Farke, Andrew A


CERATOPSIDS, HORNED herbivorous oniithischians from the Cretaceous of North America, are unique among dinosaurs in the form and expression of their cranial ornamentation. All ceratopsid genera possess a bony parieto-squamosal frill that projects over the neck region and some degree of development of nasal and orbital horns. These horns vary in form and location, from low bosses (e.g., Pachyrhinosaurus Sternberg, 1950) to long, conical horns (e.g., Triceratops Marsh, 1889 postorbital horns). Internal structural changes occurred along with these extreme external modifications of the skull (Forster, 1996).

A frontal sinus complex occupies space variably involving the postorbital, frontal, parietal, supraoccipital, and exoccipital bones of the skull roof in ceratopsid dinosaurs (Fig. 1.1; Forster, 1996). These sinuses are relatively small, simple chambers in most members of the two ceratopsid clades Chasmosaurinae and Centrosaurinae (Lehman, 1990). However, the frontal sinuses reach large proportions in some species, with a portion of the sinuses extending into the base of the postorbital horncores (Forster, 1996; Sampson et al., 1997; Lehman, 1998). Forster (1996) termed this part of the sinus complex the “cornual” sinus, analogous to the cornual diverticulum of the frontal sinuses in bovid mammals (e.g., domesticated goats; Nickel et al., 1973).

Marsh (1887) was the first to note cornual sinuses in a large chasmosaurine (USNM 1871, probably a specimen of Triceratops or Torosaurus Marsh, 1891; personal observation). Sampson et al. (1997) described small cornual sinuses in the centrosaurines Achelousaurus Sampson, 1995, Einiosaurus Sampson, 1995, and Pachyrhinosaurus. Other authors noted similar cornual sinuses in the chasmosaurines Pentaceratops Osborn, 1923, Torosaurus, and Triceratops, but cornual sinus morphology has not been described fully for any chasmosaurine (Hatcher et al., 1907; Forster, 1996; Lehman, 1998). This note describes chasmosaurine cornual sinuses in detail, using undescribed as well as previously documented specimens. All of these specimens offer new information on inter- and intraspecific variation, as well as ontogeny. Most importantly, these new data constrain hypotheses on the function and phylogenetic significance of the ceratopsid frontal sinus complex, and allow a more informed comparison of sinuses between horned dinosaurs and bovid mammals.

Institutional abbreviations.-AMNH, American Museum of Natural History, New York; GP, Glenrock Paleontological Museum, Glenrocic, Wyoming; MNA, Museum of Northern Arizona, Flagstaff; MOR, Museum of the Rockies, Bozeman, Montana; NMMNH, New Mexico Museum of Natural History, Albuquerque; OMNH, Oklahoma Museum of Natural History, Norman; SDSM, South Dakota School of Mines and Technology Museum of Geology, Rapid City; SMNH, Saskatchewan Museum of Natural History, Regina; TMM, Texas Memorial Museum, Austin; TMP, Royal Tyrrell Museum of Palaeontology, Drumheller, Alberta; UNM, University of New Mexico, Albuquerque; USNM, National Museum of Natural History, Washington, DC; UTEP, University of Texas at El Paso; YPM, Yale Peabody Museum, New Haven, Connecticut.

Anatomical abbreviations.-b, bone; be, brain cavity; cs, cornual sinus; ff, frontal fontanelle; fs, frontal sinus; nc, nasal cavity; r, ridge; s, bony septum; vas, vascular impression.


Because cornual sinuses are an internal cranial structure, the sinuses can be observed directly only in incomplete or disarticulated skulls that allow access to the internal surfaces of the postorbital bones. The rest of the frontal sinus complex has been described thoroughly elsewhere (Hatcher et al., 1907; Lehman, 1989; Godfrey and Holmes, 1995; Forster, 1996; Sampson et al., 1997; Lehman, 1998), and it will not be redescribed in this paper. Sampson et al. (1997) described cornual sinuses in centrosaurine ceratopsids, so this clade will not be discussed here, either.

The taxa and specimens considered here are listed in Appendix 1; relevant measurements are given in Appendix 2. Taxonomic assignments for specimens were based upon morphological characteristics, with some reference to stratigraphie horizon and geographic location. Because of the similarities between horns of some sympatric taxa (e.g., Torosaurus and Triceratops), the identifications of isolated horncores are provisional in most cases.

Anatomical description.-The ceratopsid cornual sinus is defined here as an extension of the frontal sinus complex into the portion of the postorbital bone directly beneath the horncore. The sinus may occupy only the horncore base (e.g., Pentaceratops sternbergi Osborn, 1923, NMMNH P.21098) or enter the shaft to extend a variable distance distally (between 8% and 27% of the horncore’s length; Appendix 2). In all observed specimens, the cornual sinus is directly confluent with the rest of the frontal sinus complex. This definition is consistent with the cornual sinuses described in some centrosaurines (Sampson et al., 1997).

Cornual sinuses were observed in Anchiceratops ornatus Brown, 1914, Pentaceratops sternbergi Osborn, 1923, Torosaurus latus Marsh, 1891, Torosaurus utahensis Gilmore, 1946, Triceratops horridus Marsh, 1889, and Triceratops prorsus Marsh, 1890. The sinus complex of the Arrhinoceratops brachyops Parks, 1925 type skull is inaccessible, so its sinus morphology is unknown. None of the observed Chasmosaurus Lambe, 1914 specimens, representing both northern and southern forms, possess cornual sinuses. Instead, the frontal sinuses terminate medial to the base of the horncore (e.g., TMP 79.11.147, UTEP P.37.7.089; Lehman, 1989; Holmes et al., 2001).

In general, the cornual sinus is a simple vaulted chamber. Some specimens exhibit folds or small accessory cavities within the sinus, but these are quite small in all observed specimens and do not significantly alter the overall shape of the cornual sinus. Consistently, the distal extremity of the cornual sinus has the shape of a rounded dome (as seen in cf. Anchiceratops sp. Brown, 1914, TMP 84.12.16, Fig. 1; Torosaurus utahensis, USNM 16169; and Triceratops sp., MOR 1120).

The position of the long axis of the cornual sinus relative to the long axis of the horncore varies. In specimens with relatively large cornual sinuses (e.g., Triceratops, MOR 1120), the sinus lies directly in the center of the horncore. In specimens with relatively small cornual sinuses, the sinus may be positioned at the center of the horncore (cf. Anchiceratops, TMP 84.12.16, Fig. 1.2) or caudal to the center (Pentaceratops, NMMNH P.21098, Fig. 1.3).

The texture of the bone lining the sinus shows two morphologies. Lehman (1998, p. 896) described the cornual sinus in Pentaceratops specimen OMNH 10165 as “walled with smooth finished bone,” a condition also observed in the subadult chasmosaurine specimens examined in this study. More typically, the surface of the bone lining the sinus bears concave neurovascular impressions coupled with neurovascular foramina (e.g., cf. Anchiceratops, TMP 84.12.16). These impressions are usually quite shallow (less than 1 or 2 mm in depth), originate within the cornual sinus, and enter or exit the horncore through small foramina. In one Triceratops specimen (MOR 1120), the neurovascular impressions form a broad network of subparallel grooves running caudolaterally along the entire length of the cornual sinus, imparting a finely corrugated appearance (similar to those illustrated in Fig. 2).

SDSM 58722 is a natural cast of a cornual sinus from a large chasmosaurine, probably Triceratops (Fig. 2). This specimen preserves the form of the neurovascular network in great detail. On the medial surface of the sinus, anastomosing channels originate rostroproximally and radiate caudodistally. The largest channel begins at the lowest part of the main sinus body and measures 9 mm in rostrocaudal length by 6 mm in mediolateral width at its base. The impression extends distally for approximately 200 mm. In the middle portion of the sinus, numerous smaller, unbranched channels radiate caudally. The neurovascular impressions on the lateral surface of the sinus are smaller and not as deeply incised as those on the medial surface. As preserved, they originate rostroproximally and radiate randomly. Some of the impressions terminate within the sinus; others leave the surface of the sinus to enter the horncore.

Triceratops specimen SMNH P2525.1 displays a series of low ridges on the caudal face of the cornual sinus, but they do not appear to be vascular in nature. The ridges originate at a common point approximately halfway up the lumen of the sinus and diverge in a bifurcating pattern. Each ridge is 2-3 mm high and approximately 10 mm wide. The ridges terminate at the distal end of the sinus.

Often a low circumferential ridge of bone (e.g., cf. Anchiceratops sp., TMP 84.12.16; Triceratops, MOR 1110, SDSM 3110) or multiple ridges (Torosaurus latus. MOR 1122) mark the base of the cornual sinus, separating it from other portions of the frontal sinus complex. But, no specimen indicates if this ridge encircled the entire sinus. Anchiceratops specimen AMNH 5259 lacks ridges, and the frontal sinuses grade smoothly into the cornual sinuses. At least one Triceratops specimen, SDSM 58721, possesses a second ridge of bone completely encircling the cornual sinus approximately halfway up the lumen of the sinus (Fig. 3.1).

Growth series.-Juvenile chasmosaurine material is quite rare, so several isolated postorbitals from the Hell Creek and Frenchman formations are particularly important in understanding the ontogenetic development of the cornual sinus. The specimens described here are probably referable to Triceratops or Torosaurus.

Of the two smallest chasmosaurine postorbitals (horncore lengths less than 90 mm) neither SMNH P2299.1 nor SMNH P2613.1 has a true cornual sinus (Tokaryk, 1997). SMNH P2613.1 (Fig. 1.5) preserves a portion of the frontal sinus, immediately caudal to the frontal suture and separated from the lateral edge of the cotylus for the laterosphenoid by a sharp ridge of bone. This sinus is caudal and medial to the horncore, but does not enter it. SMNH P2299.1 lacks the corresponding region.

AMNH 5006 (horncore length, 117 mm) displays an incipient cornual sinus under the caudomedial portion of the horncore (Fig. 1.6). The distal limit of the sinus is domed as in adults, and no neurovascular impressions occur on the sinus. AMNH 5057 (horncore length, 295 mm) is morphologically similar, except that two low, mediolateral ridges divide the sinus into three sections.

Based on these specimens, a hypothetical ontogenetic trajectory can be inferred for the cornual sinus (Fig. 1.5-1.8). In the smallest specimens (e.g., SMNH P2613.1, Fig. 1.5), the frontal sinuses do not underlie the horncores, but are confined along the caudomedial margins of the postorbital. With increased horn size, the frontal sinuses expand rostrally and laterally, extending beneath the bases of the horncores to create an incipient cornual sinus (e.g., AMNH 5006, 5057, Fig. 1.6, 1.7). As the animals matured, the cornual sinus extended further into the body of the horncore, culminating in the adult state.


Interspecific and intraspecific variation.-Lehman (1998) suggested that the occurrence of cornual sinuses was not of phylogenetic importance, but possibly related to large individual body size. Thus the occurrence of incipient cornual sinuses in the juvenile specimen AMNH 5006 (possibly referable to Triceratops sp.) is significant, particularly because this individual is smaller than Chasmosaurus specimens that completely lack the sinus (e.g., UTEP P37.7.082). In fact, no cornual sinuses occur in Chasmosaurus at all, including the comparatively large and long-horned Chasmosaurus mariscalensis Lehman, 1989. Cornual sinuses occur in all other chasmosaurines examined, although sinus size varies within taxa (Fig. 1. Appendix 2). This pattern of occurrence supports the argument that the occurrence of cornual sinuses is phylogenetically driven in at least some taxa. However, the independent evolution of cornual sinuses in centrosaurine and chasmosaurine ceratopsids also indicates that cornual sinus presence is a homoplastic character.

Additional observations suggest that cornual sinus size is related to horn size in some taxa, as illustrated by two Pentaceralops stembergi specimens. NMMNH P.2I098, with a horncore length of 378 mm, has a small cornual sinus that occupies just 13% of the horncore’s length (Appendix 2). However, OMNH 10165 has a large cornual sinus that occupies 22% of a 900 mm long horncore. More specimens are needed to investigate the relationship between horn and sinus size; in particular, information on sinus volume (rather than just length) is needed.

Cornual sinus function.-Chasmosaurine cornual sinuses and their associated frontal sinus complex are relatively uncomplicated structures, yet a myriad of possible biomechanical and physiological functions have been proposed.

Shock absorption is the most commonly cited function for the frontal sinus complex in chasmosaurines. based upon analogy with bovid mammals (Molnar, 1977; Forster, 1996). This hypothesis posits that the sinuses absorbed shocks (or more probably, disseminated and directed stress away from the brain) during horn use. However, chasmosaurine sinuses (and those of all other ceratopsids) lack bony septa, a possible shock-absorbing feature found in bovids (Schaffer and Reed. 1972; Fig. 3.2). Moreover, the shock-absorbing potential of bovid sinuses, although possible, has not been demonstrated. Thorough biomechanical modeling and experimentation is necessary to evaluate the shock-absorption hypothesis.

Mass reduction is another possible function (or a functionless byproduct of bone remodeling processes) for the cornual sinuses and the accompanying frontal sinus complex (Hatcher et al.. 1907). The cornual sinus of one Triceratops specimen (SDSM 3110) has a volume of 620 cm^sup 3^. Assuming a live cancellous bone density of approximately 1.3 g/cm^sup 3^ (Currey. 1984), one cornual sinus eliminated 0.81 kg of bone. The sinus of SDSM 58722, with a volume of 3.500 cm^sup 3^, eliminated 4.5 kg of bone. Considering both postorbital horns, the cornual sinuses effect a total mass reduction of 1.6 and 9 kg of bone for SDSM 3110 and 58722. respectively. Using volumetric measurements from a scale model of the skull and lower jaw, the total bony skull mass of Triceratops prorsus specimen YPM 1822 was estimated as approximately 220 kg. The cornual sinuses of SDSM 3110 and SDSM 58722 eliminated 0.73% and 4.1% of total cranial mass respectively, assuming the individuals had skull size and proportions similar to YPM 1822. Thus, mass reduction offered by the cornual sinuses is arguably negligible. However, the mass reduction itself may not be as important as the metabolic savings resulting from the elimination of excess tissue (Forster, personal commun.).

The occurrence of neurovascular impressions lining the cornual sinus has not been reported previously, and this feature is widespread among chasmosaurines. The pattern of the impressions (running along the surface of the sinus and entering the bone of the horncore) indicates that the blood vessels within the impressions may have fed the horncore’s bone tissue, a hypothetical keratinous sheath on the outside of the horncore, and/or soft tissue within the cornual sinus. In addition to supplying the bone tissue and horn sheath, vessels entering the horncore may be related to thermoregulation, as observed for similar vascularization in modem goats (Taylor, 1966). In this scenario, blood transported to the horns was cooled or warmed within the horncores (by virtue of their greater exposure to the external environment) before returning to the brain. A thermoregulatory role for ceratopsid horns, originally proposed by Rigby (1989). was corroborated by Barrick et al. (1998). who found that oxygen isotope variation indicates a decrease in temperature towards the distal end of the horn in Triceratops. It is doubtful that blood was cooled within the sinuses themselves, because the secondary skull roof of most ceratopsids would prevent contact between the sinus and the ambient air.

The reported neurovascular impressions also allow more informed speculation on the type of tissue within the sinus. As enumerated by Sampson et al. (1997). adipose, vascular, pneumatic, muscle, or neural tissue could have existed within the sinuses. Vascular impressions are incompatible with muscle tissue in the sinus, but compatible with the other tissue types. Some bovids (e.g., Capra hircus Linnaeus, 1758) display a vascularized bone surface texture in the sinuses, similar to that seen in chasmosaurines (Figs. 2, 3.2). Some of these vascular traces enter the body of the horncore, also as in chasmosaurines. In life, the sinuses of bovids are connected directly to the nasal cavity, filled with air. and lined by a tightly adherent mucous membrane (Nickel et al., 1973). A similar adherent membrane (mucous or otherwise) probably lined the comual sinuses of chasmosaurines, based upon the similarities in vascular impressions. However, no obvious connection to the respiratory system occurs in ceratopsids. In fact, a pneumatic source for the ceratopsid frontal sinus complex remains unknown despite numerous investigations (Forster, 1996; Sampson et al., 1997; Witmer, 1997).

Like the horns of modern animals, which have myriad functional roles (combat, display, thermoregulation, etc.). it is equally possible that the cornual sinuses of ceratopsids also had multiple roles within the skull.


Anatomical descriptions of chasmosaurine comual sinuses are an important first step toward understanding their functional and phylogenetic significance. Functional hypotheses for the cornual sinuses, like those for the rest of the frontal sinus complex, remain highly speculative. Biomechanical modeling studies will be especially important in understanding the role that the sinuses played within the skull, such as stress distribution. Bone histology data may offer insight into the type of soft tissue within the sinuses. Other related directions for future work include the evolutionary development of the frontal sinus complex and the ontogeny of the postorbital homcores (Goodwin and Horner. 2001).

Cornual sinuses have evolved at least twice, in bovid mammals and ceratopsid dinosaurs, and perhaps a common behavioral or functional characteristic links the sinuses in these two groups. As a whole, cranial sinuses are of particular interest in the study of vertebrate cranial morphology (e.g., Witmer, 1997), and the documentation of sinus variation across an assortment of clades is essential for interpreting sinus evolution and function. Only through further comparative analysis can the sinuses of ceratopsids be fully understood.


I thank M. Brett-Surman (USNM), B. Burger and C. Collins (AMNH). J. Gardner (TMP). J. Horner (MOR). S. Lucas (NMMNH). C. Herbel (SDSM). P. Owen (TMM), S. Smith (GP), and T. Tokaryk (SMNH) for granting access to specimens in their care. T. Tokaryk also kindly provided casts of SMNH P2299.1 and SMNH P2613.1. Discussions with R. Chapman. P. Dodson, M. Farney. C. Forster. R. Gay. M. Greenwald. C. Herbel. J. Horner, G. Knauss. R Larson, T. Lehman, J. Martin, J. Nelson. S. Sacrison, S. Sampson, and D. Tanke were helpful in the formulation of many of the ideas presented here. C. Forster, R. Irmis, and S. Nichols critiqued early drafts of the manuscript. Reviews by B. Chinnery-Allgeier. R Dodson, T. Lehman, S. Sampson, and an anonymous reviewer greatly improved this paper. Douglas County Memorial Hospital (Armour, South Dakota) donated use of radiographie equipment, and J. Larson operated this equipment. This study is based upon work supported under a National Science Foundation Graduate Research Fellowship and a grant from Museum of the Rockies.


BARRICK, R. E., M. K. STOSKOPF, J. D. MARGOT, D. A. RUSSELL, AND W. J. SHOWERS. 1998. The thermoregulatory functions of the Triceratops frill and horns: Heat flow measured with oxygen isotopes. Journal of Vertebrate Paleontology, 18:746-750.

BROWN, B. 1914. Anchiceratops, a new genus of horned dinosaurs from the Edmonton Cretaceous of Alberta. With discussion of the origin of the ceratopsian crest and the brain casts of Anchiceratops and Trachodon. Bulletin of the American Museum of Natural History, 33:539- 548.

CURREY, J. 1984. The Mechanical Adaptations of Bone. Princeton University Press, Princeton, New Jersey, 294 p.

FORSTER, C. A. 1996. New information on the skull of Triceratops. Journal of Vertebrate Paleontology, 16:246-258.

GILMORE, C. W. 1946. Reptilian fauna of the North Horn Formation of central Utah. U.S. Geological Survey Professional Paper, 210C:29-53.

GODFREY, S. J., AND R. HOLMES. 1995. Cranial morphology and systematics of Chasmosaurus (Dinosauria: Ceratopsidae) from the Upper Cretaceous of western Canada. Journal of Vertebrate Paleontology, 15: 726-742.

GOODWIN, M. B., AND J. R. HORNER. 2001. How Triceratops got its horns: New information from a growth series on cranial morphology and ontogeny. Journal of Vertebrate Paleontology, 21(Supplement to Number 3):56A.

HATCHER, J. B., O. C. MARSH, AND R. S. LULL. 1907. The Ceratopsia. United States Geological Survey Monograph, 49, 300 p.

HOLMES, R. B., C. FORSTER, M. RYAN, AND K. M. SHEPHERD. 2001. A new species of Chasmosaurus (Dinosauria: Ceratopsia) from the Dinosaur Park Formation of southern Alberta. Canadian Journal of Earth Sciences, 38:1423-1438.

LAMBE, L. M. 1914. On Gryposaurus notabilis, a new genus and species of trachodont dinosaur from the Belly River Formation of Alberta, with a description of the skull of Chasmosaurus belli. The Ottawa Naturalist, 27:145-155.

LEHMAN, T. M. 1989. Chasmosaurus mariscalensis, sp. nov., a new ceratopsian dinosaur from Texas. Journal of Vertebrate Paleontology, 9: 137-162.

LEHMAN, T. M. 1990. The ceratopsian subfamily Chasmosaurinae: Sexual dimorphism and systematics, p. 211-229. In K. Carpenter and P. J. Currie (eds.), Dinosaur Systematics: Approaches and Perspectives. Cambridge University Press, New York.

LEHMAN, T. M. 1998. A gigantic skull and skeleton of the horned dinosaur Pentaceratops sternbergi from New Mexico. Journal of Paleontology, 72:894-906.

LINNAEUS, C. 1758. Systema Naturae per Regna Tria Naturae (editio decima, reformata), 1. Regnum Animale. Laurentii Salvii, Stockholm, 824p.

MARSH, O. C. 1887. Notice of new fossil mammals. American Journal of Science, 34:323-324.

MARSH, O. C. 1889. Notice of gigantic horned Dinosauria from the Cretaceous. American Journal of Science, 38:173-175.

MARSH, O. C. 1890. Description of new dinosaurian reptiles. American Journal of Science, 39:81-86.

MARSH, O. C. 1891. Notice of new vertebrate fossils. American Journal of Science, 39:265-269.

MOLNAR, R. E. 1977. Analogies in the evolution of combat and display structures in ornithopods and ungulates. Evolutionary Theory, 3:165- 190.

NICKEL, R., A. SCHUMMER, E. SEIFERLE, AND W. O. SACK. 1973. The Viscera of the Domestic Mammals. Springer-Verlag, New York, 401 p.

OSBORN, H. F. 1923. A new genus and species of Ceratopsia from New Mexico, Pentaceratops sternbergii. American Museum Novitates, 93: 1-3.

PARKS, W. A. 1925. Arrhinoceratops brachyops, a new genus and species of Ceratopsia from the Edmonton Formation of Alberta. University of Toronto Studies (Geological Series), 18:1-35.

RIOBY JR., J. K. 1989. Thermoregulation in latest dinosaurs. Journal of Vertebrate Paleontology, 9(supplement to number 3):36A.

SAMPSON, S. D. 1995. Two new horned dinosaurs from the upper Cretaceous Two Medicine Formation of Montana; with a phylogenetic analysis of the Centrosaurinae (Ornithischia: Ceratopsidae). Journal of Vertebrate Paleontology, 15:743-760.

SAMPSON, S. D., M. J. RYAN, AND D. H. TANKE. 1997. Craniofacial ontogeny in centrosaurine dinosaurs (Ornithischia: Ceratopsidae): Taxonomic and behavioural implications. Zoological Journal of the Linnean Society, 121:293-337.

SCHAFFER, W. M., AND C. A. REED. 1972. The co-evolution of social behavior and cranial morphology in sheep and goats (Bovidae, Caprini). Fieldiana Zoology, 61:1-88.

STERNBERO, C. M. 1950. Pachyrhinosaurus canadensis, representing a new family of the Ceratopsia, from southern Alberta. Bulletin of the National Museum of Canada, 118:109-120.

TAYLOR, C. T. 1966. The vascularity and possible thermoregulatory function of the horns in goats. Physiological Zoology, 39:127-139.

TOKARYK, T. T. 1997. First evidence of juvenile ceratopsians (Reptilia: Ornithischia) from the Frenchman Formation (late Maastrichtian) of Saskatchewan. Canadian Journal of Earth Sciences, 34:1401-1404.

WITMER, L. M. 1997. Craniofacial air sinus systems, p. 151-159. In P. J. Currie and K. Padian (eds.), Encyclopedia of Dinosaurs. Academic Press, San Diego.



Department of Anatomical Sciences, Stony Brook University, Stony Brook, New York 11794,

Copyright Paleontological Society Jul 2006

Provided by ProQuest Information and Learning Company. All rights Reserved