Nedoceratops: To Be, or Not to Be?

Should Nedoceratops and Torosaurus be sunk into Triceratops? The debate continues, and it’s not just a bit of paleontological arcana

Feedloader (Clickability)

When the “Toroceratops” controversy broke in the summer of last year, I felt sorry for Nedoceratops. Hardly anyone said a word about this unusual horned dinosaur. Fans of Triceratops wept, wailed, and gnashed their teeth at their misapprehension that Museum of the Rockies paleontologists John Scannella and Jack Horner had exterminated the beloved horned dinosaur while paleontologists wondered if this dinosaurian mainstay of the Late Cretaceous could have grown into what had previously been called Torosaurus. But no one shed a tear at the proposition that Nedoceratops, too, might have been just a growth stage of Triceratops.

Known from a solitary skull on display at Smithsonian’s National Museum of Natural History, Nedoceratops has perplexed paleontologists since it was first described more than 100 years ago. The skull was found from the same end-Cretaceous strata that yielded Triceratops and Torosaurus, yet the dinosaur eventually labeled Nedoceratops was different from both. The skull had the general Triceratops-Torosaurus shape, but was distinguished by the lack of a nasal horn, a small opening in the preserved parietal portion of the frill, and two asymmetrical holes in the wing-shaped squamosal bones which made up the border of the frill. (These holes were thought to be old battle scars from some Cretaceous clash, but later studies showed these unusual perforations to be natural parts of the animal’s bone growth.) Scannella and Horner disagreed. Some of the unusual features, such as the apparent absence of a nasal horn, fell within the expected variation of Triceratops, and they interpreted the small hole in the parietal to be an early stage of the larger, rounded openings seen in the elongated frills of Torosaurus. Therefore, Scannella and Horner proposed, the Nedoceratops skull was a dinosaur virtually caught in the act of transitioning between the traditional Triceratops and Torosaurus forms, linking all three animals together into a single, late-life growth series.

A comparison of Triceratops (left) and Nedoceratops (right). From Scannella and Horner, 2011.

Andrew Farke, a ceratopsian expert at the Raymond M. Alf Museum of Paleontology, came to a different conclusion when he published a reanalysis of the Nedoceratops skull earlier this year. The mix of features exhibited by Nedoceratops distinguished the dinosaur from both Triceratops and Torosaurus, Farke argued, which would remove the form with transitional features from the growth series. More than that, Farke offered up additional criticisms of the growth series Scannella and Horner proposed—Torosaurus might not be sunk, after all.

Now Scannella and Horner have published a response to Farke’s response. To an outsider, this might look like an echo of the 19th century “Bone Wars,” when the cantankerous naturalists Edward Drinker Cope and Othniel Charles Marsh battled each other in print over the proper identification and interpretation of dinosaurian remains. The headline for LiveScience’s report on the new paper states that the “debate rages,” though the argument is probably better cast of a difference of opinion that has generated some friendly competition. Farke and Scannella are close colleagues, and as Farke mentioned in a behind-the-scenes post on his Nedoceratops work, the paleontologists have helped critique and strengthen each other’s arguments prior to publication. The paleontologists are not about to assault each other at the next Society of Vertebrate Paleontology meeting, either.

Despite the collegiality between the parties, however, Scannella and Horner object to Farke’s critique. For one thing, the Montana-based researchers argue, each of the seemingly unique features of Nedoceratops can be found within the variation of Triceratops (which they count as including Torosaurus-type animals). Though Triceratops is classically portrayed as being a “three-horned face,” when I asked about the apparently absent horn of Nedoceratops, Scannella pointed out that “there are many Triceratops specimens which show similar low, subtle nasal ornamentation—not quite to the degree seen in ‘Nedoceratops’ but certainly approaching that state.” Alternatively, the nasal horn of Nedoceratops might have been broken off or lost after death since the horn does not actually fuse to the nasal bones until late in life. At the moment no one knows for sure whether the horn was lost or was simply never there, but Scannella emphasizes that none of these scenarios hinders the idea that Nedoceratops might be better categorized as a Triceratops.

And that’s not all. Some of the features thought to mark the Nedoceratops skull as an old individual that had finished growing are ambiguous, Scannella and Horner say. The rough bone texture and fusion between certain skull bones—thought to be indicators of maturity, and even old age—are variable in Triceratops and don’t necessarily represent the age range of the animal accurately. They uphold their original interpretation of the dinosaur as a Triceratops, and I have to admit that I was amused that Scannella and Horner pointed out that Nedoceratops translates to “insufficient horned face” in their paper. Though this refers to the apparent lack of a nasal horn, there is a certain poetic justice to it in a paper which seeks to sink the name. “I think ‘insufficient horned face’ is a very appropriate name given that the genus likely represents variation within Triceratops,” Scannella said.

Scannella and Horner offer an explanation for the slit-like opening on one side of the specimen’s frill. (The completed Nedoceratops skull on display was partially reconstructed, so we don’t know for sure if there was a matching hole on the other half.) The projected sequence of transformation from Triceratops to a Torosaurus-type form predicts that there would be a stage in which the solid frill of Triceratops would develop depressions or holes that would eventually open to create large, circular fenestrae. Scannella explains the transformation happening like this:

As Triceratops matured, the parietal developed increasingly thin areas which eventually formed the holes previously thought to be characteristic of “Torosaurus.” If you take a typical Triceratops with a thick, solid frill and have it undergo this transformation to “Torosaurus,” there’s going to come a point where the parietal is going to begin to develop openings. These openings will likely start off rather small and continue to grow as resorption continues and the parietal expands. This is what we see in “Nedoceratops“—it’s a fairly mature specimen, the squamosals are slightly elongate (approaching the morphology observed in “Torosaurus“), and the parietal has a small opening in the same place where in Triceratops we see thinning occurring and in “Torosaurus” we see holes. So—one possibility is that this is a distinct genus of dinosaur that has tiny holes in its parietal. Another is that this is simply a Triceratops caught in the act of becoming “Torosaurus.” Jack and I favor the hypothesis that “Nedoceratops” is actually a transitional morphology, between Triceratops and “Torosaurus.”

One of the areas of debate has been the number of triangular, bony ornaments called epiossifications around the border of the Triceratops frill, which is composed of the parietal and squamosal bones. Previous studies have established that these bones start off being prominent, pointed ornaments, but as Triceratops aged these bones flattened out until they were barely visible. The question is whether the number of some of these epiossifications could change during growth, thus bridging the gap between the different number of these ornaments on the parietals of Triceratops and Torosaurus.

While Triceratops typically has five or six of these bones, called epiparietals, Torosaurus have been found with spots for 10 to 12, requiring the number to double if Scannella and Horner are right. This kind of addition has not been seen in well-sampled populations of horned dinosaurs before, but Scannella and Horner propose that such changes were indeed possible. As evidence, they cite a single epiossification marked by two peaks, which they hypothesize is an ornament in the process of splitting into two. Additional specimens will be needed to determine whether this double-peaked adornment truly was splitting during a transformative growth stage or is an unusual and unique variant. While Farke cautions that he hasn’t seen the specimen in question himself, he does offer an alternative interpretation. The double-peak shape “could also just be resorption of the tip without splitting a single element into two,” he says. “This is relatively common in ceratopsids—many of them tend to resorb the tips of the ‘high points’ on the skull, and that may be what is happening here.” If this is the case, then the epiossification would be part of the typical transformation into flatter adornments and not indicative of splitting.

This aspect of debate brings up the question of how useful epiossification counts might be for identifying distinct ceratopsids in the Hell Creek Formation. Individual variation, changes in growth and possibly even variation from one slice of time to the next might complicate matters. “As we are finding more and more Triceratops in the Hell Creek Formation of Montana,” Scannella says, “we are seeing specimens with quite a bit of variation in both the number and position of frill epiossifications–a finding which urges caution before considering epiossification number and position a set in stone indicator of taxonomic identity, at least in taxa closely related to Triceratops.” Farke takes a different view. “ almost certainly correct that there is stratigraphic variation in epiossification count (presumably related to evolutionary change in a lineage),” he says, but points out that “This would strengthen the argument that epiossification count has phylogenetic significance … f early Torosaurus have one count and late Torosaurus have another count, this would suggest that this trait changes through time and we can use epiossification count to distinguish different species.” Though all this argument over ceratopsid ornaments might seem esoteric, it is a key part of the discussion over what Nedoceratops and Torosaurus truly were. Did some ceratopsid dinosaurs add—and even double— frill ornaments as they matured? The answer to that question will have a major influence on the future of this debate.

What was Nedoceratops? That still depends on who you ask, and there is more than one possible answer. Farke, while noting that “Scannella and Horner raise some valid critiques of my diagnosis of Nedoceratops” in the new paper, still doesn’t see the dinosaur as an intermediate growth stage. “e do still disagree on the taxonomic relevance of things like the parietal fenestrae,” Farke says. “hey cite as transitional morphology between the Triceratops-morph and the Torosaurus morph of a single animal’s growth trajectory, whereas I would posit it as the end-member morphology for whatever Nedoceratops is.” And these aren’t the only options. “Of course, Nedoceratops might be an unusual or pathological individual of Triceratops. I’m not particularly married to any hypothesis at this point,” Farke says.

If Nedoceratops is an intermediate growth stage between the classic Triceratops and Torosaurus body types, further sampling of the Hell Creek and Lance Formations should eventually turn up still-growing Triceratops with similar features. Then again, if Nedoceratops is a distinct genus we would expect to eventually find juvenile individuals which share specific features with the single known skull to the exclusion of Triceratops and Torosaurus. Or maybe Nedoceratops is just a weirdo Triceratops.

This isn’t just a bit of paleontological arcana. The scientific conversation about Triceratops growth emphasizes the difficulties of recognizing prehistoric species and understanding their biology. What were once considered to be different species may just be growth stages or variants of one dinosaur, and these revisions affect our understanding of dinosaur evolution, biology, and ecology. I asked Scannella for his thoughts on the implications for his hypotheses, particularly given the fact that many dinosaurs are known from single, and often partial, specimens:

Increasingly, we are learning that many skeletal features in a wide variety of dinosaurs change throughout development.  There is also individual variation to consider. If all the differences between specimens are considered taxonomically informative, then it is easy to see how 16 species of Triceratops were named based on small differences in cranial morphology. Dinosaurs changed as they grew—and so, we need to evaluate which features are the most taxonomically informative. This can be hard to do if there is only one specimen of a particular dinosaur. We can start by examining developmental trends in dinosaurs thought to be closely related to that one specimen – as we’ve done with “Nedoceratops.” Examination of the bone microstructure is also important, in order to get an idea of relative maturity.

Paleontologists have recognized the problems of identifying slightly different specimens as new species before, but the debate over Triceratops—as well as Tyrannosaurus, Pachycephalosaurus, and other Hell Creek dinosaurs—has helped reinvigorate interest in how little dinosaurs grew up. Paleontologists are still in the relatively early phases of this investigation, and there are far more questions than there are definitive answers. The clues that will resolve the question of whether Triceratops was the lone ceratopsid of the Hell Creek still wait in museum collections and the expansive fossil graveyard that is the badlands.


Farke, A. (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

Scannella, J., & Horner, J. (2010). Torosaurus Marsh, 1891, is Triceratops Marsh, 1889 (Ceratopsidae: Chasmosaurinae): synonymy through ontogeny Journal of Vertebrate Paleontology, 30 (4), 1157-1168 DOI: 10.1080/02724634.2010.483632

Scannella, J., & Horner, J. (2011). ‘Nedoceratops’: An Example of a Transitional Morphology PLoS ONE, 6 (12) DOI: 10.1371/journal.pone.0028705

Get the latest Science stories in your inbox.