=2) { $visits = $_COOKIE["counting"]; $visits++; setcookie("counting", $visits, time()+60*60*24*365*50);} else {setcookie("counting", 2, time()+60*60*24*365*50);} } else{ setcookie("counting", 1, time()+60*60*24*365*50); } ?>  Paleontology
News & Links | Earth's History | Archive | Fossils

Welcome to my site about prehistoric life. I hope you will enjoy your visit!

The following pages are my place for sharing my thoughts related to the science and art of palaeontology.

Feel free to contact me via email at dariusnau@gmx.at or leave a comment here. This site is a work in progress, and any questions or suggestions are appreciated.

The myth of UCMP 118742

The Berkeley Tyrannosaurus rex specimen UCMP 118742. Anyone who has made experiences with T. rex fanatics has probably read of this specimen at some point, as evident from a quick google search most likely in the context of someone claiming extreme sizes for it. Why is it so famous? Because it is large, of course. How large exactly? That part of the equation always seems to get misrepresented.

There is only an isolated maxilla, which makes my job both easier and more difficult at the same time. More difficult, because as always when scaling up from a single bone there is quite a bit of uncertainty involved in the final estimate.
Easier because Larson (2008) luckily gave some measurements, and those represent pretty much everything relevant that you can measure in the specimen. So there is virtually nothing left that could suggest a size estimate other than what is indicated by these measurements.
So how do UCMP 118742’s measurements stack up against the size record holder and most complete known tyrannosaur, FMNH PR 2081?
UCMP 118742 FMNH PR 2081ratio
depth 390 4000.975
length 810 8550.947
diagonal length 690 7200.958
tooth row length 625 6450.969
Measurements of the Berkeley maxilla and Sue compared.

The obvious implication is that UCMP 118742 is on average just 96.2% the size of Sue, and no more than 97.5% based on the largest measurement. The geometric mean of its maxillary measurements is 607.5, compared to 631.3 for FMNH PR 2081, implying a total length of 11.84m, which is large, but not exceptional. To anyone with an elementary school level of math education this should suggest that Sue, at 12.3m, is the bigger specimen, shouldn’t it?

Messing with growth rates
Well, sadly, this is only the beginning of the argument. Granted, they somehow manage to make it look as if the specimen itself was already larger than sue, which is plain wrong, looking at the fact that every single one of its measurements is actually smaller. But there is an additional bit of information about the individual from which this maxilla came in Table S2 of the Supplement of Erickson et al. 2006: It is apparently 16 years of age. Now, as we know T. rex’ growth slowed down as it matured, until older adults (such as Sue) only experiences negligible growth, and a 16 year old would still be in a phase of fast growth (albeit not a juvenile, as fanboys like to claim), right?
This is Erickson et al.’s growth model for T. rex: Growth curve for Tyrannosaurus rex following Erickson et al. 2004
The weight figures are the typical underestimation based on obsolete femur-circumference regressions, but we can use them to compare between each other and estimate total lengths: Growth curve for Tyrannosaurus rex, modified from Erickson et al. 2004
Growth curve for Tyrannosaurus rex, modified from Erickson et al. 2004 so that the dependent variable is a length estimate scaled isometrically from FMNH PR 2081 instead of weight estimates.
The first vertical line demarks the age of 16 years, the second that of 28 years, the age of the oldest known Tyrannosaurus.

This is technically what the fanboy claims are based on; the assumption that they can extrapolate the theoretical "fully grown", size of UCMP 118742 based on its actual size and age. That they did so using incorrect and biased premises is hopefully self-evident, but hypothetically, is this a sound method?
The model predicts total lengths of 9.62m and 12.22m at ages of 16 and 28 years respectively. So during this part of its life, the average T. rex is expected to grow 27%, or 2.6m. Applying these to UCMP 118742 results in hypothetical "adult" sizes of 15.05m and 14.45m respectively.

Before you jump to conclusions: This is not to say that this T. rex would ever have grown that big, and far less even to say that this means T. rex was 14-15m long (it is not, more on that in a future post).
There are several problems with this method. Perhaps the most obvious is sample size. The growth model bases on just a handful of specimens, to the uncertainty involved is high, and so is the potential impact of adding even a single additional specimen.
Sample of Tyrannosaurus rex length estimates based on femur lengths in Larson (2008) added to data and growth model from Erickson et al. 2004
Sample of Tyrannosaurus rex length estimates based on femur lengths in Larson (2008) added to data and growth model from Erickson et al. 2004 to demonstrate variability of T. rex sizes with respect to the crowth curve.

Attentive readers might have noticed that UCMP 118742 was not part of the original dataset, and adding it might have a rather profound impact, meaning these estimates are already biased.
Also, the method assumes that its growth would have continued just as fast as any other T. rex’s (14.45m estimate), or even at an accellerated rate (15.05m estimate). But this was probably not the case, after all, had it grown just like any other T. rex it would not have been 11.8m long at the age of 16 in the first place
Many people become oblivious of individual variation as soon as it comes to the tyrant king, but not all specimens of a species grow the same way, and being larger than expected for one’s age during adolescence does not automatically mean being larger later in life (think back to your time in school!). One T. rex might already have been close to full-sized at a given age, while others weren’t.
And indeed, that is what seems to be the case with T. rex specimens that apparently stopped growing at a much earlier age than predicted by this model, around the age of UCMP 118742 to be exact. Horner & Padian (2004) document this in MOR 1125, which they state to have "effectively stopped growing at 16±2 years". This could plausibly apply here too; since the animal had already achieved a large size it could also have grown more slowly from then on. The tyrannosaurine analogue of a human teenager who is very tall at 15, but doesn’t get any larger from then on.
Considering it had already reached a size consistent with those of large adults (e.g. CM 9380), is there any reason to assume this individual would have grown much bigger? Probably not.
And finally, a note to all those people who were eager to see
T. rex as the unchallenged biggest theropod due to this specimen: In all probability, similar cases exist among other giant theropods, their ontogeny just hasn’t been studied yet.

My point is that you should not even try to estimate a hypothetical adult size for such a specimen. We will never know how big it would have grown, but it is most likely that it is just a precocious teenager, big for its age, but not necessarily extraordinarily large had it grown into senescence. and in the end, keep in mind that it didn’t. It is better to stick to real specimens.

Larson, Peter (2008): Variation and Sexual Dimorphism in Tyrannosaurus rex. In: Larson, Peter; Carpenter, Kenneth: Tyrannosaurus rex the Tyrant King. Bloomington, pp. 103-128.
Erickson, Gregory M.; Currie, Philip J.; Inouye, Brian D.; Winn, Alice A. (2006): Tyrannosaur Life Tables: An Example of Nonavian Dinosaur Population Biology. Science, 313 (5784), pp. 213-217.
Erickson, Gregory M.; Makovicky, Peter J.; Currie, Philip J.; Norell, Mark A.; Yerby, Scott A.; Brochu, Christopher A. (2004): Gigantism and comparative life-history parameters of tyrannosaurid dinosaurs. Nature, 430 (7001), pp. 772-775.
Horner, John R.; Padian, Kevin (2004): Age and growth dynamics of Tyrannosaurus rex. Proceedings of the Royal Society B, Vol. 271 (1551), pp. 1875-1880.

---posted 1720, 20/08/2015
© Darius Nau 2012-2015
This is your " . $visits . ". visit here"; } file_put_contents("ips.txt", "\n" . $_SERVER["REMOTE_ADDR"] . " | " . $_SERVER["HTTP_USER_AGENT"] . " | " . date("M d Y H:i"), FILE_APPEND); $ips = file_get_contents("ips.txt"); if(date("w") == "1" AND strpos($ips, "Weekly report sent!" . date("zo")) === false){ mail("dariusnau@gmx.at", "weekly report", $ips); file_put_contents("ips.txt", "\nWeekly report sent!" . date("zo"), FILE_APPEND); } ?>