Scare Quotes and Recent Ceratopsians

Scare quotes are a figurative use of quotation marks to imply the object(s) being enclosed (be it a theory, phrase, or word) are not proscribed by the quoting author(s) [n1]. In this case, they are used around taxonomic names that the authors (Longrich, 2010 [1] & Scannella and Horner, 2010 [2]) regard as unusable. There’s a further trend, in that some authors place these quotes to imply the taxa used are nomina dubia, or otherwise unavailable taxonomically, i.e., nomina nuda or some other type of informality. In which case, the usage is confusing.

Use of the Scare Quotes

1. Longrich (2010) uses quotes whenever referring to Chasmosaurus kaiseni [3] or Eoceratops canadensis [4], and stresses use of one of these taxa and “not valid.” This implies the same use for the second, from page 681, and that the “not valid” phrase would be to deem the taxa nomina dubia.

2. Scannella and Horner (2010) on the other hand also use quotes when using Torosaurus, Nedoceratops and Triceratops hatcheri. The only time these quotes aren’t being used is in the “synonyms” subsection of the Systematic Paleontology section.

Between these two papers, the use tends to denote a “doubtful name,” and seemingly more explicitly an “invalid” on in the first paper. The meaning behind these phrases is lost when the authors refer them to other taxa. The rules for nomina dubia are, generally, provided by proxy of traditional methods, once but no longer encoded by the ICZN, which I discussed here. Practical use of the “rules” of nomina dubia, as employed by various workers, implies that they cannot (or rather, should not) compete for taxonomy.

Use of the Nomina Dubia or Quoted Taxa

Further, the typical practice for rendering a nomen dubium as such is that the material the taxon is based on is undiagnostic or too incomplete to reliably refer to a known species, or lacks apomorphies to either relate or separate it from one species or another. These features are used in diagnoses to “discriminate” taxa, and without them, one cannot typically find out what said taxon is.

1a. When Longrich takes the potentially synonymous Chasmosaurus kaiseni, calls into question its diagnostic utility by pointing out how little of the parietal there is, and then “provisionally” synonymizes it, he creates a taxonomic quandary. This is because for Chasmosaurus kaiseni to be synonymous with Mojoceratops perifania, it must possess the same diagnostic features in the latter; and according to Longrich, it doesn’t [1]. If Chasmosaurus kaiseni were synonymous with more diagnostic material (i.e., had potentially one or two of the three autapomorphies Longrich uses to diagnose Mojoceratops perifania), it would be possible to refer it to the same taxon, but based solely on the work of Longrich, this is not possible; and if it were, the taxon name would need to be changed to Mojoceratops kaiseni with perifania as a synonym.

1b. When Longrich looks at Eoceratops canadensis, the taxon is rendered potentially synonymous, but as a juvenile is certainly undiagnostic of adult morphology (see also [2]) which all other taxa must be compared to. Eoceratops canadensis [4] is a nomen dubium, and like Chasmosaurus kaiseni would not be available for synonymy, and as such cannot even be referred to another taxon; however, while Longrich could suggest they are the same taxon, there would be very little way to prove this. Looking at the Brachyceratops, Eoceratops and known Triceratops juveniles, they are all mostly indistinguishable, and preserve the same shapes of premaxillary, circumorbital, and squamosal regions. So while it is plausible to claim Eoceratops canadensis as a synonym, if one were, it would be as a senior subjective synonym: Mojoceratops is a synonym of Eoceratops, and perifania is a synonym of canadensis. But this is not necessarily the case: it simply cannot be referred without having the same diagnostic features, and they do not exist [1].

2a. By the same token, the types of two of the Torosaurus species seem to belong to two very different animals: Torosaurus latus appears to have no autapomorphic features, and is virtually identical to Triceratops spp.; on the other hand, Torosaurus utahensis still can’t seem to find a perfect home, and should revert for now back to Arrhinoceratops, where it started. Torosaurus latus is directly comparable to specimens that are comparable to Triceratops horridus, and thus can be ascribed to that species, although the authors [2] do not split the specimens recovered for Torosaurus latus to either Triceratops horridus or Triceratops prorsus, and seem to offer yet another species of Triceratops, Triceratops utahensis in its stead. A further work is indicated helping to define the apparent stratigraphic separation of Triceratops horridus and Triceratops prorsus. Thus, mapping the stratigraphy of Torosaurus specimens should recover which specimens would plausibly belong to which species. All of this wouldn’t matter if the authors consider the material of Torosaurus latus nondiagnostic: if it is, as the scare quotes suggest, Torosaurus latus is exactly what it seems to be: a large bodied chasmosaur that might be Triceratops. This seems to be the case, until the authors can determine which species latus is referable. Until then, Torosaurus latus may only be Triceratops latus, and thus the species is not synonymous with another known one.

2b. The study at hand (assessing the continuity of an ontogenetic trend) appears to be useful for other taxa, most of them chasmosaurs. Of them, only one (Agujaceratops mariscalensis) is known from a monospecific bonebed containing juveniles, subadults and apparent adults; all others are known from applied association (stratigraphic or locality based, but otherwise unassociated proximity), and that includes the current taxon. While it is likely that the material represents a single taxon, that single taxon appears to be higher than the specific level; it is difficult without associating other ontogenetic trends to determine what portion of this trend is specifically viable. They might all belong to a single species, but then again, there may be more than one: this appears to be the case for some ceratopsids, which may co-occur in the same level of formation, e.g., Arrhinoceratops and Anchiceratops.


The works above illustrate a paradoxical use of the “nomina dubia problem”: While it seems logical to render dubious problem taxa for various reasons (e.g., to discard taxa based on poor material in an age of increasing paleontological collections, favoring complete or adult specimens) it seems that other practices get caught up in this process: Taxa may be discarded because their material seems “less viable” than is ideal, without observing their actual practicality (e.g., Cristatusaurus lapparenti has features diagnostic of Suchomimus tenerensis in the premaxilla, differentiating it from Baryonyx walkeri [see here]); they may also be raised up by supplemental material with the assumption of monospecific occurence of a taxon in the level or region (e.g., the material is based on structures that are present in a broad range of taxa and are indistinguishable from multiple forms as in Carcharodontosaurus saharicus, but then are diagnosed on features not present in the holotype [see here], followed by designation of further taxa from which similar teeth are located, but remain indistinguishable from the earlier holotype). We can be prone to these, for various effects, such as the need to raise new taxa or subsume them, but these tend to follow a “liberal” versus “conservative” trend in taxonomy, one favoring plausible species while the other favoring exceptional species, where any variation in a bone can be diagnostic at any level or is highly vulnerable to plasticity. Both may be true.

In my view, then, Mojoceratops perifania is not synonymous with Eoceratops canadensis or to Chasmosaurus kaiseni based on the principles raised by various authors and the ICZN; referral of the specimens seems an oversight, but the term “provisional” makes the referral somewhat acceptable, yet “provisional referral” is still “referral.” And without being able to determine absolutely which species Torosaurus latus is a synonym of, it is likely that it is still a sister taxon to Triceratops horridus or Triceratops prorsus or bears features diagnostic to both of those taxa (affirmed without using the Torosaurus latus material in its diagnosis). I am certain the latter conclusion is problematic, since it undoes much of what Scannella and Horner set to do in their paper [2], but affirming taxonomic referral must occur at the species level, not some other containing level, while leaving the species untouched.

Even using the combination Triceratops latus may be problematic in this regard. This extends further into problems with referring ontogenetic series of closely related species to a single “concept organism,” if those series are not also complete (juvenile to adult) for each representative species; as Scannella and Horner [2] do not affirm the specific relationship of the material, using the containing taxon Triceratops instead [2], it becomes less viable to consider the work a matter of specific ontogeny than to consider it a general chasmosaur ontogenetic trend to which some taxa fit particular portions. This tells me then that Torosaurus latus represents a growth stage beyond that which is represented by Triceratops horridus et al. They may also be synonymous, but that depends on whether one can discriminate the material referred to Torosaurus latus on a specific basis to either Triceratops prorsus or Triceratops horridus — morphologically. This project may require more data to show that the ontogenetic trend can extrapolate beyond points where any species preserves material more affirmatively. It should also be noted that in [1: table 1, p1159], postulated ontogenetic trends and apparent status of the fossils are tabulated: breaks in the series seem to occur securely between some data points (e.g., ontogenetic trends A and B “thinning on lateral margins of parietal” and “depressions along lateral margins on ventral side of parietal” [respectively], with subadult status applied at point B, where the parietal length is also over ~50mm) but not others (e.g., ontogenetic trends E and F “squamosals elongate” and “fenestrae open” [respectively], with adult status [Torosaurus] occurring while other young adults [Triceratops] also possess both or just one of these trends, while varying in the shape of the squamosal or the length of the parietal).

In a related trend, the authors also seem to discard the idea of sexual dimorphism to account for changes in frill shape, fenestral presence, horn shape or orientation, or horn bone compacta, which is higher in Torosaurus than in Triceratops. Size based differences, mapped against trends, do not appear to securely map “adult” status, and the maturized bone appears to be applied prima facie as a mature morphology, instead of an elaborated and potentially sexually dimorphic feature, or even specific. These ideas are discarded because at least some of them might be “less parsimonious” if it would result in another “genus.” (We can have three species in a genus, but not three species in two “genera”? And what’s a genus anyway?)

Finally, I am uncertain about the trend being used to provide all chasmosaurs with the same eventual growth series, when only two taxa have been sampled (Agujaceratops mariscalensis and Triceratops/Torosaurus), and not in the same manner. In fact, in the former, orientation of the horns, shape and orientation of the frill, indicate apparent dimorphism due to high closely associated sampling. So while Torosaurus latus may belong to some collections of taxa including Triceratops horridus, it is unclear how much other taxa are included here, and whether the trend implicated is localized or generalized. It seems that there is an intention to collapse the taxonomy of end-Cretaceous, late Maastrichtian taxa, and even were their taxonomy true, it would not be possible to securely affirm it based on their data. We are promised two (at least) further papers, so hopefully these can close the gaps I discussed.

[1] Longrich, N. R. 2010. Mojoceratops perifania, a new chasmosaurine ceratopsid from the late Campanian of western Canada. Journal of Paleontology 84(4):681-694.
[2] Scannella, J. B. & Horner, J. R. 2010. Torosaurus Marsh, 1891, is Trceratops Marsh, 1889 (Ceratopsidae: Chasmosaurinae): Synonymy through ontogeny. Journal of Vertebrate Paleontology 30(4):1157-1168.
[3] Lambe, L. M. 1902. On vertebrata of the mid-Cretaceous of the Northwest Territory, Pt. 2. New genera and species from the Belly River Series (mid-Cretaceous). Canada Geological Survey, Contributions to Canadian Paleontology 3:25-81.
[4] Brown, B. 1933. A new longhorned Belly River ceratopsian. American Museum Novitates 669:1-3.

[n1] Wkipedia: Scare Quotes.

This entry was posted in Biology, Paleontology, Science Reporting and tagged , , , . Bookmark the permalink.

3 Responses to Scare Quotes and Recent Ceratopsians

  1. “The rules for nomina dubia are, generally, provided in the ICZN: A nomen dubium cannot compete for synonymy in taxonomy, and therefore cannot be a synonym.”

    Are you sue that’s the case? There’s only one instance I can find in the whole code, other than the definition in the glossary, that mentions nomina dubia at all. Article 75.5 says that status as a nomen dubium can be used when petitioning for a neotype when the existing type specimen is non-diagnostic. It doesn’t ay anything about nomina dubia being out of the running when it comes to normal synonymy. This is probably for the best since nomena dubia are totally subjective, and can change from find to find. If, for example, a second Anchiornis specimen is tested for color pattern, and is found to be different enogh in color from the prior specimen, and is therefore recognized as a separate species diagnosed by unique coloration, any specimen no matter how complete that doesn’t preserve feathers and melanosomes is suddenly non-diagnostic, including the type, and Anchiornis would become a nomen dubium. This kind of practice can only result in a constant churn of names and types as more discoveries are made, not really an ideal situation…

    • I may have become confused. As I discussed here, there is a practical set of rules used when people use the term, regardless of its meaning in the Code, which did use to say more on the subject before leaving the concept completely up to the workers (a bad idea, perhaps).

      However, your example does not require only color be used, even if a specimen is defined only based on said color; one could argue that plumage is not sufficient for classifying taxa, as it changes in season, ontogeny and in sex in many birds. An example where different material can be coordinately supportive of their own validity is Chirostenotes pergracilis, which may be the same as Caenagnathus collinsi, but either is differentiable from other taxa even when the only coordinating material, the ROM skeleton described by Sues, is comparable but also differentiable.

      However, I will emend the post above to be less explict on the ICZN’s involvement.

  2. Pingback: The Saga of Raptorex | The Bite Stuff

Leave a Reply

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

You are commenting using your 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