What Use A Perfect Specimen?


A holotype specimen is important in systematics when the taxonomist desires to use an ideal specimen upon which a new animal is created. It is the first, and best specimen, for future systematists to refer to when comparing other taxa. It bears the diagnostic criteria from which other specimens can be compared, and thus whether those specimens do or do not belong to the new name. Occasionally, the taxonomist has used a partial specimen, say fragments of the skull (Quaesitosaurus mongoliensis Kurzanov and Bannikov, 1983), a vertebra or a series of such (Titanosaurus indicus Lydekker, 1877 — where two vertebrae with varying morphologies form the holotype and paratype), or even just a tooth (Astrodon johnstoni Leidy, 1865). In other examples, the holotype is a juvenile (such as Apatosaurus ajax Marsh, 1877), and this raises another issue.

For each of these species, questions have arisen because of either the completeness of the material, or the value of the material in question. For teeth, it is the prevalence of similar teeth in different animals, raising the specter of convergence (two different animals developing similar teeth, possibly for similar diets, from different original morphology) and conservation (similar animals with similar teeth developing unique features from one another while the teeth remain similar to one another). But also for teeth, it is the issue of the incomplete element, the partial material that reveals nothing about the shape of a structure, just elements of it. For the partial skull, it is the lack of association of postcrania, or the non-overlap of partial cranial material (a snout here, a jaw there, a braincase way over yonder, etc.). For partial series of bones, it is a fundamental problem of knowing how the series is arranged, or even if they belong to the same animal or different individuals, or different taxa. And for a juvenile, it is the problem of ontogeny.

Reasons for using such specimens as types are varied, mostly of them being historical: When there were few animals with which to positively compare, or even remotely similar, it was more likely what you see is a new type of animal. But as with all things historical, this distorts with time; inversely to modern records versus witnessing events now, paleontological history improves with time, as we collect more material with which to compare. Thus there is a historical paradox: The more partial the material and the further from time of description and diagnosis this animal was originally named, the more likely it is that the material may be found to be lacking in scientific quality.

In modern systematics, the taxonomist may take some solace in affirming the identity through this wealth of data by establishing a new name on the basis of a partial specimen, secure that no other specimen that can be compared is identical. It may even have an autapomorphy (a feature unique to it in relation to close relatives — although not necessarily completely unique). But I am cautious: Such material suffers the historical paradox, for it is certain that we continue to fail in finding complete material for all taxa we have named, and yet continue to name taxa for incomplete material.

Certainly, sometimes the best you can do is with partial material, and the perfect ideal (that of naming taxa on only complete specimens) is a slippery slope argument. It shouldn’t necessarily stop there, if we should also establish it based on known growth series, or a population, along with a complete evolutionary series. This is an ideal. Far less ideal, but adequate, it a complete skeleton. Below that,  a partial skeleton, preferably including features of the cranium, limbs, jaws, dentition (if possible), vertebrae, and so on. Below that, relatively complete sections of limbs, or series of vertebrae, or a partial skull, tend to be common, while skeletons 20% or so may be just as common or less so. Then there is the Cope-Marsh-Era level of material, such as a pair of supraorbital horns, a partial vertebra, teeth, and so forth. Some of these last level of material still survives today and still manages to get a lot of material referred to it (I keep picking on poor Troodon formosus Leidy, 1865, poor guy).

I think, fundamentally, the question I ask with this post is the title, but also its inverse: What use, if any, is a perfect specimen? If a partial specimen can tell us much, if not as much, as a complete one (say, if we’re missing only a few vertebrae, or the teeth of the lower jaw, or the entire right half of the skeleton, cut clean through the sagittal) then the value of the complete skeleton is not as much. But what if we are missing a grand deal of the skeleton, all but dorsal vertebrae or the humerus or a scapula? What if each of these was clearly immature?

The above question can be escalated up a category, from an individual to multiple individuals, where the question of age becomes relevant: Ontogeny rewards us with knowledge that between the juvenile condition and the adult condition, the shapes, sizes, and relative proportions of bones change. If so, in a given specimen, does this increase our knowledge when we lack a comparison of an adult, or decrease it? Normally, I’d say increase, but I have no context for this new information. The value this information gives to the totality of my understanding is problematic, since adding a juvenile condition when the taxa to which I am comparing are all adults tells me very very little. We can make a few arguments, however, and these are testable.

1. We can presume that ontogeny is not relevant — that the specimen’s age does not matter, and that because this material is unique for any reason, it is valid if differentiable from any taxon.

2. We can assume that the limited breadth of knowledge will increase in time (it will) and that this will allow further studies to prove or disprove the hypothesis (that a juvenile is a unique animal).

3. That the circumstances of the material (where it was found, its place in Deep Time) support the fossil’s uniqueness, regardless of its morphology.

4. We assume that a phylogenetic analysis’ results (regardless of the quality, but assume of the best) supports separation of the taxa from possible congeneric adults not through any unique feature but through a combination features.

5. We assume that imperfect preservation, including distortion or perimortem damage, have not created artefacts from which we have created apomorphies to diagnose our taxon.

So take a given specimen, and apply all or most of these conditionals to it, and affirm that we have a new taxon, and name it. When we do, what new information have we gained? Has it advanced the scientific discourse? Has it added (it alone, not supplemental data in the paper describing it saying other things) to the fullness of our knowledge of this specimen and perhaps closely related specimens? If not, why even name the specimen? The specimen by itself has all of this information just as well as the named taxon, even a specimen number, but merely lacks the nomenclature. This is generally an argument against binomial nomenclature, and for other types of identifier-number nomenclature (although not as elegant).

As an extreme example, this can be used to substantiate the use of tadpoles as unique taxa from their frog/toad adult forms, as well as other metamorphic species that have larval and pupal stages. It can differentiate taxa that undergo extreme cranial or postcranial modifications during ontogeny (ceratopsians are a recent popular example). And it can differentiate taxa that have been argued to be sexually dimorphic, such as the two coelophysids of the Whitaker Quarry blocks (currently Coelophysis bauri {Cope, 1887}) of Ghost Ranch, New Mexico (USA).

Of course, several of these arguments have been used to substantiate NON-juvenile-based taxa, but they all apply to juvenile type specimens as well. We can appraise ourselves with the logical flaws these arguments represent as well, as several can be examples of confirmation bias, telling us what we already think we know, and thus invalidating what we might be told otherwise. Or its wishful thinking, hoping further support with shore up a seeming weak defense; yet perhaps it will be allowing ourselves the chance at naming a new taxon on a novelty (a unique feature, or a famous specimen) and letting others do the hard work of trying to handle it in the future, and that’s just laziness.

So is there ever a good time to use an obvious juvenile, even if it has unique features? or an incomplete specimen, especially if it have a feature heretofore unseen? These questions hinge on the quality of the feature. Once again, we must be able to rule out the effects of distortion, or even disease, deformation premortem (such as a healing fracture mistaken for an apomorphic callous or knob), extreme fusion or ossification of cartilages of the skeleton due to advanced age or (again) diseases. These aspects distort our total data, and require detailed work to uncover, or correct once already in the literature. This means extra, additional and harder work, possibly to fix something that could have been avoided.

But this issue extends in a somewhat unintuitive direction: What about bonebed taxa? A bonebed comes in a variety of flavors, and depending on the environment, can contain multiple species. The bonebeds of the Chinle Formation, especially of the Whitaker Quarry mentioned above, contain numerous archosaurs, from crocs to dinosaurs, and of multiples of each; while the bonebeds of Alberta tend to contain groups of ceratopsians, sometimes two different types in close association. In the former case, specimens have only recently been reported as belonging to new taxa formerly confused with the ubiquitous Coelophysis, while in the latter, there was a confusion between Medusaceratops and Albertaceratops (Ryan et al., 2010), despite one of these taxa being a chasmosaurine and the other a centrosaurine (respectively). But sometimes bonebeds come in disarticulated form, and the mass is a jumble, or the specimens are relatively separated. In such cases, it becomes much more difficult to associate individual bones, and the ability to discriminate taxa decreases. Despite this, a great deal of taxa have been named from such complexes, and this has extended to inferring bones from certain stratigraphic horizons or geographic areas must belong to all one taxon (e.g., Alamosaurus sanjuanensis Gilmore, 1922; see Lehman and Coulson, 2002).

It has been with this in mind that I have developed a low regard for taxa established from such complexes, or support for them, without substantial overlap of the various specimens via different sets of articulated associations of material, or an affirmation of the consistency of the material (different elements showing either a strong phylogenentic or apomorphic similarity). Obviously, complete or articulated skeletons help establish a complex’s association of bones, so it becomes difficult why we should rush to identify taxa immediately instead of waiting, knowing that while we may have something new, it is hard to support. Alternatively, there has been little established on the relative value of apomorphies: if a specimen, no matter how small or incomplete, bears an autapomorphy, it may be useful to science. The question here, though, is the value of naming that specimen. And there is no reason to name a specimen, in my opinion, that serves us just as well with a specimen number barring better, more complete material.

Cope, E. D. 1887. The dinosaurian genus Coelurus. American Naturalist 21:367-369.
Gilmore, C. W. 1922. A new sauropod dinosaur from the Ojo Alamo Formation of New Mexico. Smithsonian Miscellaneous Collections 72(14):1-9.
Kurzanov, S. & Bannikov, A. 1983. [A new sauropod from the Upper Cretaceous of Mongolia]. Paleontologicheskii Zhurnal 2:90-96. [in Russian]
Lehman, T. M. & Coulson, A. B. 2002. A juvenile specimen of the sauropod Alamosaurus sanjuanensis from the Upper Cretaceous of Big Bend National Park, Texas. Journal of Palaeontology 76(1):156-172.
Leidy, J. 1865. Cretaceous reptiles of the United States. Smithsonian Contribution to Knowlegde 192:1-135.
Lydekker, R. 1877. Notices of new and other Vertebrata from Indian Tertiary and Secondary rocks. Records of the Geological Survey of India 10(1):30-43.
Marsh, O. C. 1877. Notice of new dinosaurian reptiles from the Jurassic formation. American Journal of Science and Arts 14:514-516.
Ryan, M. J., Russell, A. P. & Hartman, S. 2010. A new chasmosaurine ceratopsid from the Judith River Formation, Montana. p181-188 in Ryan, Chinnery-Allgeier & Eberth (eds.) New Perspectives on Horned Dinosaurs: the Royal Tyrrell Museum Ceratopsian Symposium. Indiana University Press (Bloomington).

About these ads
This entry was posted in Paleontology, Philosophy, Taxonomy and tagged , . Bookmark the permalink.

9 Responses to What Use A Perfect Specimen?

  1. Andrea Cau says:

    It’s easier to remember that Dynamosaurus imperiosus is junior synonym of Tyrannosaurus rex instead of remembering that BMNH R7994 belongs to the same species of CMN 9380 (note that it’s the same concept but using specimen numbers instead of names).
    We give names to things because we are palaeontologists (Logos=word) and not “palaeontonumerologists”.
    Contingent example: I know what is Brontomerus (it’s a taxonomic hypothesis discussed by Taylor et al., 2011), but I would never remember all the specimen numbers of all the specimens forming the “Brontomerus hypothesis”.
    I know almost 1000 taxonomic hypotheses on fossil dinosaurs because I remember their names: it’s impossible to remember 1000 specimen numbers, because our human memories use (and work with) names and not series of numbers (=specimen numbers).

    • Why should we not then name every specimen, regardless of whether they might be synonyms?

      • Andrea Cau says:

        I think that the context is relevant in taxonomy: not all specimens, but all “relevant” specimens would need a name. “Relevant” here means: expanding qualitatively the knowledge of that clade. The name is not only a word fixing the specimen, but a new “meme” that may be useful in the scientific progress.
        For example, imagine to find a fragmentary ceratopsian bone with at least an unique combination of features (if not autapomorphies) in the Morrison Formation. In my opinion, being it the “first Morrison ceratopsian” that bone is worth of being named, because it increases our knowledge of ceratopsian evolution. Probably, the same identical bone in a Cretaceous Formation should only be named by its specimen number, because it would not increase qualitatively our knowledge of ceratopsians.

        • To expand qualitatively the knowledge of the clade, shouldn’t context be provided? Lately, many new taxa are supported with systematic analyses, but too often these result in placing the new taxon into a polytomy, which in the analysis results is no qualitative gain. Or that the author of the analysis simply adds the analysis in because it’s recommended, without using it as a tool to supplement the main study. Sometimes, the analysis points to the new taxon, but for the most part, this is very rare.

          As for “the first Morrison ceratopsian,” I would be just as doubtful as “the Cretaceous Australian dicynodont” or “the first Australian tyrannosauroid.” Neither of these contentious specimens have a name, although they may deserve one for uniqueness purposes, and neither of them may have features that are differentiable from other taxa. The reduced level of the material prevents secure differentiation, allowing a lot of doubt into the conclusion of the analysts involved.

          • Andrea Cau says:

            The naming of a species is a combination of “good sense” (several distinct lines of evidence) and data.
            E.g.: since the “aussy tyrannosauroid” may instead be a neovenatorid (my large analysis does not support anymore a tyrannosauroid status for that specimen) I think it’s not worth of being named (pending new data useful in a differential diagnosis with the other known Australian neovenatorids).
            Expanding my example (the Morrison ceratopsian), a Titonian North American rostral bone is a potentially unambiguous new taxon worth of being named, more plausible than a neotheropod pubis in the Cretaceous of Australia.

  2. Andy says:

    There is considerably more incentive to name that animal than not name it. A paper coining a new dinosaur name can gain far more press than a paper simply saying that a scrappy specimen is interesting but not worth naming. Another factor is the feeling that “If I don’t name it, someone else will.” I’m not saying I agree with either of these reasons, but there they are. Paleontologists (and other taxonomists) are only human.

    • But isn’t this more of a “What use a popularity contest,” instead? If the study needs press in and of itself, why not focus on that press-worthy topic in the paper? A systematic analysis, a biomech study, etc., are certainly worthy of attention, but the nomenclature itself?

      I can see too many problems with the Cope-Marsh-Era type press-for-nomenclature the two men competed for to argue against such “press for nomenclature.”

      • Andy says:

        I agree for the most part, but the headline of “Scientists identify new species of dinosaur” sells much better than “Scientists find unassignable scrap of bone”.

        [I should state for the record that I think a fair portion of what receives global press for dinosaur research is not necessarily worth such attention - not always because it's bad research (although this sometimes happens), but because it often implies major advances where there are none and thus gives a skewed view of science to the public]

        • I agree with you on the issue of the headline that catches attention, and how some reports may not be deserving. In some cases, the scope of the paper seems to be extended for the sake of the press, while the conclusions used may not be, or are on very shaky ground. Every single “how birds evolved” paper receives this attention, or development of pterosaur flight, obscuring the contention and uncertainty involved in much of the subject matter.

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s