Category Archives: marginocephalians

Museums and the Triceratops Posture Problem – Part 1

The Triceratops in the Hall of Extinct Monsters, circa 1911. Photo from NMNH on flickr.

The world’s first Triceratops mount at the United States National Museum, built in 1905. Photo courtesy of the Smithsonian Institution Archives.

We know more about dinosaurs today than previous generations of researchers would have ever thought possible. Who would have guessed that in the 21st century, we would have direct evidence for the color of some species, or a detailed understanding of the life history and ontogeny of others? Modern paleontologists can delve deeper into the biology and ecology of extinct animals than ever before, so it comes as a surprise when a very basic question about dinosaur physiology has gone without a definitive answer for well over a century.

For 125 years, paleontogists have struggled to understand how large ceratopsids like Triceratops held their forelimbs. Usually, someone with a good understanding of anatomy can assemble a tetrapod skeleton without much difficulty. Vertebrates are all built along the same basic body plan, and bones fit together in the same general way. However, the forelimb bones of Triceratops and its relatives are quite perplexing. The head of the humerus, which articulates with the scapula, is off-center and extends backward from the shaft. Meanwhile, the lesser tubercle, a tiny nubbin on a human humerus, is enormous and boxy. Taken together, these two traits make it so that if Triceratops held its arm erect and under its body, like most dinosaurs did, the humerus would either puncture the rib cage or be completely dislocated from the shoulder. The simplest way to solve this is to orient the humerus so that the arms project at right angles from the torso, like the sprawling limbs of a lizard. But this just looks wrong. First, ceratopsid hindlimbs are plainly meant to stand straight up. Sprawling forelimbs make Triceratops look mismatched, like the front end a tortoise sewn was to the back end of a rhino. Second, and perhaps more importantly, a sprawling posture would drastically inhibit speed and maneuverability in what is otherwise a very powerfully-built animal. The posture of Triceratops and its kin would ultimately have had a dramatic impact on the animal’s behavior, lifestyle, and ecological role.

Paleontologists haven’t spent the last century just scratching their heads over this problem. Ceratopsid forelimbs have inspired a considerable amount of research over the years, as scientists continue to develop new methods and new tools to explore the biomechanics of prehistoric animals. New technologies have been developed and refined specifically to help determine how Triceratops and its relatives walked and stood. Nevertheless, my intent with this post is not to thoroughly recount the history of ceratopsid forelimb research (if you’re interested, most of the articles referenced below are freely available online). Instead, I’d like to explore the central role museum displays have played in this debate. An artist drawing a two-dimensional image of Triceratops can fudge the orientation of the limbs (and many have), but the team building a mounted skeleton needs to know exactly how to articulate the bones. The ceratopsid posture question first arose in the process of building a mounted Triceratops skeleton for display, and museum mounts continue to be referenced by researchers looking to “ground truth” their ideas. While museum mounts usually exist primarily for education and display, in the case of the ceratopsid forelimb question these exhibits have long been central to the process of studying fossil evidence and creating knowledge.

Early Reconstructions

Marsh's 1891 restoration of Triceratops.

Marsh’s 1888 restoration of Triceratops.

O.C. Marsh published the first illustrated reconstruction of a Triceratops skeleton in 1888. Marsh was legendary in his attention to detail, and the restoration holds up reasonably well today – better, in fact, than his illustrations of Stegosaurus and “Brontosaurus.” Contemporary scientists had no complaints, even though Marsh had given the Triceratops vertical forelimbs. Other dinosaurs had erect limbs, as does the superficially similar modern rhino, so why shouldn’t Triceratops? Marsh’s reconstruction was brought to three-dimensional life in 1901, when the Smithsonian Institution commissioned a life-sized papier mache replica of a Triceratops skeleton for the Pan-American Exposition in Buffalo. Since the model was hand-sculpted, not casted from original fossils, artist F.A. Lucas had no trouble making Triceratops stand up straight, exactly as portrayed by Marsh. The model appeared again at a Smithsonian exhibit in St. Louis, but was apparently lost or destroyed shortly afterwards. In its place, newly hired United States National Museum preparator Charles Gilmore began work on a mounted Triceratops skeleton composed of original fossils.

St. Louis Expo

Straight-legged Triceratops model at the Pan American Expo in St. Louis. Source

Gilmore’s 1905 Triceratops mount was the first real skeleton of a ceratopsid ever assembled for display (first image). Like virtually all dinosaur mounts of the era, the skeleton was a composite of several specimens and a few sculpted pieces. All the Triceratops fossils at Gilmore’s disposal were collected by John Bell Hatcher in the late 19th century, and inherited by the Smithsonian as part of the Marsh collection. USNM 4842, a partial skeleton consisting mostly of a torso and pelvis, formed the basis for the mount, but at least six other individuals were also incorporated. Gilmore selected the skull because it was more complete and less distorted than the other Triceratops skulls available, but it was also on the small side compared to the body. Likewise, the left humerus was about 40% smaller than the right, and conspicuously three-toed Edmontosaurus hindfeet were used (no Triceratops feet had been found at the time). In the process of building his Triceratops, Gilmore had to make several changes to the idealized Triceratops envisioned by Marsh, most notably the orientation of the forelimbs. Not only was it apparently impossible to articulate the humerus in an upright position, but as Gilmore explained it, “a straightened form of leg would so elevate the anterior portion of the body as to have made it a physical impossibility for the animal to reach the ground with its head.”

The American Museum of Natural History produced their own Triceratops mount in 1923. Like its USNM predecessor, the AMNH Triceratops was a composite of several specimens. AMNH 5033, discovered by Barnum Brown in Montana and consisting of most of the dorsal vertebral column, ribs, and pelvic girdle, made up the largest portion of the mount. The skull was recovered by Charles Sternberg in Wyoming, and many of the appendicular bones were sculpted or cast from Smithsonian specimens. Preparator Charles Lang spent over 263 working days on the project, and much of that time was reportedly spent puzzling over the forelimbs. Lang studied living and preserved specimens of a variety of tetrapods, including rhinos, lizards, crocodiles, and tortoises, trying to find a living analogue for the strangely shaped ceratopsid bones. He ended up articulating the forelimbs so that they were even more widely splayed than Gilmore’s reconstruction, to the point that the back of the Triceratops slopes dramatically forward, and the head is almost dragging along the ground. In an accompanying paper, Henry Osborn asserted that “nothing short of a horizontal humerus and completely everted elbow would permit proper articulation of the facets.” By way of explanation, Osborn offered that this posture might have been helpful in withstanding a frontal impact.

triceratops

American Museum of Natural History Triceratops mount, circa 1959. Photo courtesy of the AMNH Research Library.

Together, the Washington and New York Triceratops mounts, with their mismatched tortoise-in-the-front, rhino-in-the-back posture, would come to define both popular and scientific conceptions of ceratopsids for the better part of a century. Other museums followed Gilmore and Lang’s lead and built sprawling ceratopsids of their own, including Richard Lull’s 1929 Centrosaurus at the Peabody Museum of Natural History and Kenneth Carpenter’s 1986 Chasmosaurus at the Academy of Natural Sciences. Even as recently as 1995, AMNH curators chose not to change a single bone on the historic Triceratops mount while modernizing their exhibit.

Voices of Dissent

Robert Bakker was one of the first to challenge the ceratopsid forelimb orthodoxy. In 1986, Bakker criticized Gilmore and Lull’s museum mounts and resurrected Marsh’s original interpretation of a straight-legged Triceratops. His reasoning was that the ceratopsid glenoid fossa (the concavity on the scapula that holds the head of the humerus) was more like the narrow cup of a horse or rhino than the wide trough of a lizard. Bakker went as far as to suggest that Triceratops and its kin might have been able to run or even gallop. Gregory Paul and others piled on, arguing that earlier researchers had run into trouble articulating Triceratops forelimbs because they had made the ribcage too broad. If the ribs were articulated so that the animal had flat flanks, the elbow apparently wouldn’t get in the way. Additional evidence for an upright stance came from a set of ceratopsid trackways described by Martin Lockley and Adrian Hunt. The trackways showed forefeet in line with the hindfeet, suggesting that front and back legs were not mismatched, after all.

This cast of the AMNH Triceratops at the Field Museum replicates the sprawling posture. Photo by the author.

This cast of the AMNH Triceratops at the Field Museum replicates the sprawling posture of the original. Photo by the author.

However, paleontologists like Peter Dodson were unmoved by these new arguments. Dodson proposed that the trackways had been misinterpreted: since ceratopsids are wider at the hips than at the shoulders, evenly spaced front and back prints should imply that the animal was holding its forelimbs out farther than its hindlimbs. Dodson was concerned that the rhino analogy was being taken too far: Triceratops looked like a rhino, so reasearchers were trying their hardest to make it move and behave like a rhino.

As Kenneth Carpenter explained in a comment last year, dinosaurs can do anything on paper, but physically assembling a skeleton forces you to confront the reality of what the bones can and cannot do. In the last decade, two new Triceratops mounts provided paleontologists the opportunity to re-explore this process, with more complete specimens and modern technology at their disposal. Next time, we’ll take a look at what the new Triceratops displays at the National Museum of Natural History and the Los Angeles County Natural History Museum can tell us about ceratopsid posture and lifestyle.

References

Bakker, R.T. 1986. The Dinosaur Heresies: New Theories Unlocking the Mystery of Dinosaurs and Their Extinction. New York, NY: Citadel Press.

Dodson, P. 1996. The Horned Dinosaurs: A Natural History. Princeton, NJ: Princeton University Press.

Fujiwara, S. 2009. A Reevaluation of the Manus Structure in Triceratops (Ceratopsia: Ceratopsidae). Journal of Vertebrate Paleontology 29:4:1136-1147.

Fujiwara, S. and Hutchinson, J.R. 2012. Elbow Joint Adductor Movement Arm as an Indicator of Forelimb Posture in Extinct Quadrupedal Tetrapods. Proceedings of the Royal Society 279: 2561-2570.

Gilmore C.W. 1905.The Mounted Skeleton of Triceratops prorsus. Proceedings of the U.S. National Museum 29:1426:433-435.

Makovicky, P. 2012. Marginocephalia. The Complete Dinosaur, 2nd Edition. Eds. Brett-Surman, M.K., Holtz, T.R. and Farlow, J.O. Bloomington, IN: Indiana University Press.

Osborn, H.F. 1933. Mounted Skeleton of Triceratops elatus. American Museum Novitates 654:1-14.

Paul, G.S. and Christiansen, P. 2000. Forelimb Posture in Neoceratopsian Dinosaurs: Implications for Gait and Locomotion. Paleobiology 26:3:450-465.

2 Comments

Filed under AMNH, anatomy, dinosaurs, fossil mounts, history of science, marginocephalians, museums, NMNH, reptiles

Extinct Monsters: Brachyceratops

Click here to start the Extinct Monsters series from the beginning.

Most of the mounted dinosaur skeletons at the National Museum of Natural History (NMNH) represent taxa that are well-known to casual paleontology enthusiasts. But nestled amongst household names like Triceratops, Stegosaurus and Diplodocus is an easily-overlooked horned dinosaur that was historically called Brachyceratops montanensis (it’s currently labeled Styracosaurus sp). Tucked away in a glass case behind the giant Triceratops, this pocket-sized ceratopsian may not be the most spectacular display in the exhibit, but it is nevertheless an important one for the Museum. Discovered in 1913 by the Smithsonian’s own Curator of Fossil Reptiles Charles Gilmore, Brachyceratops represents one of only a few dinosaur species excavated, prepared, described and exhibited entirely in-house at NMNH. It is therefore unfortunate that modern researchers have banished the name Brachyceratops to the realm of taxonomic obscurity. What’s more, the days of the Brachyceratops mount, on exhibit since 1922, are numbered: when the NMNH paleontology halls closed for renovation in April 2014, this specimen was be retired to the collections, and is not planned for inclusion when the exhibit reopens in 2019.

The Brachyceratops mount today. Photo by the author.

The Brachyceratops mount today. Photo by the author.

During his tenure at NMNH, Gilmore was an inexhaustibly productive writer, publishing at least 170 scientific articles, including numerous important descriptions and reassessments of fossils discovered by O.C. Marsh’s teams in the 19th century. However, Gilmore was much happier studying fossils in his lab than excavating new finds in the field, taking part in a scant 16 NMNH-sponsored field expeditions over the course of his career. A 1913 trip to the Cretaceous Two Medicine Formation in Northeast Montana was therefore unusual for Gilmore. He was following in the footsteps of Eugene Stebinger of the US Geological Survey, who had reported the previous year that the region was only minimally explored but clearly awash in vertebrate fossils.

On this inaugural fossil prospecting trip, Gilmore’s team located abundant remains of fish, small reptiles and dinosaurs, especially hadrosaurs and ankylosaurs. The most notable find, however, was a small bone bed (about six feet square) of ceratopsian fossils, representing at least five individuals. Gilmore described this find in a 1917 monograph, naming the dinosaur Brachyceratops montanensis. Today we know that ceratopsians were quite diverse, particularly during the Campanian, but in the early 20th century the true extent of the group was only just being revealed. Still, it was clear to Gilmore that at an estimated six feet in length, Brachyceratops was an unusually small ceratopsian. He proposed that it may have fed on different plants or occupied a different niche than larger contemporaries like Centrosaurus and Styracosaurus.

pretty art

Reconstruction of Brachyceratops holotype skull. Plate from Gilmore 1917.

In 1917, most of the dinosaur mounts on display at NMNH came from fossils collected by Marsh for the US Geological Survey, and many represented species also on display in New York, Pittsburgh and New Haven. Accordingly, Gilmore was doubtlessly enthused by the prospect of displaying a dinosaur exclusive to Smithsonian. He awarded the task of creating a Brachyceratops mount to preparator Norman Boss, who would spend 345 working days on the project. Of the five individuals found in Montana, USNM 7953 was selected as the basis for the mount because it was the most complete, with the sacrum, pelvis, femora and complete set of caudal vertebrae found articulated in situ. Helpfully, Gilmore published a list of precisely which parts of the mount came from which individual specimen (see below). This was a marked contrast from some of his contemporaries at other museums, who would not bother to record such information, or even actively obscure how many disparate specimens they were using to build their mounts.

Gilmore's helpful list

A helpful breakdown of the Brachyceratops mount from Gilmore 1922.

Boss based his restoration of Brachyceratops closely on the complete, articulated Monoclonius (=Centrosaurus) specimen (AMNH 5351) discovered by Barnum Brown in 1914. In particular, Boss replicated the angle of the scapulae and the number of vertebrae (22) on the American Museum of Natural History skeleton. Missing bones and portions thereof were sculpted in plaster, easily recognized by their solid color and smooth texture. Just as Gilmore and Boss had done with their 1905 Triceratops mount, the Brachyceratops was given strongly flexed elbows. According to Gilmore, a very large olecranon process on the ulna would have forced all ceratopsians into this somewhat awkward stance. Of particular note is the restoration of the skull, which was found shattered into dozens of pieces, many smaller than one inch. A close look at the specimen reveals how Boss painstakingly reassembled these fragments. Unfortunately, this is difficult in the exhibit hall because the mount is posed with the side of the skull that is mostly plaster facing visitors.

Norman Boss Brachyceratops courtesy Smithsonian archives

Norman Boss puts the finishing touches on the Brachyceratops mount. Photo courtesy of the Smithsonian Archives.

The completed Brachyceratops mount was placed on exhibit in 1922, on the same pedestal in the Hall of Extinct Monsters as the Triceratops. The substrate beneath the mount was colored and textured to match the Two Medicine Formation sandstone in which the fossils were found. Gilmore also prepared one of his charming models of Brachyceratops, mirroring the pose of the mount, but it is unclear whether it was ever exhibited.

woo triceratops

Brachyceratops on exhibit with Triceratops. Plate from Gilmore 1922.

The Brachyceratops has remained on view through each subsequent renovation of the fossil halls, always placed close to Triceratops. This close association has prompted many visitors to mistake the diminutive Brachyceratops for a baby Triceratops, and in fact these visitors are on the right track. While Gilmore always described Brachyceratops as an unusually small but full-grown ceratopsian, Scott Sampson and colleagues confirmed in 1997 that all five specimens were juveniles. A century’s worth of new fossil discoveries has provided modern paleontologists with a thorough understanding of ceratopsian ontogeny, and characteristics like the unfused nasal horn clearly mark the mounted Brachyceratops as a young animal. Unfortunately, Gilmore’s Brachyceratops specimens lack any good diagnostic features that could link it to an adult form. According to Andrew McDonald, the most likely candidate is Rubeosaurus ovatus, which was, incidentally, discovered by Gilmore on a 1922 repeat trip to the Two Medicine site. Nevertheless, without the ability to recognize other growth stages of the same species, the name Brachyceratops is unusable and is generally regarded as a nomen dubium.

It is not difficult to surmise why the Brachyceratops would end up near the bottom of the list of mounts to include in a renovated gallery. It is not especially large or impressive, it doesn’t have a recognizable name (or any proper name at all, really) and it doesn’t tell a critical story about evolution or deep time. With limited space available and new specimens being prepped for display, little Brachyceratops will have to go. It’s not all bad, though. Taking these fossils off exhibit will make them more accessible to researchers, and allow them to be closely examined in all aspects for the first time in decades. Perhaps one day soon we will have a clearer idea of the identity of one of Gilmore’s great finds.

References

Gilmore, C.W. (1917). Brachyceratops, a Ceratopsian Dinosaur from the Two Medicine Formation of Montana, with Notes on Associated Fossil Reptiles. Washington, DC: US Geological Survey.

Gilmore, C.W. (1922). The Smallest Known Horned Dinosaur, Brachyceratops. Proceedings of the US National Museum 63:2424.

Gilmore, C.W. (1930). On Dinosaurian Reptiles from the Two Medicine Formation of Montana. Proceedings of the US National Museum 77:2839.

McDonald, A.T. (2011). A Subadult Specimen of Rubeosaurus ovatus (Dinosauria: Ceratopsidae), with Observations on other Ceratopsids from the Two Medicine Formation. PLoS ONE 6:8.

Sampson, S.D., Ryan, M.J. and Tanke, D.H. (1997). Craniofacial Ontogeny in Centrosaurine Dinosaurs: Taxonomic and Behavioral Implications. Zoological Journal of the Linnean Society 12:1:293-337.

1 Comment

Filed under dinosaurs, exhibits, Extinct Monsters, field work, fossil mounts, history of science, marginocephalians, museums, NMNH, reptiles

A Triceratops for Lawrence

In an earlier post, I described how the Smithsonian’s Triceratops was the first free-standing mount of this animal ever created, and the eccentricities of its proportions and posture would define how the species would be depicted in artwork and other reconstructions for the better part of a century. I should have clarified, however, that this was not the first time Triceratops fossils were put on public exhibit. That honor goes to the University of Kansas Natural History Museum, which was founded in 1864 and is still operational today.

Among the Museum’s basement paleontology exhibits is a case of dinosaur fossils that has literally gone unchanged since the 1950s. Although this time capsule of mid-century museum design is of some historical interest, it would be nice if those sauropod limb bones weren’t labeled “Brontosaurus” (to be clear, the whole museum doesn’t look like this, the staff has been slowly but surely modernizing the exhibits). Of particular importance is the Triceratops skull (specimen 422) on the left side of the case. While there is no historical information on its label, this specimen has been with the museum for over 115 years, having been unearthed and put on display only six years after O.C. Marsh first named and described Triceratops.

The dinosaur case at the KU Natural History Museum, untouched since the 1950s.

The dinosaur case at the KU Natural History Museum, untouched since the 1950s.

The story of the KU Triceratops is not well-known, although it is the subject of the somewhat hard-to-find book (neither the Museum nor the KU library has a copy) A Triceratops Hunt in Pioneer Wyoming. In the summer of 1895, a team from the University ventured into the frontier lands of of eastern Wyoming with the explicit goal of finding a Triceratops for display at the young Natural History Museum. The team was led by Samuel Wendell Williston, founder of the University’s geology department. Although Williston’s specialty was entomology, he had previously worked under Marsh at Yale and was well-acquainted with the plethora of dinosaurs on which his mentor had published. Also on the expedition were KU Regent James Polk Sams, and two individuals whose names are quite familiar to anyone with an interest in the history of paleontology, Barnum Brown and Elmer Riggs. Brown would, of course, go on to be the star fossil hunter at the American Museum of Natural History, while Riggs would become a curator at the Field Museum of Natural History. In 1895, however, both were students, and not especially interested in fossil collecting or paleontology. 

The team found the Triceratops they were looking for on July 9, near the confluence of Lightning and Lance Creeks. By July 22, the skull was fully excavated and crated for a journey by train back to Lawrence. The fossil apparently garnered a fair amount of attention on the journey; while Triceratops is well known today, in 1895 few had any idea that animals such as this had ever walked the Earth. And yet, here was clear, physical evidence of an extinct animal like nothing alive in the modern world, and it would soon be displayed for all to see in the University of Kansas Museum.

Triceratops skull retrieved by the 1895 Wyoming expedition.

Triceratops skull retrieved by the 1895 Wyoming expedition.

The 1895 expedition was not tremendously productive scientifically, the fossils found that summer resulting in only three short papers. However, the journey, and the Triceratops skull that was brought back, did end up being quite important for paleontology. The expedition inspired Brown and Riggs to pursue careers in paleontology, and as Brinkman and colleagues write in Triceratops Hunt in Pioneer Wyoming, “they would hunt bones for the rest of their lives, to the great benefit of science, and might never have done so had not circumstances landed them in the circle of Williston’s influence in the summer of 1895.”

Furthermore, by setting out with the clear goal to find a dinosaur for display, the University of Kansas team were trailblazers in a movement that would lead to the inseparable connection between dinosaurs and museums in popular culture today. In the first decade of the 20th century, newly burgeoning large urban museums openly competed to find and display the largest and most spectacular dinosaurs. Exhibits like the AMNH Brontosaurus, the Carnegie Museum of Natural History Diplodocus and, of course, the Smithsonian Triceratops brought millions of Americans into museums, ensuring that to this day, when we think of museums, we think of dinosaur skeletons (and vice versa). In spite of the somewhat dingy basement display it currently finds itself in, the University of Kansas Triceratops was an early trendsetter and a profound example of the intersection of science and history.

Reference

Kohl, M.F., Martin, L.D. and Brinkman, P., eds. (2004). Triceratops Hunt in Pioneer Wyoming: The Journals of Barnum Brown and J.P. Sams. Glendoo, WY: High Plains Press.

3 Comments

Filed under dinosaurs, field work, fossil mounts, history of science, marginocephalians, museums, reptiles