What is the nature and purpose of a type specimen?

September 13, 2016

It’s been interesting seeing the response to my comment on the ICZN petition to establish Diplodocus carnegii as the replacement type species of the genus Diplodocus. In particular, Mickey Mortimer’s opposition to the petition seems to be based primarily on this argument:

The dinosaur community has recently lost sight of the fact that the type concept was never meant to indicate the most well preserved or described specimen/species.

I find this unconvincing, on the basis that the ICZN was never designed with dinosaurs in mind in the first place. For the great majority of the species that have been named under its rules, the selection of the obvious holotype has been perfectly adequate, because extant animals — by far the majority — are nearly all represented by complete and well-preserved specimens.

Alphina nigrosignata (type specimen; photo courtesy Geert Goemans, Department of Ecology and Evolutionary Biology, The University of Connecticut; specimen loaned from Herbert Zettel from The Museum of Natural History Vienna (NHMV). From University of Delaware, College of Agriculture & Natural Resources.

Alphina nigrosignata (type specimen; photo courtesy Geert Goemans, Department of Ecology and Evolutionary Biology, The University of Connecticut; specimen loaned from Herbert Zettel from The Museum of Natural History Vienna (NHMV). From University of Delaware, College of Agriculture & Natural Resources.

Dinosaurs — which in many cases are represented by eroded and distorted fossils of a tiny part of the animal — are already an aberration from the perspective of the ICZN, and that is why they sometimes need special treatment.

What are type specimens for, after all? The Code itself says “The fixation of the name-bearing type of a nominal taxon provides the objective standard of reference for the application of the name it bears” (Article 61.1); and comments that type specimens “are the international standards of reference that provide objectivity in zoological nomenclature” (Article 72.10). That is a role that YPM 1920 is simply not capable of fulfilling — and, more to the point, a role that it is not filling. The Diplodocus carnegii holotype CM 84 is the international standard of reference that provides objectivity in Diplodocus nomenclature. Slavishly following the usual provisions of the Code to retain the fiction that YPM 1920 fulfils this role simply does not reflect reality.

Some people occasionally object to the nomination of neotype specimens or replacement type species on the grounds that the Code does not require this. Of course it doesn’t: if it did, there would be no need for petitions. The fact that the Code allows for petitions constitutes explicit recognition that its usual provisions do not always suffice to produce the “sense and stability for animal names” that the Commission’s web-site used to have as its banner before the last redesign. Petitions exist precisely to allow the setting aside of the usual rules when sense and stability is served by doing do.

15 Responses to “What is the nature and purpose of a type specimen?”

  1. Mickey Mortimer Says:

    A good topic. First, I’d say your characterization of the state of type specimens for living taxa is overly optimistic. Neontology isn’t my field, but I gather many holotypes are skins lacking any osteological or organ data, preserved specimens in alcohol which lose genetics and color and are difficult to examine, etc.. But like Arjan said in the last post, unless there’s some nomenclatural issue, no one petitions the ICZN to change the type to some specimen collected in modern times that is frozen, genetically sampled and CT scanned, etc. to provide maximum information.

    Second, the D. longus holotype IS fulfilling the type role because everyone agrees it is Diplodocus. This is a major beef I have with defenders of these petitions. You speak of instability and Tschopp and Mateus claim “retention of D. longus as type species would create insecurities and confusion concerning the use of Diplodocus as a genus”, but HOW? Really, how? If there were some genus that had a tail just like Diplodocus but was different elsewhere, so we really couldn’t tell where longus went, then sure. But that’s not the case. Someone tell me how D. longus is confusing or causing insecurity given our current data.

    Finally, please tell me why this is different than the many cases I listed in the prior post of Mesozoic dinosaurs with comparatively crappy type specimens. It’s common among Sauropoda as well. The standard people use for Camarasaurus is the articulated and juvenile C. lentus referred specimen described by Gilmore (1925), not the C. lentus holotype let alone the holotype of the type species C. supremus. The Apatosaurus standard is A. louisae holotype CM 3018 (Gilmore, 1936), not any specimen of the type species A. ajax. The exemplar species of Mamenchisaurus isn’t the poorly preserved M. constructus, but rather M. hochuanensis. No one uses the fragmentary holotype of Omeisaurus junghsiensis, they use the Wujiaba specimens described by Dong et al. (1983). No one uses the holotype of Shunosaurus (IVVP V9065; Dong et al., 1983), instead using nearly complete referred specimens described by Zhang (1988). The prime example of Brachiosaurus was brancai’s holotype until you made that Giraffatitan. The holotype of Epachthosaurus is an incomplete dorsal vertebra, so everyone uses articulated specimen UNPSJB-PV 920 (Martinez et al. 2004). Etc., etc… If this is the new standard, there are going to be a LOT of petitions.

  2. Mike Taylor Says:

    Someone tell me how D. longus is confusing or causing insecurity given our current data.

    When the next phylogenetic analysis — adding new taxa, or coding new characters, or using a different algorithm — finds YPM 1920 closer to Apatosaurus or Barosaurus or what have you. You of all people know how unstable phylogenies are.

    As for other dinosaurs: at this stage, I have no dog in those fights. I’m concerned with the present petition. But I wouldn’t be unhappy to see proper types established in some of the other cases you mention. Not all, obviously — as I also showed in my 2009 paper, Giraffatitan is very different from Brachiosaurus, which is a perfectly well-defined taxon of its own.

  3. Mike Taylor Says:

    Bottom line, if someone finds a specimen which they think might be Diplodocus and asks himself “Is this the same species as YPM 1920?” the honest answer is “Heck if I know”. So what is a holotype for?

  4. dale Says:

    I don’t mean to put MY “dog” into this fight but, hypothetically speaking …. what if you have a holotype that you want to designate a type because the original is a very crappy fragmentary specimen [and this new holotype is osteologically complete] and well preserved [undistorted] …. and along comes another specimen of same, found in a state of preservation wherein the internal organs [highly carbonized] along with muscles and tendons are actually preserved !?! What if other specimens in this new state of preservation start turning up ? What happens then ? I know this would be quite a lagerstatten to find ….. BUT …..

  5. dale Says:

    In short, what happens if you go and designate the osteologically complete specimen as the new type, AND THEN recover specimens of same from that new lagerstatten after the formal designation has been completed by the ICZN. I wasn’t too clear with that last comment. Sorry. But you see where I’m going with this.

  6. Mike Taylor Says:

    That would be a nice problem to have :-)

    But the issue here — for me, anyway — is: can the type specimen be used to determine whether or not a given other specimen belongs to the taxon? If so, then it’s good enough; if not, then it isn’t. Now there is obviously a grey area between “good enough” and “not good enough”. My sense is that YPM 1920 falls into the latter bucket.

  7. Chase Says:

    If everyone agrees YPM 1920 is Diplodocus, but D. longus doesn’t apparently have any distinguishing features from the other Diplodocus species, it seems fit that YPM 1920 should be designated as Diplodocus sp. My question is whether a “sp.” specimen should be and has been used as a type specimen for a non-avian dinosaur. It would make sense to me that it could set defining features for a taxon, but the issue arises as to how possibly new species can be distinguished from the “sp.” type.

  8. YPM 1920 falls into the “good enough” category, as far as I can tell, albeit just barely. While the situation of a perfect sauropod mummy is a little bit unlikely, I agree with the general principle that since we don’t need a new type designation right now, I don’t see why we shouldn’t wait until we do on the off chance a different solution has raised its head (e.g., say, a skeleton with a complete, articulated axial column and skull, with identical mid-caudals to YPM 1920, that could be made the neotype of D. longus).

    On the other hand, while it does still seem to be unique, both Leinkupal and Tornieria have some interesting similarities to Diplodocus, and Kaatedocus has no caudals for comparison.

    Wait–Kaatedocus siberi and Diplodocus longus are both members of the Barosaurus lentus+Diplodocus carnegii clade, from low in the Morrison Formation, without any overlapping material. Tschopp et al. noted the weakness of the Kaatedocus+Barosaurus node, so…could Kaatedocus siberi be Diplodocus longus?

  9. Despite what Tschopp et al. say, Diplodocus longus has a unique combination of characters and thus should be considered valid (unlike D. hallorum, it has mid-caudals with posteriorly-inclined neural spines which overhang the postzygapophyses, short prezygapophyses, and a neural arch placed on the middle of the centrum; unlike D. carnegii its mid-caudals are not mediolaterally expanded). While it lacks autapomorphies of its own, that might be to be expected, as it’s older than the other two species and so could potentially be their ancestor.

  10. Halbred Says:

    I, for one, am all about shifting type designation to better and/or more complete specimens of the same species are found.

  11. Mickey Mortimer Says:

    “When the next phylogenetic analysis — adding new taxa, or coding new characters, or using a different algorithm — finds YPM 1920 closer to Apatosaurus or Barosaurus or what have you.”

    But not only has that not happened yet, no one seems to think it’s probable. Indeed, the petitioners themselves are the only ones to have included D. longus’ holotype in an analysis and found it was “confidently identifiable as belonging to the same genus as the type specimens of D. carnegii and ‘Seismosaurus’ hallorum.” Shouldn’t we at least wait until there is some controversy, then examine the evidence behind the controversy, and if truly conflicting character data is confirmed, THEN petition the ICZN? You haven’t even tried to compute how many more steps it would take to remove YPM 1920 from Diplodocus in their matrix, or perform a chi squares analysis to get a P value on whether or not that topology is rejected. It’s just all been unevidenced speculation. I find it lamentable how quick dinosaur workers are lately to petition before we even know if we have a problem.

    “if someone finds a specimen which they think might be Diplodocus and asks himself “Is this the same species as YPM 1920?” the honest answer is “Heck if I know”. So what is a holotype for?”

    The holotype of a type species exists for the purpose of linking the first validly published name to a specimen which can be soundly argued to belong to the concept of the type genus. It is thus the standard of comparison, but this in no way implies that you should or even could be able to depend on it in a vacuum to refer additional specimens. At the very least, paleontology has always worked so that partial specimens with no shared elements can be referred to taxa by using intermediaries which preserve the elements of both. So sure, armless IGM 100/128 couldn’t be referred to Deinocheirus with much certainty by itself, but it can via the intermediary of IGM 100/127 which preserves both arms that compare to the holotype and the hind area that compares to 100/128. Even if you had a complete holotype, comparison in a vacuum is useless without other specimens to contrast with, establish the level of individual/ontogenetic/sexual variation/etc..

    Thus the honest path to an answer (assuming you meant genus, not species) is to place your new specimen in Tschopp et al.’s matrix and see where it emerges. If your new specimen still cannot be placed confidently in Diplodocus or not (e.g. skull CM 11161), that’s not YPM 1920’s fault. Because of the principle of intermediaries I outlined above, the analysis allows anchoring Diplodocus on YPM 1920 while still augmenting its known anatomy with more complete specimens. That’s what a type is for- anchoring, not being the most complete (ditto for families, clades, etc.- we use Dromaeosauridae and Oviraptoridae, not Velociraptoridae and Khaanidae).

    In truth, the Diplodocus situation is actually better evidenced than most of the theropod and sauropod examples I listed, where no analysis has ever been performed. At least we have in Tschopp et al. a very detailed and well designed analysis that supports the traditional conclusion. You could use your preemptive ‘who knows what future analyses will show’ argument to greater effect in most of those cases. But that’s not been how dinosaur paleontologists have treated taxa and the ICZN until very recently, and I’m a big fan of consistency. If that’s our new standard, expect petitions for most Mesozoic dinosaurs known from more than one substantive specimen. And aim to name new more complete species in known genera, because your path to be the author of the type species in a historic genus is just a petition away. Good job stealing Azendohsaurus, Flynn et al., A. madagaskarensis becoming the type species is inevitable…

  12. Mickey Mortimer Says:

    Chase- You can’t just make a named species into ‘Genus sp.’- that specimen retains the name it was given. But yes there is precedent for retaining a specifically undiagnostic type species in a genus, alongside the diagnostic species of that genus. See Libycosaurus as an example. The new species are diagnostic vs. each other, while only the type specimen is referred to the type species pending future analysis. Of course as John says, there isn’t actually agreement that D. longus is undiagnostic, so it’s a moot point.

    Reference- Lihoreau, F., Boisserie, J.-R., Blondel, C., Jacques, L., Likius, A., Mackaye, H.T., Vignaud P., & Brunet, M. 2014. Description and palaeobiology of a new species of Libycosaurus (Cetartiodactyla, Anthracotheriidae) from the Late Miocene of Toros-Menalla, northern Chad. Journal of Systematic Palaeontology, 12(7): 761-798.

  13. Ken Carpenter Says:

    Couple of points:
    First, fossils were indeed a consideration when the Code was in development. The 1842 report from the Zoological Nomenclature Committee of the British Association for the Advancement of Science pointed out specifically that manuscript names for “fossils in the Scarborough and other public collections, have received MS. names, which will be of no authority until they are published
    with characters.” (p. 102). Unlike today, the naturalists of the early 1800s studied both living and fossil organisms.

    Regarding fossils, the Code specifically states:
    Chap. 1, Art. 1.2. Scope. 1.2.1. The scientific names of extant or extinct animals include names based on domesticated animals, names based on fossils that are substitutions (replacements, impressions, moulds and casts) for the actual remains of animals, names based on the fossilized work of organisms (ichnotaxa), and names established for collective groups (see, in particular, Articles 10.3, 13.3.2, 23.7, 42.2.1, 66.1, 67.14), as well as names proposed before 1931 based on the work of extant animals.

    Chap. 16, Art. 72.5. “Eligibility as name-bearing types. Only the following are eligible to be a name-bearing type, or part of a name-bearing type, of a nominal species-group taxon:
    72.5.1. an animal, or any part of an animal, or an example of the fossilized work of an animal, or of the work of an extant animal if the name based on it was established before 1931;”

    Second, IF D. carnegii is accepted as a valid taxon from D. longus, then that is a tact admission that the holotype D. longus is diagnostic. If D. carnegii cannot be separated from D. longus, then that is an admission that D. carnegii is a junior subjective synonym of D. longus. Nomenclature stability is not enhanced by the T&M proposal.

  14. Ken Carpenter Says:

    In answer to the title Mike posted, “What is the nature and purpose of a type specimen? “, O.C. Marsh addressed that in 1898, “The value of type specimens and importance of their preservation” American Journal of Science ser. 4, vol. 6: 401-405. Since that article is no longer covered by copyright, you can find that article on line. If Mike wants, I’ll paste the entire article as a comment.


  15. […] Back at the start of September, I noted that Tschopp and Mateus (2016) had published a petition to the ICZN, asking them to establish Diplodocus carnegii as the type specimen of the genus Diplodocus — a role that I argued it already fulfils in practice. […]

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 )

Connecting to %s

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

%d bloggers like this: