Styracosaurus Skull

Dental Battery

40 cm

Figure 9.4. Cross-section through the upper and lower jaws of Triceratops, illustrating, (a) high-angle slicing-and-dicing motion of the teeth and (b) internal view of the dental battery in the lower jaw of Triceratops.

20 cm

40 cm

Figure 9.4. Cross-section through the upper and lower jaws of Triceratops, illustrating, (a) high-angle slicing-and-dicing motion of the teeth and (b) internal view of the dental battery in the lower jaw of Triceratops.

like beak. The hooked rim of the rostral bone (covered by a sharper cornified rhamphotheca) suggests the capability for careful selection of the plants for food. Because the earliest members of the clade all had the pointed hooked beak, some degree of selectivity must have been present at the very beginning of the history of ceratopsians and was maintained by members of the group until the end. Likewise, all ceratopsians clearly were chewers, grinding leaves and stems between the occlusal surfaces of their sturdy dentition. Worn teeth were constantly replaced, so that the active surface of the dental battery was continually refurbished. The grinding action was bequeathed from the common ancestor of all ceratopsians, and through the phylogenetic history of the group, the angle of occlusion became more and more vertical, until, in the latest forms, a near-vertical slicing and dicing along the side of the tooth, characteristic of the later large ceratopsids, was achieved (Figure 9.4).

The force behind this high-angle mastication derived from a great mass of jaw-closing musculature, which in the frilled forms crept through the upper temporal opening and onto the base of the frill. The other end of this muscle attached to a great, hulking coronoid process on the mandible. All in all, the chewing apparatus in ceratopsians was among the most highly evolved in all vertebrates.2

2 Interestingly, if the kinds of chewing we've described here were not enough for the likes of Psittacosaurus. then a packet of gastroliths lodged in the gizzard doubly pulverized its meal. Only in Psittacosaurus among ceratopsians (in fact among all ornrthischians) are gastroliths known. Perhaps such a unique occurrence represents a special (that is, derived) feeding strategy in this most primitive of ceratopsians.

On beyond the mouth, the remainder of the digestive tract does not appear to have been disproportionately large in ceratopsians, a likely consequence of their masticatory prowess. Nevertheless, it must have been sufficiently voluminous to accommodate the continual passage of the great quantities of foliage that formed the diet of these animals.

Both small and bipedal ceratopsians, as well as even the largest quadrupedal forms never browsed particularly high above the ground. At their largest (Triceratops and Torosaurus), browse height was probably no more than 2 m and no one has seriously entertained the possibility that ceratopsids were able to rear up on their hind legs to forage at higher levels (as has been suggested for some sauropods and stegosaurs). Nevertheless, they may have been able to knock over trees of modest size in order to gain access to choice leaves and fruits.

Which plants were preferred by ceratopsians remains a mystery. Once thought to be feeders on the fibrous fronds of cycads and palms, the majority of ceratopsians are rarely found in the same areas as these kinds of plants. The principal plants whose statures match browsing heights of ceratopsians were a variety of shrubby angiosperms, ferns, and perhaps small conifers. In fact, it has been argued by two paleo-botanists, S. Wing and B. Tiffney, that ceratopsians, along with their cohort of other large yet low-browsing, generalist-feeding herbivorous dinosaurs were doing a reciprocally advantageous evolutionary waltz with early flowering plants (see Chapter 16). Suffice it to say for the present that herbivorous dinosaur feeding habits may have contributed to the extraordinary rise of flowering plants during the Late Cretaceous. This hypothesis and a related one are discussed in Chapters 15 and 16.

Locomotion The legs of horned dinosaurs are unlike those of any mammal - living or extinct - and thus exactly how ceratopsians cruised over their Cretaceous landscapes remains a matter for some conjecture (Figure 9.5). Did they thunder along like enraged rhinos, or were their legs (and perhaps metabolism, as well) built in such a way that this kind of locomotion was impossible? Although the matter remains unresolved, we are closer to developing a picture of the locomotor skills of these horned dinosaurs.

As befits its primitive position within Ceratopsia, Psittacosaurus appears to have retained the fully bipedal limb posture found among other ornithischians. The same condition is found in Leptoceratops and Microceratops, but thereafter ceratopsians assumed a quadrupedal posture. How fast these animals may have traveled is not known with much precision, especially as there are no known trackways for any of these animals (why this should be so remains a mystery). However, those estimates that have been made - based principally on biomechanical analyses of the limbs - range from 20 to 50 km/h. A problem with this kind of analysis, however, is that the orientations of the front limbs in quadrupedal ceratopsians are not fully understood. R. T. Bakker has argued strongly for what is essentially a mammal-like posture; that is, fully erect front limbs that mimic the well-known posture of the back

Agathaumas Skeleton

50 cm

Figure 9.5. Lateral view of the skull and skeleton of (a) Psittacosaurus, (b) Protoceratops, and (c) Centrosaurus.

50 cm

Figure 9.5. Lateral view of the skull and skeleton of (a) Psittacosaurus, (b) Protoceratops, and (c) Centrosaurus.

limbs. Others have argued that this cannot be so, and that a more sprawling posture of the front legs is indicated (see Chapter 15).

For the present, R. A. Thulborn has presented the most detailed speed estimates. His work indicates that average walking speed for both large and small ceratopsians was somewhere between 2 and 4 km/h, while maximum running speeds ranged from 30 to 35 km/h (see also Box 15.3).

Horns, frills, and At virtually any speed, a Triceratops juggernauting across the Late ceratopsian behavior Cretaceous countryside was apt to pack one serious wallop if something got in the way, especially when fully equipped with horns. Long presumed to function for individual defense against predators, horns and their relationship to frills and defense were examined by E. H. Colbert in 1951. He began his investigation by observing that, in living African bovids (e.g., impalas, antelopes, and gnus), it is not so much the shape of the horns of these animals that is important but rather their presence that is critical to their survival. In this way, he interpreted the various cranial ornaments as being independently evolved, but more or less equally satisfactory solutions to the same problem: resisting predators during interspecific (inter - between; between different species) combat. Thus we have bovid horns as weapons in defense against predators. By analogy, ceratopsian horns were thought to have functioned to ward off predators at close quarters and in the classic Late Cretaceous All-Star Game between Tyrannosaurus and Triceratops, Triceratops could come up the winner by virtue of its prominent brow and nose horns. The frill presumably added additional insurance against lethal bites to the neck region.

Subsequent interpretations have also drawn on analogy with horned mammals such as bovids, but have instead centered on the intraspecific (intra - within; among members of the same species) functioning of horns; that is, their behavioral context in display, ritualized combat, defense of territories, and establishment of social ordering. Thus variation in horns and antlers do count, first at the level of making sure that species recognize each other and hence prevent matings between different species and second by aiding in the establishment of within-species dominance hierarchies (that is, who within a population gains in reproductive and/or resource rights) or, possibly, in defending territories.

The link between dominance, defense, and horns comes from important studies of mammals in their natural habitats, research that has gained increasing prominence ever since humans began taking stock of how much they have disturbed virtually all terrestrial (and aquatic) ecosystems. For example, we know that in the case of almost all horned mammals, larger males tend to have a reproductive advantage over smaller males. Simply put, they tend to breed more often than do small individuals, most likely because females choose them more often than not. Dominance in these mammals (and in other tetrapods) is accentuated by the development of structures that "advertise" the size of the animal; these obviously include horns and antlers, as well as the hornlike ossicones of giraffe and the nasal horns of rhinoceroses. Indeed, these structures increase visibility and hence the probability that the owners of the horns and antlers fend off competitors and impress females (in most cases). On the basis of sexual selection (selection that favors display-related structures in one or both sexes within a given species), these features should come to be highly linked with reproductive success. Put on a good show and your genes have a greater chance of making their way into subsequent generations.

In short, the variety of horn and antler shapes are now known to reflect (1) species-recognition mechanisms that aid in preventing interspecies matings and (2) intraspecific differences in displays and ritualized fighting behavior. Can such interpretations of intraspecific behavior and sexual selection in mammals shed light on the development of horns and frills in ceratopsians? Several dinosaur paleontologists, among them L. S. Davitashvili, J. O. Farlow, P. Dodson, R. E. Molnar, N. B. Spassov, J. H. Ostrom, and P. Wellnhofer, think so. As does Scott Sampson of the University of Utah, who has provided the most comprehensive and detailed discussion of the behavioral significance of cranial ornamentation in ceratopsid ceratopsians.

Let's begin by examining the common thread of these investigations into the evolution of ceratopsian horns and frills from the perspective of their intraspecific behavioral significance. No one has ever doubted that ceratopsian horns were used for combat; the question has been at whom they were aimed. Using modern horned mammals as analogues, the large nasal and brow horns of ceratopsians are thought to have functioned primarily during within-species combat (as in territorial defense and establishing dominance hierarchies). Similarly, the development of elaborate scallops and spikes along the frill margin in many of the more highly derived ceratopsians separates one species from another. However, what we dinosaur paleontologists use to recognize species aren't necessarily the same criteria that ceratopsians used to size each other up: the size and shape of horns and frills. Can we provide any more evidence that the behavioral hypothesis is likely to be true?

Before answering this question, we must first look to another hypothesis that attempts to explain the evolution of ceratopsian frills (but not horns). In 1966, Ostrom suggested that the frill provided a platform for the attachment of the major mass of jaw-closing muscles; these were thought to have extended through the upper temporal fenestra and onto the upper surface of the frill. Thus, in long-frilled ceratopsians, the jaw musculature must have been very extensive and consequently exceedingly powerful. The phylogenetic changes in the size and shape of the frill among ceratopsians, Ostrom suggested, reflected changes in the attachment and action of muscles running between the frill and the lower jaw, ultimately increasing the power of the bite.

Dodson and others have remained suspicious of such an explanation, because it focuses only on the significance of frills in aiding chewing. For example, as the frill of Protoceratops grows during the lifetime of the animal, from hatchling to old adults, the jaw muscles show a decrease in their mechanical ability to produce high bite forces. Likewise, there is a marked sexual dimorphism in the size and shape of adult frills, suggesting that jaw mechanics is unlikely to be the sole factor governing frill morphology. Instead, it is probable that ceratopsian frills answered to other important functions to account for such aspects of ceratopsian biology as sexual dimorphism, ontogenetic patterns, and taphonomic occurrences.

As has been mentioned, the earliest dinosaur egg nests to be found were then ascribed to Protoceratops, but recent discoveries in the Gobi Desert clearly indicate that these belong to oviraptorid theropods. However, Protoceratops hatchlings are known from complete skeletons that occur in what can only be interpreted as nests (Figure 9.6); these hatchlings are about 25 cm long. Virtually all of the remaining growth stages have been documented, making the ontogenetic development of Protoceratops one of the best understood. This is particularly relevant to the question at hand ("What is the function of frills?"), because it allows us to ask "When during development does the frill begin to grow?" and "When does it become most expansive?" Thanks to statistical studies of growth in Protoceratops, it is now clear that frill development takes off when individuals, apparently both males and females, are approaching fully adult body sizes, reaching their maximum shortly thereafter. In the

Triceratops Growth
Figure 9.6. A nest of hatchling Protoceratops from the Late Cretaceous of Mongolia.

group arbitrarily designated "males," the frill becomes inordinately larger and showier than in the "females" (Figure 9.7). This pattern strongly suggests that the onset of frill growth occurs with sexual maturity and therefore that there is a reproductive connection to frill size and shape. Sounds like sexual selection?

This pattern, in which frills take on their greatest prominence once sexual maturity was reached, is now thought to occur also in other cer-atopsians, among them Centrosaurus and Chasmosaurus. In many of these forms, the development of scallops and spikes on the frill margin would enhance the dimorphic nature of the frill.

Taken as individual features, frills apparently lend themselves to explanations involving within-species display. They even seem to develop in synchrony with the attainment of sexual maturity. But what can be said about the social context of ceratopsians, the arena in which the frill functioned? At the very least, it can be claimed with much justification that many if not all ceratopsians lived in large herds, at least during part

Ceratopsian Skull Anatomy
Figure 9.7. Sexual dimorphism in Protoceratops. Note in (a), a presumed female, the frill is less showy and the nasal ridge is less prominent, quite the opposite of (b), a presumed male.

if not all of the year. This justification comes from our ever-increasing catalog of ceratopsian bonebeds. These mass accumulations of single species of ceratopsians are known for at least nine separate species, including several in which the minimum number of individuals may exceed 100. Herding ceratopsians is consistent with these bonebeds. Moreover, such gregariousness also makes sense when one is putting frills and horns into their behavioral context. Territoriality, ritualized combat and display, and the establishment of dominance hierarchies are to be expected in animals that are thrown together in highly social circumstances such as herds. Perhaps in ceratopsians, we are seeing an example of the most complex of dinosaur intraspecific behavior.

Still, we are left with a series of ruminations about frills, horns, population density, and behavior. Is there nothing like a "smoking gun" from the fossils themselves that might give us a clue that we're on the right track? What might be expected of threatening, displaying, and combating ceratopsians that could be recognized in the fossil record are injuries, such as puncture wounds inflicted on faces, frills, and bodies of competing "males"? In fact, puncture wounds are preserved in at least five forms. These pathologies, not only on the cheek region, but also in the frill provide strong evidence of the blood-letting that comes from head-on engagements between competing members of the same species.

Given the possibility that ceratopsians may have had complex social behaviors involving display, ritualized combat, the establishment of dominance hierarchies, and defense of territories, it comes as a bit of a surprise that their brain size is not at all large (see Box 15.4). For example, despite being near opposites in terms of body size and display-related anatomy, both Protoceratops and Triceratops had brains less than the size expected of a similarly sized crocodilian or lizard. Cerebrally, they were above average as compared with sauropods, ankylosaurs, and stegosaurs, but commanded proportionally less gray matter than either ornithopods

Squamosal Styracosaurus
Figure 9.8. "Back off": frill display in Chasmosaurus.

or theropods. As J. A Hopson has suggested, perhaps these brain size measures indicate that ceratopsians had relatively unhurried and uncomplicated lifestyles.

Even without high levels of brainpower, there is strong evidence that the peaceful life of ceratopsians was interrupted - at least intermittently - by resounding clashes. Farlow and Dodson outlined the evolution of intraspecific behavior among ceratopsians. To begin with, display- and combat-related behavior seems to have been present primitively among neoceratopsians and perhaps even among all ceratopsians. With their frills serving as a visual dominance rank symbol and their small, yet sharp nasal horns acting as weaponry, display in Leptoceratops, Protoceratops, and Montanoceratops perhaps involved swinging the head from side to side. Should this ritual have failed to impress, these animals may have rammed their horns full tilt into the flanks of their opponent.

The more derived ceratopsids share more elaborate frills and either nasal or brow horns. Among the long-frilled chasmosaurines (e.g., Chasmosaurus, Pentaceratops, and Torosaurus), the display function of the frill may have been exaggerated. The very long frills of these dinosaurs could have provided a very prominent frontal threat display exhibited not only by inclining the head forward (Figure 9.8) but also by nodding or

Centrosaurus
Figure 9.9. "Crossing of the horns": combat between male Centrosaurus.

shaking the head from side to side. Should such a display have failed to send the message to one or the other of the opponents, combat may have involved frontal engagement of the nasal and brow horns, with shoving and wrestling determining the winner and loser of the contest. In contrast, most of the short-frilled centrosaurines (such as Centrosaurus, Avaceratops, and possibly Pachyrhinosaurus) were rather rhinoceros-like in their appearance and probably in their behavior as well (Figure 9.9). We presume that opponents tried to catch each other on their nasal horns, thus reducing to a degree the amount of damage inflicted to the eyes, ears, and snout. Nevertheless, the possibility of injury assuredly would have been very much greater in short-frilled ceratopsians than in other taxa.

In the context of frills, horns, and behavior discussed above, two anomalous forms stand out in the herd. First, Styracosaurus was a short-frilled centrosaurine whose frill appears inordinately large because of long spikes along its margin. Given its otherwise rhinoceros-like profile, Styracosaurus also was provided with the exceptionally distinctive display qualities of the frill. Should frill-wagging have proved to be an insufficient threat or deterrent, Styracosaurus could then have relied on the kinds of horn-locking and head-pushing that may have characterized other centrosaurines.

In contrast, Triceratops was a chasmosaurine ceratopsid whose frill is secondarily shortened, suggesting that its threat display may have been less utilized than in other chasmosaurines. Instead, combat may have

The evolution of Ceratopsia been similar to what is primitively found in chasmosaurines (engagement of brow and nasal horns, followed by shoving and wrestling), but in which the solid frill served as a shield against the parry and thrust of the opponents' horns.

The cephalization of weaponry and display seems almost to drive the evolution of ceratopsian dinosaurs. In this diverse group, we witness a world where display and competition were all important, where - when push came to shove - it may have been better to nod vigorously than to cross horns.

As was indicated earlier, Ceratopsia is a monophyletic taxon that consists of the common ancestor of members of Psittacosauridae and Neoceratopsia and all the descendants of this common ancestor (Figure 9.10). Among the rich array of unambiguous derived features shared by this ceratopsian clade, the most important include a rostral bone, a skull that is narrow at the beak end and flaring and deep in the cheek region (Figure 9.11), a frill composed principally of the paired parietal bones, and a strongly vaulted palate beneath the beak.

Ceratopsians and pachycephalosaurs (Chapter 8) appear to share a unique common ancestor, thus forming a larger clade that P. C. Sereno has called Marginocephalia (see introductory text to Part II: Ornithischia). More distant relationships are with Ornithopoda (Chapter 10) and Thyreophora (Chapters 6 and 7).

Psittacosaurus Cladogram

Figure 9.10. Cladogram of Cerapoda, emphasizing the monophyly of Ceratopsia, Psittacosaurus, and Neoceratopsia. Derived characters include: at I rostral bone, a high external naris separated from the ventral border of the premaxilla by a flat area, enlarged premaxilla, well-developed lateral flaring of the jugal; at 2 short preorbital region of the skull, very elevated naris, loss of antorbital fossa and fenestra, unossified gap in the wall of the lacrimal canal, elongate jugal and squamosal processes of postorbital, dentary crown with bulbous primary ridge, manual digit IV with only one phalanx, manual digit V absent; at 3 enlarged head, keeled front end of the rostral bone, much reduced quadratojugal, primary ridge on the maxillary teeth, development of humeral head, gently decurved ischium.

Figure 9.10. Cladogram of Cerapoda, emphasizing the monophyly of Ceratopsia, Psittacosaurus, and Neoceratopsia. Derived characters include: at I rostral bone, a high external naris separated from the ventral border of the premaxilla by a flat area, enlarged premaxilla, well-developed lateral flaring of the jugal; at 2 short preorbital region of the skull, very elevated naris, loss of antorbital fossa and fenestra, unossified gap in the wall of the lacrimal canal, elongate jugal and squamosal processes of postorbital, dentary crown with bulbous primary ridge, manual digit IV with only one phalanx, manual digit V absent; at 3 enlarged head, keeled front end of the rostral bone, much reduced quadratojugal, primary ridge on the maxillary teeth, development of humeral head, gently decurved ischium.

Styracosaurus Skull

50 cm 50 cm

Figure 9.1 I. Dorsal view of the skull of (a) Psittacosaurus, (b) Protoceratops, (c) Styracosaurus, and (d) Chasmosaurus.

50 cm 50 cm

Figure 9.1 I. Dorsal view of the skull of (a) Psittacosaurus, (b) Protoceratops, (c) Styracosaurus, and (d) Chasmosaurus.

Ceratopsia primitively consists of Psittacosauridae and the much more diverse monophyletic Neoceratopsia (Figure 9.12). Psittacosauridae is presently the most species-rich, genus-poor clade yet known among dinosaurs: it consists of a single genus (Psittacosaurus) and 10 species. Thus far, only the bare rudiments of the evolutionary relationships of these species are known. Still, members of this clade share as many as 12 derived features, among them a short snout, highly positioned external nares (the opening in the skull accommodating the nostrils), loss of the antorbital opening, and loss of the fifth digit on the hand. Also characteristic of all species of Psittacosaurus are their adult size (about 2 m), making them one of the smallest of all ceratopsians. Juveniles of the Psittacosaurus clade are known; hatchlings were no more than 23 cm long, about the size of an adult pigeon. In addition, a recently discovered Psittacosaurus skeleton provides evidence of its epidermal covering. In addition to the irregular pavement of large and small scales surrounding most of the body, more than 100 long, bristle-lilce structures extend vertically from the top of the base of the tail. Some regard these filaments as homologous with the integumentary filaments seen in theropods such as Sinosauropteryx and true feathers present in Caudipteryx, Microraptor,

Bovid Horn Core

Figure 9.12. Cladogram of basal Neoceratopsia, with the more distantly related Psittacosaurus and Pachycephalosauria. Derived characters include: at I elongated preorbital region of the skull, an oval antorbital fossa, triangular supratemporal fenestra, development of the syncervical (fusion of cervical vertebrae); at 2 greatly enlarged external nares, reduced antorbital fenestra, nasal horn core, frontal eliminated from the orbital margin, supraoccipital excluded from foramen magnum, marginal undulations on frill augmented by epoccipitals, more than two replacement teeth, loss of subsidiary ridges on teeth, teeth with two roots, ten or more sacral vertebrae, laterally everted shelf on dorsal rim of ilium, femur longer than tibia, hoof-like pedal unguals,

Figure 9.12. Cladogram of basal Neoceratopsia, with the more distantly related Psittacosaurus and Pachycephalosauria. Derived characters include: at I elongated preorbital region of the skull, an oval antorbital fossa, triangular supratemporal fenestra, development of the syncervical (fusion of cervical vertebrae); at 2 greatly enlarged external nares, reduced antorbital fenestra, nasal horn core, frontal eliminated from the orbital margin, supraoccipital excluded from foramen magnum, marginal undulations on frill augmented by epoccipitals, more than two replacement teeth, loss of subsidiary ridges on teeth, teeth with two roots, ten or more sacral vertebrae, laterally everted shelf on dorsal rim of ilium, femur longer than tibia, hoof-like pedal unguals, and true birds. Others have argued that they are not structurally similar to the filaments in theropods, so that they are probably not homologous. In either case, however, they were probably used in some sort of display behavior, particularly if they were colored, as has been speculated.

Another basal ceratopsian from Asia - Chaoyangosaurus from the Late Jurassic or Early Cretaceous of Liaoning, China - may be positioned directly above or below Psittacosaurus. Although the upper part of the skull is not yet known, it does preserve a rostral bone and a widely projecting cheek region, attributes of Ceratopsia. At the same time, numerous features ally it either directly with Psittacosaurus, as the closest known relative of Neoceratopsia, or as a taxon more basal than Psittacosaurus within Ceratopsia. At present, we place it in an unresolved, basal position within Ceratopsia.

Neoceratopsia, the remaining clade of ceratopsians, is clearly mono-phyletic, based on at least 10 important shared, derived characters. These include a sharply keeled rostral bone and a predentary that both end in a point, reduction or loss of premaxillary teeth, loss of the external opening in the lower jaw (the external mandibular fenestra), and a very short projection of the lower jaw beyond the jawjoint (the retroarticular process).

Thanks to recent studies by B. Chinnery and D. Weishampel, P. J. Makovicky, and H.-L. You and P. Dodson, we have a good appreciation for the structure of taxa among basal neoceratopsians (Figure 9.12). At its base, this clade consists of Archaeoceratops and an unnamed clade that can be broken down into two further clades. One of these, formed of Asiaceratops, Montanoceratops, Udanoceratops, and Leptoceratops has been called Leptoceratopsidae. The other clade, Microceratops, Protoceratops, Bagaceratops, and Zuniceratops, represents the closest relatives of Ceratopsidae, and within this group Protoceratops and Bagaceratops are each other's closest relatives. In this context only is there a monophyletic Protoceratopsidae formed of these two taxa.

Ceratopsidae Now that we have dealt with the question of primitive neoceratopsians, we can finally turn to Ceratopsidae (Figure 9.13). This monophyletic clade, which includes Centrosaurinae (those ceratopsids with short squamosals) and Chasmosaurinae (those with long squamosals), is supported by upward of 50 important diagnostic features, among

Centrosaurinae

Chasmosauridae

Ceratopsidae

Figure 9.13. Cladogram of Ceratopsidae, Derived characters include: at I enlarged rostral, presence of an interpremaxillary fossa, triangular squamosal epoccipitals, rounded ventral sacrum, ischial shaft broadly and continuously decurved; at 2 premaxillary oral margin that extends below alveolar margin, postorbital horns less than 15% skull length, jugal infratemporal flange, squamosal much shorter than parietal, six to eight parietal epoccipitals, predentary biting surface inclined steeply laterally.

them enlarged external nares set into well-developed excavations on the snout, folding of bones on the top of the head to form a secondary skull roof, reduced upper temporal opening, dental batteries, and an everted (sticking out) dorsal border on the ilium.

As a monophyletic clade, chasmosaurines all uniquely share a suite of modifications of the snout, nasal horn, and external nares. In addition, the frill becomes enlarged and there is a distinctive change in the pattern of ornamentation of the margin of the frill. Within this clade, the most primitive members are Chasmosaurus and Pentaceratops, which together appear to be each other's closest relatives. They share a number of derived features, among them huge openings in the parietal part of the frill and large, flat epoccipital bones along the frill margin.

All remaining, more derived members of Chasmosaurinae consist of the clade containing Anchiceratops, Arrhinoceratops, Torosaurus, Diceratops, and Triceratops. Sharing as many as seven derived features (among them, further modification of the snout, the development of additional sinuses in the head, and expansion of the ischium), this group has as its primitive members the small clade of Chasmosaurus and Pentaceratops, followed by another small clade that comprises Anchiceratops and Arrhinoceratops (themselves united by having a square frill that has very numerous traces of blood vessels on its undersurface).

That leaves us with a triumvirate of chasmosaurines: Torosaurus, Diceratops, and Triceratops. This clade, which shares a number of new modifications of the snout and nasal horn, has Triceratops as its primitive member and Diceratops and Torosaurus as the "top of the tree."

The other great ceratopsian clade is Centrosaurinae. This group ancestrally acquired a number of important features, among them a large nasal horn, long and narrow opening in the top of the skull roof, short brow horns, short squamosals, and broad, rounded epoccipitals decorating the rim of the frill. Centrosaurinae has at least five members: Achelousaurus, Einiosaurus, Centrosaurus, Pachyrhinosaurus, and Styracosaurus. The evolutionary interconnectedness of these taxa is as yet not well resolved. On the one hand is a small clade of Achelousaurus and Pachyrhinosaurus, with Styracosaurus and Centrosaurus on the other; the fifth taxon, Einiosaurus, remains unresolved at the base of Centrosaurinae.

Ceratopsians meet history: a short account of their discovery

As the Western Interior of the USA began yielding its fossil riches to the paleontologists of the day (or rather, to their hard-working collecting parties; see Box 6.2), the first ofwhat came to be known as ceratopsians was discovered. The initial finds, however, were not auspicious. The first was a partial pelvis discovered near Green River, Wyoming, from Upper Cretaceous rocks. Named Agathaumas (aga - very; thauma - wonder) by E. D. Cope in 1872, this material would prove to be of little value in understanding the affinities and biology of these great dinosaurs. Monoclonius (mono - single; klon - sprout, referring to the root of the tooth), recovered from Montana and named in 1876 by Cope on the basis of a fragmentary pelvis, some vertebrae, a few teeth, and a parietal bone, also provided little sense of Ceratopsia-ness. Eleven years later, O. C. Marsh announced the discovery of a pair of very large bony ceratopsian horn cores near Denver, Colorado, that he referred to as Bison alticornis!

It wasn't until the late 1880s that the true identity of the horn cores, pelvis, and collection of Monoclonius bones and teeth was finally revealed. The enlightenment was due to none other than Triceratops (treis - three; kerat - horn; ops - face), originally uncovered from Upper Cretaceous rocks in Wyoming. Named by Marsh in 1889 on the basis of an imperfect skull from a very huge animal, Triceratops is now known from innumerable complete and partial skulls, as well as a number of well-preserved skeletons that have been collected from Alberta and Saskatchewan in the north to Colorado in the south.

Two years later, Marsh named another Late Cretaceous ceratopsian -Torosaurus (toreo - perforate, referring to the holes in the frill) - again from Wyoming. Today it is the most widespread of dinosaurs, having been documented from additional sites in Saskatchewan, Montana, South Dakota, Colorado, Utah, New Mexico, and Texas.

The first quarter of this century established that ceratopsians were dinosaurs to be reckoned with. Discovered with a vengeance, particularly in Alberta during the years of the great Canadian dinosaur rush, most of the spoils went to the National Museum of Canada (now the Canadian Museum of Nature) in Ottawa, the Royal Ontario Museum in Toronto, and the American Museum of Natural History in New York. In 1904, Centrosaurus (kentron - sharp-point) gave us - and L. M. Lambe - a glimmer of the wealth of ceratopsians when it was first discovered. This was followed shortly thereafter by the discovery of Diceratops, collected from Upper Cretaceous strata of Wyoming and named by J. B. Hatcher (see below). Over the years, this ceratopsian came to be thought of as Triceratops, but C. Forster suggests that it is a valid taxon, distinct among ceratopsian genera.

These early discoveries as well as aspects of ceratopsian anatomy and paleoecology were first presented in comprehensive fashion in a large study begun by Marsh, and later taken over by Hatcher, Marsh's principal collector and an accomplished researcher in his own right, after Marsh died. The project eventually fell to Richard Swan Lull (also at Yale) to complete after Hatcher's tragic death from typhoid fever at the age of 43. Despite its morbid history, this monograph - finally published in 1907 -stands today as one of the most important references on those ceratopsians that were known at the turn of the century.

Yet it was not until the 1910s and 1920s that ceratopsian riches of Canada were to be fully realized. In 1913 and 1914 alone, Styracosaurus (styrak - spike), Chasmosaurus (chasma - wide opening), Anchiceratops (anchi - close), and Leptoceratops (leptos - slender) were discovered, described, and named by Lambe and by Barnum Brown. All the while but across the U.S. border, C. W. Gilmore was providing Montana with another ceratopsian, this time Brachyceratops (brachys - short).

The world of ceratopsians opened widely in the 1920s. Hailing from New Mexico, Pentaceratops (penta - five) was described by H. F. Osborn in 1923, while on the other side of the globe the Gobi Desert was beginning to yield its dinosaurian treasures through the toil of the American Museum of Natural History's Central Asiatic Expeditions. The most famous of all was Protoceratops (protos - first), announced to the world in 1923 by W. Granger and W. K. Gregory. Also from the Gobi Desert and no less important were the remains of Psittacosaurus (psittakos - parrot), described by Osborn in 1923. Osborn, a formidable figure in the history of vertebrate paleontology and head of Vertebrate Paleontology (and later Director) at the American Museum of Natural History, provided the first evidence of the early history of ceratopsian dinosaurs. Psittacosaurus is now also known from a variety of localities in China and even from Gorno-Altayask Autonomous Region in eastern Russia. A return to North America completes the research cycle of the 1920s. In 1925, W. A. Parks described yet another long-frilled ceratopsian, Arrhinoceratops (a - without; rhin - nose), from the great fossil beds of Alberta.

In the years that intervened between the 1920s and the present, new studies of ceratopsians came out in what might best be described as fits and starts. In 1933, Lull published his revision of ceratopsian dinosaurs, emphasizing the anatomy of Monoclonius (now known to be Centrosaurus), as well as the burgeoning record of horned dinosaurs from the great Canadian dinosaur rush. Yet it was not until 1942 that another new ceratopsian, Montanoceratops (from Montana) was announced to the world. Originally described by B. Brown and E. Schlaikjer as a new species of Leptoceratops from the Late Cretaceous of Montana, it took nine years until C. M. Sternberg, paleontologist for the National Museum of Canada, recognized it as a new kind of ceratopsian. He then baptized it with the name of its host state. Meanwhile (1950), he was also was busy with the announcement of another ceratopsian from Upper Cretaceous rocks of Alberta. Pachyrhinosaurus (pachy - thick) was one of the ugliest (or most magnificent, depending upon one's perspective) of ceratopsians. Pachyrhinosaurus came replete with masses of exceedingly roughened bone extending from the top of the snout to over both eyes. Even without the adornment of prominent nose or brow horns, here was a ceratopsian to be reckoned with.

Three years after C. M. Sternberg's description of Pachyrhinosaurus, B. Bohlin - paleontologist with the joint Sino-Swedish Expedition to the Gobi Desert - published his account of Microceratops (micro - small). Much like Leptoceratops and Protoceratops, this small ceratopsian was without horns, but sported a modest frill behind the head. Microceratops is now known from several localities in northern China and southern Mongolia.

It was another 22 years before these Asian ceratopsians were again in the limelight, this time as the result of a comprehensive study of new (and some old) material from the Gobi Desert. In 1975, T. Maryanska and H. Osmolska, the two principal dinosaur specialists in the Polish-Mongolian Palaeontological Expedition (1963-1971) provided the first comprehensive description of the anatomy, paleobiology, and evolution of Protoceratops, Microceratops, and their brethren, including a new form that they called Bagaceratops (Mongolian: baga - small).

Recently, a variety of new kinds of ceratopsians have been described. The first, Avaceratops (to honor Ava Cole, the wife of the collector of these fossils), comes from a 1986 study by Dodson and represents the first advanced horned ceratopsid to have been discovered since Sternberg's discovery of Pachyrhinosaurus. Nearly ubiquitous Protoceratops-like ceratopsians continued to tumble out of the rocks. One, discovered in Kazakhstan by L. Nessov and co-workers, was named Turanoceratops (from Turan, Kazakhstan) in 1989. Another - Breviceratops (brevis - short) - was originally thought by Maryanska and Osmolska in 1975 to be a new species of Protoceratops. It was S. Kurzanov who rechristened it Breviceratops in 1990.

Ceratopsian discoveries continued to be made throughout the rest of the 1990s and into the new millennium in both Asia and North America. In Mongolia, Udanoceratops (from Udan Sair) was the product of on-going field research conducted by Kurzanov, and Graciliceratops {gracilis - slender) represents a renaming of material from the Gobi Desert by P. C. Sereno in 2000 that was originally thought to belong to Microceratops. On the other hand, Archaeoceratops ("ancient ceratops") and Chaoyangsaurus (named for the founder of vertebrate paleontology in China, C.-C. Young) are newly discovered Chinese forms, the former named by Z. Dong and Y. Azuma in 1997 and the latter by X. Zhao, Z. Cheng, and X. Xu in 1999; both have proved pivotal in our understanding of ceratopsian relationships.

In 1995, S. D. Sampson described the latest ceratopsids to be discovered - Achelousaurus and Einiosaurus - both from the upper beds of the Two Medicine Formation (Upper Cretaceous) of Montana. These unusual-looking animals bear some of the strangest horns ever seen on a ceratopsian. While these same strata have produced the spectacular hadrosaurid nesting sites, North American ceratopsian eggs, babies, and nests remain virtually unknown.

readings Dodson, P. 1976. Quantitative aspects of relative growth and sexual dimorphism in in Protoceratops. Journal of Paleontology, 50, 929-940. Dodson, P. 1992. Comparative craniology of the Ceratopsia. American

Journal of Science, 293-A, 200-234. Dodson, P. 1996. The Horned Dinosaurs. Princeton University Press,

Princeton, NJ, 346pp. Dodson, P., Forster, C. A. and Sampson, S. D. 2004. Ceratopsidae. In Weishampel, D. B„ Dodson, P. and Osmolska, H. (eds.), The Dinosauria, 2nd edn. University of California Press, Berkeley, pp. 494-513. Farlow, J. O. and Dodson, P. 1975. The behavioral significance of frill and horn morphology in ceratopsian dinosaurs. Evolution, 29, 353-361. Hatcher, J. B., Marsh, O. C. and Lull, R. S. 1907. The Ceratopsia. U.S.

Geological Survey Monograph, 49,1-300. Lull, R. S. 1933. A revision of the Ceratopsia or horned dinosaurs. Peabody

Museum of Natural History Memoires, 3,1-175. Maryanska, T. and Osmolska, H. 1975. Protoceratopsidae (Dinosauria) from Mongolia. Palaeontologia Polonica, 33,133-182.

Ostrom, J. H. 1964. A functional analysis of the jaw mechanics in the dinosaur Triceratops. Postilla, 88,1-35.

Ostrom, J. H. 1966. Functional morphology and evolution of ceratopsian dinosaurs. Evolution, 20, 290-308.

Ostrom, J. H. and Wellnhofer, P. 1986. The Munich specimen of Triceratops with a revision of the genus. Zitteliana, 14,111-158.

Sereno, P. C. 1986. Phylogeny of the bird-hipped dinosaurs (Order Ornithischia). National Geographic Society Research, 2,234-256.

You, H.-L. and Dodson, P. 2004. Basal Ceratopsia. In Weishampel, D. B., Dodson, P. and Osmolska, H. (eds.), The Dinosauria, 2nd edn. University of California Press, Berkeley, pp. 478-493.

0 0

Responses

  • Jouko
    What desert can the centrosaurus be found?
    8 years ago
  • elena reilly
    Who was mono ansetor in prehistory mongolia?
    3 years ago

Post a comment