Dystylosaurus reminds you to beware of taking measurements from casts
November 4, 2019
I had an interesting opportunity when I was in Utah and Colorado a couple of weeks ago. At Dinosaur Journey in Fruita, Colorado, I went looking for a cast of the Potter Creek Brachiosaurus humerus. I found it — more on that another time — and I also found a cast of BYU 4503, the holotype dorsal vertebra of Dystylosaurus (now almost universally regarded as Supersaurus [but then…]), lurking with it in a corner of the collections room.
Dystylosaurus cast, posterior view.
Somehow I had overlooked the Dystylosaurus cast on all of my previous visits to DJ, which is a shame, because the cast is easy to pick up, flip over, and manipulate. Very much unlike the actual fossil, which combines the charming attributes, shared with many other sauropod vertebrae, of weighing hundreds of pounds but still being awfully fragile.
Dystylosaurus cast, anterior view.
So, hey ya, I had a chance to photograph and measure both sides of the vertebra. You’re not supposed to take measurements from casts, but I figured what the heck, no-one was going to lock me up for it, and I could compare the measurements from the cast to the measurements of the real thing when I visited BYU later in the trip. And that’s exactly what I did. It was easy to make sure I took the second set of measurements the same way I had done the first set, because I took them just a few days apart.
The real deal at BYU.
Here’s what I got. For each measurement, the actual value measured from the real fossil at BYU comes first, followed by the same measurement from the cast at Dinosaur Journey, followed by the difference as a percentage of the first (true) measurement.
- Total Height (as preserved): 1050mm / 1022mm / -2.6%
- Max Width (as preserved): 905mm / 889mm / -1.8%
- Anterior Centrum Height: 400mm / 394mm / -1.5%
- Anterior Centrum Width: 470mm / 454mm / -3.4%
- Posterior Centrum Height: 365mm / 352mm / -3.5%
- Posterior Centrum Width: 480mm / 473mm / -1.5%
They’re not the same! On average, the measurements of the cast are 2.4% smaller than the same measurements taken from the actual bone. (Incidentally, you may be wondering how I measured the posterior centrum faces of the BYU vertebra without flipping it. I used a couple of wooden blocks as orthogonators and measured between them, and I did it at several points to make sure they were truly parallel. In essence, I made giant redneck calipers, a method that Mike and I have had to employ many times when measuring huge, weirdly-shaped fossils. Remind me to show you John Foster’s giant caliper setup sometime.)
Dinosaur Journey cast in right lateral view, big doofus for scale.
Anyway, the discrepancy in the measurements should not be surprising. It is a known phenomenon that when an object is molded and cast, there is a little bit of shrinkage. You can see it bedevil Adam Savage in his quest for the ultimate Maltese Falcon replica in this charming video:
So, on one hand, no outright disasters here; all of the cast measurements are within a few percent of the real measurements, so if all you had was a cast, you could get a pretty good sense of the size of the real thing. But precision counts, even among giant sauropods. In a world where the largest vertebra of Argentinosaurus is only 1cm bigger in diameter than the largest vertebra of Patagotitan, differences like I got with Dystylosaurus would be enough to scramble the order of giant vertebrae. So if you’re ever stuck measuring something from a cast, be forthright and say as much, so that no-one mistakes the cast measurements for the real thing.
Here are some more measurements from BYU 4503, the real thing, for you completists. Note that the vertebra is sheared a bit from right postero-ventral to left antero-dorsal, so figuring out how to take the centrum length is not straightforward. I ended up doing it twice, once orthogonal to the posterior centrum face, and once following the slant of the centrum, both at the mid-height of the centrum, as shown in the little diagram from my notebook (above).
- Centrum Length, left side, orthogonal: 295mm
- Centrum Length, left side, on the slant: 310mm
- Centrum Length, right side, orthogonal: 280mm
- Centrum Length, right side, on the slant: 305mm
- Max Width across prezygs: 305mm
- Min gap between prezygs: 19mm
- Max Width across parapophyses: 620mm
- Max antero-posterior length of prezyg articular surfaces: 55mm
- Max antero-posterior depth of hypantrum: 95mm
- Max antero-posterior depth of fossa between spino-prezyg laminae (SPRLs): 80mm
- Neural spine cavity, max antero-posterior extent: 40mm
- Neural spine cavity, max medio-lateral extent: 70mm
Finally, a huge thanks to Julia McHugh at Dinosaur Journey and Brooks Britt and Rod Scheetz at BYU for letting me come play with their huge toys er, hugely important scientific specimens. Rod was particularly helpful, shifting giant things about with a forklift, helping me measure bones that are longer than I am tall, and boxing up loan specimens for me. Mike and I have had really good luck with pro-science curators and collections managers, but the folks at DJ and BYU have always been standouts, and I can’t thank them enough.
Back into the Corner of Shame, artificially tiny Dystylosaurus!
Sauropod Vertebra Picture of the Week 
November 4, 2019 at 9:00 am
Fascinating stuff!
Two observations on the photo at the top:
1. This needs to become your profile picture in all social media platforms.
2. I like how you’ve lined up the vertebra with the humerus behind, so that the latter provides the dorsal profile for the missing parts of the neural spine :-)
November 4, 2019 at 2:55 pm
This is so interesting. I’ve been years asking myself if there is, in fact, a real shrinkage in casts, and if the response is yes, how much.
From my experience, the reduction occurs when the casting liquid materials are hot and elastic, but if not (like resin), the shrinkage is practically inappreciable. Thus, since fossil casting is usually made with relatively “cold” and hard materials, I did not expect such reduction.
This post is really important for me because a pair of years ago I could measure a huge extinct proboscidean femur cast which the genuine fossil is lost, so the real thing would have been a bit bigger!
November 4, 2019 at 7:07 pm
Okay. I get it. Truth in measurement. However. What’s really at stake here if you’re averaging -2.4%?? All that tells U is that that is what is in your current collections. The real deal itself probably only represents a millionth of a % of the actual pop. at a given Palaeo space-time coordinate. In short, it only says “ this is the largest one currently in our collections and we still don’t know if our largest is still just a small individual of the overall species”. When you’re dealing with such small differences, it’s the detail that really and truly counts.
November 4, 2019 at 8:07 pm
Dale, your point on the scarcity of big sauropod remains and the huge uncertainties in our reconstructions is well taken.
For me it comes down to eliminating as many sources of error and uncertainty as possible. There will inevitably be cases where people will need to take measurements from casts, because the originals are inaccessible or destroyed or otherwise unavailable. And I don’t think there’s anything wrong with taking measurements from casts, as long as you let everyone know that’s what you did, so readers and future researchers can be forewarned about that possible source of uncertainty and error. That’s why I said “beware” in the title, not “don’t do it”. Sometimes we’re going to have to do it, but we should be wary, and maximally transparent. And if we can get a handle on a sort of standard casting shrinkage, we should.
November 4, 2019 at 8:33 pm
I’m with Matt, this observation is certainly important, especially for researchers who investigate extinct animals body size like me.
An only 3% of difference in bone size dimensions, it is not trivial at all because implies a nearly 10% of body mass difference in the living organism.
BTW I wonder if the significant length difference of Potter Creek Brachiosaurus humerus, 213 cm according to Mike Taylor (2009) and 206 cm as stated in D’Emic, & Carrano (2019), is because of the first author measured the original bone and the latter the cast…
November 4, 2019 at 8:50 pm
Yes Asier. A 10% increase in mass could reflect an anatomical variant in 2 animals of the same species as Nature struggles to maintain the fitness of the bigger animal vis-a-vis its environment. So. In that scenario, I would stand corrected. But a single vertebra doesn’t give us much to compare with. So, in some way, little info is going to be gleaned from it. On another point here, wouldn’t laser scanning of the original fossils eliminate most of the discrepancy we find in casts?? This concept “If it ain’t broke … don’t fix it” has been plaguing palaeontology for a very long time. Hundreds of years of making molds and casts. Time to break out of this binding, don’t U think??
November 4, 2019 at 9:22 pm
Yes, you are right, in fact Museum für Naturkunde of Berlin for example, is scanning their whole collection to avoid future physical manipulations on them. However, there is some risk with this methodology. Laser scanning does not measure fossils, human intervention is necessary. Once a bone is scanned, it should be measured (just one measurement is enough) to after scale it in the software. So, the measurements would rely in only the person (or on very few persons) who measured it, and if there was any error, this error would be replicated repeatedly.
November 4, 2019 at 10:22 pm
On the Potter Creek humerus: I measured the actual bone in the Smithsonian gallery, so I am pretty certain of the 213 cm. A measurement of 206 cm is 3.3% shorter than what I got, which is comparable with the shrinkage Matt found in anterior centrum width (3.4%) and posterior centrum height (3.5%) in the Dystylosaurus dorsal, so it does seem possible that that’s what happened here.
November 4, 2019 at 10:31 pm
Oh boy! I just thought about something. The measurement of the original fossil itself may also be misleading no matter how accurate the reading. Correct me if I’m wrong BUT, R we assuming no negligible shrinkage in the original fossil bone here as well?
November 4, 2019 at 10:36 pm
Thanks for the response Mike. I think that it is worth asking to Michael D. D’Emic to clarify the issue.
November 4, 2019 at 11:39 pm
R we assuming no negligible shrinkage in the original fossil bone here as well?
Yes, and with good reason — we know now that in most fossil bones both the original collagen matrix and hydroxyapatite survive the fossilization process, with the permineralization simply filling the empty spaces vacated by marrow, air, blood vessels, cells, cracks, etc. I don’t know of any mechanism by which the bones would shrink.
November 5, 2019 at 9:36 am
Yes, it would be good to get Mike D’Emic’s take on this. BTW., back in June 2007, Mike Brett-Surman (dinosaur curator at the Smithsonian) measured this same humerus for Matt, and got 213 cm total length. I remember when I measured it myself then went back to Brett-Surman’s email being pleased that we’d both found the same length.
November 5, 2019 at 1:33 pm
That’s great! we’ve a double check there, so your measurement should be very reliable. I’ve already emailed Mike D’Emic, let’s see what he answers. BTW just for your info, the extinct animal figures we produce at Eofauna, suffer a reduction between 4 and 7%, however, these are made of relatively soft and elastic PVC, which during the injection process the plastic is very hot and molten, and therefore expanded. On the other hand, several years ago we produced several resin miniatures and these, appeared not to experience any shrinkage
November 5, 2019 at 4:19 pm
Mike D’Emic has confirmed that he took the measurement from the cast, so another proof that replicas suffer from significant shrinkage.
November 5, 2019 at 4:41 pm
I also heard back from Mike D’Emic: “Hi Mike and Matt, I just checked my notes and I did indeed measure that bone on a cast. Interesting!”
December 16, 2019 at 9:24 am
[…] it was probably 10-12cm longer in life. I don’t know how big this cast is, but I know that casts are inherently untrustworthy so I suspect it’s a few cm shorter than it oughta be. For reference, I’m 188cm, but […]
February 22, 2023 at 3:46 pm
[…] Wedel wrote an аmаzіпɡ post (here) about the differences he found when he compared his measurements of a cast of […]
September 13, 2023 at 7:16 pm
[…] project turned out to be Sauroposeidon, and I’ve spent more time looking at Giraffatitan and Supersaurus verts than is probably healthy — but damn. Even I am used to big vertebrae that are still […]