- Any ratios must be calculated for individual specimens, not from Frankensteinian averages of elements from different specimens. It is OK to average ratios across specimens, just not the raw measurements.
- Extremely young juveniles should be excluded, as they may differ in body proportions from adults of the same or closely related species.
With this in mind, I’ve started to go through the data set to flag obviously small juveniles. I would propose that we use body size rather than sexual maturity or LAGs as an indicator. Recent studies (e.g., this one on pachycephalosaurs by Horner and Goodwin) have shown that some near-adult size dinosaurs are probably not fully mature. I suggest that we assume that the limb proportions approximate those of adults sufficiently in these individuals (even if the skull morphology doesn’t). This will allow us to keep a few interesting taxa in. For instance, Fruitadens is not fully adult, but probably pretty close to it (according to histology; see the original paper). It’d be a shame to ignore the world’s smallest known ornithischian!
This is pretty straightforward for taxa known from only single specimens of obviously young juveniles (e.g., Avaceratops, Nipponosaurus, etc.). These taxa can be pretty safely removed.
What about animals like Stegosaurus armatus, where we have many specimens of various sizes, from tiny to gigantic? Here, I suggest cutting the specimens that are smaller than 2/3 the size of the largest specimen in one or more measurements (modifying a suggestion from William Miller). This 2/3 is a completely arbitrary cut-off, but I feel that it removes the smallest individuals while recognizing that indeterminate growth means adults can have a range of sizes.
As an example, the longest Stegosaurus femur, from YPM 1853, is 1,334 mm in length. Two-thirds of this is 889 mm, so any specimen with a femur smaller than this would be cut. This includes YPM 1856 (883 mm), DNM 2438 (308 mm), and many more. Of course, we can use our judgment for specimens close to the line. Because some specimens lack a femur, we would look at other elements (e.g., the humerus) to narrow the sample down more.
This elimination method will give us a rarefied data set, with the most egregious juveniles out of the picture. I suggest that we then use the average of the ratios from the trimmed sample for further analysis. In cases where we need actual bone lengths, I suggest using the largest specimen with sufficiently complete data necessary for the analysis.
For Everyone to Check
Following these criteria, I have posted a new Google spreadsheet here. It has two worksheets. The first, titled “To Analyze” (you can find the tabs along the bottom of the screen), has specimens that should be considered for removal marked with a greenish color. Do you agree? Disagree? See another specimen that should be removed? Let me know in the comments.
The second worksheet, titled “Deletion Candidates,” has specimens and taxa that I think should be excluded from the final analysis. The reasoning is given in the final column. Examples include, “can’t match with actual specimen” (for cases where it looks like derivative data given without citation that couldn’t be tied to a particular specimen), “non-diagnostic taxon of uncertain affinity” (should be obvious), etc. Once again, please check it over. Do you agree? Disagree? Say something in the comments here.
A nagging problem in our dataset has been the issue of the Sauropelta edwardsi specimen AMNH 3032. Its tibia seems way, way too short (even for ankylosaurs), and it’s thus a bizarro outlier.
So, I finally got my behind into gear to check out the problem. Was the measurement a typo? Was the tibia incompletely preserved? Was it from a smaller individual? On p. 115, I found the answer. The tibia for AMNH 3032 is here listed as 57.5 cm (575 mm) in length. This makes much, much more sense, so we’ll go with that number. Apparently, the value given in the table was simply an error. On further looking, I was able to confirm that the table heading for Sauropelta measurements on p. 120 was actually transposed with a Tenontosaurus table (as Rob Taylor had ably picked up on some time ago)! It looks like the measurements here are all from AMNH 3032, not YPM 5456 and 5459. Mystery (finally) totally solved, and the database is fixed!
It also turns out that the tabular error has been perpetuated across a few more publications. The Tenontosaurus-turned-Sauropelta YPM 5456 makes an appearance in Ford and Kirkland 2001 as well as Maidment et al. 2008. It in no way harms the conclusions of these papers (after all, the error is in the specimen number, not the measurements). But, I have removed all references for a YPM 5456 Sauropelta from our data tables.
Ford TL, Kirkland JI (2001) Carlsbad ankylosaur (Ornithischia, Ankylosauria): an ankylosaurid and not a nodosaurid. In: Carpenter K, editor. The Armored Dinosaurs. Bloomington: Indiana University Press. pp. 239–260.
Maidment SCR, Norman DB, Barrett PM, Upchurch P (2008) Systematics and phylogeny of Stegosauria (Dinosauria: Ornithischia). Journal of Systematic Palaeontology 6: 367-407.
Ostrom JH (1970) Stratigraphy and paleontology of the Cloverly Formation (Lower Cretaceous) of the Bighorn Basin area, Wyoming and Montana. Bulletin of the Peabody Museum of Natural History 35: 1–234.
Image Credit: Restoration by John Conway, from Wikimedia Commons. This file is licensed under the Creative Commons Attribution ShareAlike 3.0 License. In short: you are free to share and make derivative works of the file under the conditions that you appropriately attribute it, and that you distribute it only under a license identical to this one. Official license
As the past two posts have attested, we are on the cusp of doing some actual, real-life analysis. If at all possible, I want to run the analyses using freely available, open source software. Fortunately, all of the major software packages for phylogenetically-informed, quantitative analaysis are free! Here are the ones I’m looking at:
- For constructing trees and some simple analyses: Mesquite. (all of the tree plots you’ve seen have been done in this program)
- For general statistical analysis: R, along with the smatr, ape and ggplot2 packages. I will make a serious effort to post all scripts that are used to generate graphics and analyses; I would encourage all of you to do the same.
- For additional analysis of quantitative data in a phylogenetic context: COMPARE.
As a follow-up to our last post on thyreophorans, here are marginocephalians (ceratopsians and pachycephalosaurs) with some of their outgroups. Moving or removing the uncertainly-placed Stenopelix has little effect on character reconstruction.
Now that we’ve gotten a reasonable phylogeny hammered out, it’s time to start putting it to use! Just for fun, I used the (open source and free to download) program Mesquite to plot the femur:tibia ratio (as John Dziak had talked about not so long ago) as it changed from basal ornithischians up through Thyreophora, the clade including ankylosaurs and stegosaurs. The cladogram shown here only illustrates taxa for which we know the ratio, and species with multiple individuals had their data averaged.
Note that everything up through and including Scutellosaurus is presumed to be bipedal; after that, they’re quadrupedal. There’s a rather major change! Rather than the tibia being longer than or approximately equal in length to the femur, the femur ends up much longer than the tibia (and Stegosaurus is just insane in this respect)! Now the interpretation of this isn’t as easy as you might think. Is it due to being quadrupedal? Or is it due to being a big animal? The bipedal guys are all small, and the quadrupedal guys are all relatively large, so it’s tough to separate these two factors. Either way, we can say something interesting about locomotor evolution! And, note that the branches between nodes are reconstructed, with a reasonably wide error bar. So, please don’t consider them absolute truth.
And while I’m thinking of it. . .why didn’t theropods ever go quadrupedal? Some of them are as big as the largest quadrupedal ornithischians! Quadrupedal vs. bipedal locomotion can’t strictly be a size thing, then. Is there something about being a carnivorous dinosaur that discourages quadrupedalism? Perhaps the use of the forearms for prey acquisition? This is something we’ll want to touch on in the paper.
P.S.: The latest version of the combined spreadsheet is available here (Excel spreadsheet).
I think we’re now converging on a cladogram that reasonably captures the current state-of-the-art. Thank you to everyone who has contributed so far. The topology has incorporated most of the suggestions, and I think the nomenclature is finally up-to-date. As always, post something in the comments if you see anything that needs work.
The phylogeny, in general, is intended to be used for numerical analyses (e.g., phylogenetically independent contrasts) that incorporate information on the evolutionary relationships of organisms (that topic is crying out for a post, had I the time). Studies have shown that these methods are generally pretty robust, even if the phylogeny is incorrect in some details. So. . .it means that we shouldn’t sweat too much about uncertainties or the inevitable error/new data. In a handful of cases (hadrosaurs, perhaps?), we may want to run analyses with alternative trees, but we’ll worry about that when we get there. The position of some annoyingly fickle taxa (like Micropachycephalosaurus) will end up being a moot point, because we don’t have decent postcrania for them.
That said, this tree is a mutt (but a loveable one). We’ve done the best we can with the published analyses and our personal opinions. It’s pretty darned good for the most part (in my opinion), but it will never be perfect. Anyone is welcome to use it, but caveat emptor. If using the tree for other purposes, please refer to the notes given in previous posts so you have an understanding of the how’s and why’s of the tree. I am happy to send the NEXUS file to anyone who requests it (and we’ll want to include it as supplemental information when we publish).
The Current Version
The version below shows all of the taxa, with most corrections incorporated as suggested.
I’ve also generated a version where taxa that aren’t in our database are trimmed out. This is what we’ll use for most of our analyses. See below:
Notes from the previous version still apply.
Psittacosaurus xinjiangensis given as sister taxon to P. sinensis, following Sereno 1987, p. 267. This taxon, the measurements for which are based on a juvenile (Sereno 1988, original description), has not been included in any published phylogenetic analysis.
The species of Camptosaurus and Dryosaurus are placed so as to preserve monophyly for each genus.
Animantarx is given as sister to Edmontonia, following Hill et al. 2003, Figure 9
Both species of Postosuchus need to be shown
Edmontosaurus saskatchewanensis needs to be dealt with
Eoraptor needs to be added into the trimmed cladogram.
It’s been a long time since we’ve featured one of our project participants. . .so, here’s a chat with a relative newcomer to the ODP, David Dreisigmeyer.
Tell us a little bit about yourself. Where are you from? What do you do (professionally)? Any other interesting facts?
I’m originally from Pennsylvania, where I attended Juniata College for my undergraduate degree in Pre-law. After taking a year off, I relocated to Boulder, CO and then started grad school at Colorado State University (in Fort Collins, CO). After a year in the Economics Department I switched over to the Mathematics Department. I was first introduced to data/pattern analysis here while working with Michael Kirby. After my M.S., I moved over to the Electrical Engineering Department for my Ph.D., specializing in signal processing and mathematical physics.
After grad school, I moved back to the Math Department at CSU for a postdoc in image processing and dimensionality reduction. After this, we (myself, my wife Lisa and our son Dyson) moved to Los Alamos for a second postdoc. During this time I did signal and image processing and some protein modeling. Now we’re in Pennsylvania while I’m at Pitt on a visiting position in the Medical School and the Math Department. Currently, in addition to the ODP, I’m doing some modeling of the influenza virus (epidemiology and evolution), ecology and some fault detection methods. I’m also just starting to help out on a developmental aid database.
My wife and I are really into climbing. We loved New Mexico, especially the trail running (and seeing those bears nice and close up like). Ghost Ranch was absolutely amazing (the hiking, not those other things you may be thinking of). And if you have the chance to visit Chaco Canyon you should definitely do so. My other interest are the American Revolution and history (a la Susan Wise Bauer).
Why did you decide to participate in the ODP?
First of all I always wanted to be a paleontologist until I was about twelve. A couple months ago I decided to look back into it (after spending some time reading about the field again). But, how could someone like me actually do research in this field? A google search on ‘data analysis dinosaur’ led me to the ODP. This what I had been looking for. And it was open source — double bonus!
So far, what has been the best part of the ODP for you, and why?
Being able to deal with the data, both the analysis and the ‘cleaning’. Since I’m a data analyst I love… data. And, I can’t imagine neater data than dino data.
What have you learned from your participation in the ODP?
Working with paleontologists is a great experience. It’s awesome to learn about the questions the field is interested in examining.
What advice would you give those who might be interested in helping out with the ODP?
Check everything. In a database this large, put together ‘by hand’, there’s bound to be some errors, it’s unavoidable. By having everyone help everyone else out the final product is only improved.