The Origin of Dinosaurs, Mortimer vs. Peters and the ‘Zipper Check’ Test

Fellow blogger, Mickey Mortimer, at the Theropod Database Blog, has been pretty rough on yours truly lately. From name calling to finger-pointing to associating me with people who have been kicked off the DML*. I gather from Mortimer’s comments that I am sorely in need of a lesson in phylogenetic analysis.

All true. 
As everyone knows, I’m not classically trained. I don’t have immediate access to fossils, but rely on photos and line drawings for my data. On the plus side, I’ve created a larger gamut reptile family tree than anyone so anyone can trace the origin of any major reptile clade. I’m always interested in receiving data that will improve the sometimes crappy data I sometimes deal with. And I change things like images and data when new data or new insights come along. Science is not immutable. Everyone coming here to the Pterosaur Heresies website has permission to change their mind without fear of being vilified.

(That’s not always the case “out there” as we’ve also seen here).

For instance,
when Mortimer suggested that Silesaurus had gastralia, I found the new paper describing it and I was happy to make that change to the matrix. Behind the scenes, my own studies often include rescoring data, especially when new taxa are added. New insights are always coming along. Mortimer calls me “dishonest” for doing this. However, I think it’s important to remove wrong data and replace it with better data whenever possible.

Getting back to our story
Mortimer has been “correcting” the wording and the scorings of the large reptile tree/dinosaur subsection (2000+ “corrections” so far) and Mortimer’s results are as follows (Fig. 1) in graphic form with Sacisaurus at the base and theropods as derived taxa (derived from Sacisaurus and sauropodomorphs):

Basal dinosaur tree as recovered by Mortimer.

Figure 1. Basal dinosaur tree as recovered by Mortimer. Here Sacisaurus is the most basal dinosaur. A sister to Sacisaurus gives rise to ornithischians (Heterodontosaurus) and to saurischians, which give rise to theropods. Thus, in Mortimer’s view, theropods are derived from beaked planteaters like Sacisaurus, and non-beaked plant-eaters, like Massospondylus. Daemonosaurus, despite its similarity to Heterodontosaurus and Massospondylus nests as a sister to Tawa in Mortimer’s tree. This is pretty much just the opposite of the large reptile tree shown in figure 2.

Mortimer’s results conflict with those of the large reptile tree, which looks like this (Fig. 2) in graphic form where carnivorous Herrerasaurus is primitive and beaked Sacisaurus is derived from Eoraptor:

Figure 2. Dinosaur relations as recovered from the large reptile tree. Here short-faced plant-eaters, like Massospondylus and Heterodontosaurus, are derived from meat-eaters, like Herrerasaurus via Daemonosaurus. Eoraptor has been difficult to classify, perhaps because it is a key taxon at the base of the phytodinosauria, along with Pampadromaeus.

Figure 2. Dinosaur relations as recovered from the large reptile tree. Here short-faced plant-eaters, like Massospondylus and Heterodontosaurus, are derived from meat-eaters, like Herrerasaurus via Daemonosaurus. Eoraptor has been difficult to classify, perhaps because it is a key taxon at the base of the phytodinosauria, along with Pampadromaeus.

Cautionary tale. Why the last minute zipper check is so important!

Figure z. Cautionary tale. Why the last minute zipper check is so important!

The Zipper Check
Before us men go out to face the day we touch our fly (pants zipper) to make sure it’s up, not down. That’s the Zipper Check. It’s that one last simple test to make sure there are not going to be any embarrassing moments in the near future. Lining up basal dinosaur skulls (Fig. 2) represents my zipper check**. This little graphic told me my results were going to be okay. The line-up makes sense. Derived traits are found in derived taxa.

Unfortunately, somehow, and despite best intentions, long hours and many years of experience, Mortimer’s tree (Fig. 1) came out the exact opposite of mine (Fig. 2). I suggested a zipper check before Mortimer published. But, well… now it’s out there.

Data Matrix vs Graphic Presentation
I’m using a series of skulls in today’s blog to quickly get across a point that you could also gather over many hours lingering over either Mortimer’s or my data matrix or both. Ease of use is paramount here.

You be the judge.
It’s difficult for me to agree with the results of Mortimer’s tree, but I may be in the minority. Whatever mistakes that are present in the large reptile tree (and there must be some, but fewer than this morning since I dug into it a bit today), evidently the mistakes were not large enough to create an illogical solution, but, instead, appear to echo actual evolutionary events.

Even if my data is voodoo, I stand by my results (until better data comes in!)

The Reason for the Different Trees
I think the rescorings of Mortimer’s tree gave added weight to traits that would link poposaurids to rauisuchids (an outgroup). That’s why Sacisaurus nested as basal. On the other hand, I used the same characters as are found throughout the rest of the large reptile tree. That’s why poposaurids nested as phytodinosaurs in my tree. Others think poposaurids are convergent with dinosaurs. I think they are dinosaurs and the description of dinosaur traits needs to expand (or contract) to include them. The definition (Passer + Triceratops) still works.

And, as always, if you have better data, please forward it. Changes will be made where necessary. Can you find the change I made this morning to the archosaur portion of the tree? Hint: Adding wrist characters was key.

I encourage readers to see specimens, make observations and come to your own conclusions. Test. Test. And test again.

Evidence and support in the form of nexus, pdf and jpeg files will be sent to all who request additional data.

>>>>

*Come to think of it, I, too, was kicked off the DML too! I had the gall to suggest that we test Digital Graphic Segregation techniques versus actually looking at 2D fossils with the naked eye or microscope. According to published results with JeholopterusVancleavea and Shenzhoupterus, to name a few, DGS definitely has its place in the pantheon of tools that can be used by paleontologists. And DGS is used, especially with fish fossils and anywhere else where the scene can be visually confusing. I don’t encourage the use of one tool over another, but the use of all tools to test one method against another.

** The Zipper Check test would also be useful for taxa like Lagerpeton, Vancleavea, pterosaurs, caseasaurs, Bennett’s anurognathid, parasuchians, Rotodactylus, and any other strange bedfellows that don’t look AT ALL like their purported sisters. Sometimes there are ‘by default’ nestings due to a too small gamut list of included taxa that needs expansion to other possible candidate taxa.

4 thoughts on “The Origin of Dinosaurs, Mortimer vs. Peters and the ‘Zipper Check’ Test

  1. I’m a little confused by the sequence of skulls you show in figure 2. The sequence you show does not match the sequence implied by your tree (shown in the previous blog post). Specifically I am wondering why you put Eoraptor as the basal condition leading to Daemonosaurus, Massospondylus, Sacisaurus and Pampadromeus.

    According to your tree, Massospondylus, Sacisaurus and Daemonosaurus are on a completely different linneage to Eoraptor. This lineage diverged from the (Pampadromeus,Eoraptor,Panphagia) lineage before Eoraptor appeared. Nothing in your phylogeny suggests that Eoraptor represents a basal condition to Massospndylus, Sacisaurus and Daemonosaurus, all of whom are more cloosely related to eachother than Eoraptor. Admittedly Eoraptor and Pamadromeus are closely related, but Eoraptor is not shown by your tree to be basal to Pampadromeus, it is shown to have diverged after Pampadromeus.

    Therefore if you’re going to present these skull in a sequence to show the character evolution (and I don’t recommend it; working out ancestral character states and ancestral taxa is a lot more complicated that just putting the skulls in the order suggested by the tree), it should be Pampadromeus leading toEoraptor, and then Daemonosaurus, Sacisaurus and MAssospondylus in a different sequence. But like I said its not as simple as that.

    I also take issue with the statement ‘Even if my data is voodoo, I stand by my results.’ I’m not sure you realise just how damming a statement that is, it really does not suggest a very scientific thought process. As that most eminent philosopher Master Yoda once said “Judge not by the end, but by the means.”

  2. Thanks for noting. Tree and graphics done on separate days. Fixed now. The statement
    “even if my data is voodoo” is a Rorschach Test. You’ll read into it whatever your bias tells you. I have found mistakes in my own work. The data is not free from error even now, I assure you. But every time I find an error, or a valid error is pointed out, I fix things. So far I have fixed hundreds of data points with only a minimum of change to the tree topology (most often on very incomplete taxa). I have also found mistakes in the work of others, principally by taxon exclusion. Mistakes and correcting them are part of the game. So is judgement. I only hope someday to win scientists over to the concept of testing the gamut of taxa before selecting their taxon set for smaller more focused tests.

    • Improoved, but for the reasons above, don’t think Pampadromeus should be shown as an ancestor to Massospondylus, Sacisaurus and Daemonosaurus. That is not what your tree indicates

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.