Wednesday, December 31, 2014

Implications of Spare Parts

A fossil skeleton cast of the house-cat-sized mammal Didelphodon vorax graces the new exhibit at the Smithsonian’s National Museum of Natural History.

The subject of a previous post, this exhibit, titled The Last American Dinosaurs, offers a glimpse of the flora and fauna that were part of the complex ecosystem of the Hell Creek Formation during the last couple of million years of the Cretaceous period.  D. vorax, the largest mammal living at the time, was a member of the most diverse and abundant of mammalian groups in Late Cretaceous North America – the metatherians.  This group, originally named by Thomas Huxley in 1880, includes “all mammals more closely related to living marsupials (such as kangaroos and opossums) than to living placentals (such as humans and hedgehogs) and monotremes [egg-laying mammals].”  (Thomas E. Williamson et al., The Origin and Early Evolution of Metatherian Mammals:  The Cretaceous Record, ZooKeys, Volume 465, 2014, page 8.)

As I’ve learned about metatherians in the Cretaceous, some questions about this skeleton cast and, by extension, all such casts, have bubbled to the surface.  More on that in a bit.

Paleontologist Thomas Williamson and his colleagues have written a comprehensive article on metatherians in the Cretaceous (see citation above).  They consider in detail the evolutionary development of the metatherians, review the fossil record of these animals, describe the ecological environment within which they lived, and hypothesize about the impact of the extinction event at the end of the Cretaceous Period on these organisms.  Their treatment of these subjects is surprisingly accessible to a general reader, though, certainly, my interest in one of the metatherian species helps.

Blogger Brian Switek (Laelaps) posted a good overview of the main findings of Williamson’s piece.  Short of reading the article itself, his post serves nicely.

Perhaps the key aspect of the article for me, and the one that started me wondering about the Didelphodon skeleton cast, is how often Williamson and his colleagues come back to that fundamental reality for mammalian fossils in the Cretaceous and earlier, and for the metatherians in particular:  their fossils are few and very fragmentary, nearly always just isolated teeth and pieces of jaws.

Here are but two of the many instances in the article when they allude to the nature of these fossils:
Most Mesozoic mammal fossils consist of fragmentary jaws and teeth, which largely explains the intense emphasis that paleontologists place on the evolution of the mammalian dentition.  (p.6)
The crania and postcrania of Cretaceous metatherians are poorly known.  Most taxa are represented only by teeth and jaw fragments.  (p. 24)
Coincidentally, in recent weeks, I’ve been engrossed in paleontologist Michael Novacek’s memoir titled Time Traveler:  In Search of Dinosaurs and Ancient Mammals from Montana to Mongolia (2002).  It’s quite a rip-roaring account of one paleontologist’s life (including serious injuries and illnesses) in the field, from New Mexico to Montana to Chile to Mongolia – with some additional stops.  All in all, a great read.

At one juncture in a description of his paleontological work in the Gobi desert, Novacek observes that the truly important finds by the paleontologists there might not have been dinosaur remains, but small mammal skulls.
Some of these skulls were no bigger than an almond, but they were extremely important. Up to that time very little was known about mammals of the Cretaceous that lived alongside dinosaurs, and most of the evidence hailed from the American West, from Wyoming and Montana, where these tiny creatures left their remains as enigmatic fragments of teeth and jaws – as we call them, “spare parts.” (p. 283)
Spare parts!  What a brilliant characterization, but what a sobering one as well.  There is precious little beyond “spare parts” for Cretaceous metatherians.  Williamson et al. caution that in all of the research literature, “[o]nly two specimens that include articulated and partial skeletons have been described:  the stem marsupialiform Asiatherium from the Campanian of Mongolia and the early-diverging metatherian Sinodelphys from the Early Cretaceous (Barremian) of China.”  (p. 24)

The implications of this reality of metatherian fossils?  Paleontologists have to rely almost exclusively on metatherian teeth to glean insight into the shape, size, and life of these creatures.  Williamson and colleagues stress, “Reconstructing the postures, locomotor abilities, and habitat preferences of most fossil metatherians is exceedingly difficult, because most extinct taxa are known only from isolated jaws and teeth.”  (p. 45)

But wait, there’s a very specific implication of this dearth of articulated fossils . . . what about that beautiful Didelphodon skeleton cast?  Fabricated solely on the basis of just teeth and fragments of jaws?

Well, it turns out, no.  Some exploration of the information presented in the label for the skeleton cast makes that clear.

(Note that it's the cast that has had a Natural History Museum identification number assigned to it.)

The collector, Mike Triebold, is a commercial fossil hunter and preparator.  He is founder of the Rocky Mountain Dinosaur Resource Center and president of Triebold Paleontology, Incorporated.  According to the description on its website, TPI is a full-service company for vertebrate fossils, from collecting to preparing to mounting fossil skeletons.  I take it that the Smithsonian’s cast is one made by TPI, and based on a specimen found by Triebold and held in the RMDRC’s collection.

This specimen is more than teeth and jaw fragments.  RMDRC’s description states that the fossil discovered by Triebold constitutes some 30 percent of the entire skeleton.  In this context, that’s a remarkably large portion.  Indeed, RMDRC asserts, “This is the only North American mammal skeleton ever found from the late Cretaceous period.”

But, to date, as far as I can determine, this D. vorax skeleton has not yet been the subject of a formal published description.  Williamson et al., as quoted above, observe that there are only two articulated, partial, metatherian skeletons described in the literature, neither of them a Didelphodon.

So, the skeleton standing in the case at the Natural History Museum’s exhibit is based on a specimen that is about a third complete.  What are the implications of that?  For me, it reinforces my inclination to approach this cast, and, by extension, all casts, with a questioning attitude.  For this specific skeleton, I’d really like to know how the full skeleton was “fleshed out.”  Did the preparators have access to enough isolated fossil bones from other disparate D. vorax individuals to assemble an accurate full skeleton?  A reading of Williamson et al. would suggest they perhaps did not, but I don’t know.  To what extent did extinct or living analogues serve as models for the casts of specific bones missing from the fossil record?  I assume that occurred.  When did informed and educated assumptions about aspects of the skeleton enter into the equation?

Further, when the label (as in this case) for a skeleton cast identifies who collected the original specimen (a nice piece of information), would it not make sense to state what portion of the displayed skeleton is actually based on that fossil find?

These aren’t questions intended to convey skepticism of the validity and value of casts.  I embrace their use.  Rather, I am in search of a fuller picture of how any skeleton cast reflects the reality of its fossil record, particularly when that record is composed primarily of spare parts.


  1. This is a really interesting problem, and one that seems to particularly horrify museum workers in fields outside paleontology. Thing is, one of the core principles of vertebrate paleo is that all chordates (well, tetrapods mostly) are built on the same body plan, and most bones and body parts are directly comparable. If we have a recognizably mammalian jaw, we know it came from an animal with four limbs, a tail, and fur, at the very least.

    But even though we can be pretty confident in reconstructions based on near relatives from a scientific standpoint, the display context adds an additional wrinkle. Displaying an illustration or model is one thing, since it's clearly a reproduction, but people tend to assume mounted bones are real (why else would you display a creepy walking skeleton?). I wouldn't say displaying a skeleton that consists of 70% sculpted/hypothetical material is deceptive, but explaining it to the uninitiated can be...complicated.

    It also varies from case to case. I have no doubt that the C. megalodon at the Calvert Marine Museum is a perfectly reasonable reconstruction, for example, but then you have things like this Spinosaurus in Japan:
    It was pretty much entirely sculpted, and new fossils discovered in the past year show that it was pretty wide of the mark.

  2. Thanks for the thoughtful and thought-provoking comment.

    You're right that there are certain attributes of vertebrates (particularly bilateral symmetry, which I should have mentioned in the post) that facilitate building a reasonably valid skeleton cast without having an overwhelming percentage of the original fossil bones of the specimen. I certainly wasn't challenging the validity of skeleton casts and I hope I didn't come across as doing so. Rather, I was appealing for a bit more context for such museum displays, even if the cost is a bit more complication in the story visitors have to deal with. Sure, it might muddy things for museum visitors to learn what portion of a skeleton cast is based on the original find, but my sense is that it's a relevant piece of information, particularly when a collector is identified for a cast (a collector means a specific find). What did he or she actually find? How much of what we have before us reflects what was found? Such complication might be preferable to visitors coming away from a skeleton cast assuming that the cast in its entirety reflects the fossils the collector found.

    Thanks again,


Nature Blog Network