Nedoceratops - Random Thoughts

The last two posts here have focused on my most recently published paper, fully describing the skull of the horned dinosaur known as Nedoceratops hatcheri and critiquing the hypothesis that it, along with Torosaurus latus, is simply an older individual of what we call Triceratops. Because I've already talked about the science of the paper, and some collegial interactions, I'm going to spend this final post in the series talking about a few odds-and-ends that just didn't fit anywhere else. Most of these are little windows into the process behind the paper - from writing to review to revision. And we'll start with. . .

Open Source Composition

I'm proud to say that every single step of the authoring process for my paper happened in open source software. I wrote the manuscript in OpenOffice.org Writer, formatted most of the references in Zotero, did initial image editing (contrast adjustment and background removal) in GIMP, assembled the figures in Inkscape, and submitted the manuscript through the journal website on the browser Firefox, all of which were running on various releases of Ubuntu. Score one for open source software (and open access publishing)!

On Organizing the Paper

I'm under no illusion that everyone (?anyone?) will agree with my conclusion that Nedoceratops is a valid taxon. In fact, I'm quite accepting of the possibility that I may be wrong. But even if this is the case, I still want my paper to be useful. So, I made my best effort to separate data from interpretation in the description section of the paper. Of course, I couldn't be completely successful on this point - after all, I had to compare Nedoceratops with Triceratops and Torosaurus (the most likely candidates for synonymy) - but I like to think that I mostly achieved my goal. If nothing else, I have pretty pictures. And. . .

Speaking of Pictures

My figures went through some pretty drastic changes during the evolution of this paper. In the first round of reviews, it was pointed out that in the text I kept referring to various structures illustrated in the figures, but only a ceratopsian geek could figure out what I was talking about.

For example, we have this lovely sentence:

The narial strut is inclined rostrally towards the dorsal end of the element, and enough original bone surface is preserved to indicate that a posterior internarial flange did not project from the caudal surface of this structure (Figure 4).

My original Figure 4 looked like this:

The photo is relatively pretty, but only a die-hard ceratopsian nerd could locate the narial strut or know where to look for a posterior internarial flange if such a thing even existed in this animal. So, for my next iteration I added some labeling:

Of course, all of the abbreviations are explained in the caption (not shown here). "ns" refers to the narial strut I was talking about above. Finally, the editor mentioned that I should do a better job of indicating the "cpf" (canal at the edge of the premaxillary fossa). It wasn't just at the tip of the arrow, but over a somewhat broader area. Thus, that brings us to version 3:

This, with the extra arrows showing the position of the canal, was the version that appeared in the paper.

If I learned anything from this experience, it was about the importance of good labeling and interpretive drawings for non-expert readers. Most of the labeled interpretive drawings alongside photographs (with the exception of parts B and D in Figure 1) were added at the direct request of the editor. Looking at the end product, this addition was a major improvement to the paper. Of course, I must also admit that having relatively unlimited space in an online journal allows this luxury!

Editorial Ethics

It's probably not a surprise to many of you that I am a volunteer academic editor at PLoS ONE. And those of you who have been paying attention probably noticed that the Nedoceratops paper was just published in that very same journal. This sounds pretty problematic on the face of it.

Thankfully, PLoS ONE has pretty strict editorial controls when one of their own editors submits a paper (in addition to a competing interests policy that covers this and similar situations). My experience as a submitting author was exactly the same as for any other author. Once the "submit" button was pressed, I had to wait just like everyone else. I couldn't control which editors handled it, who reviewed it, or even have a sneak peek at the reviews on-line. In other words, the system functioned exactly as it should.

My authorial feet were held to the fire by Leon Claessens, the handling editor for my submission. Leon, in my opinion, did a very professional job and didn't let me get away with anything (even sending the manuscript back to me a second time for a few last corrections and improvements). The reviewers - Michael Ryan and Peter Dodson - also did their jobs (in my opinion). And, as mentioned in my last post, comments by John Scannella and Jack Horner offered additional constructive feedback.

One thing I really like about PLoS ONE is that my competing interest - as an editor at the journal - is stated up-front in the paper. Although it's somewhat scary seeing it there, I think such notices are certainly appropriate.

Final Thoughts

It's nice to finally have this paper out there - these ideas have been floating around in my head for awhile, and I've always had a secret desire to be the person to describe Nedoceratops. I'm relatively pleased with the final product (of course, there are always one or two typos that slip through, and why couldn't some of the figures in the PDF have been bigger?), and look forward to the discussion that this paper generates. Thank you to all who helped out (see the acknowledgments for a comprehensive list)!

Nedoceratops - Fun with Science

In my last post, I mentioned that I have had more fun with my recent project on Nedoceratops than anything else I've done lately. Just as a refresher, this paper described an oft-neglected horned dinosaur skull known as Nedoceratops hatcheri, and presented counter-evidence to a hypothesis (published by John Scannella and Jack Horner) that Nedoceratops, Triceratops, and Torosaurus were all the same animal. This all sounds kinda boring and academic, so where's the fun in that?

As a scientist in a small field like vertebrate paleontology, it can be awkward when you disagree with a colleague. I've heard third-hand accounts of shouting matches at scientific conferences, and have occasionally seen very heated discussions during the Q&A time at presentations. Thankfully, this sort of behavior is pretty rare. Yet, I was a little worried about what might happen when I publicly presented a counter-argument to Scannella and Horner's hypothesis.

My nightmare of a confrontation with John Scannella (left) at SVP. Maybe there wouldn't be fists involved, but at a minimum a wrestling match. He's in better shape than I am, so I would be in trouble.

Of course, I was less concerned about a shouting match, and more focused on not being a jerk in print. I've known both John Scannella and Jack Horner for a number of years, and wanted to stay on at least semi-cordial terms with them. Of course I was going to disagree with them (based on my interpretations of the available data), but I wanted to do so in a way that was fair, collegial, and honest.

So, I did something that some people might consider stupid. I sent John and Jack a copy of my unpublished, unaccepted, in-review manuscript. At the very least, I figured it was only fair that I should give them a heads-up that I would be presenting counter-arguments to their hypothesis about Triceratops and Torosaurus. More importantly, I wanted to make sure that I was representing their work fairly and accurately.

This initiated a lengthy and wide-ranging email conversation. Although I had done a decent job of representing most of their points, there were a few areas where I had inadvertently set up a straw man. I fixed those as best I could. In some (thankfully minor) areas, I needed to update information or account for some specimens I had neglected. For instance, I had grossly understated the amount of variation in the frills of various adult Triceratops (something that has only been adequately documented thanks to John's Ph.D. work). With their honest feedback, I was able to craft a much-improved version of my manuscript. It is not just a smarmy platitude to state that I genuinely appreciate their input.

Of course, I won't claim that Scannella and Horner find my counter-arguments (that Torosaurus and Triceratops are different animals) entirely convincing. And, my current opinion on the matter is not unchangeable. John and I had a nice long chat at the SVP meeting this year, comparing notes and talking about our future research plans. He has some really cool data (some of which he has presented at SVP and other conferences), and I look forward to seeing it in print. Undoubtedly, we will both modify our interpretations as new data are published.

Now why am I finding this to be so enjoyable? It's the joy of discovery, the entertainment of questioning long-held ideas (especially my own), and the pursuit of new data. After all, science shouldn't be about scoring rhetorical points, but working towards an accurate view of our world. I know beyond a doubt that we all are playing on the same team. My dialog with John (and Jack) has been engaging, challenging, and stimulating in a unique way. I've learned more about Triceratops in the past six months than I had in the past six years (to be ultra-nerdy, for instance, some specimens lack the mid-line epiparietal - neat!!!). It's just darned fun to be working on a research problem like this!

No matter how this whole issue shakes out, I think there is one thing we can all agree on right now. Horned dinosaurs are AWESOME!

John Scannella and I, with our favorite dinosaur

Coming Up: A few final thoughts on the process behind this paper.

Nedoceratops - A Full Description at Last

Every group of animals has at least one notable yet neglected specimen. In horned dinosaurs, a particular example is a large skull at the Smithsonian discovered in Wyoming during the closing years of the 19th century. Unfortunately, this specimen has suffered a twisted and sometimes tragic history.

The skull of Nedoceratops hatcheri, modified from Farke 2011

The collector of the fossil, John Bell Hatcher, wrote a paper about the specimen, but died before he could publish it. So, the task fell upon Yale's Richard Lull, who gave this nearly complete skull the name of Diceratops hatcheri. It looked much like a Triceratops (the famous three-horned face), but differed from the standard "Trike" in having a tiny nose horn, several holes in the frill, and a handful of other characteristics. Later on, other scientists decided that these differences were probably just the result of individual variation, injury, or other illness. So, Diceratops became just another Triceratops to most workers (a 1986 paper by John Ostrom and Peter Wellnhofer was influential in this regard). Still, there wasn't unanimity in that thought - Cathy Forster, for one, published the opinion (in 1996) that Diceratops was indeed distinct from Triceratops.

Nedoceratops hatcheri, as restored by Nobu Tamura.

In 2000, the skull (which was on exhibit at the Smithsonian) was damaged when some rowdy museum visitors crashed through a barricade and broke the snout. Fortunately, the museum's preparators were able to fix it. As if to add insult to injury, it turned out that the name Diceratops wasn't unique. A German entomologist (coincidentally named Förster) had applied the name to an insect way back in 1868, so a new name had to be found for the dinosaur. Unfortunately, this didn't happen in the most organized way. Two researchers independently published the replacement names of Diceratus (in 2008) and Nedoceratops (in 2007). The second one, although less elegant (in my opinion), had priority because it was published first.

But wait - there's more! The story of Nedoceratops took an interesting twist last year, when John Scannella and Jack Horner suggested that it represented a life stage of Triceratops, halfway through its transformation into Torosaurus (see the figure below). This was not an evolutionary transformation, of course, but ontogenetic (one that happened as an individual animal got older). So, our three animals - Triceratops, Nedoceratops, and Torosaurus - were all just the same thing! Such revelations happen frequently in paleontology. For instance, the duck-billed dinosaur Procheneosaurus turned out to be young Corythosaurus, Lambeosaurus, Hypacrosaurus, and the like. But, not all horned dinosaur experts are convinced that this was what was going on with Nedoceratops and Torosaurus.

From left, life restorations of Triceratops, Nedoceratops, and Torosaurus (all modified after originals by Nobu Tamura). The arrows indicate the relative age of each animal, as proposed by Scannella and Horner. If they all are the same thing, Triceratops is the "young" life stage, and Torosaurus is the "old" life stage, with Nedoceratops being a transitional form. The big question: are these the same animal, or different species?

Named, renamed, renamed again, broken, pieced together, and declared invalid, Nedoceratops has had a checkered past. Yet, the skull has never received a fair treatment in the scientific literature. I'm not just talking about people's opinions of the specimen. Instead, I'm talking about a full description.

Descriptions - detailed accounts of a specimen's characteristics - are the data upon which much of paleontology relies. But, the skull of Nedoceratops was never fully described. A few paragraphs have been written about it here or there, but it turns out that many aspects of these were inaccurate or incomplete. Given the controversy over this skull, an accurate and complete description of the animal was particularly important. So, I set out to fix the situation. In my recent PLoS ONE paper, I published the first comprehensive description and illustration of Nedoceratops hatcheri.

At risk of boring you readers with endless details, I'll just mention a few minor points. For instance, it turns out that many of the early drawings of the specimen were inaccurate (missing bone was shown as present, for instance). I was able to correct these errors, and talk about areas of the skull that were well-preserved but never discussed in the literature before. My paper also includes detailed and never-before-published photographs of the skull in various views, which I hope will be useful for folks who can't see the skull first-hand.

Finally, and probably of the broadest interest, I go out on a limb and say that Nedoceratops hatcheri is a unique species - not the same as Triceratops or Torosaurus. In my opinion (and it is but an opinion), there are just too many features that are different between these animals, and few features can be chalked up to injury or growth changes. Will this opinion stand the test of time? Maybe, maybe not. My opinion on the validity of Nedoceratops is probably the most tentative conclusion I've ever published, so my feelings won't be terribly hurt if I turn out to be wrong (although of course, I'd rather be right).

And what about the idea of Triceratops being a junior version of Torosaurus? I argue that Torosaurus and Triceratops are indeed distinct species, not just old and young versions of the same animal. Why is this?

• Triceratops and Torosaurus have vastly different numbers of bony bumps - called epiparietals and episquamosals - on the edges of their frills. If Torosaurus is the younger version of Triceratops, this means that Triceratops added a bunch of these bumps to the frill during growth. Yet, there is no good evidence that any other horned dinosaur did this.
• Triceratops has a solid frill, and Torosaurus has big holes in its frill. In all other horned dinosaurs we know (such as Protoceratops and Centrosaurus), if adults have holes, the young ones have holes. Thus, it doesn't make a lot of sense that Triceratops/Torosaurus would only add these holes when it got really big. [of course, I will admit that just because something doesn't make sense doesn't mean it couldn't happen - just that it is much less likely]
• It was previously claimed that there were no good examples of "young" Torosaurus. But, a skull at Yale (collected by Hatcher, the same person who discovered the Nedoceratops skull) fits all of the characteristics of a young animal. Its skull sutures are all open, or unfused, and the bone has a smooth texture typical of young dinosaurs. In my mind, this is probably the best evidence that Torosaurus is not a grown-up Triceratops.

Undoubtedly, many other paleontologists will have something to say about these issues. Some will agree, some will disagree, some will show parts of my paper are incorrect, and others will present more supporting data (at least I hope, on all counts). I suspect the next few years will feature much, much more discussion on these fascinating horned dinosaurs!

Coming Up: It is safe to say that I have had more fun with this project than with anything else I've done recently. Why is that? In part, it's been due to some very stimulating discussions with John Scannella and Jack Horner, who recently published the "Toroceratops" hypothesis. See my next post for more!

Citations

Farke, AA (2011) Anatomy and taxonomic status of the chasmosaurine ceratopsid Nedoceratops hatcheri from the Upper Cretaceous Lance Formation of Wyoming, U.S.A. PLoS ONE, 6 (1) DOI: 10.1371/journal.pone.0016196

Forster CA (1996) Species resolution in Triceratops: cladistic and morphometric approaches. J Vertebr Paleontol 16: 259–270.

Förster A (1869) Synopsis der Familien und Gattungen der Ichneumonen. Verhandlungen des Naturhistorischen Vereins der Preussischen Rheinlande und Westfalens 25: 135–221.

Hatcher JB (1905) Two new Ceratopsia from the Laramie of Converse County, Wyoming. Am J Sci, series 4 20: 413–422.

Mateus O (2008) Two ornithischian dinosaurs renamed: Microceratops Bohlin, 1953 and Diceratops Lull, 1905. J Paleontol 82: 423.

Ostrom JH, Wellnhofer P (1986) The Munich specimen of Triceratops with a revision of the genus. Zitteliana 14: 111–158.

Scannella JB, Horner JH (2010) Torosaurus Marsh, 1891, is Triceratops Marsh, 1889 (Ceratopsidae: Chasmosaurinae): synonymy through ontogeny. J Vertebr Paleontol 30: 1157–1168.

Ukrainsky AS (2007) A new replacement name for Diceratops Lull, 1905 (Reptilia: Ornithischia: Ceratopsidae). Zoosystematica Rossica 16: 292.