonaparte, 1976; Sereno, 1991).
Sereno (1991) and Irmis et al. (2007) both indicate that Bonaparte’s (1976) presumed
posteroventral position of the pubis (dashed line from ventral portion of the fragmented pubis in
Figure 12A) is unwarranted due to the actual preservation of the pelvic impression. Sereno
(1991) and Irmis et al. (2007) agree that the pubis was directed anteroventrally (Fig. 12B) and
lacking a prepubic process, thus making these ancestral traits of the Ornithischia. As indicated
by Sereno (1991), the distal portion of the tibia above the fibular flange is deeper anteroposteriorly than it is broad mediolaterally (Fig. 13B, C); a feature unknown in
ornithischians or basal saurischians, likely making it an autapomorphy for Pisanosaurus.
Collectively, these seemingly paradoxical characters have led some authors (Sereno,
1991; Norman et al., 2004a) to opine that the holotype may be a chimera. According to Irmis et
al. (2007) and Bonaparte’s in situ map of the preserved semi-articulated skeletal elements
(1976:fig. 1), the material assigned to the holotype (PVL 2577) is from one individual. With this
mixture of plesiomorphic and derived features, it is clear why the evolutionary placement of
Pisanosaurus has been confusing (see Norman et al., 2004a); however, its status among the
ranks of the Ornithischia can only be validly tested phylogenetically.
Eocursor parvus
Known from a single individual comprised of several cranial elements as well as a partial
postcranial skeleton, including large well-preserved appendicular elements, Eocursor is the most
well-known definitive Triassic ornithischian dinosaur (Fig. 14; Butler et al., 2007). The type and 32
only known specimen of Eocursor (SAM-PK-K8025) includes cervical and caudal vertebrae that
preserve visible neurocentral sutures, indicating that the specimen is likely a subadult (Brochu,
1996; Irmis, 2007). However, ontogenetic processes do not appear to have affected the apomorphies that diagnose Eocursor (Butler et al., 2007). Additionally, the general morphology
of Eocursor closely resembles that of other basal ornithischians such as Stormbergia,
Lesothosaurus, and Scutellosaurus; and the enlarged manus of Eocursor is similar to that of the
heterodontosaurids (Sereno [1986] indicated that an enlarged manus was a synapomorphy of the
Heterodontosauridae; however, in light of Butler et al.’s [2007] recovery of Eocursor more closely related to Lesothosaurus and all other ornithischians sans the heterodontosaurids and
Pisanosaurus, this may be a plesiomorphic feature of the Dinosauria, considering that it is present in basal saurischians such as Herrerasaurus).
Figure 14. Reconstruction of Eocursor parvus. The cranial and postcranial material shown is that of the only known specimen (SAM-PK-K8025). Scale bar = 10 cm. (From Butler et al., 2007.)
According to Butler et al. (2007), there are three autapomorphies that distinguish
Eocursor from all known ornithischian dinosaurs. Those autapomorphies are: fossa on the lateral surface of the basisphenoid that lies posterior to the canal for the carotid artery; maximum transverse expansion of the distal end of the humerus is only 50% of the maximum transverse 33
expansion of the proximal end; and the obturator foramen of the pubis is subcircular and
enlarged with respect to the proximal pubic shaft.
Eocursor possesses several unambiguous ornithischian synapomorphies (Butler et al.,
2007), indicating that the Ornithischia was well established prior to the close of the Triassic.
Several of the ornithischian synapomorphies present in Eocursor that are either ambiguous or absent in Pisanosaurus include: four definitive sacral vertebrae (likely five); well-defined rod- like prepubic process; opisthopubic pelvis; distinct anterior trochanter; pendant fourth trochanter; and expanded distal portion of the tibia.
Lesothosaurus diagnosticus
Lesothosaurus is considered one of the basalmost ornithischian dinosaurs and historically
has been considered the sister taxon to all ornithischian dinosaurs (Genasauria sensu Sereno,
1986). One character that is considered to be the ancestral state for all ornithischians that
Lesothosaurus possesses is the absence of the buccal emargination (tooth rows are marginal).
Originally, Ginsburg (1964) described a partial dentary with three preserved teeth as the
holotype for a new species of dinosaur Fabrosaurus australis (MNHN LES 9). He considered it
most closely related to Scelidosaurus harrisonii based on the similarity of the morphology of the dental characters, but regarded Fabrosaurus as a primitive member due to the smaller size of the specimen and the vertical nature of the teeth as opposed to the recumbent appearance of the
Scelidosaurus teeth.
In a series of papers, Thulborn (1970a, 1971b, 1972) described far more complete
specimens that he referred as the cranial and postcranial remains of Fabrosaurus. Based on his
analyses, he noted that the affinity to Scelidosaurus was inaccurate based on, among other 34
things, the lack of dermal armor (scutes) and the uncertain relationship of Scelidosaurus to other
ornithischians (Thulborn, 1970a) and placed Fabrosaurus in the family Hypsilophodontidae
(Thulborn, 1972). Galton (1972) noted the primitive nature of Fabrosaurus as well as several
other taxa including Echinodon becklesii and Nanosaurus rex (= Othnielia rex) in that the lateral maxilla is flat or only incipiently indented (no cheek region), thus separating them from the rest of the ornithischian dinosaurs. In recognizing this, he erected the family Fabrosauridae as the basalmost family of ornithischian dinosaurs.
Charig and Crompton (1974) noted that the only feature of the holotype of Fabrosaurus
that was characteristic of Ornithischia was the plesiomorphic appearance of the teeth. They
suggested that the genus Fabrosaurus and the species F. australis should be considered nomina
dubia and that no other specimens be referred to either the genus or species. Upon further
review of the Fabrosaurus material (both type and referred material), Galton (1978) removed
Thulborn’s (1970a, 1971b, 1972) specimens and placed them in a new genus and species
Lesothosaurus diagnosticus and kept the holotype of Ginsburg’s (1964) specimen within the
taxon Fabrosaurus australis. The major difference between Fabrosaurus and Lesothosaurus,
according to Galton (1978), was the presence of special foramina (for the access of replacement
teeth) in the material from Fabrosaurus, but not present in Lesothosaurus. However, Gow
(1981) and Sereno (1991) indicated that these special foramina were present in Lesothosaurus
(Fig. 15E) and so this feature cannot be used as an autapomorphic diagnosable character for
Fabrosaurus. Weishampel and Witmer (1990b) and Sereno (1991) considered Fabrosaurus and
F. australis nomina dubia. Sereno (1986, 1991), Gauthier (1986), and Weishampel and Witmer
(1990b) considered the family Fabrosauridae a nomen dubium based on its paraphyletic grouping
(taxa such as Echinodon and Scutellosaurus are more closely related to other taxa and 35 ornithischian clades than to Lesothosaurus) and the only material referable to it (MNHN LES 9) is here considered Archosauriformes incertae sedis (MNHN LES 9 was originally considered as
Ornithischia indet. by Gauthier [1986], Weishampel and Witmer [1990b], and Sereno [1991], but in light of new studies by Parker et al. [2005] and Irmis et al. [2007], it is here more appropriately assigned to Archosauriformes incertae sedis). In response to Galton (1978),
Weishampel and Witmer (1990b), and Sereno (1991), Thulborn (1992) reaffirmed his conclusions about Fabrosaurus australis and considered Galton’s (1978) Lesothosaurus diagnosticus an invalid junior synonym. With the exception of Peng (1992, 1997), no other known author has since used Fabrosaurus australis as the formal taxonomic name for specimens referred to Lesothosaurus diagnosticus. Peng (1992, 1997), in accordance with Thulborn (1992), disagrees with Sereno and colleagues in that he considers Fabrosaurus australis a valid taxon as well as the Fabrosauridae. This will be addressed further with the description of Agilisaurus louderbacki. Basal ornithischians are still referred to casually as ‘fabrosaurids’; however, this term is generally not used in formal nomenclature. 36
Figure 15. Reconstructed left mandible of Lesothosaurus diagnosticus. Predentary in A, lateral; B, dorsal; and C, ventral views. Mandible in D, lateral; E, medial; and F, dorsal views. Abbreviations: a, angular; ad, articular surface for dentary; ar, articular; co, coronoid; d, dentary; emf, external mandibular fenestra; imf; internal mandibular foramen; pd, predentary; pra, prearticular; sa, surangular; sp, splenial; sp.fa, special foramina. Scale bar = 1 cm. (Modified from Sereno, 1991.)
In Sereno’s (1991) revised description of Lesothosaurus, he indicated six putative autapomorphies: a slot in the maxilla for insertion of the lacrimal at the dorsal margin of the antorbital fossa (Fig. 16A, B); anterior premaxillary foramen (Fig. 16B); lateral exposure of the ventromedially angled brevis shelf (what Sereno terms the “brevis surface” [1991:172]) on the postacetabular process of the ilium (Fig. 17); dorsal groove on the proximal shaft of the ischium
(Fig 17); reduced pedal digit I (Fig. 18D); and forelimb is less than 40% the length of the hindlimb (Fig. 19). These putative autapomorphies have been shown to exist in several other basal ornithischian dinosaurs. Agilisaurus may exhibit the same maxilla-lacrimal contact (Peng,
1997) and therefore cannot be considered an unambiguous autapomorphy for Lesothosaurus.
The anterior premaxillary foramen may be present in Hypsilophodon foxii; however, this cannot 37
be corroborated due to the fragmentary nature of most ornithischian premaxillae and may yet
serve as an autapomorphy for Lesothosaurus (Butler, 2005). Peng (1992, 1997) indicates the
presence of the lateral exposure on the postacetabular process of the ilium in Agilisaurus; Butler
(2005) indicates its presence in Stormbergia dangershoeki and in Eocursor; and Sereno (1991) notes that it is present in Scelidosaurus (although he claims that the features were acquired by
Lesothosaurus and Scelidosaurus independently). As with most of these Lesothosaurus autapomorphies, the dorsal groove on the proximal ischial shaft is known in Agilisaurus (Peng,
1992, 1997), Stormbergia, and Eocursor (Butler, 2005). Butler notes that this is only weakly evident in a paratype of Stormbergia (BMNH R11000) and absent in the holotype (SAM-PK-
K1105). The reduction of pedal digit I is spread among several basal ornithischians including
Agilisaurus (Peng, 1992, 1997) and Stormbergia (Butler, 2005). A short forelimb (<40%
relative to the hindlimb) is also present in Agilisaurus (Peng, 1992, 1997), Stormbergia, and
Eocursor, and thus is not autapomorphic for Lesothosaurus (Butler, 2005).
Regardless of the character state occurrences in other basal ornithischians noted above,
Lesothosaurus is a well-established taxon. It is the combination of these characters (Sereno’s
[1991] autapomorphies) as well as the retention of several plesiomorphies (e.g., the external
mandibular fenestra) and more derived traits (e.g., the opisthopubic pelvis) that make
Lesothosaurus a distinct basal ornithischian dinosaur. 38
Figure 16. Reconstructed cranium of Lesothosaurus diagnosticus. A, left lateral; B, dorsal; C, posterior; and D, ventral views. Abbreviations: antfe, antorbital fenestra; antfo, antorbital fossa; apmf, anterior premaxillary foramen; bo; basioccipital; bs, basisphenoid; ec, ectopterygoid; eo, exoccipital; f, frontal; j, jugal; l, lacrimal; m, maxilla; n, nasal; op, opisthotic; p, parietal; pap, palpebral; pl, palatine; pm, premaxilla; pmf, premaxillary foramen; po, postorbital; prf, prefrontal; ps, parasphenoid; pt, pterygoid; q, quadrate; qj, quadratojugal; scr, sclerotic ring; so, supraoccipital; sq, squamosal; v, vomer. Scale bar = 2 cm. (Modified from Sereno, 1991.) 39
Figure 17. Reconstructed pelvis of Lesothosaurus diagnosticus. Abbreviations: ace, acetabulum; bs, brevis shelf; il, ilium; is, ischium; isg, dorsal groove on the ischial shaft; obf, obturator foramen; p, pubis. Scale bar = 1 cm. (From Sereno, 1991.) 40
Figure 18. Hindlimb material of Lesothosaurus diagnosticus. Right femur in A, lateral; B, posterior; and C, proximal views. D, left metatarsus in dorsal view. Metatarsal numbers are indicated. Abbreviations: ft, fourth trochanter; gt, greater trochanter; lt, lesser (anterior) trochanter. Scale bar = 2 cm. (From Sereno, 1991.) 41
Figure 19. Reconstruction of Lesothosaurus diagnosticus. Scale bar = 10 cm. (From Norman et al., 2004a.)
Stormbergia dangershoeki
Stormbergia is a new genus of ornithischian dinosaur discovered in the Upper Elliot
Formation (Lower Jurassic) of South Africa. The first and only published account of
Stormbergia was presented by Butler (2005). The referred specimens for Stormbergia do not include any cranial material, but do include postcranial elements from several individuals that diagnose it. According to Butler, there are no unambiguous autapomorphies that diagnose
Stormbergia; instead, there is a combination of character states present that are not known in any other basal ornithischian dinosaur. Those characters include: elongate pubic peduncle on the ilium; well-developed supraacetabular flange; ventromedially directed brevis shelf; ventral flange of the ilium partially closing the acetabulum; ischial symphysis is restricted to the distal portion; untwisted ischial shaft; well-developed tab-shaped obturator process; and a short mediolaterally flattened prepubic process (Fig. 20). 42
Figure 20. Left ilium of Stormbergia dangershoeki. The paratype (BMNH R11000) in lateral view. Abbreviations: bs, brevis shelf; isp, ischiadic peduncle; pop, postacetabular process; pp, prepubic process; prp, preacetabular process; saf, supraacetabular flange; vf, ventral flange. Scale bar = 2 cm. (From Butler, 2005.)
Lesothosaurus can be differentiated from Stormbergia by several character state differences. Lesothosaurus lacks the obturator process present in Stormbergia (Sereno, 1991;
Butler, 2005; contra Thulborn, 1972:fig. 9). Lesothosaurus ischia consistently exhibit torsion, which likely accounts for the symphysis continuing for at least 50% of the length of the ischium
(Butler, 2005). These features are phylogenetically significant enough to distinguish
Lesothosaurus and Stormbergia.
The presence of a prepubic process on Stormbergia as well as a posteroventrally deflected pubis and lateral flange on the distal tibia distinguish it from Pisanosaurus. The ventromedially angled brevis shelf differs from most other ornithischians such as the heterodontosaurids where the brevis shelf is reduced and positioned horizontally. Stormbergia can also be set apart from other basal ornithischians such as Agilisaurus by the presence of the 43
elongate pubic peduncle, supraacetabular flange, and abbreviated prepubic process (Butler,
2005). This suite of plesiomorphic and derived character states suggest an affinity with basal ornithischian dinosaurs.
Heterodontosaurus tucki
Crompton and Charig (1962) described a primitive ornithischian dinosaur from the
Stormberg Series of Basutoland (Lesotho). The Stormberg Series is the same package of rock units that bore the fossils of Eocursor, Lesothosaurus, Stormbergia, Abrictosaurus, and
Lycorhinus. This new dinosaur was termed Heterodontosaurus tucki. The generic name is based
on the heterodont dentition that the fossils exhibit. Large caniniform teeth are present in both the
premaxilla and anterior dentary. Crompton and Charig liken these caniniform teeth to “the
canines of therapsids and mammals” (1962:1075). They also indicate that the coronoid process
is well developed, unlike Lesothosaurus. Hopson (1975) considered three heterodontosaurid taxa, Heterodontosaurus, Abrictosaurus (see below), and Lycorhinus, based primarily on tooth morphology to be most closely related within the Heterodontosauridae. Thulborn (1970b) described a specimen that he concluded was so similar to Lycorhinus angustidens that it was conspecific and further suggested that these were so similar to Heterodontosaurus tucki that
Heterodontosaurus should be considered invalid and assigned the new material to L. tucki. This was later shown to be an incorrect assignment and Heterodontosaurus tucki was resurrected
(Galton, 1973; but see Abrictosaurus consors below).
Crompton and Charig (1962) listed several characters that place Heterodontosaurus
within the Ornithischia. Those characters include but are not limited to: a supraorbital
(palpebral); a predentary; a coronoid process; medially recessed maxillary teeth (buccal 44
emargination); and ridged or denticulate margins of the cheek teeth. They diagnosed
Heterodontosaurus on the following characters: large antorbital fenestra; palpebral projects into
the orbit; “carnivorous” premaxillary teeth, the third tooth distinctly larger then the first two;
expansion of the lateral process of the jugal (jugal boss); a deep diastema (gap); highly
specialized maxillary teeth; first tooth on the dentary is large and caniniform and projecting into
the diastema; and the remaining dentary teeth closely packed and similar to the maxillary teeth.
With the presence of caniniform teeth, there were some questions regarding the diet of
Heterodontosaurus (i.e., herbivorous, carnivorous, omnivorous, etc.), but the serrations do not
resemble those of the carnivorous archosaurs and along with the herbivorous nature of the cheek
teeth (leaf-shaped and denticulate), it was likely that its diet was not carnivorous; however, as
Smith (1997) points out, this does not rule out insectivory. The Heterodontosaurus manus also provides some evidence into the type of diet; it is long and gracile with laterally compressed claws that would have been useful in digging up insect burrows and/or tubers and roots
(Weishampel and Witmer, 1990a; Smith, 1997).
As mentioned above, Hopson (1975) considered three taxa within the
Heterodontosauridae. He suggested that Abrictosaurus was the most primitive of the three (see below), Lycorhinus an intermediary, and Heterodontosaurus was the most derived of the three.
Along with the autapomorphies listed by Crompton and Charig (1962), among which several have been shown to exist in other heterodontosaurids, Hopson (1975) listed several other characters to diagnose Heterodontosaurus. Those characters include, but are not limited to, the following: caniniform tooth of the dentary is serrated mesially and distally and the crowns of the dentary cheek teeth are asymmetrical (Fig. 21). Postcranial characters that diagnose
Heterodontosaurus according to Weishampel and Witmer (1990a) and Norman et al. (2004c) 45
include: a tibiofibulotarsus, which is a fusion of the tibia and fibula that forms a tibiofibula and a
fusion of the astragalus and calcaneum that forms an astragalcalcaneum (the tibiofibula and
astragalcalcaneum are subsequently fused together to form the tibiofibulotarsus); ossified epaxial
tendons on the posterior dorsals but not extending through to the caudals; absence of an obturator
process (Fig. 22); short and anteriorly squared off prepubic process (Fig. 22); and a well-
developed cnemial crest on the tibia. The tibiofibulotarsus provides the strongest evidence that
Heterodontosaurus was a habitual biped (Fig. 23; Weishampel and Witmer, 1990a; Smith,
1997).
Figure 21. Skull of Heterodontosaurus tucki. Reconstruction in right lateral view. Abbreviations: a, angular; d, dentary; f, frontal; m, maxilla; n, nasal; pap, palpebral; pd, predentary; pm, premaxilla; po, postorbital; prf, prefrontal; qj, quadratojugal; sa, surangular; sq, squamosal. Scale bar = 2 cm. (From Norman et al., 2004c.) 46
Figure 22. Pelvis of Heterodontosaurus tucki. Note the absence of the tab-shaped obturator process. Abbreviations: il, ilium; is, ischium; p, pubis. Scale bar = 10 cm. (From Weishampel and Heinrich, 1992.)
Figure 23. Reconstruction of Heterodontosaurus tucki. Scale bar = 20 cm. (From Norman et al., 2004c.)
Heterodontosaurus, due to several of its primitive characters, is considered as a basal ornithischian dinosaur. It belongs to the eponymous clade Heterodontosauridae, which, according to Norman et al. (2004c), includes three other taxa, Abrictosaurus consors, Echinodon 47
becklesii, and Lycorhinus angustidens. Heterodontosauridae is considered the basalmost group
of the Ornithopoda (Norman et al., 2004c) and may be the stem taxon to all other ornithischians
(Irmis et al., 2007, Butler et al., 2007). A recent cladistic study by Xu et al. (2006), however,
described a new ceratopsian from China, Yinlong downsi, and in the process concluded that the
Heterodontosauridae is actually the stem taxon to the Marginocephalia. They concluded that
several of the dental characters used by Norman et al. (2004c) to define the Heterodontosauridae
– enlarged premaxillary teeth, chisel-shaped crowns of the maxillary teeth with denticles
restricted to the apical third, prominent mesial and distal ridges on the maxillary teeth – can be
used to diagnose the clade comprised of Heterodontosauridae and Marginocephalia. Xu et al.
(2006) termed this clade Heterodontosauriformes and indicated that it was a node-based clade
defined as Heterodontosaurus, Triceratops, their most recent common ancestor, and all its
descendants. This potential affinity will be phylogenetically tested in this study.
Abrictosaurus consors
Thulborn (1974) described a heterodontosaurid dinosaur from the Stormberg Series of
Lesotho that he suggested was most closely associated with Lycorhinus angustidens and
therefore named the individual L. consors. The material that was described constituted portions
of the skull (Fig. 24) and some postcranial material. The specific name L. consors was given because the skull did not possess the caniniform teeth and so was presumed to be a female
(consors is derived from Latin for “spouse”). Thulborn (1974) diagnosed L. consors on the following characters: narrow nasals and frontals confined primarily to the skull roof; presence of a diastema; angular coronoid process; heterodont dentition, with the teeth in a linear series; two premaxillary teeth with no caniniform; twelve maxillary teeth; fourteen dentary teeth with no 48
caniniform; cheek teeth crowns are compressed mediolaterally and denticles are divergent at the
occlusal margin; and four unfused sacral vertebrae.
Figure 24. Skull of Abrictosaurus consors. Abbreviations: a, angular; d, dentary; f, frontal; j, jugal; l, lacrimal; m, maxilla; n, nasal; pap, palpebral; pd, predentary; pm, premaxilla; po, postorbital; prf, prefrontal; q, quadrate; qj, quadratojugal; sa, surangular; sq, squamosal. Scale bar = 1 cm. (Modified from Thulborn, 1974.)
Based on the similarity between Heterodontosaurus tucki, Lycorhinus angustidens, and
L. consors, Thulborn (1974) considered Heterodontosaurus tucki a junior synonym and reassigned Heterodontosaurus to Lycorhinus tucki, an assertion he made previously (1970b) although it was refuted by Galton (1973). Charig and Crompton (1974) compared the two holotypes of Lycorhinus angustidens and Heterodontosaurus tucki and the specimen that was referred to L. consors and concluded that L. angustidens and H. tucki were not congeneric and neither was L. consors with either of them. Hopson (1975) echoed the conclusions by Charig and Crompton (1974) and assigned L. consors to the genus Abrictosaurus (from the Greek 49
abriktos meaning “awake” to refute Thulborn’s [1974] assertion that heterodontosaurids
hibernated during dry seasons, based on the lack of evidence for replacement teeth, now known
not to be the case [Weishampel and Witmer, 1990a]). Hopson (1975) diagnosed Abrictosaurus on the following characters: cheek teeth possess relatively symmetrical crowns with narrow basal cingula and unswollen roots and the buccal side of the tooth has an apicobasally oriented ridge with the mesial and distal denticles relatively equal (Fig. 25).
Figure 25. Dentary tooth of Abrictosaurus consors. Note denticles on the mesial and distal edges. Scale bar = 5 mm. (From Weishampel and Heinrich, 1992.)
As mentioned above, Hopson (1975) indicated that Abrictosaurus is the most primitive of
three known heterodontosaurids based on the morphology of the lower teeth. Caniniform teeth
are present in one of the referred specimens of Abrictosaurus and it was shown that the
caniniforms are serrated on only the mesial portion of the tooth. He also indicated that
Lycorhinus, the intermediary among the three heterodontosaurids, has caniniform teeth that are
serrated on both the mesial and distal sides of the tooth and the cheek teeth have symmetrical 50 crowns. Heterodontosaurus, as mentioned above, is the most derived of the three in that it possesses mesial and distal serrations as well as asymmetrical cheek tooth crowns. It should be noted that the lack of caniniform teeth and the unfused sacrals that Thulborn (1974) described in the heterodontosaurids, particularly in Abrictosaurus (L. consors), may be a juvenile feature rather than sexually dimorphic (Weishampel and Witmer, 1990a).
Agilisaurus louderbacki
Peng (1992) gave a complete description of Agilisaurus, a small bipedal ornithischian dinosaur from the Shaximiao Formation of the Sichuan Province, China. It is known from one nearly complete skeleton (ZDM T6011) diagnosed by the following characters (Peng, 1992;
Barrett et al., 2005): a short and high skull (Fig. 26A; which Norman et al. [2004c] suggest may be a juvenile condition or perhaps a stage in tooth replacement shared by individuals of
Agilisaurus and Abrictosaurus); parietals constricted at the center; fossae on the maxilla and mandible; well-developed palpebral with a posterior process that attaches laterally to the postorbital; absence of an external mandibular fenestra; five premaxillary teeth with caniniforms that occlude with the three anteriormost dentary teeth, which are conical and differ in morphology from the remainder of the mandibular teeth (Fig. 26B; Barrett et al., 2005); five sacrals; ossified tendons on the dorsals and sacrals; ilium is long and possesses a supraacetabular flange and ventral flange on the anteromedial portion of the acetabulum (Fig. 27A); obturator process on the ischium; anterior trochanter on the femur is lower than the greater trochanter and separated by a cleft (Fig. 27B); and tibia is longer than the femur. 51
Figure 26. Skull of Agilisaurus louderbacki. A, illustration of the skull of Agilisaurus; B, lateral view of right mandible of Agilisaurus (ZDM T6011; note three anteriormost dentary teeth that differ from the remainder of the mandibular dentition). Abbreviations: aof, antorbital fenestra; external nares; j, jugal; l, lacrimal; lto, lateral temporal fenestra; m, maxilla; me; maxillary embayment; n, nasal; p.pr, paroccipital process; pap, palpebral; pd, predentary; pm, premaxilla; po, postorbital; prf, prefrontal; pt, pterygoid; q, quadrate; qj, quadratojugal; scl, sclerotic ring; sorb, supraorbital (fractured from palpebral?); sq, squamosal. Scale bar = 2 cm. (From Barrett et al., 2005.) 52
Figure 27. Left ilium and femur of Agilisaurus louderbacki. Lateral view of A, left ilium and B, left femur of Agilisaurus (based on ZDM 6011). Abbreviations: ft, fourth trochanter; gt, greater trochanter; lt, lesser (anterior) trochanter; saf, supraacetabular flange; vf, ventral flange. Scale bar = 4 cm. (From Peng, 1992.) 53
Peng (1992) concluded that the characters present in the holotype of Agilisaurus were most closely aligned with those of the Fabrosauridae (as defined by Galton [1972]) rather than the hypsilophodontids and so placed Agilisaurus in the Fabrosauridae. As noted above, the
Fabrosauridae has since been considered invalid due to the lack of referred taxa, but Peng (1997) resurrected the clade by including Agilisaurus and a poorly known Chinese taxon Gongbusaurus wucaiwanensis (which may not be a valid taxon [Norman et al., 2004c]) to be more closely related to each other and to Fabrosaurus australis (= Lesothosaurus diagnosticus) than to any other ornithischian. Peng (1997) based this on the following putative synapomorphies: lacrimal inserts into the apex of the maxilla; finger-like retroarticular process of the mandible; forelimb length less than 40% of the length of the hindlimb; supraacetabular flange on the ilium; ventromedially angled brevis shelf; dorsal groove on the ischial shaft; and reduced pedal digit I.
As mentioned before, these are characters that are considered ornithischian symplesiomorphies that do not support a monophyly of the Fabrosauridae (Norman et al., 2004a; Barrett et al.,
2005). Agilisaurus does exhibit characters that may exclude it from the Cerapoda, such as the supraacetabular flange, absence of a diastema, and the ventromedial acetabular flange; however,
this will be addressed phylogenetically.
Hypsilophodon foxii
Galton (1974b) published a complete account of Hypsilophodon foxii and indicated that it
had several characters in common with iguanodontids. Galton further opined that other
ornithopods such Thescelosaurus and Parksosaurus were successively more closely related to
the Iguanodontidae, thus creating a paraphyletic assemblage at the base of the iguanodontids. 54
Later studies (Sereno, 1986, 1998; Sues and Norman, 1990; Weishampel and Heinrich, 1992;
Coria and Salgado, 1996; Sues, 1997) indicated that Hypsilophodontidae is a monophyletic taxon
with Hypsilophodon as its eponymous genus with which the majority of the characters for the
clade were scored. Monophyly has since been cast in doubt (Weishampel et al., 2003; Norman
et al., 2004c), but this putative paraphyletic assemblage has not been corroborated.
Several cranial autapomorphic features identified by Galton (1974b; Fig. 28) are: five premaxillary teeth; diastema between the premaxilla and maxilla; frontals transversely narrow; vaulted palate; jugal is excluded from the margin of the antorbital fenestra; quadratojugal is large and as a result decreases the size of the lateral temporal fenestra; thickly enameled buccal side of maxillary teeth and thickly enameled lingual side of dentary teeth (Fig. 29). Some postcranial characters observed by Galton (1974b) are: length of the humerus is at least as long as the scapula; at least partially ossified sternal segments of the anterior dorsal ribs; fifteen or sixteen dorsals and five or six sacrals; slender rod-like prepubic process; small cnemial crest on the tibia; anterior trochanter is separated from the greater trochanter by a cleft; obturator process is present about halfway down the ischial shaft (Fig. 30); forelimbs are about 50% the length of the hindlimbs; and the smaller manus (relative to other basal ornithischians such as
Heterodontosaurus) suggests a habitual, or obligatory, bipedal stance (Fig. 31); Figure 31 also shows the presence of ossified hypaxial tendons beneath the proximal and central caudals, as well as covering the distal caudals. 55
Figure 28. Skull of Hypsilophodon foxii. Abbreviations: a, angular; d, dentary; j, jugal; l, lacrimal; m, maxilla; n, nasal; pap, palpebral; pd, predentary; pm, premaxilla; po, postorbital; prf, prefrontal; q, quadrate; qj, quadratojugal; sa, surangular; sq, squamosal. Scale bar = 2 cm. (From Norman et al., 2004c.)
Figure 29. Teeth of Hypsilophodon foxii. Buccal view of an unworn maxillary tooth (left) and lingual view of a worn dentary tooth (right). Scale bar = 5 mm. (From Norman et al., 2004c.) 56
Figure 30. Pelvis of Hypsilophodon foxii. Abbreviations: il, ilium; is; ischium; obp, obturator process; p, pubis; prp, prerpubic process. Scale bar = 4 cm. (From Sues and Norman, 1990.)
Figure 31. Reconstruction of Hypsilophodon foxii. Scale bar = 20 cm. (From Norman et al., 2004c.)
Thescelosaurus neglectus
Galton (1974a) redescribed the genus Thescelosaurus and attempted to resolve ornithopod phylogeny based primarily on the hindlimb proportions. Thescelosaurus was 57 originally split into two species, T. neglectus and T. edmontonensis, based primarily on variation between the cervicals (expansion and flattening of the ventral surface) and in the thickness of the preacetabular process of the ilium. Galton’s observations concluded that the variation seen in the cervicals was likely due to lack of differentiation between the anterior and posterior cervicals – posterior cervicals possessing a higher degree of ventral expansion than the anterior cervicals – and the variation noted in the preacetabular process of the ilium is probably a character that varies intraspecifically, as it does in Hypsilophodon. After Galton’s reevaluation, he regarded T. edmontonensis as a junior synonym to T. neglectus (T. neglectus was published first, and thus has priority).
Several of the cranial characters listed by Galton (1974a; Fig. 32) that diagnose
Thescelosaurus are: wide frontals relative to Hypsilophodon; postorbital is firmly attached to the frontal; and at least four premaxillary teeth. Postcranial characters include: humerus is longer than the scapula; deltopectoral crest is less robust than that of Hypsilophodon; rod-like prepubic process as in Hypsilophodon; broadly horizontal brevis shelf; inner condyle of the femur is larger than the outer condyle; fourth trochanter on the femur is located farther down the femoral shaft than it is in Hypsilophodon; tibia is shorter than the femur and possesses a robust cnemial crest. 58
Figure 32. Reconstruction of the skull of Thescelosaurus neglectus. Abbreviations: a, angular; d, dentary; f, frontal; pap, palpebral; po, postorbital. Scale bar = 2 cm. (From Norman et al., 2004c.)
As Galton (1974a) noted, most previous authors had placed Thescelosaurus within the
Hypsilophodontidae. However, upon reevaluation, Galton concluded that it belonged within the
Iguanodontidae as the most basal member due to its robust limb structures and affinities with other graviportal ornithopods that lack the cranial specialties of the hadrosaurids. However, Sues and Norman (1990) indicate that the affinities of Thescelosaurus are more closely aligned with hypsilophodontids than any other clade and indicate that the prepubic process is not rod-like as in Hypsilophodon (contra Galton, 1974a). Furthermore, Weishampel et al. (2003) concluded that there are no synapomorphies to support a monophyletic Hypsilophodontidae, indicating that
Hypsilophodon is the most primitive member of the Euornithopoda (sensu Weishampel, 1990); leading successively to Iguanodontia are Gasparinisaura cincosaltensis (contra Coria and
Salgado, 1996), Thescelosaurus neglectus, and Zalmoxes sp. Norman et al. (2004c) concurred 59
with Weishampel et al.’s (2003) assertion that there are no synapomorphies to support a
monophyletic hypsilophodontid clade and that Thescelosaurus is more closely related to the
Iguanodontia than it is to Hypsilophodon.
Iguanodontia
Iguanodontia is the well-supported stem-based taxon name given to all euornithopods
more closely related to Edmontosaurus than to Thescelosaurus (Norman, 2004; amended from
Sereno, 1998) and diagnosed by the following synapomorphies (Norman, 1984; Sereno, 1984,
1986; Forster, 1990; Norman and Weishampel, 1990; Weishampel and Horner, 1990; Norman,
2004, Horner et al., 2004; Novas et al., 2004): eversion of the labial margin of the premaxilla; absence of premaxillary teeth; smooth margin of the predentary in dorsal view; denticulate predentary; parallel dorsal and ventral borders of the dentary ramus (Fig. 33); antorbital fenestra within the antorbital fossa is small or absent (Fig. 33); enamel absent from lingual side of the maxillary teeth and buccal side of the dentary teeth; reduced to three phalanges in digit III of the manus (Fig. 34); low postacetabular process with a prominent brevis shelf (Fig. 35); and the anterior intercondylar groove on the femur is weak and the posterior intercondylar groove is deep. 60
Figure 33. Reconstructions of the skulls of A, Tenontosaurus tilletti; B, Zalmoxes robustus; C, Camptosaurus dispar. Abbreviations: a, angular; d, dentary; f, frontal; j, jugal; l, lacrimal; m, maxilla; n, nasal; pap, palpebral; po, postorbital; prf, prefrontal; q, quadrate; qj, quadratojugal; sa, surangular; sq, squamosal. Scale bar = 5 cm. (From Norman, 2004.)
Figure 34. Manus of Tenontosaurus tilletti. Left manus in dorsal view. A, metatarsals and B, phalanges. Numerals refer to digit count. Abbreviation: un, ungual claw. Scale bar = 5 cm. (Modified from Forster, 1990.) 61
Figure 35. Pelvis of A, Tenontosaurus tilletti and B, Dryosaurus altus. Abbreviations: il, ilium; is, ischium; p, pubis. Scale bar = 10 cm. (Modified from Norman, 2004.)
The basalmost known iguanodontian, Tenontosaurus tilletti, possesses the aforementioned synapomorphies (Forster, 1990), and therefore will be the main focus for scoring characters for the Iguanodontia. Tenontosaurus possesses several autapomorphies, including: enlarged nares relative to orbit; prepubic process that is deeper dorsoventrally than it is mediolaterally; expanded postacetabular process; and increased cervical and caudal vertebral count. Tenontosaurus does still possess ossified hypaxial tendons (as with the basal ornithopods such as Hypsilophodon and Thescelosaurus) but shares more characters with the remaining iguanodontians and has been well supported in numerous cladistic analyses as the basalmost iguanodontian (e.g., Sereno, 1986, 1997, 1999; Forster, 1990, 1997; Weishampel and Heinrich,
1992; Coria and Salgado, 1996; Weishampel et al., 2003; Norman et al., 2004c; Norman, 2004).
As shown in Figure 33, the lacrimal does not contact the premaxilla in either Tenontosaurus or
Zalmoxes (Fig. 33A, B), but does so in all other iguanodontians (Forster, 1990, Weishampel et al., 2003). In most derived iguanodontians the palpebral is either lost or is incorporated into the 62
orbital margin; however, the palpebral in the basal iguanodontians shown in Figure 33 is located ventrolaterally and enters the orbit, as in basal ornithischians (Horner et al., 2004).
Based on the phylogenetic affinities of Tenontosaurus tilletti with the remainder of the
iguanodontian members, and its fairly derived form relative to Hypsilophodon, Thescelosaurus,
and the other aforementioned basal ornithischians (absence of premaxillary teeth, three
phalanges in manual digit III, asymmetrically enameled cheek teeth, etc.), it is appropriate to
code Tenontosaurus within the Iguanodontia. Iguanodontia will here be coded as a composite
OTU for this analysis.
Pachycephalosauria
Pachycephalosauria is one half of the clade Marginocephalia. The Marginocephalia was
first described by Sereno (1986) and diagnosed by the following synapomorphies (Sereno, 1984,
1986; 1989; Currie and Padian, 1997): parietosquamosal shelf; short posterior premaxillary
palate; short postpubic process (pubic shaft), which results in the loss of the pubic symphysis.
Since its formal recognition, the Marginocephalia has been relatively well supported. However,
recently this monophyletic clade has come under scrutiny. Sullivan (2006) claims that the main
diagnosing character – the parietosquamosal shelf that obscures the occiput in dorsal view – may
be secondarily derived in pachycephalosaurians; however, he does not test this hypothesis
phylogenetically and it is here considered purely speculation. Conversely, in a phylogenetic
study of a new basal ceratopsian from China, Xu et al. (2006) indicate that the clade
Marginocephalia is strongly supported based on characters that were formerly used to diagnose
the Pachycephalosauria. Those characters, present in the basalmost known ceratopsian Yinlong 63
downsi, include: an angular jugal; broad flat dorsal margin of the temporal bar; postorbital- squamosal tubercles; and anteroposteriorly compressed basal tubers (Xu et al., 2006).
The obligatorily bipedal Pachycephalosauria was first formally described by Maryańska
and Osmólska (1974) as being taxonomically synonymous with the Pachycephalosauridae.
However, Sereno (2000) indicated that there are basal taxa that are placed outside of the
Pachycephalosauridae such as Homalocephale calathocercos, Goyocephale lattimorei, and
Wannanosaurus yansiensis. The Pachycephalosauria therefore is a stem-based taxon defined as
all marginocephalians more closely related to Pachycephalosaurus than to Triceratops.
Synapomorphies that diagnose the Pachycephalosauria (Maryańska and Osmólska, 1974; Sereno,
1989, 2000; Maryańska, 1990; Maryańska et al., 2004; Sullivan, 2006) include: cranial (Fig. 36)
– thickened skull roof; flat or incipiently to fully-domed frontoparietals; loss of posterolateral
flanges on the parietals; frontal excluded from the orbital margin; two supraorbitals (palpebrals);
broad expanded squamosal shelf; and postcranial (Fig. 37) – pubis nearly to fully excluded from
acetabulum; elongate sacral ribs; distal expansion of preacetabular process of the ilium; medial
process on the iliac blade; “basketwork” of ossified tendons on the distal portion of the caudals.
Pachycephalosauria also retains several plesiomorphies for ornithischians (Maryańska et al.,
2004) such as: short premaxilla with associated teeth (Fig. 36); heterodont dentition; low coronoid process; and lack of an obturator process on the ischium (Fig. 37). 64
Figure 36. Reconstructions of the skulls of A, Prenocephale prenes and B, Stegoceras validum. Note the thickened, fused frontoparietals, posterior extension of the parietosquamosal shelf, and the presence of a second supraorbital (palpebral). Abbreviations: ect, ectopterygoid; exo, exoccipital; f, frontal; j, jugal; l, lacrimal; m, maxilla; n, nasal; p, parietal; pm, premaxilla; po, postorbital; prf, prefrontal; pt, pterygoid; q, quadrate; qj, quadratojugal; so1, first supraorbital; so2, second supraorbital; sq, squamosal. Scale bar = 10 cm. (From Maryańska et al., 2004.)
Figure 37. Sacropelvic region of Homalocephale calathocercos. A, ventral view of the sacrum and B, left lateral view of the pelvis. Abbreviations: ca1, first caudal vertebra; dv, dorsal vertebra; il, ilium; ilmp, medial process on the iliac blade; is, ischium; p, pubis; prp, prepubic process; rs1, first sacral rib; s1, first sacral vertebra; s6, sixth sacral vertebra. Scale bar = 5 cm. (From Maryańska et al., 2004.)
65
Though not entirely well known due to their very fragmentary fossil record, from the
characters that are preserved and have been coded, the pachycephalosaurians are considered a
well-established monophyletic clade and will be coded as a composite OTU here. They are also
sufficiently derived from and biostratigraphically distant from any putative basal ornithischian
(Sereno, 1986, 1989, 2000; Maryańska, 1990; Maryańska et al., 2004; Sullivan, 2006) to have
any impact on the phylogenetic affinities of basal ornithischians.
Ceratopsia
The other half of the clade Marginocephalia is the Ceratopsia. The Ceratopsia is a stem-
based taxon defined as all marginocephalians closer to Triceratops than to Pachycephalosaurus
(Sereno, 1998) and consists of two major clades, the Psittacosauridae and Neoceratopsia, plus the basal members Yinlong and Chaoyangsaurus (Xu et al., 2006).
The psittacosaurids are a primitive stem taxon to the ceratopsians. They are bipedal,
possibly facultatively bipedal, and possess a rudimentary parietosquamosal frill, a feature that
becomes greatly enlarged in more derived ceratopsians (Fig. 38; Sereno, 1986, 1989, 1990;
Dodson, 1990, 1997a, 1997b; You and Dodson, 2004). The Psittacosauridae is a monogeneric
clade consisting of at least eight different species (Zhou et al., 2006). The autapomorphies of the clade include (Sereno, 1990): rounded rostral bone; preorbital portion of the skull is less than
40% of the total skull length; elevated external nares; lateral surface of the snout is formed by the
premaxilla; premaxilla, jugal, lacrimal, and maxilla converge on one sutural point; conspicuous
absence of the antorbital fenestra and antorbital fossa. Sereno (1990) notes that the postcrania of
psittacosaurids are very primitive in nature and closely resemble that of Hypsilophodon. 66
Figure 38. Reconstruction of A, Psittacosaurus mongoliensis and B, Pentaceratops sternbergii. Scale bar = 20 cm for A and 1 m for B. (A, from Sereno, 1989; B, from Dodson, 1997b.)
The more derived neoceratopsians, with familiar names such as Triceratops, have a more pronounced frill than the psittacosaurids dominated by the parietals and also become obligate quadrupeds (Dodson and Currie, 1990; Dodson, 1997b; Dodson et al., 2004). Neoceratopsia is a 67
stem-based taxon defined as all ceratopsians more closely related to Triceratops than to
Psittacosaurus (Sereno, 1998). Neoceratopsia is diagnosed by the following autapomorphies
(Dodson and Currie, 1990; Dodson, 1997b; Sereno, 2000; Xu, et al., 2002; You and Dodson,
2003): skull length greater than 20% of the postcranial skeleton; keeled rostral; reduced
retroarticular process (relative to psittacosaurids and pachycephalosaurians); premaxilla wider
than it is high; robust forelimbs; loss of manual digit V.
The ceratopsians (Yinlong + Chaoyangsaurus + psittacosaurids + neoceratopsians) are
diagnosed by the following synapomorphies (Maryańska and Osmólska, 1975; Sereno, 1984,
1986, 1989, 1990, 2000; Dodson, 1990, 1997a, 1997b; Xu et al., 2002; You and Dodson, 2003,
2004; Zhou et al., 2006; Xu et al., 2006): a neomorphic rostral bone; high vaulted palate;
pentangular skull in dorsal view; tall snout emphasized by a large premaxilla; at least an
incipiently formed parietosquamosal frill; immobile mandibular symphysis; maxilla at least two-
thirds as tall dorsoventrally as it is long anteroposteriorly; well-developed outward lateral
expansion of the jugal (jugal boss); six to eight sacrals; thin strap-like preacetabular process of
the ilium; absence of obturator process on the ischium (Fig. 39 illustrates some of the cranial
features of basal ceratopsians). As mentioned above, the basalmost known ceratopsian, Yinlong,
shares several dental synapomorphies with the Heterodontosauridae (Xu et al., 2006). Other
affinities with heterodontosaurids that Xu et al. listed are: a midline fossa in the nasal; jugal contributes to the margin of the antorbital fossa; and a postorbital ridge running along the jugal
process. 68
Figure 39. Skull of Psittacosaurus mongoliensis. A, right lateral view and B, dorsal view. Abbreviations: a, angular; d, dentary; f, frontal; j, jugal; l, lacrimal; m, maxilla; n, nasal; p, parietal; po, postorbital; prf, prefrontal; q, quadrate; qj, quadratojugal; r, rostral; sq, squamosal. Scale bar = 2 cm. (Modified from You and Dodson, 2004.)
As with Pachycephalosauria, the Ceratopsia has been well established by myriad phylogenetic analyses. The Psittacosauridae retain some plesiomorphic features, especially within the postcrania, as noted by Sereno (1990). Regardless, the psittacosaurids have been shown to be a stem taxon to the neoceratopsians, together along with the basal members such as
Yinlong, forming the monophyletic clade Ceratopsia. Ceratopsia will be considered a composite
OTU for this analysis.
Scutellosaurus lawleri
A primitive bipedal (possibly facultatively bipedal) ornithischian, Scutellosaurus is considered the basalmost member of the clade termed Thyreophora that includes the more exclusive clades Ankylosauria and Stegosauria. When Scutellosaurus was first described (Fig.
40; Colbert, 1981) it was placed within the Fabrosauridae, as Lesothosaurus was the closest known dinosaur to resemble (phenetically) the material. Norman (1984), Sereno (1984, 1986),
Cooper (1985), and Gauthier (1986) were the first to place Scutellosaurus in the Thyreophora 69
cladistically, indicating that the sister taxon to the Thyreophora (including Scutellosaurus) was
Heterodontosaurus (Rosenbaum and Padian, 2000). Scutellosaurus has been well established as
a basal thyreophoran since that time.
Figure 40. Reconstruction of Scutellosaurus lawleri. Scale bar = 20 cm. (From Colbert, 1981.)
One of the more salient features of thyreophorans is the presence of osteoderms, or scutes. However, noting that some aetosaurian archosaurs possess scutes as well as some derived titanosauriform sauropod and ceratosaurian theropod dinosaurs, scutes cannot be a sole diagnosing apomorphy for the thyreophorans. Colbert (1981), in his initial description of
Scutellosaurus, correctly recognized the presence of these scutes as well as other dinosaurian affinities; however, he did not recognize the thyreophoran relationship. The Thyreophora is diagnosed by eight unambiguous synapomorphies (Sereno, 1986): dentary tooth row sinuous in lateral view; frontal is excluded from the orbital margin due to the incorporation of one of the supraorbitals (palpebral); medial quadrate condyle significantly more robust than the lateral condyle; basisphenoid significantly shorter relative to the basioccipital; median palatal keel with the pterygoid and vomer markedly tall anterodorsally; jugal orbital bar with a transversely broad orbital rim; parasagittal row of keeled scutes on dorsal body surface; and lateral row of keeled scutes on body surface. The last three synapomorphies listed are present in Scutellosaurus; the absence of the first five is due to incomplete remains of Scutellosaurus skull material. 70
Rosenbaum and Padian (2000) report that skull material from the Kayenta Formation (UCMP
130580) exhibits similar features to the skull of the basal thyreophoran Emausaurus described by
Haubold (1990); particularly the medial and lateral condyles of the quadrate, the jugal orbital
bar, and the size of the basisphenoid relative to the basioccipital. Rosenbaum and Padian (2000)
indicate that the sinuous curve to the dentary tooth row and the incorporation of the supraorbital
into the orbital margin are definitely absent in Scutellosaurus, uniting more derived
thyreophorans (Thyreophoroidea sensu Sereno, 1986) to the exclusion of Scutellosaurus.
Scutellosaurus, due to its primitive nature, is considered here as an OTU to be coded and used in
this phylogenetic analysis.
Emausaurus ernsti
Haubold (1990) described the partial remains of Emausaurus from Germany. The
material referred to Emausaurus consists of a well-preserved skull (Fig. 41) and some fragmentary associated postcrania including scutes. Several characters that unite Emausaurus
with the other thyreophorans, in particular Scelidosaurus, are (Haubold, 1990; Norman et al.,
2004b): box-like posterior portion of the skull; five premaxillary teeth; laterally extensive
antorbital fossa with a small antorbital fenestra located posteriorly within it; well-developed
buccal emargination due to a laterally expanded maxillary shelf; twenty-one tooth positions in
the maxilla; large palpebral; long jugal; medial quadrate condyle is more robust than the lateral
condyle; ‘W’-shaped contact between jugal and quadratojugal; slightly sinuous curve to the
dentary in dorsal view; and the presence of an external mandibular fenestra. 71
Figure 41. Skull of Emausaurus ernsti. Abbreviations: a, angular; aofe, antorbital fenestra; antorbital fossa; d, dentary; emf, external mandibular fenestra; f, frontal; j, jugal; l, lacrimal; m, maxilla; mandibular groove; n, nasal; pap, palpebral; pd, predentary; pm, premaxilla; po, postorbital; prf, prefrontal; q, quadrate; qj, quadratojugal; sa, surangular; sq, squamosal; sym, mandibular symphysis. Scale bar = 5 cm. (From Haubold, 1990.) 72
There are no diagnostic features of the postcranial remains described by Haubold (1990)
with the exception of the scutes. However, based on the morphology of the skull, Emausaurus
traditionally has been considered as a basal member of the thyreophorans due to the several
plesiomorphies that it shares with basal ornithischians and the derived nature of several of its
cranial characters and the presence of scutes. Emausaurus will be coded as an OTU for this
analysis.
Scelidosaurus harrisonii
Scelidosaurus was the first reasonably complete dinosaur ever uncovered and described
(Norman, 2000, 2001). Owen (1861, 1863) was the first to describe the Early Jurassic dinosaur
and it has since been considered a relative of the stegosaurs and/or ankylosaurs. However, in a
reevaluation of the referred material, Thulborn (1977) reclassified Scelidosaurus within the
Ornithopoda. Aside from Owen’s (1861, 1863) initial description, Thulborn’s (1977) account is
the best known description of all the characters of Scelidosaurus. The characters that diagnose
Scelidosaurus (Thulborn, 1977; Galton, 1985; Coombs et al., 1990; Martill et al., 2000; Norman,
2000, 2001; Norman et al., 2004b) include: cranial – box-like posterior portion of the skull
(similar to Emausaurus); relatively long premaxilla; maxilla with an expanded lateral shelf that creates a distinct buccal emargination; the maxilla possesses nineteen tooth positions; jugal contacts the quadratojugal with two finger-like projections; medial quadrate condyle larger than the lateral quadrate condyle; fused parietals; zigzag suture pattern for the frontoparietal region; a
relatively low coronoid process with respect to other, more derived ornithischians; sinuous curve
to the dentary in dorsal view; seventeen tooth positions in the dentary; lack of an external 73
mandibular fenestra; and postcranial – humerus is thick and has a large deltopectoral crest;
hindlimb is more robust than that of Scutellosaurus (Fig. 42); as in Scutellosaurus, the ischium is
rod-like and lacks an obturator process (Fig. 43); anterior trochanter is well separated from the
greater trochanter; and the presence of scutes.
Figure 42. Hindlimbs of Scutellosaurus lawleri and Scelidosaurus harrisonii. Anterior view of A, Scutellosaurus, and B, Scelidosaurus. Numbers refer to: 1, femur; 2, crus; 3, pes. Scale bar = 20 cm. (Modified from Norman et al., 2004b.) 74
Figure 43. Pelvis of A, Scutellosaurus lawleri, and B, Scelidosaurus harrisonii. Abbreviations: il, ilium; is, ischium; p, pubis. Scale bar = 5 cm (A), 20 cm (B). (From Norman et al., 2004b.)
Scelidosaurus is a relatively large (up to four meters) graviportal basal thyreophoran (Fig.
44) and is considered the most derived of the three known basal members. Its position within the
Thyreophora is a matter of confusion. The placement of Scelidosaurus in the Thyreophora but not within the Stegosauria has been well established (Galton, 1985; contra Thulborn, 1977); however, its affinities with the Ankylosauria have been debated (Thulborn, 1977; Galton, 1985;
Sereno, 1986; Coombs et al., 1990; Norman, 2000, 2001; Carpenter, 2001; Norman et al., 2004b;
Maidment et al., 2006). Scelidosaurus is generally considered a more derived thyreophoran than either Scutellosaurus or Emausaurus, but whether it is the sister taxon to the Eurypoda
(Stegosauria + Ankylosauria) or the sister taxon to Ankylosauria has yet to be well established. 75
Figure 44. Reconstruction of Scelidosaurus harrisonii. Scale bar = 1 m. (From Norman et al., 2004b.)
Stegosauria
The stegosaurs are known as the plated dinosaurs due to the parasagittal rows of scutes that have expanded dorsally into either plates or spines (Fig. 45). The stegosaurs are grouped with the ankylosaurs in the Eurypoda (“broad foot”) due to the elephantine-like manus and pes
(Galton, 1997). The stegosaurs typically have small heads relative to their bodies (compared with other ornithischians), short and massive forelimbs, and long columnar hindlimbs, indicating their likely graviportal nature (Galton, 1990). The stem-based Stegosauria is defined as all thyreophorans more closely related to Stegosaurus than to Ankylosaurus (Sereno, 1998). 76
Figure 45. Reconstructions of various stegosaurs. A, Stegosaurus sp., B, Huayangosaurus taibaii, C, Kentrosaurus ethiopicus. Scale bar = 50 cm. (From Norman et al., 2004b.)
77
Sereno and Dong (1992) reevaluated the basal stegosaur Huayangosaurus taibaii and performed a cladistic analysis on the Stegosauria. They revised the diagnosis of the basalmost known stegosaur on the following autapomorphies: overall skull is deeper than other stegosaurs
(Fig. 46); oval depression between the premaxilla and maxilla; small horn core on the dorsal surface of the postorbital; high tooth count in the maxilla; anterior dorsal ribs with intercostal flanges; and a coossified carpal block. Huayangosaurus has seven premaxillary teeth, more than any other known ornithischian dinosaur (Lesothosaurus has six); all other stegosaurs lack premaxillary teeth. Sereno and Dong (1992) also used the reported absence of ossified epaxial tendons in Huayangosaurus as a uniting feature of Huayangosaurus + Stegosauridae and thus a synapomorphy for Stegosauria. However, Maidment et al. (2006) indicate that aside from the autapomorphies listed by Sereno and Dong (1992), Huayangosaurus, unlike all other stegosaurs
(Stegosauridae), retains several plesiomorphies such as: premaxillary teeth; poorly developed supra- and postacetabular iliac processes; and ossified epaxial tendons. These symplesiomorphies more exclusively unite the Stegosauridae to the exclusion of
Huayangosaurus, further solidifying its status as the basalmost known stegosaur.
With the recent revision of Stegosauria, the following characters are considered
synapomorphic (Galton, 1990; Sereno and Dong, 1992; Galton and Upchurch, 2004b; Maidment
et al., 2006): oval fossa on the pterygoid ramus of the quadrate is absent; compressed proximal
head of the quadrate; dorsal neural arches are at least 1.5 times as high the associated centra;
middle dorsal transverse processes are angled at least 50 degrees to the horizontal (Fig. 47A, B,
C); end of the anterior caudal neural spine is flattened anteroposteriorly; scapular plate is larger than the coracoid (Fig. 47D); presence of a parascapular spine; triceps tubercle and descending ridge posterolaterally to the deltopectoral crest (Fig. 48); ulnare and intermedium are fused; 78
proximal carpals are large and block-shaped; distal carpals are either absent or fail to ossify;
prepubis:pubis ratio is higher than 0.4 (Fig. 49); pedal digit I is lost; pedal digits III and IV have
no more than three phalanges; and the presence of parasagittal plates and/or spines.
Figure 46. Skull of Huayangosaurus taibaii. Abbreviations: a, angular; antfe, antorbital fenestra; antfo, antorbital fossa; asor, anterior supraorbital; d, dentary; emf, external mandibular fenestra; j, jugal; l, lacrimal; lf, lacrimal foramen; m, maxilla; msor, medial, supraorbital; n, nasal; p, parietal; pd, predentary; pm, premaxilla; po, postorbital; poh, postorbital horn; popr, paroccipital process; prf, prefrontal; psor, posterior supraorbital; pt, pterygoid; q, quadrate; qj, quadratojugal; sa, surangular; sp, splenial; sq, squamosal. (From Sereno and Dong, 1992.) 79
Figure 47. Postcrania of Stegosaurus armatus. Dorsal vertebrae in A, anterior, B, lateral, C, posterior views. D, left scapula in lateral view. Scale bar = 10 cm. (From Sereno and Dong, 1992.)
Figure 48. Left humerus of Stegosaurus armatus. Humerus in A, anterior, B, lateral, and C, posterior views. Abbreviations: dr, descending ridge; tt, triceps tubercle. Scale bar = 10 cm. (From Sereno and Dong, 1992.) 80
Figure 49. Pelvis of Stegosaurus sp. Abbreviations: il, ilium; is, ischium; p, pubis; prp, prepubic process. Scale bar = 20 cm. (From Norman et al., 2004b.)
Stegosauria has been well established on the diagnosis of the above mentioned
synapomorphies through many cladistic analyses. The group as a whole, including
Huayangosaurus, is sufficiently derived from the basal ornithischian bauplan to not warrant
further dissection here, as it is beyond the scope of this study. Because of this, Stegosauria is
here coded as a composite OTU.
Ankylosauria
Ankylosaurs are the quadrupedal armored dinosaurs that have scutes arranged not only
parasagittally, but also laterally on the dorsal surface of the body. The ankylosaurs have been
long thought of as a monophyletic group and likely closely allied with the stegosaurs, though this
hypothesis was not formally corroborated until the employment of cladistics within dinosaur paleontology (e.g., Sereno, 1986, 1999; Sereno and Dong, 1992; Carpenter, 2001; Vickaryous et 81 al., 2004). As defined by Sereno (1998), the Ankylosauria is a stem-based taxon defined as all thyreophorans closer to Ankylosaurus than Stegosaurus. As stated earlier, the placement of
Scelidosaurus among the thyreophorans is a matter of debate and has been since its description by Owen (1861, 1863). Some authors have placed Scelidosaurus as the sister taxon to the eurypods (e.g., Sereno, 1986, 1999; Vickaryous et al. 2004) and others have placed it within
Eurypoda as the sister taxon to the ankylosaurs (e.g., Norman, 1984; Carpenter, 2001).
Maidment et al. (2006) claim that two characters that Carpenter (2001) used to include
Scelidosaurus in the Ankylosauromorpha (Scelidosaurus + Ankylosauria) are also present in
Huayangosaurus. Those two characters are the horizontal expansion of the ilium and the postpubic shaft backing the obturator foramen. Another character, the presence of ‘U’-shaped rings in the cervicals, was shown to not have a phylogenetic impact on the putative ankylosaurian affinities of Scelidosaurus and so must either be plesiomorphic for the eurypods and lost in the stegosaurs, or gained independently in both the ankylosaurs and Scelidosaurus.
As mentioned above, Scelidosaurus affinities will be tested in this analysis.
The Ankylosauria can be diagnosed based on the following synapomorphies (Coombs and Maryańska, 1990; Carpenter et al, 1998; Carpenter, 2001; Xu et al., 2001; Vickaryous et al.,
2004): skull low and wide; ornamentation of the skull across the rostral region and lateral surface of the angular (Figs. 50, 51); pterygoid obscures passage between spaces above the palate and below the braincase; quadratojugal contacts the postorbital; lack of antorbital, supratemporal, and external mandibular fenestrae; multiple parasagittal and lateral rows of scutes; imperforate acetabulum; synsacral region comprised of coossified dorsal, sacral, and caudal vertebrae (Fig.
52); pubis nearly excluded from acetabulum; and absence of the prepubis. 82
Figure 50. Skull of Euoplocephalus tutus. A, left lateral view of the cranium, and B, right lateral view of mandible. Scale bar = 10 cm. (From Rybczynski and Vickaryous, 2001.) 83
Figure 51. Skulls of various ankylosaurs. Dorsal view of the skulls of A, the nodosaurid, Panoplosaurus mirus, and the ankylosaurids B, Saichania chulsanensis, and C, Ankylosaurus magniventris. Scale bar = 10 cm. (From Vickaryous et al., 2004.) 84
Figure 52. Synsacrum of nodosaurid and ankylosaurid ankylosaurians. A, the nodosaurid, Struthiosaurus languedocenis, and B, the ankylosaurid, Euoplocephalus tutus. Abbreviations: ac, acetabulum; il, ilium; is, ischium; posac, postsacral rod; presac, presacral rod. Scale bar = 10 cm. (From Vickaryous et al., 2004.) 85
Figure 53. Reconstructions of ankylosaurid and nodosaurid ankylosaurians. A, an ankylosaurid, Tarchia gigantea, and B, a nodosaurid, Sauropelta edwardsorum. Note that Sauropelta does not maintain a bony club on the end of the tail. Scale bar = 1 m. (From Carpenter, 1997.)
Historically, the Ankylosauria has been divided into two major groups, the
Ankylosauridae and the Nodosauridae (Fig. 53; Coombs, 1978). The most basic distinction between the two clades is that the ankylosaurids have bony clubs comprised of scutes on the distal caudals and the nodosaurids do not. The nodosaurids are generally considered the more primitive of the two groups and reached their highest diversity earlier in the Cretaceous than did the ankylosaurids. Conversely, the ankylosaurids reached the apex of their diversity later in the
Cretaceous (Coombs and Maryańska, 1990; Carpenter, 1997; Vickaryous et al., 2004; Taylor,
2006; Butler et al., 2006). The two clades of ankylosaurs are unambiguously more closely 86 related to one another than any other taxon and thus constitute a monophyletic clade and will be treated as such here. The Ankylosauria will be coded as a composite OTU. 87
4. METHODS
4.1. Characters and Character States
Appendix 1 lists each of the 97 characters and character states selected for the phylogenetic analysis. All characters listed have been culled, modified, and/or rescored from the following previously published analyses: Norman (1984, 2004); Sereno (1984, 1986, 1991, 1997,
1999, 2000); Cooper (1985); Forster (1990); Sereno and Dong (1992); Weishampel and Heinrich
(1992); Coria and Salgado (1996); Carpenter (2001); Weishampel et al. (2003); Dodson et al.
(2004); Galton and Upchurch (2004b); Horner et al. (2004); Maryańska et al. (2004); Norman et al. (2004b, c); Novas et al. (2004); Vickaryous et al. (2004); You and Dodson (2004); Butler
(2005); Barrett et al. (2005); Maidment et al. (2006); Langer and Benton (2006); Xu et al. (2002,
2006); and Butler et al. (2007). Each character, where applicable, is scored numerically as: (0), the presumed plesiomorphic or ancestral state; (1), the more derived state; and (2) or (3) for any character that has more than one derived state (i.e., characters 1, 4, 5, 52, 69, and 84). Characters not scored numerically (i.e., ?, -, M) are described below. The characters were divided into five separate categories corresponding to the associated anatomical regions. Those five categories are: characters 1-15, dental – all characters listed refer to any morphological variation among the teeth of the taxa chosen; characters 16-54, cranial – each character is located in the skull or mandible; characters 55-60, postcranial (axial) – each character is located on the vertebral column or is an extension of it (i.e., character 60, ossified sternal ribs); characters 61-94, postcranial (appendicular) – characters that include all morphological variation among the pectoral and pelvic girdles and the fore- and hindlimbs; and characters 95-97, other characters – these include accessory ossifications such as epaxial (character 95) and hypaxial (character 96) 88
tendons and dermal scutes (character 97). It should be noted that of the appendicular characters,
only four (61-64) refer exclusively to the forelimbs. This may be due to preservation issues with
the known taxa or, more likely, to the lack of phylogenetic information that the forelimbs contain.
4.2. Taxon-Character Matrix
The 19 OTUs listed in Section 3.1 and the 97 characters and character states noted in
Appendix 1 are presented in the data matrix in Appendix 2. The character states were culled and/or rescored from the literature associated with each OTU, including: Thulborn (1970a,
1971b, 1972, 1974, 1977, 1992); Galton (1974a, b, 1978, 1985); Hopson (1975); Bonaparte
(1976); Norman (1984, 2004); Sereno (1984, 1986, 1991, 1997, 1999, 2000); Cooper (1985);
Gauthier (1986); Forster (1990); Haubold (1990); Weishampel and Witmer (1990a, b); Novas
(1992); Peng (1992, 1997); Sereno and Dong (1992); Weishampel and Heinrich (1992); Sereno and Arcucci (1994); Coria and Salgado (1996); Carpenter et al. (1998); Benton (1999, 2004);
Carpenter (2001); Weishampel et al. (2003); Langer (2004); Dodson et al. (2004); Galton and
Upchurch (2004a, b); Horner et al. (2004); Maryańska et al. (2004); Norman et al. (2004b, c);
Novas et al. (2004); Vickaryous et al. (2004); You and Dodson (2004); Butler (2005); Barrett et al. (2005); Maidment et al. (2006); Langer and Benton (2006); Xu et al. (2002, 2006); Zhou et al.
(2006); Butler et al. (2007); and Irmis et al. (2007).
Appendix 2 contains the data matrix that was compiled using the OTUs and character
scorings mentioned above. The matrix was constructed using the NEXUS Data Editor (NDE)
software. The NDE software was developed by Roderic D. M. Page (1999) to aid in the creation
of NEXUS files for their use in the PAUP* (Swofford, 2002) program. NEXUS is a file format 89 that can be used in various computer programs such as MacClade (Maddison and Maddison,
2002) and PAUP* (Swofford, 2002) for evaluation of various forms of systematic data (e.g., protein codes, morphological variation, molecular data, etc.) to determine phylogenetic relationships (Maddison et al., 1997). NDE was developed because MacClade is not compatible with personal computers that use Microsoft Windows operating systems (Page, 1999).
The character states listed in the data matrix in Appendix 2 contain the codings listed above (e.g., 0, 1, 2, 3) as well as three other designations: (?), (-), and (M). The (?) indicates that the character state is unknown in that particular OTU either because the material that would possess that character is not present in any known specimen or the character state is unclear due to several factors, including poor preservation. For example, the first 54 characters for
Stormbergia are coded as (?) because those characters refer to dental and cranial features. The only known material referred to Stormbergia consists of postcrania and it is therefore not possible to assign specific character states. A coding of (-) indicates that the character state in question is not applicable to that OTU because that character is absent. For example, in the
Iguanodontia characters 2 and 3 are scored with (-) because they are concerned with the shape of the premaxillary teeth and the occlusal counterpart of the premaxillary teeth, respectively. This
OTU does not possess premaxillary teeth; therefore, characters 2 and 3 are not applicable. A (M) coding indicates a multistate, or polymorphic, character for that OTU. For example, character
30, the position of the palpebral relative to the orbit – projecting into the orbit (0) or incorporated into the orbital margin (1), is coded as (M) for the Iguanodontia because several members of that clade possess palpebrals that project into the orbit (e.g., the basal members Camptosaurus dispar and Tenontosaurus tilletti) and the palpebrals of others are incorporated into their orbital margins
(e.g., the more derived Iguanodon atherfieldensis and Parasaurolophus walkeri). In this study, 90
(M) indicates variation only among the composite-coded supraspecific OTUs; however, it should
be noted that intraspecific variation does occur, possibly due to ontogenetic variation, but that did not factor into this analysis.
4.3. Phylogenetic Analysis
The phylogenetic analysis was carried out by importing the NEXUS file into the PAUP*
program. PAUP* is an acronym for “Phylogenetic Analysis Using Parsimony” and the asterisk
is used to designate “and Other Methods,” including maximum likelihood and distance methods
(Swofford, 2002). All of the 97 characters were considered unordered and unweighted. None of
the characters warrant an ordered sequence (similar to that of Butler’s coding of the external
mandibular fenestra [2005:character 39]). The characters are unweighted because there is not sufficient evidence to consider a priori one character more significant than another. By keeping
all characters unordered and unweighted, the number of assumptions is kept to a minimum. The
optimality criterion in PAUP* was kept as the default, which is parsimony. Parsimony, as in
unordered and unweighted characters, keeps assumptions to a minimum when using
morphological data and indicates that, with the available information, the resultant phylogeny
requires the fewest evolutionary transformations (shortest evolutionary pathway) that could have been produced through evolution. This does not imply that the phylogeny is ultimately the true evolutionary pathway, but it is the simplest (the shortest tree) that can be produced. PAUP* indicated that of the 97 characters, 88 are parsimony-informative characters that affect the resultant phylogeny. Characters 2, 16, 18, 24, 29, 32, 40, 46, and 90 were identified by PAUP* as parsimony-uninformative (Table 2) because the character is either an autapomorphy for one 91
OTU (i.e., characters 2, 16, 24, 32, 40, and 90) or is ambiguous in one of the outgroup OTUs and known for only one ingroup OTU (i.e., characters 18, 29, and 46).
Table 2. Parsimony-uninformative characters. The table below contains the nine parsimony- uninformative characters identified by PAUP* in the phylogenetic analysis. The characters are taken from Appendix 1.
Character Character Name Reason for Uninformative Status Number
Currently, only known to be present in 2 Premaxillary tooth size Heterodontosaurus
A neomorphic bone present in only the 16 Rostral Ceratopsia
Present in all ornithischians where material is Edentulous portion of the 18 known, but is uncertain in the outgroup, premaxilla Marasuchus
Ventral margin of the external Currently, only known to be present in the 24 naris outgroup, Saurischia
Present in all ornithischians where material is Ossified accessory orbital 29 known, but is uncertain in the outgroup, element (palpebral/supraorbital) Marasuchus
Currently, only known to be present in 32 Slot in maxilla for lacrimal Lesothosaurus
Currently, only known to be present in 40 Large quadratojugal foramen Hypsilophodon
Present in all ornithischians where material is 46 Predentary known, but is uncertain in the outgroup, Marasuchus
Fibula reduced to a splint Currently, only known to be present in 90 distally Heterodontosaurus
92
The analysis was performed by using a branch-and-bound search. Typically, the
preferred method would be to perform an exhaustive search, which will fully evaluate all
possible branching topologies and find the most parsimonious tree or trees (MPTs). PAUP*
cannot perform an exhaustive search if the dataset in question contains more than 12 taxa due to
computational limitations. However, a branch-and-bound search can be carried out by
calculating tree lengths each time another taxon is added. The prospective tree is compared with the shortest tree already identified and is immediately discarded if it exceeds that shortest length, thus shortening the computational process while still providing the MPTs (Hammer and Harper,
2006).
The branch-and-bound search produces a tree or trees with statistical values, including:
tree length (TL); consistency index (CI); retention index (RI); and a rescaled consistency index
(RC). TL refers to the total number of character state changes that occurred along the entire
length of the tree produced. The CI can be calculated by dividing the collective smallest number
of possible steps (character changes) for all characters by the number of steps actually observed
(Hammer and Harper, 2006). The CI is a measure of the amount of homoplasy in the tree
topology that was produced. Homoplasy is the umbrella term for the existence of both
evolutionary reversals and evolutionary convergences. Parsimony-uninformative characters
skew the CI value higher than it actually is; therefore, the CI reported for the four MPTs found in
this phylogenetic analysis excludes those nine uninformative characters. The RI is the number of
actual steps observed on a given tree subtracted from the largest possible number of all character
state changes divided by the smallest number of all possible character states changes subtracted
from the largest possible number of state changes (Hammer and Harper, 2006). The RI is a 93 measure of synapomorphic characters that support the resultant phylogeny. The RC is the product of the CI and RI (Hammer and Harper, 2006). The RC is used to evaluate the degree of information that a character can provide for a given phylogeny.
A bootstrap analysis of the data matrix was carried out with 1,000 replicates.
Bootstrapping is a statistical test that randomly resamples a given dataset. With the 97 characters in this analysis, the bootstrap analysis randomly sampled the matrix and rebuilt it with 97 characters, each of which could have been chosen more than once, thus excluding some characters and giving others a higher weighting. The default search option for PAUP* is set for a heuristic search; however, to ensure that the shortest trees were produced during the bootstrap process, the search option was set for branch-and-bound. After all 1,000 replicates were calculated, PAUP* produced a majority rule consensus tree in which only nodes that were present in at least 50% of the replicates were retained. Those nodes that were not present in at least 50% of the replicates collapsed down onto the next most stable node, creating polytomies.
The higher the bootstrap value, the more robust the clade is with regard to the random weighting of the characters. 94
5. RESULTS
The branch-and-bound search generated four MPTs, each with a TL of 198 steps, a CI of
0.51, a RI of 0.66, and a RC of 0.35; however, PAUP* only reports the RC computed with all
characters, including the parsimony-uninformative (the actual RC is 0.34). Figure 54 illustrates
all four of the MPTs found by PAUP*. The only differences among the four MPTs are: the
placement of the Iguanodontia with respect to the clades consisting of Hypsilophodon +
Thescelosaurus and Pachycephalosauria + Ceratopsia; and the relationships of Scutellosaurus
and Emausaurus with respect to the more derived clade consisting of Scelidosaurus +
(Stegosauria + Ankylosauria). Figure 55 is the strict consensus tree of all four MPTs. The
numbers beneath each node of Figure 55 (all 100) are percentages indicating that those clades
were recovered in each of the four MPTs generated. Figure 56 is the majority rule consensus
tree generated by the bootstrap analysis. The numbers beneath each node are the percentages
recovered for all 1,000 replicates.
The individual consistency indices (ci), individual retention indices (ri), and individual
rescaled consistency indices (rc) were evaluated for each parsimony-informative character to
determine the amount of homoplasy exhibited by each character (Table 3). There are 27
characters (characters 3, 7, 11, 13, 17, 22, 36, 38, 42, 50, 60, 61, 66, 67, 68, 73, 77, 78, 79, 80,
83, 84, 85, 91, 93, and 97) with a rc of 1.00, indicating that these characters provide robust support for the clade with which they are associated. These characters evolved once on the phylogeny evaluated (in this case the first MPT given by PAUP*) and did not change in any of the more derived subclades higher up on the tree. Nine characters (characters 14, 19, 20, 23, 44,
51, 57, 58, and 87) have a rc of 0, indicating that these characters are homoplastic and do not 95
provide much support for the clades with which they are associated. A low value such as this
suggests that these characters either arose separately in two or more different taxa/clades or the
character states present are evolutionary reversals. All other characters, with the exception of
character 69 (rc=0.50), have rc values less than 0.50, suggesting that there is a fair amount of homoplasy in the resultant phylogeny.
Table 3. Character indices. The table below lists the statistical values for each of the 88 parsimony-informative characters used in this analysis. See Appendix 1 for the character names.
Character Number ci ri rc 1 0.50 0.50 0.25 3 1.00 1.00 1.00 4 0.50 0.71 0.36 5 0.50 0.67 0.34 6 0.25 0.40 0.10 7 1.00 1.00 1.00 8 0.33 0.67 0.22 9 0.33 0.50 0.17 10 0.25 0.57 0.14 11 1.00 1.00 1.00 12 0.50 0.50 0.25 13 1.00 1.00 1.00 14 0.33 0 0 15 0.50 0.67 0.34 17 1.00 1.00 1.00 19 0.50 0 0 20 0.50 0 0 21 0.50 0.67 0.34 22 1.00 1.00 1.00 23 0.50 0.50 0.25 25 0.50 0 0 26 0.50 0.80 0.40 27 0.50 0.67 0.34 28 0.50 0.50 0.25 30 0.50 0.75 0.38 31 0.33 0.50 0.17 33 0.50 0.50 0.25 34 0.50 0.50 0.25 96
35 0.50 0.67 0.34 36 1.00 1.00 1.00 37 0.50 0.50 0.25 38 1.00 1.00 1.00 39 0.50 0.50 0.25 41 0.50 0.50 0.25 42 1.00 1.00 1.00 43 0.50 0.67 0.34 44 0.33 0 0 45 0.50 0.80 0.40 47 0.33 0.33 0.11 48 0.50 0.50 0.25 49 0.33 0.33 0.11 50 1.00 1.00 1.00 51 0.33 0 0 52 0.33 0.50 0.17 53 0.25 0.40 0.10 54 0.33 0 0 55 0.25 0.25 0.06 56 0.50 0.75 0.38 57 0.50 0 0 58 0.50 0 0 59 0.50 0.60 0.3 60 1.00 1.00 1.00 61 1.00 1.00 1.00 62 0.33 0.33 0.11 63 0.33 0.33 0.11 64 0.50 0.50 0.25 65 0.50 0.67 0.34 66 1.00 1.00 1.00 67 1.00 1.00 1.00 68 1.00 1.00 1.00 69 0.67 0.75 0.50 70 1.00 1.00 1.00 71 0.50 0.83 0.42 72 0.50 0.80 0.40 73 1.00 1.00 1.00 74 0.50 0.67 0.34 75 0.33 0.50 0.17 76 0.33 0.50 0.17 77 1.00 1.00 1.00 78 1.00 1.00 1.00 79 1.00 1.00 1.00 80 1.00 1.00 1.00 97
81 0.50 0.67 0.34 82 0.33 0.60 0.20 83 1.00 1.00 1.00 84 1.00 1.00 1.00 85 1.00 1.00 1.00 86 0.33 0.50 0.17 87 0.50 0 0 88 0.33 0.50 0.17 89 0.50 0.67 0.34 91 1.00 1.00 1.00 92 0.50 0.80 0.40 93 1.00 1.00 1.00 94 0.33 0.50 0.17 95 0.50 0.50 0.25 96 0.50 0.50 0.25 97 1.00 1.00 1.00
98
Figure 54. Four MPTs generated by PAUP* in the phylogenetic analysis. Note the slightly differing positions of Iguanodontia and Scutellosaurus and Emausaurus in each of the four trees. 99
Figure 55. Strict consensus of the four MPTs. The numbers beneath each node indicate the percent recovered in all four MPTs. Polytomies indicate unresolved relationships. See text for discussion of the numbers above the nodes. 100
Figure 56. Bootstrap majority rule consensus tree. 1,000 bootstrap replicates were compiled to generate this tree. Numbers beneath each node indicate percent recovered in the 1,000 replicates. Polytomies indicate percentages less than 50%.
101
The results of the phylogenetic analysis are presented here and correspond to the numbers
above the nodes in Figure 55 (node numbers 1-11). The nodes that have been numbered are
those that were corroborated by the 50% majority rule consensus tree the bootstrap analysis generated. The “RECONSTRUCT” command of PAUP* illustrates the character state transformations for each character along the resultant phylogeny. Only those characters from this phylogenetic analysis that unambiguously diagnose a particular clade are presented in the node descriptions below. Unless otherwise noted in parentheses following the character number, each character state change was observed to be from (0) to (1).
5.1. Node 1 – Ornithischia
Node 1 was recovered in all 1,000 bootstrap replicates generated. This clade, to the
exclusion of the two outgroup OTUs, Marasuchus and Saurischia, forms the monophyletic
Ornithischia. The 18 characters that diagnose the Ornithischia are: character 5,
maxillary/dentary tooth crown shape is not blade-like and apicobasally tall; character 9, the
presence of an asymmetric swelling at the base of the cheek teeth crowns (‘cingulum’); character
11, the loss of recurvature in the maxillary/dentary teeth; character 12, the adjacent crowns of the
maxillary/dentary teeth overlap; character 13, the central or posterocentral cheek teeth achieve
maximum size; character 28, buccal emargination of the maxilla; character 49, spout-shaped
symphysis of the dentary; character 51, buccal emargination of the dentary; character 54; the jaw
articulation is offset ventrally relative to the maxillary tooth row; character 59, at least four sacral
vertebrae; character 66, the preacetabular process of the ilium is strap-like and the distal end is
anterior to the pubic peduncle; character 67, the preacetabular process of the ilium expands
distally in dorsal view; character 78, the pubic shaft (postpubic process) is rod-like; character 79, 102
the pubic symphysis is restricted to the distal portion or absent; character 84, the anterior
trochanter is separated from the greater trochanter by a cleft and is subequal to it in width;
character 86, a pendant fourth trochanter; character 93, metatarsal 5 is greatly reduced relative to
metatarsal 3; and character 95, ossified epaxial tendons.
The characters listed here that diagnose the Ornithischia have been well established in
previous analyses. In previous analyses, however, Lesothosaurus has generally been considered
basal to all other ornithischians because it lacks buccal emargination. Buccal emargination of
the maxilla has generally been considered a more derived feature. However, Lesothosaurus is
the only OTU of this clade that does not possess this character and, therefore, this appears to be
an evolutionary reversal. This character does not seem to have a significant impact on the
phylogenetic placement of Lesothosaurus within the Ornithischia.
5.2. Node 2 – Genasauria
The Genasauria was recovered in 78% of the 1,000 bootstrap replicates. This
monophyletic clade includes all ornithischians to the exclusion of Pisanosaurus. The five
characters that diagnose the Genasauria are: character 15, the presence of special foramina
medial to the maxillary and dentary tooth rows; character 77, the pubis is directed
posteroventrally; character 80; the presence of a prepubic process; character 89, proximal and distal portions of the tibia are subequal; and character 91, presence of a tibial facet on the calcaneum.
The Genasauria literally means “cheeked lizards” based on the presence of buccal
emargination. However, as noted above, Lesothosaurus was recovered within the Genasauria
despite its lack of buccal emargination. This analysis corroborates that of Butler (2005) and 103
Butler et al. (2007) and places Lesothosaurus within the Genasauria. In all four MPTs, the clade
consisting of Lesothosaurus + (Stormbergia + Agilisaurus) was recovered as the sister taxon to
the Thyreophora (contra Butler [2005] and Butler et al. [2007]). This clade, however, collapses down to the base of the Genasauria in the bootstrap analysis. Eocursor was recovered in the four
MPTs as the sister taxon to the Cerapoda; however, in the bootstrap consensus tree, the Cerapoda
collapses to the base of Genasauria. Therefore, the basal relationships of the Genasauria are
unresolved and there is a polytomy consisting of Lesothosaurus, Eocursor, Stormbergia +
Agilisaurus, Cerapoda, and Thyreophora. Based on this analysis and those of Butler (2005) and
Butler et al. (2007), buccal emargination can no longer be used as a diagnosing character of the
Genasauria, as it does not exhibit a clear phylogenetic signal.
5.3. Node 3 – Unnamed Clade (Stormbergia + Agilisaurus)
Stormbergia and Agilisaurus are more closely related to each other than any other OTU.
This unnamed clade was recovered in 65% of the bootstrap replicates. There are no
synapomorphies unique to this clade that unite it to the exclusion of Lesothosaurus, indicating that the characters that do unite this clade also arose independently in another clade or clades.
However, three characters unite Stormbergia + Agilisaurus with Lesothosaurus, including:
character 62, humeral length is less than 55% of the femoral length; character 65, the forelimb is
reduced to 40% of the hindlimb length; and character 74, a dorsal groove on the ischium. This is
the first analysis that places these two taxa more closely related to each other than to any other
ornithischian. In all four MPTs, this clade was recovered with Lesothosaurus as its sister taxon.
This is an interesting grouping as it runs counter to the majority of published analyses that
considered Lesothosaurus as a basal ornithischian taxon. Although the clade of Lesothosaurus + 104
(Stormbergia + Agilisaurus) collapsed in the bootstrap majority rule tree (less than 50% support) to a basal position within the Genasauria, its recovery in all four MPTs provides some evidence of a weakly supported Fabrosauridae proposed first by Galton (1972) and recently championed by Thulborn (1992) and Peng (1992, 1997). These analyses, however, did not place the
Fabrosauridae as a more derived ornithischian clade as it is recovered here (as the sister taxon to the Thyreophora); rather, they placed the Fabrosauridae as the basalmost group of ornithischian dinosaurs and the sister taxon to the Genasauria.
5.4. Node 4 – Cerapoda
The Cerapoda (sensu Sereno, 1986) was recovered in 71% of the bootstrap replicates.
This clade includes the marginocephalians, heterodontosaurids, hypsilophodontids, and
iguanodontians. The 12 characters that diagnose this clade are: character 3, posterior
premaxillary teeth oppose only the edentulous predentary; character 4, twelve or less teeth in the maxilla; character 7, asymmetrically enameled cheek teeth; character 21, the ventral margin of the premaxilla is ventrally offset from the maxillary tooth row; character 22, presence of a diastema between the premaxilla and maxilla; character 47, the predentary is subequal to longer than the premaxilla; character 58, sixteen or more dorsals; character 59 (1-2), six or more sacrals; character 68, brevis shelf is horizontal; character 70, absence of a ventral acetabular flange; character 71, absence of a supraacetabular flange; and character 85, the anterior trochanteric height is equal to that of the femoral head.
The Cerapoda has been a well-established clade in the majority of published phylogenetic
analyses. The position of the Heterodontosauridae, however, has been more problematic.
Sereno (1986) originally placed the Heterodontosauridae as the basal member of the Ornithopoda 105
(Hypsilophodontidae + Iguanodontia), whereas Butler (2005) recovered the heterodontosaurids as the basal member of the cerapodans, Xu et al. (2006) recovered it as the sister taxon to the marginocephalians, and Butler et al. (2007) recovered it as the sister taxon (in a polytomy with
Pisanosaurus) to all other ornithischians. Norman et al. (2004c) recovered the single species of
Heterodontosaurus tucki as either the basal member of the Ornithopoda or the sister taxon to the
Marginocephalia. This analysis recovered the Heterodontosauridae at the base of the Cerapoda, in agreement with Butler (2005).
5.5. Node 5 – Heterodontosauridae
The Heterodontosauridae (Heterodontosaurus + Abrictosaurus) was recovered in 75% of
the bootstrap replicates. The seven characters that diagnose the Heterodontosauridae are:
character 1 (1-2), three premaxillary teeth; character 9 (1-0), the absence of the ‘cingulum’ at the
base of the cheek teeth crowns; character 15 (1-0), the absence of special foramina for
replacement teeth; character 23, arched premaxilla-maxilla diastema; character 48, reduced
ventral process of the predentary; character 49 (1-0), V-shaped dentary symphysis; and character
89 (1-0), distal tibia is less expanded than the proximal tibia.
As mentioned above, the Heterodontosauridae has been placed as the basal member of
the Ornithischia, and Cerapoda, and as the sister taxon to the Marginocephalia, but is recovered here as the basal member of the Cerapoda. There are several plesiomorphic traits that the heterodontosaurids possess that may make it seem, at least superficially, that the
Heterodontosauridae is the basalmost member of the Ornithischia. However, as shown in this analysis, these plesiomorphies are the product of evolutionary reversals and are countered by synapomorphies placing the clade within the Cerapoda. 106
5.6. Node 6 – Unnamed Clade (Hypsilophodontidae + Iguanodontia + Marginocephalia)
The unnamed clade that consists of the Hypsilophodontidae + Iguanodontia +
Marginocephalia was supported in 96% of all bootstrap replicates. The eight characters that diagnose this clade are: character 27, subcircular shape to the antorbital fossa; character 33, posterior process of the maxilla-lacrimal contact excludes the jugal from the margin of the antorbital fossa; character 73, the ischial peduncle of the ilium is broadly swollen and projects ventrolaterally; character 81, rod-like prepubic process; character 83, a distinct constriction separating the greater trochanter from the femoral head; character 84, a narrow anterior trochanter and the cleft between it and the greater trochanter is absent; character 92, the medial distal tarsal articulates with both metatarsals 2 and 3; and character 96, ossified hypaxial tendons on the caudals.
Based on the robust bootstrap value and the eight identified synapomorphies, this clade is extremely well supported and consists of more derived cerapodans than the heterodontosaurids.
The position of Iguanodontia was not resolved in either the four MPTs or the bootstrap majority rule tree. All known recent analyses have placed the Iguanodontia as derived ornithopods with respect to the hypsilophodontids. The unresolved affinities of the Iguanodontia in this analysis appear to cast the monophyly of the Ornithopoda in doubt.
5.7. Node 7 – Hypsilophodontidae
The Hypsilophodontidae (Hypsilophodon + Thescelosaurus) was recovered in 57% of the bootstrap replicates. The two characters that diagnose it are: character 60, the presence of ossified sternal ribs; and character 61, a weakly developed deltopectoral crest on the humerus. 107
As mentioned above, there has recently been some suspicion about the monophyly of the
Hypsilophodontidae (Weishampel et al., 2003; Norman et al., 2004c). This analysis has
recovered a weakly-supported clade consisting of just two taxa; however, both Hypsilophodon and Thescelosaurus have been recovered as members of the Hypsilophodontidae in previous analyses (e.g., Weishampel and Heinrich, 1992; Norman et al., 2004c); therefore, this group is here considered a monophyletic assemblage.
5.8. Node 8 – Marginocephalia
The Marginocephalia has been a well-established and uncontroversial clade since it was
formally introduced (Sereno, 1986). This analysis has recovered a well-supported clade (81% of
the bootstrap replicates) consisting of the Pachycephalosauria and the Ceratopsia. The seven
characters that support this clade are: character 17, preorbital skull length is less than 40% of the
total skull length; character 21 (1-0), ventral margin of the premaxilla is level with the maxillary
tooth row; character 37, postorbital is plate-like; character 42, a parietosquamosal shelf that
obscures the occiput in dorsal view; character 69 (0-2), the postacetabular process of the ilium is
more than 40% of the length of the entire ilium; character 75 (1-0), absence of the tab-shaped
obturator process of the ischium; and character 76 (1-0), the ischial symphysis extends for at
least half the length of the ischial shaft.
Despite a recent challenge to the monophyly of the Marginocephalia (Sullivan, 2006), the
affinities of the Pachycephalosauria and the Ceratopsia have withstood numerous phylogenetic
analyses. This analysis is in agreement with the overwhelming majority of published
phylogenetic analyses that support a robust monophyletic Marginocephalia. Contrary to a recent 108
study by Xu et al. (2006), this analysis did not recover the Heterodontosauridae as the sister
taxon to the Marginocephalia, creating a monophyletic Heterodontosauriformes.
5.9. Node 9 – Thyreophora
The Thyreophora is another well-established clade that includes all ornithischians that
possess dermal armor (scutes). The Thyreophora was recovered in 75% of the 1,000 bootstrap
replicates and is diagnosed by the following three characters: character 4 (0-2), twenty or more
maxillary teeth; character 36, the depth of the jugal orbital ramus is broader than the transverse
breadth; and character 97, presence of scutes.
The Thyreophora consists of the successive basal members Scutellosaurus, Emausaurus,
and Scelidosaurus and the Eurypoda. This analysis provides further support for the monophyly
of the Thyreophora.
5.10. Node 10 – Thyreophoroidea
Thyreophoroidea (sensu Sereno, 1986) is the clade consisting of Scelidosaurus +
Eurypoda to the exclusion of Scutellosaurus and Emausaurus. This is a very well-supported
clade (recovered in 87% of the bootstrap replicates). The two characters that diagnose this clade are: character 69, the postacetabular process of the ilium is less than 20% of the entire length of
the ilium; and character 88, femur is proportionally longer than the tibia.
The Thyreophoroidea is well supported here; although there are only two synapomorphic
characters that diagnose this clade, the high bootstrap value indicates that a combination of these
synapomorphies as well as other characters that were acquired independently in other groups
(i.e., characters 30, 31, 53, and 63) serve to unite this clade to the exclusion of other 109
ornithischians. Emausaurus was recovered in 50% of the bootstrap replicates as the sister taxon to the Eurypoda. The singular diagnosing character that supports this unnamed clade to the
exclusion of Scutellosaurus and the remainder of the Ornithischia is character 50 (dentary tooth
row is sinuous in lateral view). This unnamed clade was not recovered in the four MPTs.
5.11. Node 11 – Eurypoda
The eurypods consist of the Stegosauria and the Ankylosauria, and have been well
established since their formal description (Sereno, 1986). The Eurypoda is very well supported
in this analysis (bootstrap value of 96%); however, there are no unique synapomorphies that
support this clade. There are seven characters that arose independently in other groups that do
not share a direct common ancestry with the eurypods to the exclusion of the basal
thyreophorans Scutellosaurus, Emausaurus, and Scelidosaurus, thus uniting the clade Eurypoda
to the exclusion of all other ornithischians. Those characters are: character 1 (0-3), less than
three premaxillary teeth; character 8, apicobasal ridges on the lingual/labial surfaces of the
maxillary/dentary teeth, respectively; character 52, a weakly-developed coronoid process formed
at least partially by the posterior process of the dentary; character 55, the lack of epipophyses on
the anterior cervical vertebrae; character 56, V-shaped zygapophyseal articulation of the dorsal
vertebrae; character 86 (1-0), the lack of a pendant fourth trochanter; and character 94, first
phalanx of pedal digit I is the longest non-ungual phalanx of the pes.
As noted above, several recent analyses have placed Scelidosaurus as the sister taxon to
the ankylosaurs (e.g., Carpenter, 2001; Butler, 2005). However, with a robust bootstrap value of
96%, it is unlikely that Scelidosaurus is nested within the Eurypoda. 110
6. DISCUSSION
The Ornithischia is a stem-based clade defined as all dinosaurs more closely related to
Triceratops than to Tyrannosaurus (Weishampel, 2004). The major subdivisions within the
Ornithischia that were identified here are Genasauria, Cerapoda, Marginocephalia,
Heterodontosauridae, Hypsilophodontidae, Thyreophora, and Eurypoda. Other major groups that are not defined here as they were utilized as composite OTUs are Stegosauria, Ankylosauria,
Pachycephalosauria, Ceratopsia, and Iguanodontia. The Ornithopoda, a clade composed of the
Hypsilophodontidae and the Iguanodontia, was not recovered in this analysis. The Ornithopoda has been a well-documented clade of cerapodans; however, in this analysis, with the focus on the basal ornithischians and the lack of derived taxa that characterize the Ornithopoda, the affinities of the Hypsilophodontidae and the Iguanodontia were not emphasized. In order to evaluate the monophyly of the Ornithopoda, a more thorough investigation should include other more derived taxa such as Hexinlusaurus, Orodromeus, Gasparinisaura, Parksosaurus, Othnielia, and
Zephyrosaurus. These taxa as well as the taxa used in this analysis (Hypsilophodon,
Thescelosaurus, and Iguanodontia) would provide more legitimate support for its status as a monophyletic clade or paraphyletic assemblage.
The basalmost ornithischian dinosaur recovered in this analysis was Pisanosaurus. This
is in agreement with every other phylogenetic analysis that has considered Pisanosaurus with
other ornithischians (e.g., Langer, 2004; Butler, 2005; Langer and Benton, 2006, Butler et al.,
2007). However, only Butler (2005) and Butler et al. (2007) have compared the within-group
relationships of Pisanosaurus and other basal ornithischians. Langer (2004) and Langer and
Benton (2006) both assumed a priori that Pisanosaurus was basalmost within the Ornithischia 111
(each compared Pisanosaurus and a composite-coded Ornithischia). This analysis corroborates
that of Butler (2005) and Butler et al. (2007) in the basalmost position of Pisanosaurus as an
ornithischian. Norman et al. (2004a) suggested that Pisanosaurus, due to its several derived
characters such as the absence of an external mandibular fenestra, buccal emargination, and
occlusal wear facets on the cheek teeth, may be nested within the Genasauria to the exclusion of
Lesothosaurus. Based on this analysis, characters such as the lack of a prepubic process,
anteroventrally-directed pubis, and unequal proportions of the proximal and distal tibial ends
have a stronger phylogenetic signal than do buccal emargination and lack of a mandibular
fenestra.
The newly described Triassic ornithischian from South Africa, Eocursor, was recovered as the sister taxon to the Cerapoda in all four MPTs; however, that clade collapsed in the majority rule consensus tree during the bootstrap analysis. This analysis therefore provides weak support for a clade consisting of Eocursor + Cerapoda. Butler et al. (2007) recovered Eocursor as the basalmost member of the Genasauria (contra Butler, 2005; this paper); however, as shown in Figure 56, the basal Genasauria collapse to an unresolved polytomy.
Lesothosaurus was recovered in all four MPTs in a clade with Stormbergia + Agilisaurus
as the sister taxon to the Thyreophora. This is the first known analysis to recover a clade that
contains Lesothosaurus as the sister taxon to thyreophorans. This is also the first known
phylogenetic analysis that provides at least weak support for a “fabrosaurid” clade (sensu Peng,
1997). Although Fabrosaurus australis is considered a nomen dubium, there is no reason to
discard Fabrosauridae as a clade name, since the material previously referred to Fabrosaurus has
been assigned to Lesothosaurus. Even though the support for a Fabrosauridae clade is weak,
there is decent support for a clade containing Stormbergia + Agilisaurus. Stormbergia is a newly 112
described taxon (Butler, 2005) and has not been used in any other analysis aside from the recent
description of Eocursor (Butler et al., 2007). Butler et al. (2007) recovered Stormbergia and
Agilisaurus as successively more distantly related taxa to the Cerapoda, although there was only
weak support for the basal position of Stormbergia to Agilisaurus + Cerapoda (bootstrap value
less than 50%) and the support for Agilisaurus and the Cerapoda was only slightly higher
(bootstrap value of 54%). It appears that there is an affinity of Stormbergia to Agilisaurus, and
that a new clade comprised of the two should be erected. However, the potential for this new
clade to be nested within a fabrosaurid clade and to be the sister taxon to the Thyreophora
requires further investigation.
Recent phylogenetic analyses have placed the enigmatic Heterodontosauridae in several
positions including: the base of the Ornithopoda (Norman et al., 2004a); the base of the
Cerapoda (Butler, 2005); sister taxon to the Marginocephalia (Xu et al., 2006); and as basal ornithischians (sister taxon to the Genasauria; Butler et al., 2007). The Heterodontosauridae is here placed at the base of the Cerapoda and was recovered in all four MPTs. It was recovered within the Cerapoda with relatively robust support (bootstrap value of 71%) and the clade itself
(Heterodontosaurus + Abrictosaurus) was recovered with similar support (bootstrap value of
75%). In the majority of phylogenetic analyses, the Heterodontosauridae (Abrictosaurus +
[Heterodontosaurus + Lycorhinus]) has been recovered with relatively robust support; its placement as whole, however, has been problematic (Norman et al., 2004a). More thorough investigation is needed to attempt to resolve its placement within the Cerapoda; however, as shown here, the Heterodontosauridae is not considered to be either the sister taxon to the
Marginocephalia or the sister taxon to all other known ornithischians. Additionally, Sereno’s
(1998) node-based definition of the Ornithopoda (i.e., the most recent common ancestor of 113
Heterodontosaurus and Pachycephalosaurus and all its descendants) is here considered invalid and the stem-based definition of Norman et al. (2004c; i.e., all cerapodans more closely related to Edmontosaurus than to Triceratops) is adopted.
Weishampel et al. (2003) considered the Hypsilophodontidae as a paraphyletic assemblage with each taxon from Hypsilophodon to Gasparinisaura to Thescelosaurus +
Parksosaurus being successively more closely related to Iguanodontia. The only other known analyses to have considered the Hypsilophodontidae as a paraphyletic assemblage are both abstracts (Scheetz, 1998; Winkler et al., 1998). All other known analyses have substantiated the existence of a monophyletic group. The support for the Hypsilophodontidae here is not robust
(57%), but it was within the majority rule consensus of the bootstrap analysis and recovered in all four MPTs. The taxa used here to specify the clade, Hypsilophodon and Thescelosaurus, are generally considered to be the most distantly related taxa within the Hypsilophodontidae. In two of the four MPTs, Iguanodontia was recovered as the sister taxon to the Hypsilophodontidae
(providing support for a monophyletic Ornithopoda), but, as mentioned above, a more thorough analysis is needed to resolve this clade.
Scelidosaurus was recovered as the sister taxon to the Eurypoda (Stegosauria +
Ankylosauria) with a robust bootstrap value (87%). The Eurypoda, Stegosauria + Ankylosauria to the exclusion of the Scelidosaurus, was also recovered with a robust bootstrap value (96%).
These two clades are in agreement with Sereno (1999) and Maidment et al. (2006), whereas it differs from that of Carpenter (2001) in the placement of Scelidosaurus. The weak support for the clade consisting of Emausaurus + Thyreophoroidea to the exclusion of Scutellosaurus is likely due to the fragmentary nature of the only known specimen of Emausaurus (a nearly complete skull with scant associated postcrania). 114
There is little issue regarding the remaining clades that were recovered here (i.e.,
Marginocephalia, Thyreophora). With the exception of Sullivan (2006), who did not perform a
phylogenetic analysis, the Marginocephalia has received strong support in all known analyses
and is corroborated here. Similarly, the Thyreophora has received strong support in all recent phylogenetic analyses and is supported in this analysis. 115
7. SUMMARY AND CONCLUSIONS
The phylogeny presented here (Fig. 57) provides strong support for several clades that
have been previously established in prior publications (e.g., Marginocephalia, Cerapoda).
Conversely, it provides evidence for less well-supported clades that have either been problematic
(e.g., Heterodontosauridae, Eurypoda) or have been purely speculative or have not been
previously proposed (e.g., Fabrosauridae, Stormbergia + Agilisaurus, respectively).
Figure 57. Ornithischia cladogram. The nodes are labeled to indicate previously-named clades.
This phylogeny provides a starting point for future investigations into basal ornithischian affinities. The polytomy at the base of the Genasauria will undoubtedly benefit from future 116
discoveries of Late Triassic and Early Jurassic dinosaurs. After decades of assumptions, it is
apparent that Lesothosaurus, while its status as a basal member of the Ornithischia is still
supported, is no longer considered the sister taxon to the Genasauria, but rather a basal member
within it. The plesiomorphies retained by Lesothosaurus such as lack of buccal emargination
and the presence of a well-developed external mandibular fenestra do not exhibit a strong enough
phylogenetic signal to place it outside of the Genasauria. Pisanosaurus, in agreement with all
known phylogenetic studies as well as in agreement with most authors’ assumptions (but see
Norman et al., 2004a), is the basalmost known ornithischian dinosaur. Plesiomorphies such as
an anteroventrally-directed pubis and the lack of a prepubic process have placed Pisanosaurus as
the sister taxon to the Genasauria.
The much maligned clade Fabrosauridae has not been recovered in any known previously-published phylogenetic analysis. There is some evidence to support such a clade containing Lesothosaurus + (Stormbergia + Agilisaurus). Peng (1992, 1997) originally considered Lesothosaurus + (Agilisaurus + Gongbusaurus) as a monophyletic assemblage, although Gongbusaurus may not be a valid taxon (Norman et al., 2004c). The clade Stormbergia
+ Agilisaurus was reasonably well supported in the bootstrap analysis and is here considered to be a monophyletic clade; however, the relationship to Lesothosaurus, although present in all four
MPTs, was not recovered in the bootstrap analysis and is tentatively not considered
monophyletic. Future discoveries and focus on these three taxa should illuminate these potential
relationships.
Eocursor was recovered in all four MPTs as the sister taxon to the Cerapoda; however, it
was not supported in the bootstrap analysis and can only be assigned as a basal genasaur. The
Heterodontosauridae was recovered with relatively robust support as the basalmost taxon within 117
the Cerapoda. This taxon has been the subject of controversy regarding its placement within the
Ornithischia. It is here regarded as the sister taxon to the monophyletic clade consisting of the
Hypsilophodontidae + Iguanodontia + Marginocephalia. The Hypsilophodontidae was recovered
as a monophyletic clade with weak support and only two taxa (Hypsilophodon +
Thescelosaurus), contrary to recent publications (Weishampel et al., 2003; Norman et al.,
2004c).
The remaining clades that were identified in this analysis have generally been regarded as
monophyletic assemblages and there has never been any serious challenge to their phylogenetic
status. The future of basal ornithischian phylogenetics should focus on the interrelationships of
the thyreophorans and the “fabrosaurids” as well as the heterodontosaurids and Eocursor. With
the Late Triassic Eocursor and the Early Jurassic heterodontosaurids of South Africa, there is an
intriguing paleobiogeographic implication for the potential origin of the Cerapoda. It appears
that the cerapodans evolved first in southern Gondwana, more specifically, the region of
southern Africa, and radiated outward. Much more work is needed, however, to resolve the
paleobiogeography of the Early Jurassic ornithischian dinosaurs Lesothosaurus and Stormbergia
(both South African), Agilisaurus (China), and the thyreophorans (Scutellosaurus of the
southwestern United States, and Emausaurus and Scelidosaurus of northern Europe). 118
REFERENCES
Arcucci, A. B. 1997. Dinosauromorpha, p. 179-183. In Currie, P. J., and Padian, K. (eds.), Encyclopedia of Dinosaurs. Academic Press, San Diego.
Bakker, R. T., and Galton, P. M. 1974. Dinosaur monophyly and a new class of vertebrates. Nature 248:168-172.
Barrett, P. M., Butler, R. J., and Knoll, F. 2005. Small-bodied ornithischian dinosaurs from the Middle Jurassic of Sichuan, China. J. Vert. Paleontol. 25:823-834.
Barsbold, R., and Maryańska, T. 1990. Segnosauria, p. 408-415. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. University of California Press, Berkeley.
Benton, M. J. 1999. Scleromochlus taylori and the origin of dinosaurs and pterosaurs. Phil. Trans. Roy. Soc. London B 354:1423-1446.
Benton, M. J. 2000. Stems, nodes, crown clades, and rank-free lists: is Linnaeus dead? Biol. Rev. 75:633-648.
Benton, M. J. 2004. Origin and relationships of Dinosauria, p. 7-19. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
Bonaparte, J. F. 1976. Pisanosaurus mertii Casamiquela and the origin of the Ornithischia. J. Paleontol. 50:808-820.
Brochu, C. A. 1996. Closure of neurocentral sutures during crocodilian ontogeny: implications for maturity assessment in fossil archosaurs. J. Vert. Paleontol. 16:49-62.
Brochu, C. A. 2001. Progress and future directions in archosaur phylogenetics. J. Paleontol. 75:1185-1201.
Brochu, C. A., and Sumrall, C. D. 2001. Phylogenetic nomenclature and paleontology. J. Paleontol. 75:754-757.
Butler, R. J. 2005. The “fabrosaurid” ornithischian dinosaurs of the Upper Elliot Formation (Lower Jurassic) of South Africa and Lesotho. Zool. J. Lin. Soc. 145:175-218.
Butler, R. J., Smith, R. M. H., and Norman, D. B. 2007. A primitive ornithischian dinosaur from the Triassic of South Africa, and the early evolution and diversification of Ornithischia. Proc. R. Soc. B 274:2041-2046.
Butler, R. J., Upchurch, P., Norman, D. B., and Parish, J. 2006. A biogeographical analysis of the ornithischian dinosaurs, p. 13-16. In Barrett, P. M., and Evans, S. E. (eds.). Ninth 119
Symposium on Mesozoic Terrestrial Ecosystems and Biota. The Natural History Museum, London.
Cantino, P. D., and de Queiroz, K. 2004. PhyloCode: a phylogenetic code of biological nomenclature, version 2b, http://www.ohiou.edu/phylocode/.
Carpenter, K. 1997. Ankylosauria, p. 16-20. In Currie, P. J., and Padian, K. (eds.), Encyclopedia of Dinosaurs. Academic Press, San Diego.
Carpenter, K. 2001. Phylogenetic analysis of the Ankylosauria, p. 455-483. In Carpenter, K. (ed.), The Armored Dinosaurs. Indiana University Press, Bloomington.
Carpenter, K., Miles, C., and Cloward, K. 1998. Skull of a Jurassic ankylosaur (Dinosauria). Nature 393:782-783.
Casamiquela, R. M. 1967. Un nuevo dinosaurio ornitisquio Triásico (Pisanosaurus mertii; Ornithopoda) de la formación Ischigualasto, Argentina. Ameghiniana 4:47-64. Translated by Matthew T. Carrano, University of Chicago, 1997. Obtained via the Polyglot Paleontologist, http://ravenel.si.edu/paleo/paleoglot/index.cfm.
Charig, A. J., and Crompton, A. W. 1974. The alleged synonymy of Lycorhinus and Heterodontosaurus. Ann. S. Afr. Mus. 64:167-189.
Chatterjee, S. 1984. A new ornithischian dinosaur from the Triassic of North America. Naturwissenschaften 71:630-631.
Clark, J. M., Maryańska, T., and Barsbold, R. 2004. Therizinosauroidea, p. 151-164. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
Colbert, E. H. 1981. A primitive ornithischian dinosaur from the Kayenta Formation of northern Arizona. Bull. Mus. N. Arizona. 53:1-61.
Coombs, W. P., Jr. 1978. The families of the ornithischian dinosaur order Ankylosauria. Palaeontol. 21:143-170.
Coombs, W. P., Jr., and Maryańska, T. 1990. Ankylosauria, p. 456-483. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. University of California Press, Berkeley.
Coombs, W. P., Jr., Weishampel, D. B., and Witmer, L. M. 1990. Basal Thyreophora, p. 427- 434. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. University of California Press, Berkeley.
Cooper, M. R. 1985. A revision of the ornithischian dinosaur Kangnasaurus coetzeei Haughton, with a classification of the Ornithischia. Ann. S. Afr. Mus. 95:281-317. 120
Cope, E. D. 1866. On anatomical peculiarities of some Dinosauria (on the anomalous relations existing between the tibia and fibula in certain Dinosauria). Proc. Acad. Nat. Sci. Philadelphia 18:316-317.
Coria, R. A, and Salgado, L. 1996. A basal iguanodontian (Ornithischia, Ornithopoda) from the Late Cretaceous of South America. J. Vert. Paleontol. 16:445-457.
Crompton, A. W., and Charig, A. J. 1962. A new ornithischian from the Upper Triassic of South Africa. Nature 196:1074-1077.
Currie, P. J., and Padian, K. 1997. Marginocephalia, p. 415. In Currie, P. J., and Padian, K. (eds.), Encyclopedia of Dinosaurs. Academic Press, San Diego.
Dodson, P. 1990. Ceratopsia, p. 578. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. University of California Press, Berkeley.
Dodson, P. 1997a. Ceratopsia, p. 106. In Currie, P. J., and Padian, K. (eds.), Encyclopedia of Dinosaurs. Academic Press, San Diego.
Dodson, P. 1997b. Neoceratopsia, p. 473-478. In Currie, P. J., and Padian, K. (eds.), Encyclopedia of Dinosaurs. Academic Press, San Diego.
Dodson, P., and Currie, P. J. 1990. Neoceratopsia, p. 593-618. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. University of California Press, Berkeley.
Dodson, P., Forster, C. A., and Sampson, S. D. 2004. Ceratopsidae, p. 494-513. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
Doolittle, W. F. 1999. Phylogenetic classification and the universal tree. Science 284:2124-2128.
Dyke, G. J. 2002. Should paleontologists use “phylogenetic” nomenclature? J. Paleontol. 76:793- 796.
Dzik, J. 2003. A beaked herbivorous archosaur with dinosaur affinities from the early Late Triassic of Poland. J. Vert. Paleontol. 23:556-574.
Forster, C. A. 1990. The postcranial skeleton of the ornithopod dinosaur Tenontosaurus tilletti. J. Vert. Paleontol. 10:273-294.
Forster, C. A. 1997. Iguanodontidae, p. 359-361. In Currie, P. J., and Padian, K. (eds.), Encyclopedia of Dinosaurs. Academic Press, San Diego.
Forster, C. A. 1999. Gondwanan dinosaur evolution and biogeographic analysis. J. Afr. Earth Sci. 28:169-185. 121
Fraser, N. C., Padian, K., Walkden, G. M., and Davis, A. L. M. 2002. Basal dinosauriform remains from Britain and the diagnosis of the Dinosauria. Palaeontol. 45:79-95.
Galton, P. M. 1972. Classification and evolution of ornithopod dinosaurs. Nature 239:464-466.
Galton, P. M. 1973. The cheeks of ornithischian dinosaurs. Lethaia. 6:67-89.
Galton, P. M. 1974a. The ornithischian dinosaur Hypsilophodon from the Wealden of the Isle of Wight. Bull. Br. Mus. (Nat. Hist.), Geol. 25:1-152.
Galton, P. M. 1974b. Notes on Thescelosaurus, a conservative ornithopod dinosaur from the Upper Cretaceous of North America, with comments on ornithopod classification. J. Paleontol. 48:1048-1067.
Galton, P. M. 1978. Fabrosauridae, the basal family of ornithischian dinosaurs (Reptilia, Ornithopoda). Paläontol. Z. 52:138-159.
Galton, P. M. 1985. British plated dinosaurs (Ornithischia, Stegosauridae). J. Vert. Paleontol. 5:211-254.
Galton, P. M. 1990. Stegosauria, p. 435-455. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. University of California Press, Berkeley.
Galton, P. M. 1997. Stegosauria, p. 701-703. In Currie, P. J., and Padian, K. (eds.), Encyclopedia of Dinosaurs. Academic Press, San Diego.
Galton, P. M., and Upchurch, P. 2004a. Prosauropoda, p. 232-258. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
Galton, P. M., and Upchurch, P. 2004b. Stegosauria, p. 343-362. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
Gauthier, J. A. 1986. Saurischian monophyly and the origin of birds. Mem. California Acad. Sci. 8:1-55.
Ginsburg, L. 1964. Découverte d’un scélidosaurien (dinosaure ornithischien) dans le Trias supérieur du Basutoland. C. R. Acad. Sci. Paris 258:2366-2368. Translated by Matthew T. Carrano, SUNY Stony Brook, 1999. Obtained via the Polyglot Paleontologist, http://ravenel.si.edu/paleo/paleoglot/index.cfm.
Gow, C. E. 1981. Taxonomy of the Fabrosauridae (Reptilia, Ornithischia) and the Lesothosaurus myth. S. Afr. J. Sci. 77:43.
122
Hammer, Ø., and Harper, D. A. T. 2006. Paleontological Data Analysis. Blackwell Publishing, Oxford, 351 pp.
Haubold, H. 1990. Ein neuer Dinosaurier (Ornithischia, Thyreophora) aus dem unteren Jura des nördlichen Mitteleuropa [In German, with an English Abstract]. Rev. Paléobiol. 9:149-177.
Heckert, A. B. 2002. A revision of the Upper Triassic ornithischian dinosaur Revueltosaurus, with a description of a new species. NM Mus. Nat. Hist. Sci. Bull. 21:253-267.
Hennig, W. 1950. Grundzüge einer Theorie der phylogenetischen Systematik. Deutscher Zentralverlag, Berlin.
Hennig, W. 1966. Phylogenetic Systematics. University of Illinois Press, Urbana and Chicago, 263 pp.
Holtz, T. R. 1994. The phylogenetic position of the Tyrannosauridae: Implications for theropod systematics. J. Paleontol. 68:1100-1117.
Holtz, T. R., and Osmólska, H. 2004. Saurischia, p. 21-24. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
Hopson, J. A. 1975. On the generic separation of the ornithischian dinosaurs Lychorhinus [sic] and Heterodontosaurus from the Stormberg Series (Upper Triassic) of South Africa. S. Afr. J. Sci. 71:302-305.
Horner, J. R., Weishampel, D. B., and Forster, C. A. 2004. Hadrosauridae, p. 438-463. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
Hunt, A. P., and Lucas, S. G. 1994. Ornithischian dinosaurs from the Upper Triassic of the United States, p. 227-241. In Fraser, N. C., and Sues, H. –D. (eds.), In the Shadow of the Dinosaurs: Early Mesozoic Tetrapods. Cambridge University Press, London.
Irmis, R. B. 2007. Axial skeleton ontogeny in the Parasuchia (Archosauria, Pseudosuchia) and its implications for ontogenetic determination in archosaurs. J. Vert. Paleontol. 27:350-361.
Irmis, R. B., Parker, W. G., Nesbitt, S. J., and Liu, J. 2007. Early ornithischian dinosaurs: the Triassic record. Hist. Biol. 19:3-22.
Langer, M. C. 2004. Basal Saurischia, p. 25-46. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
Langer, M. C., and Benton, M. J. 2006. Early dinosaurs: A phylogenetic study. J. Syst. Palaeontol. 4:309-358.
123
Langer, M. C., Abdala F., Richter, M., and Benton, M. J. 1999. New record of Dinosauria from the Santa María Formation (upper Carnian of the Paraná Basin – Brazil). Ameghiniana. 36:103.
Lucas, S. G. 2000. Dinosaurs: The Textbook. 2nd ed. McGraw-Hill, New York, 278 pp.
Maddison, D. R., and Maddison, W. P. 2002. MacClade 4: Analysis of Phylogeny and Character Evolution. Sinauer Associates, Sunderland, Massachusetts.
Maddison, D. R., Swofford, D. L., and Maddison, W. P. 1997. NEXUS: an extensible file format for systematic information. Syst. Biol. 46:590-621.
Maidment, S. C. R., Wei, G., and Norman, D. B. 2006. Re-description of the postcranial skeleton of the Middle Jurassic stegosaur Huayangosaurus taibaii. J. Vert. Paleontol. 26:944-956.
Martill, D. M., Batten, D. J., and Loydell, D. K. 2000. A new specimen of the thyreophoran dinosaur cf. Scelidosaurus with soft tissue preservation. Palaeontol. 43:549-559.
Maryańska, T., and Osmólska, H. 1974. Pachycephalosauria, a new suborder of ornithischian dinosaurs. Palaeontol. Polonica 30:45-102.
Maryańska, T., and Osmólska, H. 1975. Protoceratopsidae (Dinosauria) of Asia. Palaeontol. Polonica 33:133-182.
Maryańska, T., Chapman, R. E., and Weishampel, D. B. 2004. Pachycephalosauria, p. 464-477. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
Nesbitt, S. J., Irmis, R. B., and Parker, W. G. 2007. A critical re-evaluation of the Late Triassic dinosaur taxa of North America. J. Syst. Palaeontol. 5:209-243.
Norman, D. B. 1984. A systematic reappraisal of the reptile order Ornithischia, p. 157-162. In Reif, W. –E., and Westphal, F. (eds.), Third Symposium on Mesozoic Terrestrial Ecosystems, Short Papers. Attempto Verlag, Tübingen.
Norman, D. B. 2000. Professor Richard Owen and the important, but neglected, dinosaur Scelidosaurus harrisonii. Hist. Biol. 14:235-253.
Norman, D. B. 2001. Scelidosaurus, the earliest complete dinosaur, p. 3-24. In Carpenter, K. (ed.), The Armored Dinosaurs. Indiana University Press, Bloomington.
Norman, D. B. 2004. Basal Iguanodontia, p. 413-437. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
124
Norman, D. B., and Weishampel, D. B. 1990. Iguanodontidae and related ornithopods, p. 510- 533. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. University of California Press, Berkeley.
Norman, D. B., Witmer, L. M., and Weishampel, D. B. 2004a. Basal Ornithischia, p. 325-334. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
Norman, D. B., Witmer, L. M., and Weishampel, D. B. 2004b. Basal Thyreophora, p. 335-342. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
Norman, D. B., Sues, H. –D., Witmer, L. M., and Coria, R. A. 2004c. Basal Ornithopoda, p. 393- 412. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
Novas, F. E. 1992. Phylogenetic relationships of the basal dinosaurs, the Herrerasauridae. Palaeontol. 35:51-62.
Novas, F. E. 1993. New information on the systematics and postcranial skeleton of Herrerasaurus ischigualastensis (Theropoda: Herrerasauridae) from the Ischigualasto Formation (Upper Triassic) of Argentina. J. Vert. Paleontol. 13:400-423.
Novas, F. E. 1997. Herrerasauridae, p. 303-311. In Currie, P. J., and Padian, K. (eds.), Encyclopedia of Dinosaurs. Academic Press, San Diego.
Novas, F. E., Cambiaso, A. V., and Ambrosio, A. 2004. A new basal iguanodontian (Dinosauria, Ornithischia) from the Upper Cretaceous of Patagonia. Ameghiniana 41:75-82.
Olsen, P. E. 1997. Stratigraphic record of the early Mesozoic breakup of Pangea in the Laurasia- Gondwana rift system. Annu. Rev. Earth Planet. Sci. 25:337-401.
Owen, R. 1842. Report on British fossil reptiles. Report of the British Association for the Advancement of Science 11:60–204.
Owen, R. 1861. Monograph of the Fossil Reptilia of the Liassic formations. Part I. A monograph of a fossil dinosaur (Scelidosaurus harrisonii Owen) of the Lower Lias. Palaeontogr. Soc. Monogr. 1:1-14.
Owen, R. 1863. A monograph of the Fossil Reptilia of the Liassic formations. Part II. A monograph of a fossil dinosaur (Scelidosaurus harrisonii Owen) of the Lower Lias. Palaeontogr. Soc. Monogr. 2:1-26.
Padian, K. 2004. Basal Avialae, p. 210-231. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
125
Padian, K., Hutchinson, J. R., and Holtz, T. R. 1999. Phylogenetic definitions and nomenclature of the major taxonomic categories of the carnivorous Dinosauria (Theropoda). J. Vert. Paleontol. 19:69-80.
Page, R. D. M. 1996. TREEVIEW: An application to display phylogenetic trees on personal computers. Comp. App. Biosci. 12:357-358.
Page, R. D. M. 1999. NDE: NEXUS Data Editor for Windows. Obtained via the World Wide Web, http://taxonomy.zoology.gla.ac.uk/rod/NDE/manual.html.
Parker, W. G., Irmis, R. B., Nesbitt, S. J., Martz, J. W., and Browne, L. S. 2005. The Late Triassic pseudosuchian Revueltosaurus callenderi and its implications for the diversity of early ornithischian dinosaurs. Proc. R. Soc. B. 272:963-969.
Peng, G. 1992. Jurassic ornithopod Agilisaurus louderbacki (Ornithopoda, Fabrosauridae) from Zigong, Sichuan, China. Vert. PalAs. 30:39-51.
Peng, G. 1997. Fabrosauridae, p. 237-240. In Currie, P. J., and Padian, K. (eds.), Encyclopedia of Dinosaurs. Academic Press, San Diego.
Rauhut, O. W. M. 2003. The interrelationships and evolution of basal theropods. Spec. Pap. Palaeontol. 69:1-213.
Romer, A. S. 1971. The Chañares (Argentina) Triassic reptile fauna. X. Two new but incompletely known long-limbed pseudosuchians. Breviora 378:1-10.
Romer, A. S. 1972. The Chañares (Argentina) Triassic reptile fauna. XV. Further remains of the thecodonts Lagerpeton and Lagosuchus. Breviora 394:1-7.
Rosenbaum, J. N., and Padian, K. 2000. New material of the basal thyreophoran Scutellosaurus lawleri from Kayenta Formation (Lower Jurassic) of Arizona. PaleoBios 20:13-23.
Russell, D. A. 1997. Therizinosauria, p. 729-730. In Currie, P. J., and Padian, K. (eds.), Encyclopedia of Dinosaurs. Academic Press, San Diego.
Rybczynski, N., and Vickaryous, M. K. 2001. Evidence of complex jaw movement in the Late Cretaceous ankylosaurid Euoplocephalus tutus (Dinosauria, Thyreophora), p. 299-317. In Carpenter, K. (ed.), The Armored Dinosaurs. Indiana University Press, Bloomington.
Scheetz, R. D. 1998. Phylogeny of basal ornithopod dinosaurs and the dissolution of the Hypsilophodontidae. J. Vert. Paleontol. 18(Suppl. to No. 3):75A.
Scotese, C. R. 2001. Atlas of Earth History, Volume 1: Paleogeography, PALEOMAP Project, Arlington, Texas, 52 pp. Obtained via the World Wide Web, http://www.scotese.com/ – PALEOMAP website.
126
Seeley, H. G. 1887. On the classification of the fossil animals commonly called Dinosauria. Proc. R. Soc. London 43:165-171.
Sereno, P. C. 1984. The phylogeny of the Ornithischia: a reappraisal, p. 219-226. In Reif, W. – E., and Westphal, F. (eds.), Third Symposium on Mesozoic Terrestrial Ecosystems, Short Papers. Attempto Verlag, Tübingen.
Sereno, P. C. 1986. Phylogeny of the bird-hipped dinosaurs (Order Ornithischia). Natl. Geogr. Res. 2:234-256.
Sereno, P. C. 1989. Pachycephalosaurs and ceratopsians (Ornithischia, Marginocephalia), p. 71- 79. In Culver, S. J. (ed.), The Age of Dinosaurs, 12th Annual Short Course of the Paleontological Society, University of Tennessee, Knoxville.
Sereno, P. C. 1990. Psittacosauridae, p. 579-592. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. University of California Press, Berkeley.
Sereno, P.C. 1991. Lesothosaurus, “Fabrosaurids,” and the early evolution of Ornithischia. J. Vert. Paleontol. 11:168-197.
Sereno, P. C. 1993. The pectoral girdle and forelimb of the basal theropod Herrerasaurus ischigualastensis. J. Vert. Paleontol. 13:425-450.
Sereno, P. C. 1997. The origin and evolution of dinosaurs. Annu. Rev. Earth Planet. Sci. 25:435- 489.
Sereno, P. C. 1998. A rationale for phylogenetic definitions, with application to the higher-level taxonomy of Dinosauria. N. Jb. Geol. Paläontol. Abhandl. 210:41-83.
Sereno, P. C. 1999. The evolution of dinosaurs. Science 284:2137-2147.
Sereno, P. C. 2000. The fossil record, systematics, and evolution of pachycephalosaurs and ceratopsians from Asia, p. 480-516. In Benton, M. J., Shishkin, M. A., Unwin, D. M., and Kurochkin, E. N. (eds.), The Age of Dinosaurs in Russia and Mongolia. Cambridge University Press, Cambridge.
Sereno, P. C. 2005a. The logical basis of phylogenetic taxonomy. Syst. Biol. 54:595-619.
Sereno, P. C. 2005b. Stem Archosauria—TaxonSearch URL [version 1.0, 2005 November 7] http://www.taxonsearch.org/Archive/stem-archosauria-1.0.php.
Sereno, P. C. 2007. The phylogenetic relationships of early dinosaurs: a comparative report. Hist. Biol. 19:145-155.
Sereno, P. C., and Dong, Z. 1992. The skull of the basal stegosaur Huayangosaurus taibaii and a cladistic diagnosis of Stegosauria. J. Vert. Paleontol. 12:318-343. 127
Sereno, P. C., and Arcucci, A. B. 1994. Dinosaurian precursors from the Middle Triassic of Argentina: Marasuchus lilloensis, gen. nov. J. Vert. Paleontol. 14:53-73.
Sereno, P. C., and Novas, F. E. 1993. The skull and neck of the basal theropod Herrerasaurus ischigualastensis. J. Vert. Paleontol. 13:451-476.
Smith, J. B. 1997. Heterodontosauridae, p. 317-320. In Currie, P. J., and Padian, K. (eds.), Encyclopedia of Dinosaurs. Academic Press, San Diego.
Sues, H. –D. 1997. Hypsilophodontidae, p. 356-358. In Currie, P. J., and Padian, K. (eds.), Encyclopedia of Dinosaurs. Academic Press, San Diego.
Sues, H. –D., and Norman, D. B. 1990. Hypsilophodontidae, Tenontosaurus, Dryosauridae, p. 498-509. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. University of California Press, Berkeley.
Sullivan, R. A. 2006. A taxonomic review of the Pachycephalosauridae (Dinosauria, Ornithischia). New Mexico Mus. Nat. Hist. Sci. Bull. 35:347-365.
Swofford, D. L. 2002. PAUP*: Phylogenetic Analysis Using Parsimony (*and Other Methods). Version 4.0b10. CD-ROM. Sinauer Associates, Sunderland, Massachusetts.
Taylor, M. P. 2006. Dinosaur diversity analysed by clade, age, place and year of description, p. 134-138. In Barrett, P. M., and Evans, S. E. (eds.). Ninth symposium on Mesozoic Terrestrial Ecosystems and Biota. The Natural History Museum, London.
Taylor, M. P. 2007. Phylogenetic definitions in the pre-PhyloCode era: implications for naming clades under the PhyloCode. PaleoBios 27:1-6.
Thulborn, R. A. 1970a. The skull of Fabrosaurus australis, a Triassic ornithischian dinosaur. Palaeontol. 13:414-432.
Thulborn, R. A. 1970b. The systematic position of the Triassic ornithischian dinosaur Lycorhinus angustidens. Zool. J. Lin. Soc. 49:235-245.
Thulborn, R. A. 1971a. Origins and evolution of ornithischian dinosaurs. Nature 234:75-78.
Thulborn, R. A. 1971b. Tooth wear and jaw action in the Triassic ornithischian dinosaur Fabrosaurus. J. Zool. 164:165-179.
Thulborn, R. A. 1972. The post-cranial skeleton of the Triassic ornithischian dinosaur Fabrosaurus australis. Palaeontol. 15:29-60.
Thulborn, R. A. 1974. A new heterodontosaurid dinosaur (Reptilia, Ornithischia) from the Upper Triassic Red Beds of Lesotho. Zool. J. Lin. Soc. 55:151-175. 128
Thulborn, R. A. 1977. Relationships of the Lower Jurassic dinosaur Scelidosaurus harrisonii. J. Paleontol. 51:725-739.
Thulborn, R. A. 1992. Taxonomic characters of Fabrosaurus australis, an ornithischian dinosaur from the Lower Jurassic of Southern Africa. GeoBios 25:283-292.
Vickaryous, M. K., Maryańska, T., and Weishampel, D. B. 2004. Ankylosauria, p. 363-392. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
Weishampel, D. B. 1990. Ornithopoda, p. 484-485. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. University of California Press, Berkeley.
Weishampel, D. B. 2004. Ornithischia, p. 323-324. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
Weishampel, D. B., and Horner, J. R. 1990. Hadrosauridae, p. 534-561. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. University of California Press, Berkeley.
Weishampel, D. B., and Witmer, L. M. 1990a. Lesothosaurus, Pisanosaurus, and Technosaurus, p. 416-425. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. University of California Press, Berkeley.
Weishampel, D. B., and Witmer, L. M. 1990b. Heterodontosauridae, p. 486-497. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. University of California Press, Berkeley.
Weishampel, D. B., and Heinrich, R. E. 1992. Systematics of Hypsilophodontidae and basal Iguanodontia (Dinosauria, Ornithopoda). Hist. Biol. 6:159-184.
Weishampel, D. B, Jianu, C. -M., Csiki, Z., and Norman, D. B. 2003. Osteology and phylogeny of Zalmoxes (n. g.), an unusual euornithopod dinosaur from the latest Cretaceous of Romania. J. Syst. Palaeontol. 1:65-123.
Weishampel, D. B., Barrett, P. M., Coria, R. A., Le Loeuff, J., Xing, X., Xijin, Z., Sahni, A., Gomani, E. M. P., and Noto, C. R. 2004. Dinosaur distribution, p. 517-606. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
Winkler, D. A., Murry, P.A., and Jacobs, L. L. 1998. The new ornithopod dinosaur from Proctor Lake, Texas, and the deconstruction of the Family Hypsilophodontidae. J. Vert. Paleontol. 18(Suppl. to No. 3):87A.
129
Xu, X., Wang, X., and You, H. 2001. A juvenile ankylosaur from China. Naturwissenschaften 88:297-300.
Xu, X., Forster, C. A., Clark, J. M., and Mo, J. 2006. A basal ceratopsian with transitional features from the Late Jurassic of northwestern China. Proc. R. Soc. B. 273:2135-2140.
Xu, X., Makovicky, P. J., Wang, X., Norell, M. A., and You, H. 2002. A ceratopsian dinosaur from China and the early evolution of Ceratopsia. Nature 416:314-317.
You, H., and Dodson, P. 2003. Redescription of neoceratopsian dinosaur Archaeoceratops and the early evolution of Neoceratopsia. Acta Palaeontol. Polonica 48:261-272.
You, H., and Dodson, P. 2004. Basal Ceratopsia, p. 478-493. In Weishampel, D. B., Dodson, P., and Osmólska, H. (eds.), The Dinosauria. 2nd ed. University of California Press, Berkeley.
Zhou, C., Gao, K., Fox, R. C., and Chen, S. 2006. A new species of Psittacosaurus (Dinosauria, Ceratopsia) from the Early Cretaceous Yixian Formation, Liaoning, China. Palaeoworld 15:100-114. 130
APPENDIX 1.
Character-Character State Descriptions
The list of characters presented below is broken down into five subcategories based on
the anatomical region in which they are found, including: dental – all characters listed refer to
any morphological variation among the teeth of the taxa chosen; cranial – each character is
located in the skull or mandible; postcranial (axial) – each character is located on the vertebral
column or is an extension of it; postcranial (appendicular) – characters that include all
morphological variation among the pectoral and pelvic girdles and the fore- and hindlimbs; and
other characters – these include accessory ossifications such as epaxial (character 95) and hypaxial (character 96) tendons and dermal scutes (character 97). Section 4.1 contains a more thorough discussion of the characters and character states listed.
Dental:
1. Premaxillary teeth: Six or more (0); Four or five (1); Three (2); Two or less (3).
2. Premaxillary tooth size: Each tooth subequal in size (0); Posterior teeth significantly larger
than anterior teeth (1).
3. Posterior premaxillary teeth: Oppose anterior dentary teeth (0); Oppose edentulous
predentary only (1).
4. Maxillary teeth: Fourteen to eighteen (0); Twelve or less (1); Twenty or more (2).
5. Maxillary/Dentary tooth crown shape: Blade-like and apicobasally tall (0); Triangular and
apicobasally short – less than 50% higher than mesiodistally wide (1); Chisel-shaped and
apicobasally tall (2). 131
6. Maxillary/Dentary tooth crown denticles, restricted to the apical third of crown: Absent (0);
Present (1).
7. Maxillary/Dentary teeth, enamel: Symmetrical (0); Asymmetrical (1).
8. Maxillary/Dentary teeth, apicobasal ridges on the lingual/labial surfaces, respectively:
Absent (0); Present (1).
9. Maxillary/Dentary teeth, presence of a ‘cingulum’ (labiolingual swelling of the base of the
crown): Absent (0); Present (1).
10. Maxillary/Dentary teeth exhibit occlusal wear facets: Absent (0); Present (1).
11. Maxillary/Dentary teeth, recurvature: Present (0); Absent (1).
12. Maxillary/Dentary adjacent tooth crowns overlap: Absent (0); Present (1).
13. Maxillary/Dentary teeth achieve maximum size in central or posterocentral tooth row
positions: Absent (0); Present (1).
14. Dentary tooth row, heterodont dentition: Absent (0); Present (1).
15. Special foramina medial to maxillary and dentary tooth rows: Absent (0); Present (1).
Cranial:
16. Rostral: Absent (0); Present (1).
17. Preorbital skull length: More than 45% (0); less than 40% of total skull length defined from
the anterior tip of premaxilla/rostral to posterior end of quadrate (1).
18. Edentulous anterior portion of the premaxilla: Absent (0); First premaxillary tooth is inset by
the width of at least one tooth position (1).
19. Anterior premaxillary foramen: Absent (0); Present (1). 132
20. Premaxilla-lacrimal contact: Absent (0); Present, excludes the maxilla from contact with the
nasal (1).
21. Ventral margin of the premaxilla: Level with maxillary tooth row (0); Ventrally offset from
maxillary tooth row (1).
22. Diastema between premaxilla and maxilla: Absent (0); Present (1).
23. Premaxilla-maxilla diastema: Flat (0); Arched (1).
24. Ventral margin of the external naris: Below the ventral margin of the orbit (0); Above the
ventral margin of the orbit (1).
25. Deep elliptical fossa along the nasal suture: Absent (0); Present (1).
26. Antorbital fenestra anteroposterior length vs. orbital anteroposterior length: More than 50%
(0); Less than 50% (1).
27. Antorbital fossa shape: Subtriangular (0); Subcircular (1).
28. Buccal emargination of the maxilla: Absent (0); Present (1).
29. Ossified accessory orbital element (palpebral/supraorbital): Absent (0); Present (1).
30. Palpebral position: Projects into the orbit (0); Incorporated into the orbital margin (1).
31. Palpebral, number: One (0); Two or more (1).
32. Slot in maxilla for lacrimal: Absent (0); Present (1).
33. Posterior maxilla-lacrimal contact excludes jugal from margin of antorbital fossa: Absent (0);
Present (1).
34. Lateral extension of jugal (jugal boss): Absent (0); Present (1).
35. Lateral ridge on the surface of the jugal: Absent (0); Present (1).
36. Jugal orbital ramus depth vs. transverse breadth: Deeper (0); Broader (1).
37. Shape of postorbital: T-shaped (0); Plate-like (1). 133
38. Postorbital-squamosal bar: Subequal mediolaterally and dorsoventrally (0); Transversely
broad and dorsoventrally flattened (1).
39. Ascending process of the quadratojugal contacts the descending process of the squamosal:
Present (0); Absent (1).
40. Large quadratojugal foramen: Absent (0); Present (1).
41. Frontals, anteroposterior length exceeds mediolateral length: Absent (0); Present (1).
42. Parietosquamosal shelf that obscures the occiput in dorsal view: Absent (0); Present (1).
43. Paroccipital processes: Extend laterally and expand distally (0); Distal end pendant and
extends ventrally (1).
44. Infratemporal fenestra: Significantly smaller than the orbit (0); Subequal to larger than the
orbit (1).
45. Mandibular articulation (ventral condyles of the quadrate): Subequal (0); Medial condyle is
larger (1).
46. Predentary: Absent (0); Present (1).
47. Predentary: Shorter than the premaxilla (0); Subequal to longer than the premaxilla (1).
48. Ventral process of the predentary: Present (0); Very reduced or absent (1).
49. Dentary symphysis: V-shaped (0); Spout-shaped (1).
50. Dentary tooth row in lateral view: Straight (0); Sinuous (1).
51. Lateral ridge on the posterior dentary indicating emargination that accounts for
approximately 50% of the transverse width: Absent (0); Present (1).
52. Coronoid process formed at least partially by the posterior process of the dentary: Absent (0);
Weakly developed, posterodorsally oblique, depth of mandible at coronoid is <140% depth 134
of mandible beneath tooth row (1); Well-developed, distinctly elevated, depth of mandible at
coronoid is >180% depth of mandible at tooth row (2).
53. External mandibular fenestra: Present (0); Absent (1).
54. Position of jaw articulation: Level with maxillary tooth row (0); Offset ventrally relative to
maxillary tooth row (1).
Postcranial (axial):
55. Epipophyses on the anterior cervical vertebrae: Present (0); Absent (1).
56. Zygapophyseal articulation of the dorsal vertebrae: Flat (0); V-shaped (1).
57. Axial epipophyses: Present (0); Absent (1).
58. Number of dorsals: Fifteen (0); Sixteen or more (1); Twelve to thirteen (2).
59. Number of sacrals: Two to three (0); Four to five (1); Six or more (2).
60. Ossified sternal ribs: Absent (0); Present (1).
Postcranial (appendicular):
61. Deltopectoral crest: Well-developed flange (0); At most incipiently developed as an
anterolateral ridge on the humerus (1).
62. Humeral length: More than 60% of femoral length (0); Less than 55% of femoral length (1).
63. Metacarpals with block-like proximal ends: Absent (0); Present (1).
64. Second phalanx of the second and third manual digits: Shorter than the first phalanx (0);
Longer than the first phalanx (1).
65. Forelimb reduced to 40% of hindlimb length: Absent (0); Present (1). 135
66. Shape and position of the preacetabular process of the ilium: tab-shaped, distal end is
posterior to the pubic peduncle (0); strap-like, distal end is anterior to pubic peduncle (1).
67. Preacetabular process of the ilium expands distally in dorsal view: Absent (0); Present (1).
68. Brevis shelf and fossa on the postacetabular process of the ilium: fossa faces laterally and
creates a deep postacetabular process (0); fossa faces ventrally, forming at least a narrow
horizontal shelf (1).
69. Length of the postacetabular process of the ilium: 30% (0); 20% or less (1); 40% or more (2)
of the entire length of the ilium.
70. Ventral acetabular flange: Present (0); Absent (1).
71. Supraacetabular flange: Present (0); Absent (1).
72. Pubic peduncle of ilium relative to the ischial peduncle: Large, robust (0); Short, reduced in
size (1).
73. Ischial peduncle of ilium: Projects ventrally (0); Broadly swollen, projects ventrolaterally (1).
74. Dorsal groove on ischium: Absent (0); Present (1).
75. Tab-shaped obturator process on the ischium: Absent (0); Present (1).
76. Ischial symphysis: Along at least 50% of the length (0); Restricted to distal portion only (1).
77. Orientation of the pubis: Anteroventral (0); Posteroventral, opisthopubic (1).
78. Shape of pubic shaft: Blade-like (0); Rod-like (1).
79. Pubic symphysis: Elongate (0); Restricted to the distal portion, or absent (1).
80. Prepubic process: Absent (0); Present (1).
81. Prepubic process, shape: Blade-like (0); Rod-like (1).
82. Prepubic process: Poorly developed, stub-like (0); Well-developed, elongate (1). 136
83. Femoral head: Confluent with greater trochanter, separated by shallow groove (fossa
trochanteric) (0); Distinct constriction separating greater trochanter and femoral head (1).
84. Shape of anterior (lesser) trochanter: Salient spike or ridge (0); Subequal in width to the
greater trochanter, separated from greater trochanter by a cleft (1); narrow, cleft between it
and greater trochanter absent (2).
85. Proximal anterior trochanteric height relative to the femoral head: Lower (0); Approximately
the same level (1).
86. Fourth trochanter, pendant: Absent (0); Present (1).
87. Location of the fourth trochanter: Proximal half of the femur (0); Midlength or lower on the
femoral shaft (1).
88. Femoral length compared with tibial length: Shorter than tibia (0); Longer than tibia (1).
89. Proximal/Distal portions of the tibia: Distal tibia less expanded than proximal tibia (0);
Proximal and distal portions are subequal (1).
90. Fibula reduced to a splint distally: Absent (0); Present (1).
91. Tibial facet on the calcaneum: Absent (0); Present (1).
92. Medial distal tarsal: Articulates only with metatarsal 3 (0); Articulates with metatarsals 2 and
3 (1).
93. Length of metatarsal 5 relative to metatarsal 3: More than 50% (0); Less than 25% (1).
94. First phalanx of pedal digit I is the longest non-ungual phalanx of the pes: Absent (0);
Present (1).
Other Characters:
95. Ossified epaxial tendons: Absent (0); Present (1). 137
96. Ossified hypaxial tendons on caudals: Absent (0); Present (1).
97. Postcranial osteoderms (scutes): Absent (0); Present (1). 138
APPENDIX 2.
Taxon-Character Matrix
Below is the data matrix compiled from the characters and character states presented in
Appendix 1. Marasuchus and Saurischia are the designated outgroup taxa. The asterisks (*)
over characters 2, 16, 18, 24, 29, 32, 40, 46, and 90 indicate parsimony-uninformative characters.
Each character, where applicable, is scored numerically as: (0), the presumed plesiomorphic or
ancestral state; (1), the more derived state; and (2) or (3) for any character that has more than one
derived state (i.e., characters 1, 4, 5, 52, 69, and 84). The (?) indicates that the character state is
unknown in that particular OTU either because the material that would possess that character is
not present in any known specimen or the character state is unclear due to several factors, including poor preservation. A coding of (-) indicates that the character state in question is not applicable to that OTU because that character is absent. A (M) coding indicates a multistate, or polymorphic, character for that OTU. In this study, (M) indicates variation only among the composite-coded supraspecific OTUs; however, it should be noted that intraspecific variation does occur, possibly due to ontogenetic variation, but that did not factor into this analysis.
139
10 20 30 40 50
* * * * * * * * Marasuchus ????00?000 000??????? 00-????0?? ?????????? ?????????? Saurischia 10-0000000 0000000000 00-100100- -000000000 000100--00 Pisanosaurus ???010?0?1 11100????? ???????1?? ?????????? ?????????0 Eocursor ???1101010 1????????? ?????????? ?????????? ?0??????10 Lesothosaurus 0000100010 1110100110 00-0?00010 0100000000 0000010010 Stormbergia ?????????? ?????????? ?????????? ?????????? ?????????? Heterodontosaurus 2111211101 1011000101 1110100110 0001100000 0011010100 Abrictosaurus 2011211001 111000010? 1110?00110 000?10???? ???0?11100 Agilisaurus 1000210011 11111001?0 00-0100110 1?000000?0 0010110010 Hypsilophodon 1011211111 1110100110 1100?11110 0010000011 1010011010 Thescelosaurus ?????01??1 ??????0??? ???????110 0?100000?? 10?001??10 Iguanodontia 3--0211110 1110100101 110M01111M 00M000011M 00M0011010 Pachycephalosauria 301010?111 1111101100 M11M01-111 1011101110 0110010?00 Ceratopsia 3--1211101 111011110M 010M011111 0001101110 01M1011010 Scutellosaurus 000?100010 111010???0 00-0??01?? ???001???? 1???1????0 Emausaurus 1002100010 1110100?00 00-0?00110 0000010010 00?0110011 Scelidosaurus 0002100010 1110100??0 00-0010111 1000010010 00??011010 Stegosauria 3--2100110 1110100100 00-0010111 1000010010 00M0110011 Ankylosauria 3--2100110 1110100100 00-M01-111 1000011010 0000110111
140
60 70 80 90 * Marasuchus ????101?00 00??000000 1000000000 --00000000 010?000 Saurischia 0000000000 0111000000 0000000000 --00000000 0000000 Pisanosaurus 1?11?????? ?????????0 ?????00??0 --??????00 0?????0 Eocursor 1001?0??1? 00?1011101 0001001111 0001010010 ????1?0 Lesothosaurus 0001001?1? 0100111000 0001001111 0001010010 101?100 Stormbergia ????001?1? ????111000 0001111111 0001010010 1?1?100 Heterodontosaurus 11010?1220 0011011101 1000001111 0001110001 ?010100 Abrictosaurus 1101????1? 00??011101 100?0011?? ???11100?? ??????0 Agilisaurus 11111?1010 01??111000 0101111111 0101010010 1010100 Hypsilophodon 1111101021 1000011121 1110111111 1112110010 1110110 Thescelosaurus 1110?0?111 1?1?011101 11?0111111 1112111110 1111110 Iguanodontia 111M111120 0010011101 1110111111 01121M1110 11111M0 Pachycephalosauria 011111??2? 01??111121 1110001111 111?1100?M 11??110 Ceratopsia 1110111220 0010011121 1110001111 1M12100010 1111100 Scutellosaurus 1????0?01? 000?011?0? ??0?001111 ??01010010 ????1?1 Emausaurus 1001?????? ??0??????? ?????????? ?????????? ??????1 Scelidosaurus 1011000010 00??011010 0000001111 0001010110 1010101 Stegosauria 1111111?10 0010011?10 0000001111 010??0M110 1011001 Ankylosauria 1111111?10 00??011?10 ??00001111 000??0M110 1011101