A single fossil bone can tell so much

Print Friendly, PDF & Email

Thomas H Rich and Patricia Vickers-Rich (Australia)

Whether they had horns or not, the ceratopsian “horn faced” dinosaurs are distinctive, not only from other dinosaurs, but all other vertebrates as well, in the structure of their skulls. In addition to the horns, another element of their skeleton, the lower arm bone (called the ulna or elbow bone), unexpectedly is so distinctive that it has provided clear evidence that, 130 million years ago, these very ceratopsians were living in Australia. Prior to that discovery, the ceratopsians were known almost exclusively in the Northern Hemisphere.

Just over a century ago, a toothless lower jaw found in Patagonia, Argentina was named Notoceratops, “the southern horned face”. The last time that fossil was seen was a decade later when the world-renowned dinosaur authority, Fredrich von Huene, studied and redescribed that fossil and agreed unreservedly that it was a ceratopsian. Illustrations of that bone strongly support its correct identification as a ceratopsian. However, unfortunately, von Huene is the last person known to have laid eyes on it, and the fossil cannot now be found. Thus, the only ceratopsian previously thought to have come from the Southern Hemisphere, disappeared.

When the Victorian dinosaur ulna, which is the subject of this article, was first found at the base of the Arch near Kilcunda (Fig. 1) by Mike Cleeland, Tom’s first guess was that it was some kind of carnivorous dinosaur or theropod. This was because it was a short, stumpy bone, which is so characteristic of the forearms of T. rex. Many theropods have such reduced forelimbs, so this was a good first guess. But it was wrong as further information clarified.

Fig. 1. The Arch near Kilcunda, Victoria, Australia. The arrow points to the site where the ulna of the ceratopsian Serendipaceratops arthurcclarkei Rich and Vickers-Rich 2003 NMV P186385 was found. (Photo courtesy of Mike Cleeland, who found the fossil.)

A bit later, Tom paid a visit to the Royal Tyrell Museum of Paleontology in Drumheller, Alberta, Canada. There, he was able to examine the ulnae of a number of theropods. None of these were even close in shape.

While Tom was examining a variety of theropod ulnae, he walked past a skeleton of the ceratopsian dinosaur, Centrosaurus, numerous times. The skeleton had been laid out on the top of some adjacent storage cabinets and, in that way, were readily seen. What jumped out at him on one of these casual inspections of this skeleton was that its ulna was very similar in shape, but just about twice the length of the still as yet unidentified Victorian ulna.

Knowing that much smaller ceratopsians than Centrosaurus existed, Tom asked staff at the museum if they had any of these other, smaller ones in their collection. The answer was that a skeleton of one of these smaller ceratopsians, Leptoceratops (Fig. 2), was at that very moment being moulded and cast in downtown Drumheller. Leptoceratops had been collected not far from the Tyrell Museum in rocks about 60 million years younger than those from which the Victorian ulna came.

Fig. 2. Skeleton of Montanaceratops, formerly known as Leptoceratops cerorhynchus, on display at the American Museum of Natural History in New York.

Accompanied by intrigued members of the Museum staff, Tom was quickly driven to where the Albertan fossil was, and soon the Australian and Canadian ulnae were laid side by side. A gasp went up from all concerned. If the two fossils had come from the same hole in the ground, instead of having been found 14,000km apart and from rocks differing by 60 million years in age, one could be excused for having considered them as belonging to the same species.

Further examination of several other dinosaurs soon clearly demonstrated to Tom that the laterally flattened cross section of the ulna of ceratopsians was unique to that group (Fig. 3).

Fig. 3. Ulnae of three dinosaurs: (A) Serendipaceratops arthurcclarkei, Victoria, Australia; (B) Leptoceratops gracilis, Alberta, Canada; and (C) Megaraptorid theropod, Victoria, Australia.

Thus, the realisation gradually came to Tom that this ulna was not only unique, but so close to a ceratopsian that it had to be one. Eventually, this led to this intriguing fossil to become the type specimen for a brand-new genus and species of dinosaur, Serendipaceratops arthurcclarkei, representing a group previously unknown on the Australian continent – the ceratopsians.

Because of the unique similarity of the ulnae of Leptoceratops to that of Serendipaceratops, it was the basis for the artist Peter Trusler to reconstruct a life-like restoration of Serendipaceratops based on a skeleton of the Albertan ceratopsian (Fig. 4).

Fig. 4. Serendipaceratops arthurcclarkei. (Artist: Peter Trusler.)

With the only other Southern Hemisphere record of a ceratopsian being the lost holotype of Notoceratops, and the general unfamiliarity of palaeontologists regarding the uniqueness of the morphology of ceratopsian ulnae, there was a cool response to the idea that Serendipaceratops was truly a ceratopsian when it was formally published as a new dinosaur.

Because of this general unfamiliarity of most palaeontologists concerning the shapes of ceratopsian ulnae, it was necessary to provide convincing evidence of such similarity and its true uniqueness, when compared to that same bone of other dinosaurs. Fortunately, Brenda Chinnery had already compiled the key pieces of information needed to support this ceratopsian similarity. A year after Serendipaceratops was formally named, Brenda published a table with 11 measurements that she had made of the ulnae of 58 ceratopsians (Fig. 5).

Fig. 5. The 11 ceratopsian ulnar measurements that were made by Brenda Chinnery plotted on an image of the ulna of Triceratops porosus published as plate XIII in Hatcher, John Bell, and Othniel C Marsh. The Ceratopsia. No. 310. (US Government Printing Office, 1907.)

Comparing five of those measurements with those of ulnae of other dinosaurs and further afield (that is, other tetrapods) in five-dimensional space, the ulnae of ceratopsians clustered together to the exclusion of all other groups. And Serendipaceratops was situated right in the middle of that cluster. Because we cannot see in five dimensions, the best we can do to visualise this clustering is to plot three of those dimensions at one time (Fig. 6). All different combinations of the three dimensions at a time clearly show the same clustering.

Fig. 6. Plots of variables 5”, 7” and 10” of the ceratopsian ulnae measurements compiled by Brenda Chinnery, plus that of Serendipaceratops,together with the corresponding ulnae measurements of other tetrapods.
Stereoscopic pair of a 3D scatter diagram of the holotype of Serendipaceratops arthurcclarkei (large red cross) compared with other ceratopsians (small crosses), ankylosaurs (solid black spheres), and other dinosaurs and non-dinosaurian amniotes (light hatched spheres). Plot is of Variables 5”, Variable 7”, and Variable 10”, where these variables have been normalised in morphospace representation.
Var 5”. Minimum craniocaudal width of shaft, perpendicular to the long axis.
Var 7”. Length from sigmoid fossa rim to distal end, parallel to the long axis.
Var 10”. Mediolateral minimum shaft width.
The two other ulna measurements that were part of the 5D plot of the data were:
Var 4”. Maximum craniocaudal width at the sigmoid fossa rim, perpendicular to the long axis.
Var 11”. Mediolateral maximum width across the sigmoid fossa rim. This measurement indicates robusticity of the humeral joint.

A single bone being the foundation for the suggestion that ceratopsians were present in Australia seems to be basing a fundamental change of ideas about where the dinosaurs once lived on very slim evidence. However, the dinosaur record on this continent is meagre in the extreme compared with that of other continents; for example, Asia, North America, and South America. Even on those land masses, the distribution is not even. Most North American dinosaurs are found on the western half of the continent. 130 million years ago, when Serendipaceratops lived, Australia was far from North America and Asia where most ceratopsian fossils are known.

How interchange could have occurred between Australia and those continents, even with Europe included as another land mass with a ceratopsian record (albeit a meagre one), leaves the identification of the Australian specimen a topic rife for speculation. Australia was more isolated from those four land masses then than it is now. The slim evidence for a South American ceratopsian suggests that an interchange between there and Australia across Antarctica could have been possible, as South America was closer to the Northern Hemisphere land masses than Australia at that time.

What’s in a name?

Why the name Serendipaceratops arthurcclarkei? If one word can be said to run through this story about this fossil, it is serendipidity, “the facility for making fortuitous discoveries by accident”. The good fortune begins with the discovery of the fossil itself at a remote site, which is difficult to access, where only a few other fossils have ever been collected due to the determination of Mike Cleeland.

The next piece of good luck was Tom’s chance selection of this particular fossil to take to North America in an attempt to identify it. Then it was having a skeleton of Centrosaurus laid out in such a way at one particular museum so that a similarity of the ulnae was clearly recognised by a casual glance. Following that, the fact that the ulnae of ceratopsians are so markedly distinct from those of other dinosaurs and other terrestrial vertebrates generally was absolutely critical. And, adding to that, Brenda Chinnery having tabulated measurements of more ceratopsian ulnae than had been systematically done for any other vertebrate group, was also critical in being able to readily establish the uniqueness of the proportions of ceratopsian ulnae.

All of those coincidences explain, in part, why the generic name Serendipaceratops was appropriate, but only in part. The specific name arthurcclarkei has a role in the choice of the generic name, too.

Arthur C Clarke had been Tom’s favourite science fiction author. Shortly after learning that Sir Arthur had gotten interested in science because he was fascinated by dinosaurs, and through a mutual acquaintance, Tom found his postal address in Sri Lanka. Tom then wrote to Arthur in appreciation of all the enjoyment his writings had given Tom over the years. With that letter went a copy of the book Dinosaurs of Darkness that Tom and Pat had recently published. Back came an effusive letter from Sir Arthur that was the first of a number of exchanges of correspondence. Included in that first letter was a picture of Sir Arthur standing by a statue of a small dinosaur in his garden in Sri Lanka (Fig. 7).

Fig. 7. (Above) Sir Arthur with his dinosaur companion standing in his Sri Lankan garden. (Right) Sir Arthur in his “Ego Room”, as he had dubbed it, with his typical wry sense of humour, festooned as it was by numerous awards and mementos, and flanked by Pat and Tom.

Sir Arthur invited us to come and visit him if ever we should be happening to be passing through Sri Lanka. We had no other reason to go to Sri Lanka, but decided the chance to meet him was quite sufficient, so we went there and enjoyed the visit very much, including watching with Arthur, 2001 A Space Odyssey, sitting in his office. That all took place soon before Serendipaceratops arthurcclarkei was formally named in 2003. So, when the name was coined, the fact that the fossil had many aspects of serendipidity in its history and that Serendip was an ancient Persian name for Sri Lanka were all things that shaped the generic name. And that our friendship with Sir Arthur, to whom Sri Lanka was home, was brought about because of a common interest in dinosaurs put the genus and species together in our minds (Fig. 8).

Fig. 8. Abstract of the paper describing Arthur’s frilled dinosaur – Serendipaceratops arthurcclarkei. (Photo by P Vickers-Rich.)
Fig. 9. Arthur was so pleased to have this dino-name that he inscribed the front of this paper published as one of the Records of the Queen Victoria Museum in Launceston, Tasmania: “Now I have a dinosaur and an asteroid named after me, what else is there to live for?” (Photo by P Vickers-Rich.)

About the authors

Since the age of 12, Tom has been fascinated with Mesozoic mammals. None were known from Australia when the two of us first set foot there. So when an opportunity to immigrate there arose by chance (serendipidity again?), we did so. In pursuing that interest, dinosaurs such as Serendipaceratops were found and not ignored.

Pat studied the fossil birds of Australia for her doctorate. After having worked for years on that topic, she realised that an equally intriguing question was the nature of the multi-cellular organisms that once lived in Namibia at the very end of the PreCambrian. And so she switched to investigating them.


Rich, T.H., Kear, B.P., Sinclair, R., Chinnery, B., Carpenter, K., McHugh, M.L. and Vickers-Rich, 2014. Serendipaceratops arthurcclarkei Rich & Vickers-Rich, 2003 is an Australian Early Cretaceous ceratopsian. Alcheringa 38:456-479.   

Rich, T. H., and Patricia Vickers-Rich 2020. Dinosaurs of Darkness: In Search of the Lost Polar World. 2nd edition, Indiana University Press.

Leave a Reply