Hairy Museum of Natural History

Last summer, a comment about the size of the newly-described South American sauropod Puertasaurus led me to investigate the size of Amphicoelias fragillimus, the name E. D. Cope gave to part of one gigantic sauropod vertebra found in Colorado in the 1870s. In a comment to that post, Jerry Harris dropped the hint that more information on A. fragillimus was in the works, and now at least part of what he alluded to has come to pass. A new paper by Ken Carpenter of the Denver Museum of Nature and Science reviews what is known of A. fragillimus, provides a revised size estimate, and discusses why so many sauropods got so darned big.

This paper has inspired a couple of very good posts on Darren Naish’s Tetrapod Zoology blog, and I’ll refer the reader here and here for his excellent summary of the history and validity of A. fragillimus. I’ll focus on the new size estimates, and why A. fragillimus was even larger than I had imagined.

My initial estimate of the size of A. fragillimus was made by comparing Cope’s reconstruction of the type vertebra to the vertebrae of local giant Seismosaurus hallorum (properly Diplodocus hallorum at this point, but more on that later). Scaling up a Seismosaurus to fit Cope’s estimate of a 6 to 7 foot tall A. fragillimus vertebrae results in an animal about 50 meters (160 feet) long.

As it turns out, Cope may have significantly underestimated the height of the vertebra of A. fragillimus. Carpenter compared Cope’s drawing of the Amphicoelias fragillimus specimen with material known from another, more reasonably-sized species of Amphicoelias, A. altus. Scaling an A. altus dorsal to fit the A. fragillimus material results in a whopping 2.7 meter (8.8 foot) tall bone!

From left to right:

• The diplodocid Seismosaurus hallorum dorsal 8 (after Herne and Lucas 2006),
• The diplodocid Amphicoelias fragillimus dorsal 9/10? as reconstructed by Cope 1878,
• The diplodocid Amphicoelias fragillimus dorsal 9/10? as reconstructed by Carpenter 2006, scaled from
• The diplodocid Amphicoelias altus dorsal 10? (after Carpenter 2006, modified from Osborn and Mook 1921),
• The titanosaur Puertasaurus reuili dorsal 2 after Novas et al 2005, Figure 2). Scale bar equals 1 meter.

Carpenter scaled up a Diplodocus to fit this new super-sized vertebra, and his Amphicoelias fragillimus measures a full 58 meters (190 feet) from snout to tail. The estimated mass of this mega-sauropod would be about 122,400 kg (about 270,00 lbs or 135 tons).

Here is the size comparison I made last summer, comparing the big diplodocids Seismosaurus hallorum (33 meters) and Amphicoelias fragillimus (my 50-meter estimate) to some of the largest titanosaurs, Argentinosaurus huinculensis (at 37.5 meters, the length of the skeletal reconstruction on display at the Fernbank Museum) and Puertasaurus reuili (at 40 meters, the upper estimate reported in the media). Click on the image for a larger version:

In Figure 3 of his paper, Carpenter shows a more detailed size comparison of mega- and super-sauropods (mega-sauropods are defined as those that reach or exceed 30 meters in length)—a copy of it is posted here. In addition to an enormous Amphicoelias, it shows Seismosaurus (C) and a silhouette of Argentinosaurus (F) based on the profile of the related titanosaur Saltasaurus. This new titanosaur shape changes their dimensions considerably, shrinking Argentinosaurus from 37.5 down to 30 meters in length.

Based on this information, I resized and revised and came up with the following size comparison (click for a larger version):

From left to right:

• The diplodocid Seismosaurus hallorum (skeleton on display at the NMMNHS–33 m. (110 ft.) long),
• The diplodocid Amphicoelias fragillimus (est 58 m. (190 ft.) long),
• Homo sapiens (1.8 m. (6 ft.) tall),
• African Elephant Loxodonta africana (4 m. (13 ft.) tall at the shoulder),
• The titanosaur Argentinosaurus huinculensis (est 30 m. (98 ft.) long),
• The titanosaur Puertasaurus reuili (shortest reported estimate 35 m. (115 ft.) long).

Note that, in this latest estimate, Seismosaurus completely fits under the tail of Amphicoelias fragillimus!

Further Reading:

Carpenter, K. 2006. Biggest of the big: a critical re-evaluation of the mega-sauropod Amphicoelias fragillimus Cope, 1878. New Mexico Museum of Natural History and Science Bulletin 36: pp. 131–137. PDF.

Cope, E. D. 1878. A new species of Amphicoelias. American Naturalist 12(8): pp. 563–564. JPG. HTML.

Herne, M. C. and Lucas, S. G. 2006. Seismosaurus hallorum: osteological reconstruction from the holotype. New Mexico Museum of Natural History and Science Bulletin 36: pp. 139–148.

Novas, F. E., Salgado, L., Calvo, J., and Agnolin, F. 2005. Giant titanosaur (Dinosauria, Sauropoda) from the Late Cretaceous of Patagonia. Rev. Mus. Argentino Cienc. Nat., 7(1): pp. 37–41. PDF.

Don’t forget the two posts at Tetrapod Zoology, and Matt Wedel (a.k.a. Doctor Vector) promises to weigh in on the topic, so you know that’s worth keeping an eye out for.

—Matt Celeskey.

File under: Dinosaurs, Jurassic.

Comments on record: (8)

The cover of this past week’s issue of Nature is graced with a restoration of a newly discovered gliding mammal from the Age of Dinosaurs. The associated article inside describes Volaticotherium antiquum—”the ancient flying beast”—known from a flattened skeleton from the Daohugou beds in northeastern China. The fossil preserves traces of a big blob of hair and soft tissue that the authors interpret as a patagium—a large area of skin supported by the limbs that, when spread open in midleap, enabled Volaticotherium to glide through the treetops. Such gliding has evolved independently several times among mammals, in flying squirrels, for example, and colugos, and, (presumably) among the ancestors of bats.

Volaticotherium, however, hails from well before any of these aerial upstarts. The oldest known gliding rodent is not a squirrel but an eomyid from the Late Oligocene (~25,000,000 years ago), and the earliest known evidence for flight in bats comes from the Eocene, 51,000,000 years ago. Volaticotherium is at least 70,000,000 years older than that, and could be even older. (The age of the Daohugou beds is controversial, but lies somewhere between the Early Cretaceous (~125,000,000 years ago) and the Middle Jurassic (~170,000,000 years ago).) If the older age is correct, this would mean that Volaticotherium was experimenting with aerial locomotion before the 150,000,000-year-old Archaeopteryx, the earliest known bird.

Further reading:

Meng, J., Hu, Y., Wang, Y., Wang, X., and Li, C. 2006. A Mesozoic gliding mammal from northeastern China. Nature 444, 889–893. Editor’s summary.

A larger image of the cover art by Chuang Zhao and Lida Xing is shown at National Geographic News.

12/19 Update: Much more information on Volaticotherium (and how its discovery fits into the grand arc of Mesozoic Mammal Studies) in this article by Trevor Dykes.

—Matt Celeskey.

File under: Cretaceous, Jurassic, Mammals, Recent Discoveries.

Comments on record: (1)

A photo posted over at NewCritters.com has led me to revise my earlier post comparing the newly-described Puertasaurus reuili with the legendary Amphicoelias fragillimus. Turns out Puertasaurus is a little less massive than I had portrayed it, although I still wouldn’t want to carry one for any distance.

Apologies to all museum visitors for passing on less-than-accurate info—rest assured I’ll be cross-checking future figures before posting them.

—Matt Celeskey.

File under: Cretaceous, Dinosaurs, Jurassic.

Comments on record: (1)

In the comments to last Sunday’s post on the giant South American sauropod Puertasaurus, wolfwalker asked how this new contender for biggest dinosaur ever compares to Amphicoelias fragillimus, a giant North American sauropod named by Edward Drinker Cope in 1878.

Amphicoelias fragillimus (“Doubly hollow and very fragile”) was only known from part of a dorsal vertebra collected near Cañon City, Colorado. Cope reported that the preserved portion of the bone measured 1.5 meters in height, and estimated that the complete vertebra would have been at least 6 feet tall. (The original description is relatively short, and well out of copyright, so I’ve taken the liberty of transcribing it into the HMNH Library. A scan of the description as it appeared in the journal American Naturalist is available elsewhere.)

Unfortunately, today nobody knows where this fossil is. How, you might ask, could anyone lose the better part of a six-foot-tall vertebra? Definitive answers are lacking, but most suspect that it crumbled to pieces in storage or transit (Cope didn’t call it “fragillimus” for nothing). So all we are left with is Cope’s original description and figure (reproduced to the left), making A. fragillimus something of a dinosaurian fish-story—the big one that got away.

Provided that Cope’s description is accurate (a claim not everyone is willing to accept, given the rather colorful state of paleontology at the time), Ampicoelias fragillimus would have been by far the largest diplodocid sauropod ever described. Diplodocids were a family of lean, long and low sauropods, whose members include Apatosaurus (75 feet long), Diplodocus (83 feet long), Barosaurus (93 feet long), and Seismosaurus (110 feet long). Scaling up a Seismosaurus skeleton to make the vertebrae as large as Amphicoelias fragillimus results in an animal about 160 feet long!

How does this compare to the newly-described Puertasaurus reuili? Puertasaurus is estimated to be 115 to 130 feet long, so it couldn’t match the estimated length of Amphicoelias. But Puertasaurus was a titanosaur, a group that had a wider, heavier, and stockier build than their distant diplodocid cousins. So, even though it would have taken more steps to walk the length of Amphicoelias fragillimus, Puertasaurus would still have been a much somewhat more massive animal. This is easy to see when you compare their vertebrae side-by-side: See the revision below:

From left to right:

• The diplodocid Seismosaurus hallorum dorsal 3 (from the Jurassic exhibit at the NMMNHS),
• The diplodocid Amphicoelias fragillimus dorsal 9/10? (after Cope 1878),
• The titanosaur Argentinosaurus huinculensis dorsal 4? (after Novas et al 2005),
• The titanosaur Puertasaurus reuili dorsal 2 (ibid). Scale equals 1 meter.

To see the scale of the different animals, I worked up a rough series of silhouettes. Click on the image for a larger version:

From left to right:

• The diplodocid Seismosaurus hallorum (skeleton on display at the NMMNHS–110 ft. long),
• The diplodocid Amphicoelias fragillimus (est. 160 ft. long),
• Homo sapiens (6 ft. tall),
• The titanosaur Argentinosaurus huinculensis (skeleton on display at the Fernbank Museum–123 ft. long),
• The titanosaur Puertasaurus reuili (est. 135 ft. long),
• African Elephant Loxodonta africana (13 ft. tall at the shoulder).

Note that this image is no more than a rough guideline—none of the sauropods are known from more than a 30% complete skeleton, and the proportions of the titanosaurs are particularly sketchy. But I think it is safe to say that the enormous sauropods coming out of South America are giving even the most legendary giants of the western U.S. a run for their money.

Revision 8/18: Something isn’t quite right about the images above.
Something seemed amiss when I saw this image over at NewCritters.com. The Puertasaurus vertebra pictured there couldn’t be as large as I’d made it in the comparison above.

After double-checking the figures in Novas et al.’s paper, I realized that they had illustrated this vertebrae with two different scales. In Figure 2 of the paper, the scale shows it to be consistent with the reported height of 106 centimeters. In Figure 3, where this vertebra is compared to those of other sauropods, the scale shows it to be at least 25% larger. Since the latter figure had an Argentinosaurus vertebra to (what I assumed was) the same scale, I used that figure for the above diagram without double-checking the text. A revised vertebral comparison (showing the discrepancies in the original figures) is shown below:

From left to right:

• The diplodocid Seismosaurus hallorum dorsal 3 (from the Jurassic exhibit at the NMMNHS),
• The diplodocid Amphicoelias fragillimus dorsal 9/10? (after Cope 1878),
• The titanosaur Puertasaurus reuili dorsal 2 (after Novas et al 2005, Figure 3),
• The titanosaur Puertasaurus reuili dorsal 2 (after Novas et al 2005, Figure 2). Scale equals 1 meter.

What a difference in size! The smaller version on the right is consistent with the reported measurements and photographs, and I feel fairly confident that this is the correct size. It makes it much smaller, however, and much closer in scale to Seismosaurus than Amphicoelias.

How would this affect the size of the animal? More on that later…

Revision 8/19: Updating the animals

In putting together the silhouette comparison above, I modified an outline of Gabriel Lio’s Puertasaurus restoration for the two titanosaurs. The resulting outlines of Argentinosaurus and Puertasaurus seemed a little beefy to me, but since they were in keeping with the relative sizes of the vertebrae, I figured they were fairly close. With the revised size of the Puertasaurus vertebra, I thought I’d try another method.

A Google image search revealed a couple of good photos of the skeletal reconstruction of Argentinosaurus huinculensis mounted at the Fernbank Museum of Natural History, which comes in at 123 feet long. Using these images, I came up with a revised giant-titanosaur silhouette for Argentinosaurus and Puertasaurus which is considerably more svelte. However, plugging them into the comparison with Seismosaurus and Amphicoelias shows that they were still nothing to sneeze at (click on the image for a larger version):

From left to right:

• The diplodocid Seismosaurus hallorum (skeleton on display at the NMMNHS–110 ft. long),
• The diplodocid Amphicoelias fragillimus (est. 160 ft. long),
• Homo sapiens (6 ft. tall),
• African Elephant Loxodonta africana (13 ft. tall at the shoulder).
• The titanosaur Argentinosaurus huinculensis (skeleton on display at the Fernbank Museum–123 ft. long),
• The titanosaur Puertasaurus reuili (maximum est. 135 ft. long).

—Matt Celeskey.

File under: Cretaceous, Dinosaurs, Jurassic.

Comments on record: (8)

A few more items of interest concerning the dwarf sauropod Europasaurus:

• Afarensis goes into more detail on the bone histology that allowed researchers to determine that the six-meter sauropod was fully grown.
• A German press release from the University of Bonn contains many more images (scroll down). An imageless English translation of the press release is available here.
• GrrlScientist has the story over at Living the Scientific Life.
• And Octavio Mateus, one of the authors that described Europasaurus, has made the full article available for download from the comprehensive DinoData.net: Direct PDF link.

Happy reading!

—Matt Celeskey.

File under: Dinosaurs, Jurassic, Recent Discoveries.

Comments on record: (1)