Concerning Brontosaurus

"And I've found/It's all up to what you value" — G. Harrison

By now, the paleontological world has weighed in on the news that Brontosaurus may in fact be a valid genus after all (Tschopp et al. 2015). If you've got some time (befitting its subject matter, it's not what you'd call a short document), the publication is well worth reading. Alternately, if you are strapped for time or just break out in hives when confronted with anatomical terminology, there are many shorter and less technical explorations of the topic. The concise description is that Tschopp et al. ran a number of diplodocid and putative diplodocid specimens through a phylogenetic analysis and then attempted to apply more objective measures than the classic eye test to determine which species should be in which genera. Among the results was the absorption of Dinheirosaurus by Supersaurus, which would make the latter the newest member of the exclusive Morrison–Lourinhã club; the designation of the new genus Galeamopus for longtime problem "Diplodocus" hayi; and the headline finding, that Brontosaurus was distinct from Apatosaurus.

I never had much of a stake in the Brontosaurus fight, although my platonic ideal of the sauropod is Apatosaurus louisae. My formative dinosaur book was The Illustrated Encyclopaedia of Dinosaurs, which featured Apatosaurus, and being a born pedant I of course was happy with the whole ball of technicalities. The "beloved old genus reduced to a synonym" I imprinted on was Anatosaurus, of all things. Sinking that into Edmontosaurus bothered me. In part, this was because although every 1980s dinosaur book worth its salt covered practically every other dinosaur-related topic known to humanity, they did not delve into the arcane arts of nomenclature beyond the concept of priority. I could have really used an explanation about type species and holotypes and so forth; it would have saved literally minutes of frustration (which of course are more like hours when you are about 10). Instead, I got speculative restorations of sauropods with trunks, ceratopsians with their frills embedded in their necks, and that darn "dinosauroid", when I could have had an explanation about how Anatosaurus annectens had been reevaluated as a species of Edmontosaurus and why "A." copei, the one I liked, could not be called Anatosaurus. Later, of course, it turned out that "A." copei, by then re-genera-ed under the unintentionally comic Anatotitan, was just mature annectens (said the cowboy to the edmontosaur, "why the long face?"). No doubt within a few years someone will formally split Anatosaurus from Edmontosaurus again, perhaps on stratigraphic grounds (you're not supposed to let things like stratigraphy and geography influence you, but there you go).

The Tschopp et al. paper is the kind of thing that would have been really interesting to translate to the late Thescelosaurus. I don't know if I ever had to contend with something like it; the disintegration of Iguanodon and certain prosauropods (most notably Euskelosaurus) are vaguely comparable, but those events took place over longer periods of time, and there wasn't the same cachet. You probably don't recall any headlines about the revival of Plateosauravus (unless you're a dedicated dinosaur fan, this may be the first time you've ever heard of it), and the major reactions concerning Iguanodon were a sense of relief ("glad that's getting looking at") mixed with incredulity at the rapidly expanding iguanodont supernova, and some surprise at the realization that the quintessentially British dinosaur had picked up shop and was now based on Belgian fossils. In general, I tend to be conservative with nomenclature, both in the sense of opting for stability and in the sense of restricting names to material that can be shown to belong (note that these two directives do not always harmonize; that's where headaches and entries for seemingly every lousy dubious name arise), and I've learned that the most recent publication is not necessarily correct, just the most recent. I think that in the end, I would have gone along with Tschopp et al.'s results, more or less; they are plausible and supported by reasonable arguments. There's this to consider, though. Below is an ASCII phylogeny of part of Tschopp et al.'s tree using their preferred nomenclature:

--Diplodocidae
    |--Amphicoelias altus
    `--+--Apatosaurinae
         |    |--unnamed species
         |    `--+--+--Apatosaurus ajax
         |         |    `--A. louisae
         |         `--+--Brontosaurus excelsus
         |              `--+--B. parvus
         |                   `--B. yahnahpin
         `--Diplodocinae
              |--unnamed species
              `--+--Tornieria africana
                   |--+--Supersaurus lourinhanensis
                   |    `--S. vivianae
                   `--+--Leinkupal laticauda
                        `--+--Galeamopus hayi
                             `--+--+--Barosaurus lentus
                                  |    `--Kaatedocus siberi
                                  `--+--Diplodocus carnegii
                                       `--D. hallorum

Now check out these:

--Diplodocidae
    |--Amphicoelias altus
    `--+--Apatosaurinae
         |    |--unnamed species
         |    `--+--+--Apatosaurus ajax
         |         |    `--"Apatosaurus" louisae
         |         `--+--Brontosaurus excelsus
         |              `--+--Elosaurus parvus
         |                   `--Eobrontosaurus yahnahpin
         `--Diplodocinae
              |--unnamed species
              `--+--Tornieria africana
                   |--+--Dinheirosaurus lourinhanensis
                   |    `--Supersaurus vivianae
                   `--+--Leinkupal laticauda
                        `--+--Galeamopus hayi
                             `--+--+--Barosaurus lentus
                                  |    `--Kaatedocus siberi
                                  `--+--Diplodocus carnegii
                                       `--Seismosaurus hallorum

--Diplodocidae
    |--Amphicoelias altus
    `--+--Apatosaurus
         |    |--unnamed species
         |    `--+--+--A. ajax
         |         |    `--A. louisae
         |         `--+--A. excelsus
         |              `--+--A. parvus
         |                   `--A. yahnahpin
         `--Diplodocus
              |--unnamed species
              `--+--D. africana
                   |--+--D. lourinhanensis
                   |    `--D. vivianae
                   `--+--D. laticauda
                        `--+--D. hayi
                             `--+--+--D. lentus
                                  |    `--D. siberi
                                  `--+--D. carnegii
                                       `--D. hallorum

Nothing is changed except the generic labels. The second and third represent extremes of generic splitting and lumping, respectively, and both are technically acceptable interpretations. (In fact, I could have rotated them so that my favorite Apatosaurus louisae was the last name in each, and the results still would have been the same, but again just presented a bit differently.) So, is this all just a trick of semantics? Well, yes and no.

It is a simple but oft-overlooked fact that all facets of taxonomy and nomenclature for extinct animals are arbitrary constructs. Some constructs may be more wrong-headed than others, but in the end they are all produced by people working after the fact. Dinosaur workers prefer genera. Specialists in other fields prefer species. Researchers working in the early decades of paleontology produced a multiplicity of names, in part because they had less material for comparisons, in part because there were different standards then, in part because variation was poorly known, and in part because they wanted to create names for various reasons outside of science. The pendulum then swung toward lumping for several decades as people got embarrassed over things like fifteen species of Triceratops and nobody knows how many oreodonts. Splitting now seems to be more acceptable again (with certain exceptions due to advances in the study of growth), in part because of more detailed stratigraphy, and in part because people are more concerned with diagnostic type specimens.

However, even though they are subjective, naming and classification are important for communication. How else are you supposed to talk about something? The other major alternative for identification is using museum catalog numbers, which is all right when discussing a couple of specimens, but rapidly becomes confusing as more specimens are added, and heaven help you if you have any tendencies to transposing numbers. Names and classifications exist to serve our needs. When faced with choices like those above, eventually the practitioners will come to a consensus with the implicit understanding that it best serves their needs. In the end, the choice will say something about the state of the field at that time, about how they were handling taxonomy and nomenclature. 

Two other asides:
As noted elsewhere, what happened to Atlantosaurus montanus? Nowadays you only stumble across Atlantosaurus if you are trawling historical documents, but as recently as the early 20th century the Morrison Formation was sometimes known as the "Atlantosaurus beds". Granted, that became more of a reference to sauropods in general than Atlantosaurus proper, which just benefited from showing up first, but still, a study of type specimens that may be diplodocid omitting A. montanus is a bit like omitting Stu Sutcliffe from a history of The Beatles.

There are significant quantities of diplodocid material at the Science Museum of Minnesota, albeit perhaps not as firmly associated and articulated as would have been preferred, but the stuff is there, along with fossils of your friend and mine, Haplocanthosaurus. This is not a criticism of the paper for not including them, but a plug for the SMM.

References:

Tschopp, E., O. V. Mateus, and R. B. J. Benson. 2015. A specimen-level phylogenetic analysis and taxonomic revision of Diplodocidae (Dinosauria, Sauropoda). PeerJ 3:e857.