HONR 259C "Fearfully Great Lizards": Topics in Dinosaur Research

Fall Semester 2017
Dinosaur Diversity I: Ornithischia

Dinosaur Park Formation large ornithischians Chasmosaurus (chasmosaurine ceratopsid), Lambeosaurus (lambeosaurine hadrosaurid), Styracosaurus (centrosaurine ceratopsid), Euoplocephalus (ankylosaurid ankylosaur), Prosaurolophus (hadrosaurine hadrosaurid), and Panoplosaurus (nodosaurid ankylosaur). Detail of image by J. Csotonyi from Mallon & Anderson (2013) doi: 10.1371/journal.pone.0067182

Key Points:
•Thyreophorans represent the armored dinosaurs, and are a clade of (predominantly) quadrupedal ornithischians.
•There are characterized by the presence of osteoderms (armor plates) in their skin. Different clades of thyreophorans express these osteoderms in different patterns.
•Beyond a few basal taxa, thyreophorans are divided into the plated Stegosauria and the tank-like Ankylosauria.
•Armor in thyreophorans seem to have functions beyond simple defense: they served as display structures and (in the case of the stegosaurs and the club-tailed ankylosaurine ankylosaurs) as active weapons.
•Neornithischia is the sister-group to Thyreophora. It contains the Marginocephalia, the Ornithopoda, and a series of basal forms.
•Thescelosauridae is the most diverse group of non-marginocephalian, non-ornithopod neornithischians. Thescelosaurids remained obligate herbivores, and were minor herbivores in most regions of the world in the Cretaceous.
•Ornithopoda was one of the most successful of all dinosaur herbivore groups. Primitive members were small obligate bipeds, but many evolved into facultative quadrupeds. Some include the largest land animals other than sauropods of all time.
•Ornithopods showed extensive modification of their chewing ability, culminating in the mobile skulls and dental batteries of Hadrosauridae. •The Hadrosauridae of the Late Cretaceous was the most speciose branch of Ornithopoda. These "duck-billed" dinosaurs are known from the entire life cycle, and from entire herds. Both major clades--hollow-crested Lambeosaurinae and broad-snouted Hadrosaurinae--show extensive features for some form of visual (and/or auditory) displays, suggested complex social interactions.
•Marginocephalia is a clade of herbivorous ornithischians characterized by an expansion of the bones at the rear of the skull. There are two major clades within it: Pachycephalosauria and Ceratopsia
•Pachycephalosaurs are currently only known from the Late Cretaceous of Asia and North America. They were obligate herbivores, and recognized by a thickening of the skull roof (possibly for combat between individuals) and spike ornaments along the edge of the skull.
•Ceratopsians first appear in the earliest Late Jurassic and make it to the end of the Cretaceous. They all share a rostral bone: a single bone anterior to the paired premaxillae.
•Primitive ceratopsians were facultative bipeds. These include the Chaoyangsauridae and Psittacosauridae. The more derived Neoceratopsia is characterized by a frill at the back of the head.
•A series of basal neoceratopsian forms lie outside the large, derived, speciose Ceratopsidae. Ceratopsids (as well as some of their immediate outgroups) were obligate quadrupeds. Ceratopsidae is characterized by a shearing dental battery, the presence of horns on the postorbitals and snout, and a series of small bones along the edge of the frill.
•Ceratopsids are limited to the the end of the Cretaceous and (with a few exceptions) to western North America. It is divided into two clades: Chasmosaurinae and Centrosaurinae. In both clades there is evidence for herd living.

Simplified cladogram of Ornithischia:

Ornithischians are, at present, not definitely present in the Triassic. Since the 1970s the oldest and most primitive ornithischian was thought to be Pisanosaurus of the early Late Triassic Argentine Ischigualasto Formation. The fossil is incomplete, so many aspects of its anatomy are uncertain. But new analyses in 2016 and 2017 have shown this is probably a silesaurid, not a dinosaur.

Ornithischia is characterized by the following traits:

Based on their tooth form and the retroverted pubis, ornithischians were herbivorous. (That doesn't mean that they were exclusively plant eaters, of course! In the modern world, many "herbivorous" sauropsids and mammals eat some meat.)

Heterodontosauridae was a primitive group of ornithischians. Although one fragmentary specimen was thought to be from the Late Triassic, redating shows it was younger; the oldest heterodontosaurids are known from Early Jurassic, and persist into the mid-Early Cretaceous. They had skulls which are relatively deep and powerfully built, indicating that they ate fairly tough food. Advanced heterodontosaurids also had premaxillary tooth row that were ventral to the maxillary tooth row and jaw joints that were ventral to the dentary tooth row: the result were jaws that brought the teeth together all at once (like a nutcracker), and not slicing (as in scissors, or as in most dinosaurs). These latter jaw adaptations evolved convergently in Ornithopoda, and so for a long time Heterodontosauridae was considered a clade within Ornithopoda. However, primitive heterodontosaurs lack these convergent adaptations.

Most heterodontosaurids are quite small. Some are only about 1-1.5 m long, and Fruitadens of the Late Jurassic of western North America may have been no more than 80 cm long as an adult (most of which length is tail) and Manidens of the Middle Jurassic of Argentina only 65-75 cm; that makes these the smallest known ornithischians.

Interestingly, the early Late Jurassic Chinese heterodontosaurid Tianyulong had a fuzzy body covering over at least part of its body! If this is found to be homologous to the protofeathers of tetanurine theropod saurischians it would suggest that the concestor of all dinosaurs was fuzzy, and that dinosaurs were thus fuzzy ancestrally! At present, however, there is enough uncertainty to make the homology between Tianyulong's fuzz and tetanurine protofeathers suspicious. (But do not be terribly surprised if in the future we discover that most dinosaurs were fuzzy to some degree or another! All we need is a fuzzy primitive sauropodomorph, and it is basically a done deal!) (By the way, the initial reports placed Tianyulong in the Early Cretaceous, but the formation in which it was found has been redated to the earliest part of the Late Jurassic, around 160 Ma.)

The similarly aged neornithischian Kulindadromeus of Siberia also shows simple filimentous fuzz, as well as scales, plates, and additional bizarre tufted plates, showing that primitive ornithischians had a wide variety of integumental features.

Once thought to be from the Late Triassic, Eocursor of South Africa is from the earliest Early Jurassic. Of comparable age is the earliest Early Jurassic Laquintasaura. This is the oldest and most primitive ornithischian known from group assemblages, indicating that at least some of these lived in groups during life.

Ornithischians more derived than Heterodontosauridae, Eocursor, and Laquintasaura had greatly reduced hands, losing most of their grasping ability. This suggests a switch to jaws-only while obtaining food.

The remaining ornithischians (Genasauria, the "cheeked reptiles") include the armored Thyreophora and the highly diverse Neornithischia (especially beaked Ornithopoda and ridge-headed Marginocephalia).

Simplified cladogram of Thyreophora

More detailed phylogeny of Genasauria, with emphasis on Thyreophora


Thyreophora are united by various skeletal attributes, the most obvious of which is:

The derived thyreophorans are the plated Stegosauria (shingled lizards) and the heavily armored Ankylosauria (fused lizards). Both clades are present by the Middle Jurassic. There are several Early Jurassic taxa which fall outside either of the two advanced clades.

Lesothosaurus of the Early Jurassic of southern Africa has sometimes been found to be the oldest and most primitive known thyreophoran in several recent new analyses. If so, it is united with the later thyreophorans on various skeletal traits that are outside the scope of this course. Importantly, it would be the only known unarmored thyreophoran. (But we would expect the oldest ones to have been unarmored.) However, more recent information indicates it is more likely the oldest and most primitive neornithischian.

The oldest and most primitive definite (i.e., armored) thyreophoran is Scutellosaurus of the Early Jurassic of western North America. It was a 1.5 m long biped (possibly facultative biped) not very dissimilar to other primitive ornithischians like Lesothosaurus or Hexinlusaurus: small herbivores with small hands. The primary distinction of Scutellosaurus is the presence of a great many small osteoderms over the body. These would protect against small-bodied predators, but might not help against the new larger theropods that had begun to appear in the Early Jurassic.

In response, thyreophorans evolved larger size and heavier armor, as seen in Emausaurus and Scelidosaurus (new specimen shown here), both of Europe. The larger body size (3-4 m long) and proportionately larger osteoderms may have been more effective defense against attacking predators, but forced them onto all fours (at least for Scelidosaurus: in other words, they were obligate quadrupeds.

(Note that the hypothesis shown here is that Scutellosaurus, Emausaurus, and Scelidosaurus were progressive closer to the Stegosauria-Ankylosauria clade (Eurypoda). However, some paleontologists have considered Emausaurus to be a primitive stegosaur, and others that Scelidosaurus was the oldest and most primitive ankylosaur. However, eurypods share a number of transformation not found in these Early Jurassic taxa: these include:

among others.)

From a Scelidosaurus-like ancestor, the stegosaurs evolved armor that was less covering and more concentrated. While they had some small osteoderms in their skin (particularly around the neck and the hips), most of their armor was specialized as:

In general, the stegosaur armament suggests active defense: the dinosaur probably turned in response to attacking predators, trying to keep the tail towards the theropod so that it could use its thagomizer. Damaged thagomizer spikes and theropod bones with thagomizer-generated puncture wounds confirms their use in defense.

Early stegosaurs were only about 2.5-3 m long, but the most derived forms ranged up to 9 m or more. Their narrow snouts suggests that they were rather picky eaters (that is, instead of munching a lot of plants at once, they were selective as to which ones they chomped.) Biomechanical analysis shows that their bite was stronger than many herbivorous saurischians, but still weaker than many specialized ornithischians. Although they were obligate quadrupeds in terms of locomotion, they may have been able to rear on their hind legs in order to feed higher in trees.

Some tantalizing footprints suggest possible Early Jurassic stegosaurs, but these may be from a more basal Scelidosaurus-like thyreophoran instead. Primitive stegosaurs include Middle Jurassic Huayangosaurus and Late Jurassic Chungkingosaurus, Gigantspinosaurus and Tuojiangosaurus (all from China). More derived stegosaurs form the clade Stegosauridae, and include Late Jurassic Dacentrurus and long-necked Miragaia of Europe, Kentrosaurus of eastern Africa, and the Stegosaurinae.

Stegosaurids had dorsoventrally stretched neural arches and disproportionately short forelimbs, giving them an odd profile.

The most derived stegosaurids (Stegosaurinae) lacked shoulder spines (also missing in Tuojiangosaurus), had only plates rather than spikes along the back (again, shared with Tuojiangosaurus), and had alternating rather than parallel plates. This advanced group includes the three Late Jurassic western North American genera (Hesperosaurus, Alcovasaurus [figures A, B, E & F], and famous Stegosaurus [also known from Portugal]), and Early Cretaceous Wuerhosaurus of China, the last of the stegosaurs. (Some stegosaur fragments are known from Europe about the same age as Wuerhosaurus, but are not distinctive enough to place within the stegosaur phylogeny). (Note: some paleontologists consider Alcovasaurus, Wuerhosaurus, and Hesperosaurus to all belong within the genus Stegosaurus.)

Stegosaurs are relatively common in Middle and Late Jurassic formations (especially so in China), are present but rare in some European assemblages, and vanish before the end of the Early Cretaceous. Claims of later stegosaurs have so far turned out to be either mis-dated or misidentified.

While the stegosaurs evolved active defense, the ankylosaurs (at least at first) seem to have been selected for passive defense: the ability to stay put and absorb damage from an attack. They were even more extensively armored than Scelidosaurus, and were characterized by:

Additionally, in ankylosaurs the predentary bone is reduced, and the jaws arranged so that they would have more extreme rotation (and also be pulled back further) when chewing than in most ornithischians. This evolved convergently (and to a far more extreme form) in the advanced ornithopods. This motion allowed ankylosaurs to more effectively chew up their food.

Ankylosaur systematics remains contentious. There does seem to be a well-supported clade comprised of Early Cretaceous Sauropelta and Late Cretaceous Edmontonia and Panoplosaurus (all of North America): the Nodosauridae. There is also the well-supported clade Ankylosauridae discussed below. There are a number of other ankylosaurs, however, which are considered by some to be either primitive nodosaurids and ankylosaurids; and/or to lie outside a nodosaurid-ankylosaurid clade; and/or to form their own clade Polacanthidae. The latest published study places such forms as either basal nodosaurids or basal ankylosaurids, or just outside both. The oldest ankylosaur (Middle-Late Jurassic Tianchisaurus of China) was not included in this study and may represent an ankylosaur outside the Nodosauridae-Ankylosauridae clade. All ankylosaurs in that study were other closer to Ankylosaurus than to Nodosaurus (and thus members of Ankylosauridae) or closer to Nodosaurus than to Ankylosaurus (and thus members of Nodosauridae).

In the study followed here, Late Jurassic North American Mymoorapelta and Gargoyleosaurus, and Early Cretaceous Gastonia of North America and Polacanthus of Europe are among the most primitive nodosaurids. Sauropelta and later Cretaceous forms seem to form a clade. Advanced nodosaurids are united by a large muscle attachment projection on the scapula and often have a tall laterally-directed shoulder spine. In addition to the North American taxa listed above, nodosaurids include many additional Cretaceous North American taxa such as Animantarx and spectacularly-preserved Borealopelta, as well as radiation of primarily European forms such as Europelta of the Early Cretaceous and Struthiosaurus of the Late Cretaceous. Nodosaurids range from about 3-7 m in length. They are known (as shown above) from Europe, North America, and Asia; there are possible nodosaurid bones from other regions (including Antarctopelta from Antarctica and an unnamed form from Argentina).

Early Cretaceous Kunbarasaurus of Australia seems to be the most primitive known ankylosaurid, with Hylaeosaurus (the first discovered thyreophoran) slightly more derived. Another primitive ankylosaurid is tiny Liaoningosaurus of China (to be fair, the only known specimen is a juvenile, so we don't know how big it got.) Ankylosaurids have distally-stiffened tails

In the more derived clade Ankylosaurinae there are extremely complex air chambers in the skull (convergently evolved with Gastonia)

Additionally, ankylosaurines are further transformed relative to other armored dinosaurs in having:

Thus, ankylosaurines had a more active defense (in the form of the tail club) than other ankylosaurs. Ankylosaurids ranged from about 3 m to 8 m (the latter include (Cedarpelta of Early Cretaceous western North America) and some of the most derived (Tarchia of Mongolia and Ankylosaurus of western North America). Pinacosaurus of Asia is one of the best known primitive ankylosaurines. Other Asian forms from the Late Cretaceous include Zaraapelta, Saichania, and Talarurus. Euoplocephalus, Dyoplosaurus, Ziapelta, Scolosaurus and Anodontosaurus of the Late Cretaceous of western North America are the best studied.

Ankylosaurines are only known from the late Early Cretaceous of Asia and the Late Cretaceous of Asia and western North America at present; similar distributions are known for various other dinosaurs, as we shall see.

Ankylosaurids are more than their armor! Recent work has demonstrated that the bones which control the tongue and throat are incredibly well-developed in these dinosaurs: far more so than all other dinosaurs (outside of some birds.) They almost certainly have long, powerful tongues: possibly for insect eating, possibly for grabbing plants, and possible for both and more.

Probably the most conspicuous aspect of their evolution. Thyreophorans begin with a set of small scutes; develop larger scutes at the expense of bipedalism (and speed); than split between specialists in active (Stegosauria) vs. passive (Ankylosauria) defenses. Within the ankylosaurs, though, Ankylosaurinae independently evolves an active defensive tail weapon.

Relative success
Stegosaurs and ankylosaurs, as sister taxa, have their origins at the same time. However, stegosaurs flourish first (in the Middle and Late Jurassic), only to peter out during the Early Cretaceous and disappear before its end. Ankylosaurs are typically rare in the Jurassic (although at one location Gargoyleosaurus is very common), but become abundant in the Cretaceous.

Feeding adaptations:
Narrow-snouted basal thyreophorans and stegosaurs differ from broader-snouted ankylosaurs. The latter have a greatly reduced predentary bone, which may have allowed for more complex motion of the dentaries (for additional munching power) while feeding. The powerful tongue of ankylosaurids hint at some specialized form of feeding.

Group behavior:
Thyreophorans are only rarely found in mass death assemblages, and so (unlike some sauropods, ornithopods, and marginocephalians) probably did not live in large groups.

Display structures:
The spikes, plates, and osteoderms of thyreophorans almost certainly had a defensive function. But many are very broad (particularly stegosaur plates), and the patterns tend to be specific to each species. They may have served an additional function: as visual displays of species recognition. There may also have been a sexual display function to them, but at present it is uncertain if the variation we see in some thyreophorans is from sexual dimorphism or if it is from multiple species of the same genus living together.

Detailed phylogeny of Genasauria

The Incredible Shrinking Ornithopoda

Traditionally, Ornithopoda ("bird feet") comprised all ornithischians that weren't stegosaurs, ankylosaurs, or neoceratopisans. Eventually, pachycephalosaurs were recognized as their own distinct clade, and psittacosaurids as ceratopsians. With the development of cladistic analysis, it was recognized that Scutellosaurus and Scelidosaurus belonged with the stegosaur-ankylosaur clade, and that Pisanosaurus and Lesothosaurus were primitive ornithischians outside of all the other major groups.

But even at the dawn of the 21st Century, Heterodontosauridae was still generally considered as sharing a more recent common ancestor with the "hypsilophodonts" and iguanodontians than with any other group of dinosaur: thus, the heterodontosaurids were thought to be the oldest branch of Ornithopoda. More recently, however, heterodontosaurids have been recognized as splitting from other ornithischians at a very basal divergence, and thus are no closer to ornithopods than to marginocephalians or to thyreophorans. So there are at present no recognized Late Triassic or Early Jurassic ornithopods.

Even more recently, Middle Jurassic taxa (such as Chinese Agilisaurus and Hexinlusaurus, early Middle Jurassic Siberian Kulindadromeus, and Late Jurassic western North American Othnielosaurus (all of which were considered to be primitive hypsilophodont-grade ornithopods in many cladistic analyses) now seem to be basal members of Neornithischia. This latter clade includes marginocephalians, true ornithopods, and all taxa closer to them than to Thyreophora.

(The recognition of the non-ornithopod status of these Jurassic forms comes in great part with the discovery of Eocursor, "Stormbergia" (really, adult Lesothosaurus), and other Late Triassic/Early Jurassic ornithischians. The data from these fossils are helping to sort out the relationships of the bird-hipped dinosaurs.)

But wait; there's more! A brand-new (late 2015) extensive survey of ornithischian relations reduces Ornithopoda even further, which is the source of the phylogenies shown above and below. A cluster of (mostly) Cretaceous small (1-3 m) beaked bipedal ornithischians previously called "hypsilophodontians" were found to not represent either a clade of ornithopods (the old model) or a grade of basal ornithopods (the newer model). Instead, nearly all lie outside the Ornithopoda-Marginocephalia clade Cerapoda. (Cerapods are united by the shared presence of asymmetrical teeth with enamel on only one side, allowing them to be ever-sharpening.)

Instead, as well as the aforementioned Jurassic Agilisaurus, Hexinlusaurus, and Othnieliosaurus, the Cretaceous ultra-long tailed Australian Leaellynasaura and the Cretaceous Chinese clade Jeholosauridae (including Jeholosaurus and Yueosaurus) belong in this part of the tree.

Of currently uncertain position is Kulindadromeus; it might be a basal parksosaurid, or it might be outside the parksosaurid-cerapod clade. It is a noteworthy animal: not so much in terms of its skeleton (which is boringly standard for a neornithischian), but for its behavioral evidence and integument. Firstly it was found in bonebeds of many dozens of individuals, so it is very likely it lived in groups. More interesting than that, though, is its body covering. It is found in an environment which fine details can be preserved. It has some parts of its body (bottoms of the feet) there are simple scales; on the front of the legs and on top of the tail are more plate-like scales; on other parts of the body are simple filaments; and then there are odd plates with fuzz coming off them (unlike any structure known in other dinosaurs). So small ornithischians show complex types of integument beyond scales (and beyond fuzz).

Simplified cladogram of Thescelosauridae & Ornithopoda

More detailed phylogeny of Thescelosauridae & Ornithopoda

So what IS an ornithopod, then? Ornithopoda is defined as Parasaurolophus and all taxa closer to it than to Triceratops. In the latest study very few of the traditional "hypsilophodonts" fall within this definition: Early Cretaceous Hypsilophodon itself and a few others mentioned below. Instead a clade of Cretaceous small-bodied bipedal dinosaurs traditionally considered ornithopods was found to lie outside Cerapoda. This clade has sometimes been called "Parksosauridae" (for instance, in earlier versions of this course!), but the proper name is Thescelosauridae. At present no definite Jurassic thescelosaurid is known, but Kulindadromeus (or possibly some other Jurassic taxon) might turn out to be early thescelosaurids.

Thescelosaurids are united in having fused premaxillae and a series of other modifications of the front end of the snout, among other traits. There are two main clades of thescelosaurids:

Thescelosaurids lacked serious defensive weapons, and their limb proportions were not particularly well-adapted to fast running. Their narrow beaks suggest that they were choosy eaters in terms of the plants they selected.

Given the reorganization of the positions of these various neornithischians, many features once used to characterize Ornithopoda are now more broadly distributed among dinosaurs. Instead, one of the more obviously attributes that characterizes the greatly reduced Ornithopoda are maxillary tooth crowns that are shorter than wide. Additionally, ornithopods have a more complex chewing (grinding of upper teeth against lower ones) than other dinosaurs. A hinge is present between the premaxilla, upper part of the jaws, and braincase on the one side and the maxilla and bones of the cheek region on the other. It was once thought that this had a simple out-and-back motion to help grind the teeth while chewing. As we will see below, however, the motion is more complex. In any case, even early ornithopods seem to be able to grind up their food to a finer degree than most dinosaurs, allowing them to more quickly nutrients from that food.

(A note on the name "Ornithopoda": advanced iguanodontians do indeed have three-toed feet something like birds, as seen in these tracks. But basal ornithopods have four forward-facing toes, and no ornithopod seems to have the backwards-facing digit I of birds. In fact, it is kind of a lousy name for the clade, but rather late in the game to change it...)


Currently the basalmost ornithopod is the first discovered and best known of this range of ornithopods is little 2 m long Hypsilophodon of Early Cretaceous Europe (and possibly North America). All remaining ornithopods fall within the diverse clade Iguanodontia.

The members of Iguanodontia were transformed from their "hypsilophodont" cousins by a number of features:

All retained some bipedal ability, but many of the iguanodontians were facultative bipeds only, spending a sizable fraction of time on all fours.

The oldest iguanodontian known is the Middle Jurassic dryosaurid Callovosaurus. Iguanodontians become more common in the Late Jurassic, but really come into their own in the Cretaceous. In many ecosystems the iguanodontians are the most abundant large animals, displacing sauropods and stegosaurs.

Among the diversity of primitive iguanodontians are:

The remaining ornithopods form the specialized Cretaceous clade Styracosterna. Their snouts have become longer and broader-ended with a better developed grinding jaws, while their hands have become better adapted for absorbing weight. These transformations are more fully developed in the hadrosauriforms.

Simplified cladogram of Styracosterna

This represents the clade comprised of Hadrosauridae and all taxa closer to hadrosaurids than to Camptosaurus. The primitive styracosternans were once all grouped together as "Iguandontidae", at least some of the old "iguanodontids" turn out to be paraphyletic with respect to hadrosaurids. Styracosterna is by far the most successful radiation among the ornithischians.

Styracosternans show the following transformations from the ancestral state:

The combination of their great size, ability to walk on their hindlegs or all fours, and powerful beaks with grinding teeth allowed styracosternans to be excellent browsers of both low and high vegetation. At least some seem to have lived in herds.

Styracosternans are known from most of the Cretaceous world, but are most particularly common or diverse in Europe, North America, Asia, and (in the Early Cretaceous) northern Africa. Among the diversity of Early Cretaceous styracosternans are:

Recent studies do support a monophyletic Iguanodontidae (after a decade or so when this cluster of dinosaurs were a paraphyletic series with respect to Hadrosauridae). Iguanodontids were a sucessful group of large-bodied Early Cretaceous ornithopods. Nearly all have a very prominent thumb spike (but to be fair, so do more basal styracosternans.) Among the the iguanodontids currently recognized are:

One subset of styracosternans (Hadrosauria) in particular shows a series of transformations including an increase in the number of tooth positions in the jaws and expansion of the snout. These dinosaurs are on the lineage which leads to the duckbilled dinosaurs (Hadrosauridae). Among the precursors and cousins of the hadrosaurids are Early Cretaceous tall-snouted Altirhinus of Asia (once considered a species of Iguanodon, Equijubus of North America, Gongpoquansaurus and Probactrosaurus of China; Eolambia of western North America; Early-to-Late Cretaceous Protohadros of western North America; and Late Cretaceous Bactrosaurus and Plesiohadros of Asia. (These latter two fall out within Hadrosauridae proper in some analyses). There are many others, and more are being named every year.

True Hadrosauridae is the most speciose and specialized branch of the ornithopods. All known members of Hadrosauridae proper are from the Late Cretaceous. Although known from Europe, South America, and Antarctica, the main diversity of hadrosaurids is in Asia and North America.

The transformations of hadrosaurids relative to their ancestors include:

Hadrosaurids see the fullest expression known of the ornithopod grinding mechanism. As mentioned above, it was once thought that the motion was relative simple: the side unit would move outwards when the lower jaw was brought up, giving a side-to-side grinding of the teeth during chewing. This model (proposed during the 1980s) was called "pleurokinesis" (or "side-motion"). Here is a video of a computer animation of this interpretation:

However, more detailed study using CT scans and more complete computer models show the motion is really a LOT more complex. Pleurokinesis plays a part in it, but there are other directions as well. No name is given at present for this form of jaw mechanics, but below is a preliminary model of how it works:

Their exceedingly-complex tooth form--which were made of six different tissue types, rather than the standard two of most tetrapods--maintained a good girding surface as they wore down.

Hadrosaurids include some definite herd dwellers. The entire life cycle of hadrosaurids is preserved: nests, eggs, embryos, hatchlings, juveniles, subadults, and adults. Hadrosaurid footprints and isolated hadrosaurid teeth are among the most common Late Cretaceous fossils of North America. Skin impressions and even mineralized soft tissue are known for duckbills.

The latest on-going phylogenetic analyses show two major subclades of Hadrosauridae: crested Lambeosaurinae and broad-snouted Hadrosaurinae. The latter group has sometimes been called "Saurolophinae", because in some analyses Hadrosaurus proper of New Jersey and the closely related Eotrachdon of Alabama (and the very similar Italian Tethyshadros and Telmatosaurus of Romania) seems to lie outside the Lambeosaurinae-Hadrosaurinae clade (Euhadrosauria) (However, a note of caution: some preliminary studies suggest that "hadrosaurines" may be paraphyletic with respect to Lambeosaurinae). Both the major clades are known from a great number of excellent skeletons.

Lambeosaurines are characterized by a hollow crest covering the nasal passage. These crests, which vary between species, may have had both a visual and sound display function. Baby lambeosaurines lacked this structure.

CT scans allow for the pathways of these passages to be studied in greater detail:

Differences in crest size and shapes within some populations may reflect sexual and/or ontogenetic varations.

Among the better known lambeosaurines are Nipponosaurus, Olorotitan, Tsintaosaurus, and Charonosaurus of Asia and Parasaurolophus, Corythosaurus, Hypacrosaurus, Velafrons, Lambeosaurus, and GIGANTIC Magnapaulia of North America.

Hadrosaurines (aka "saurolophines") differ from their relatives by greatly flared snouts and greatly expanded nares. Some hadrosaurines had (relatively) short snouts: North American Gryposaurus and Brachylophosaurus, for instance. Others had longer snouts: North American Maiasaura and Prosaurolophus and transcontinental (Asia and North American) Saurolophus. The extreme development of the duckbill can be found in the the Edmontosaurini, a group containing sauropod-sized Shantungosaurus (largest of all ornithischian dinosaurs) of China and North American dwarf Ugrunaaluk and large (but not quite as big as Shantungosaurus) Edmontosaurus proper and species sometimes considered separate genera (Anatosaurus and Anatotitan), but sometimes all considered Edmontosaurus. (There is recent work to show that there are only two species, each with different growth stages: personally, I am fine using the old name Anatosaurus for the geologically-younger annectens, but am willing to follow the common usage and call them all "Edmontosaurus".) Babies of even the long-snouted hadrosaurines had relatively short faces.

Both hadrosaurines and lambeosaurine produced giants of greater than 13 m in length. These represent the largest animals other than sauropods that have ever lived on land, and the heaviest bipeds in Earth's history.

Microwear analysis of the teeth of hadrosaurids is consistent with their complex chewing patterns. In at least the broad-billed edmontosaurs there is a great degree of scratching on the teeth, suggesting that they were primarily low browsers of tough vegetation ("grazers"). (Given the wide snouts of Edmontosaurus, Shantungosaurus, and so forth, this makes a lot of sense.) Studies have not yet been published to see if most hadrosaurids were primarily low browsers, or if some of them might have been mostly high browsers. Given the diversity of bill shapes and snout lengths (and the diversity of species overall), there was probably a number of different diets among the hadrosaurids.

Feeding adaptation transformations:

Locomotory changes:

Social behavior in Ornithopoda:
Neornithischians (in particular ornithopods (in particular iguanodontians (in particular hadrosaurids (in particular lambeosaurines)))) have abundant evidence for socially-related adaptations, including: herding; visual (and possibly aural) displays; species recognition structures; possible sexual dimorphism. We will discuss these more fully in the third section of the course.

Heterochrony, size, and ornithopod history:
In general, peramorphosis seems to play an important role in neornithischian evolution. Hatchling iguanodontians tend to resemble adult "hypsilophodonts", while hatchling hadrosaurids tend resemble young primitive iguanodontians, and young hadrosaurids tend to resemble the immediate outgroups of Hadrosauridae.

Basal neornithischians, thescelosaurids, and basal ornithopods were small (comparable to basal members of other ornithischian groups). But at the base of Iguanodontia and the base of Styracosterna there are major size increases. Additionally, various different styracosternan lineages independantly achieved very large (>12 m) size.

Simplified cladogram of Marginocephalia

More detailed phylogeny of Marginocephalia

The third major clade of ornithischians (after Thyreophora and Ornithopoda) is Marginocephalia ("ridged heads"). Like ornithopods (and thescelosaurids), marginocephalians are neornithischians; collectively, Ornithopoda and Marginocephalia make the clade Cerapoda. There are two main clades of marginocephalians: thick-skulled Pachycephalosauria ("thick headed lizards") and beaked (and later frilled (and eventually horned)) Ceratopsia ("horned faces"). The bipedal pachycephalosaurs and psittacosaurid ceratopsians were once included in Ornithopoda, but are now recognized as closer to the horned dinosaurs (Ceratopsidae) and their more primitive frilled-but-hornless relatives (the rest of Neoceratopsia).

Various postcranial shared derived characters unite Pachycephalosauria and Ceratopsia, but the most distinctive specialization is the one that gives this clade its name:

In the pachycephalosaurid pachycephalosaurs and the ceratopsid ceratopsians there is the shared attribute of extensive bone growth and remodeling of the skull at the fully adult stage: however, this trait is not seen in basal ceratopsians, so it is likely convergently evolved between these two clades.

At present there are no dinosaurs known which are closer to marginocephalians than to the ornithopods, but we can infer that such proto-marginocephalians did exist. When Heterodontosauridae was thought to be a clade of ornithopods, there was the mystery of the missing Late Triassic, Jurassic, and early Early Cretaceous marginocephalians. Now that the oldest known ornithopods are only Middle Jurassic, and we now have Middle-Late Jurassic marginocephalian fossils, there is no big gap in time.


PACHYCEPHALOSAURIA Pachycephalosaurs (aka boneheads aka domeheads aka buttheads aka headbangers...) are specialized by the presence of:

as well as a few postcranial features. Pachycephalosaurs were persistenly bipedal (the only major group of herbivorous dinosaur with no trend towards quadrupedality). Their hands and arms were relatively small, so they probably didn't use them for terribly much.

Stenopelix of the Early Cretaceous of Germany is often considered to be a basal member of Pachycephalosauria based on some limb and pelvic characters. Its skull is unknown, so we cannot determine if it had evolved the thickened skull roof of later pachycephalosaurs. Interestingly, it is the only currently known Early Cretaceous pachycephalosaur, and the only member of the clade from outside Asia or western North America. Of note, however, is a recent study that finds it not to be a pachycephalosaur, but instead a basal ceratopsian (sister taxon to Yinlong): if this is upheld with additional analyses, this would remove the only non-Asiamerican, non-Late Cretaceous pachycephalosaur from the record.

The more specialized Pachycephalosauridae are characterized by a tall thickened dome formed by the frontals and parietals. Paleontologists have debated whether these dinosaurs butted heads together in the manner of modern bighorn sheep, or if the domes may have been more for visual display. There does seem to be some sexual dimorphism in the size and development of the dome.

It was once thought that flat-topped specimens (such as those called "Homalocephale" of Mongolia) were primitive pachycephalosaurs. However, recent work strongly suggests that these are simply subadults which had not grown their adult dome. (The Homalocephale specimens seem to be juvenile Prenocephale.) Recent work suggests that the is a clade of primarily Asian forms (including Prenocephale, Tylocephale, and Goyocephale), nested among a radiation of mostly North American taxa (such as Stegoceras, Gravitholus, Sphaerotholus, and the like). There is a clade (or species!) of latest Cretaceous (68-65.6 Ma) western North American large bodied (4-5 m, and thus bigger than humans) long-snouted spike-fringed and spike-nosed forms: Stygimoloch, Pachycephalosaurus, and Dracorex (which is quite like just a subadult of one or the other of the previous two.) In fact, these three may just represent juvenile ("Dracorex"), subadult ("Stygimoloch"), and adults (Pachycephalosaurus: this name has priority) of the same species. The latest work, however, indicates that the short-spined Pachycephlaosaurus specimens are all stratigraphically older than the long-spined Stygimoloch ones, so these may be distinct taxa.

While material of the pachycephalosaurs are rather limited, the ceratopsian fossil record is very good. The last decade has seen an explosion of discoveries in primitive ceratopsians, extending the stratigraphic range of this group and giving us a more complete picture of the transformations from a basal neornithischian to the derived Ceratopsidae.

Ceratopsians are united by some important cranial specializations, including:

The oldest and most primitive known ceratopsian is Yinlong from the latest Middle Jurassic (or earliest Late Jurassic) of China. Its skull shows the large size of the temporal (jaw muscle) region, indicating that this dinosaur had begun to evolve the powerful bite that characterizes ceratopsians. Other early forms are Late Jurassic Chinese Chaoyangsaurus and Xuanhuaceratops.

The next oldest ceratopsians are the Psittacosauridae ("parrot lizards") of Early Cretaceous Asia. These are actually some of the most common dinosaur fossils of Early Cretaceous Asia: in particular, the several species of Psittacosaurus. These include Psittacosaurus babies found together (but the specimen was artificially modified to put an adult with them!), and many complete skeletons including one with scales and apparently some form of hollow quills sticking out of its tails.

(Note: we do not yet know the homology of the quills of Psittacosaurus. If (and it remains an "if") they represent homologs to the fuzz of the heterodontosaurid Tianyulong, they may indicate that ornithischians in general had some expression of quill or fuzz structures ancestrally.)

Some have considered the psittacosaurids to be facultative quadrupeds, but a recent study showed that their hands could not be placed in a position to support weight. Thus, they retained the ancestral bipedal condition. However, that is true of the ADULTS; recent work shows that hatchlings were quadrupedal, but becoming bipedal as juveniles and adults (not unlike certain common modern intelligent primates...).

The remaining ceratopsians form the clade Neoceratopsia. This group is further modified from the ceratopsian condition by the presence of a frill: itself just an elaboration of the ridge of typical marginocephalians. In earlier forms the frill is short, and its expansion was probably primarily for increased muscle attachment of the supratemporal muscles. Another neoceratopsian trait is that the jugal "horns" point well posteriorly of the orbit.

A number of primitive Asian Early Cretaceous neoceratopsians have recently been described: Liaoceratops, Archaeoceratops, Auroraceratops, and Yamaceratops. In addition to the Asian taxa, there is the recently discovered Aquilops of the American West. These forms were still bipedal.

The Late Cretaceous neoceratopsians, however, seem to have been quadrupedal because of increasing relative skull size. Additionally, these dinosaurs have lower jaws in which rotation of the dentaries is prevented and motion of the lower jaw is up-and-down and backwards. Among the Late Cretaceous groups is transcontinental (Asia and North America) Leptoceratopsidae, including North American Prenoceratops, Montanoceratops, and Leptoceratops, and Asian Udanoceratops. Another is the transcontinental (Asia and European) Protoceratopsidae, including Mongolian Protoceratops (literally hundreds of skeletons (including hatchlings and embryos), eggs, and nests are known from this Mongolian dinosaur) and Bagaceratops, Chinese Magnirostris, and Hungarian Ajkaceratops.

Collectively, Protoceratopsidae + Ceratopsoidea form the clade Coronosauria. Coronosaurs have an enlarged frill and enlarged skull, and consequently were obligate quadrupeds. The frill is enlarged but not thickened, so most of the expansion was not for muscle attachment. It may have served for visual display.

The remaining coronosaurs are the Ceratopsoidea. More primitive ceratopsians were generally small: 1-3 m, and no heavier than a sheep or pig at most (Udanoceratops being an exception). The ceratopsoids were all larger: cow-to-elephant sized. Ceratopsoids had postorbital horns sticking from the eyebrow over the eyes. Cow-sized Zuniceratops of western North America and Asian Turanoceratops and Asiaceratops represent primitive ceratopsoids. The remaining forms are the Ceratopsidae proper.

The horns of ceratopsoids may have had both a defensive function (against predators) and a display function against other ceratopsids. These displays may have been species recognition, sexual, or dominance displays; and could have been merely visual or may have involved direct combat.

Simplified cladogram of Ceratopsidae

True Ceratopsidae is the most speciose and specialized branch of Marginocephalia. This clade is restricted to the last twenty million years or so of the Late Cretaceous. (Maybe. Turanoceratops from about 92 Ma of Uzbekistan may be a true ceratopsid.) Up until 2009 it seemed clear that all members of Ceratopsidae proper were from western North America, but (questionably) Turanoceratops and (unquestionably) the centrosaurine Sinoceratops document the presence of this clade in Asia. Additionally, teeth of ceratopsids have been described from the end of the Cretaceous in eastern North America. Yet in that restricted time and space they underwent a major radiation. We have complete skeletons of several species; complete skulls of even more; growth series; skin impressions; and trackways.

Ceratopsids are characterized by the following specializations:

Ceratopsids include some definite herd dwellers. Ceratopsids are rivalled (and in some formations exceed) only by hadrosaurids in their abundance in the Late Cretaceous of western North America.

There were two major subclades of Ceratopsidae: Centrosaurinae and Chasmosaurinae (The latter group is sometimes called "Ceratopsinae" to some paleontologists. However, it is no longer certain that the fragmentary Ceratops is definitely closer to the chasmosaurines than the centrosaurines, so in this course we will use "Chasmosaurinae".). It was once easy to distinguish these two groups based on their horn patterns: in centrosaurines the nasal horn was large and the postorbitals reduced or lost; in chamsosaurines the nasal horn was small and the postorbital horns were (with one or two exceptions) long.

Unfortunately, the discovery of Zuniceratops and the basal centrosaurines such as Xenoceratops, Albertaceratops and Diabloceratops (and several others) have muddled that.

Under our current understanding, long postorbital horns are characteristic for Ceratopsoidea as a whole, so the fact that Chasmosaurinae retain them is simply a primitive feature.

Instead, we know recognize Centrosaurinae by the presence of:

Avaceratops, Nasutoceratops, Xenoceratops, Diabloceratops, Wendiceratops, and Albertaceratops are among the most primitive centrosaurines. Additional forms include the closely related Centrosaurus, Spinops, and Coronoceratops plus sword-horned Rubeosaurus and spike-frilled Styracosaurus, all of which share apowerful nasal horn; the aforementioned Asian centrosaurine Sinoceratops and its North American relative Xenoceratops; hook-horned Einiosaurus; and a great radiation of lump-nosed species (the Pachyrostra), including the genera Achelousaurus; and several species of Pachyrhinosaurus, last and largest centrosaurine.

Chasmosaurinae is characterized by:

Chasmosaurinae includes the newly described basal form Medusaceratops (found in the same bonebed as the centrosaurine Albertaceratops); the various species of Chasmosaurus; Agujaceratops and its newly (2010) described relative Mojoceratops; Pentaceratops; Anchiceratops, Arrhinoceratops, and their enormously-horned Mexican relative Coahuilaceratops; Regaliceratops (convergent on the centrosaurines in its snout shape and elongated nasal horn); and the latest Cretaceous giants Titanoceratops, Ojoceratops, Eotriceratops, Torosaurus, Triceratops, Nedoceratops (formerly "Diceratops"), and little Tatankaceratops.Some authors consider all of these separate taxa, with the older Titanoceratops, Ojoceratops, and Eotriceratops close to the ancestry of the others. Alternatively, "Tatankaceratops" and "Nedoceratops" may be the juvenile and subadult (respectively) of Triceratops and/or Torosaurus. Furthermore, it has been proposed that "Torosaurus" itself may simply be the fully adult form of Triceratops, so it may be that instead of four genera (Tatankaceratops, Nedoceratops, Triceratops, Torosaurus) in the latest Cretaceous of western North America, there was simply Triceratops.

(If you have a spare hour, you can see the following debate as to whether Torosaurus is an adult Triceratops:


Ceratopsids were a major component of the large bodied herbivore fauna in western North America (and probably the wetter parts of Asia) until the very end of the Cretaceous.

Feeding adaptation transformations:

Locomotory changes:
Pachycephalosaurs, basal ceratopsians, and basal neoceratopsians were bipedal; increased skull size forced advanced neoceratopsians onto all fours, such that coronosaurs were obligate quadrupeds.

Size trends:
Most pachycephalosaurs, basal ceratopsians, and basal neoceratopsians were small (all in the 1-3 m range, smaller than humans); increased size in pachycephalosaurs only occurs at the very end of the Late Cretaceous. Advanced neoceratopsians show an increase earlier, culminating in the major size increases at the base of Ceratopsoidea, at the base of Ceratopsidae, and independantly in Centrosaurinae and Chasmosaurinae. For most of its history, Ceratopsia consisted of only small dinosaurs.

Social behavior in Marginocephalia:
Both pachycephalosaurs and ceratopsians seem to have used their heads in within-species interactions. In particular, visual display structures (marginal osteoderms, domes, frills, horns, etc.) and possible combat features (thickened skulls, horns) are present in both clades, and some of these only appear at fully adult sizes. We'll examine these more in the third section of the course.

To Next Lecture.
To Previous Lecture.
To Lecture Notes.

Last modified: 27 September 2017

The iconic Late Jurassic thyreophoran Stegosaurus, by Cheung Chung Tat