What’s the problem with Squalodon atlanticus?

The name Squalodon atlanticus is encountered fairly frequently on specimen labels in collections of Atlantic coastal fossils. This is an interesting name, with a lot of history behind it, but unfortunately one which I believe should not be used. Let me explain…

Squalodonts were a group of toothed whales found more or less globally in the Oligocene to the middle Miocene. The name was first used by Grateloup in 1840 for a rostrum fragment with four teeth from middle Miocene deposits near Bordeaux, France, although a section of a squalodont mandible was figured by Scilla as early as 1670.

The first report of a squalodontid from North America was in 1856, when Joseph Leidy described three teeth from New Jersey under the name Macrophoca atlantica (Leidy believed at the time that the teeth were from a seal). That’s how things remained until 1867, when Leidy’s former student E. D. Cope (of the Cope-Marsh “Bone Wars” fame) reported on some maxilla fragments from Charles County, Maryland. Cope’s report read, in part:

“ … a fragment of the muzzle, including the proximal portions of the maxillary bones, with molars, and the canine teeth of the Squalodon Atlanticus (Leidy).”

This was the first use of the name Squalodon for a North American specimen, but Cope clearly indicated that he is referring to a Leidy species. Later the same year, Cope clarified his description, and specified that he was (correctly) referring Leidy’s Macrophoca atlantica to Squalodon (as S. atlanticus), and referring two new specimens to the same species:

“Squalodon Atlanticus Leidy MS. Macrophoca atlantica Leidy, Proceed. Acad. 1856, 220.
“Remains of three individuals of this species before me indicate considerable variety in the forms of the two rooted molar teeth.
“The individual from which Leidy determined the species is represented by only three true molars….

He then went on to describe the other two specimens:

“The second individual is represented in the Thomas collection by the proximal portions of the maxillary bones of both sides, and seven molar teeth in place….
“The third individual is represented by one true molar, and two caniniform premolars….

And finally he stated where these specimens came from (the “Thomas collection” was from Charles County):

“No. 1 was discovered in the miocene of New Jersey; the others were in the Thomas collection.”

In the 1860’s most publications did not include figures, and neither Leidy nor Cope included any figures of these specimens. As far as I can determine, the first time an image of any of these was published was by E. C. Case in 1904:

This drawing was published with the following caption:

Figs. 1-3 Squalodon atlanticus Leidy
1. Portion of the right side of a jaw containing three teeth. Shiloh, N. J. Acad. Nat. Sci., Phila.

The caption says New Jersey, which is where Leidy’s original 1856 specimens were found. Leidy’s material also included 3 teeth, as does the specimen in the figure. It certainly seems that this specimen represents Leidy’s material, and would therefore be the type specimen of Squalodon atlanticus. This specimen is still in the collections of the Academy of Natural Sciences in Philadelphia, with the number ANSP 11221 (I’m showing it upside down, because that’s how Case oriented his figure):

It was apparently accepted after 1904 that ANSP 11221 was the S. atlanticus holotype, and later referrals to the species used this specimen as a reference. For example, in their list of ANSP type specimens, Spamer et al. (1995) included the following note (with the figure number a reference to Case, 1904):

“The holotype teeth [of Squalodon atlanticus] in jaw fragment are illustrated in [Plate 10] fig.1.”

There is, of course, a problem here. Leidy originally mentioned 3 teeth; he didn’t say that they were in a jaw fragment. Cope reiterated this point (”only three true molars”), while later specifying the presence of maxilla fragments in his new specimen from Charles County. This suggests that the specimen above is not Leidy’s specimen.

It turns out that ANSP 11221 is not alone in the ANSP collections. It is stored with ANSP 11220, shown below:

But now what do we have? Two maxilla fragments, one from each side, with seven teeth in place (one tooth has a broken crown, so it’s not visible in side view). That’s exactly what Cope described as his second specimen, the one from Maryland that was clearly not the S. atlanticus holotype! Taking away a little of my thunder, it turns out that the specimen label with ANSP 11220 and 11221 specifies that they are, in fact, from Charles County, Maryland. It turns out that the allocation of this specimen as the type was an error that traces its origin to Case’s incorrect figure caption in 1904. The error was not caught until 1998.

So what about Leidy’s original three teeth, the real type S. atlanticus? It turns out that they are also in the ANSP collections (ANSP 11217-11219), correctly labeled as coming from Shiloh, New Jersey:

So, mystery solved … except for one thing. I said at the beginning of this post that in my opinion the name S. atlanticus should not be used, at least for new material. This is an important point, because over the years people have been calling new specimens Squalodon atlanticus because they are similar to Cope’s specimen. The problem is that there is no way to prove that Cope’s specimen is the same species as Leidy’s S. atlanticus (and, by definition, if you call something Squalodon atlanticus you’re saying it belongs to the same species as Leidy’s teeth).

In my opinion different species of Squalodon cannot be reliably identified based on their teeth alone. When Leidy and Cope were describing these specimens there were no known squalodont skulls, only teeth and bone fragments, so at that time Cope’s referral was perfectly reasonable (after all, S. atlanticus was the only known North American squalodont species in 1867). But since then there have been a number of skulls found, and it turns out that the characteristics that supposedly made S. atlanticus a unique species (mostly features of the roots and enamel) in fact can vary within the mouth of a single individual. In 2003 I referred S. atlanticus to Squalodontidae incertae sedis, meaning that while I believe it is a squalodont, it’s impossible to tell for certain which genus or species it should be assigned to (I could have also designated it a nomen dubium). With this definition, newly collected squalodont remains should not be assigned to S. atlanticus, unless it can be demonstrated that it is the same species as Leidy’s tooth (and currently there is no known way to do that).

This work was part of my doctoral dissertation at LSU, and was later published in Jeffersoniana (available from the VMNH online store):

Dooley, A. C. Jr., 2003. A review of the Eastern North American Squalodontidae (Mammalia: Cetacea). Jeffersoniana No. 11, 26 p.


Case, E. C., 1904. Systematic Paleontology, Miocene, Mammalia, in Clark, W. B., Shattuck, G. B., and Dall, W. H., The Miocene
Deposits of Maryland, Maryland Geological Survey Miocene Volume, Baltimore, Johns Hopkins Press, 3-58, pls. X-XXVI.
Cope, E. D., 1867a. [Four extinct species of Mammalia, which were discovered by Jos. T. Thomas, in the Miocene deposits of the Yorktown epoch in Charles Co., Maryland.] Proceedings of the Academy of Natural Sciences of Philadelphia, 19(4):131-132.
Cope, E. D., 1867b. An addition to the vertebrate fauna of the Miocene period, with a synopsis of the extinct Cetacea of the United States. Proceedings of the Academy of Natural Sciences of Philadelphia, 19(4):138-156.
Leidy, J., 1856. Notice of remains of extinct vertebrated animals of New Jersey, collected by Prof. Cook of the State Geological Survey under the direction of Dr. W. Kitchell. Proceedings of the Academy of Natural Sciences of Philadelphia, 8(6):220-221.
Spamer, E. E., E. Daeschler, and L. G. Vostreys-Shapiro, 1995. A study of fossil vertebrate types in the Academy of Natural Sciences of Philadelphia. The Academy of Natural Sciences Special Publication 16, iv+434 pp.
This entry was posted in Chesapeake Group and tagged . Bookmark the permalink.

8 Responses to What’s the problem with Squalodon atlanticus?

  1. Brian Beatty says:

    So how do these specimens differ from the European specimens?
    Would it be feasible to see them when in England for SVP?

  2. Alton Dooley says:

    Most of the European specimens are in France and Italy; I’m not aware of any in Britain. Squalodon bariensis from France is very similar to Squalodon calvertensis (the first American taxon based on a skull), although I’m not ready to say they’re conspecific (if they are, S. bariensis has priority).

    Pilleri named a bunch of Italian taxa which I think are probably not valid (they probably represent only 1 or 2 species at most). Nearly all the specimens are quite similar to each other; S. barbara from the Caucausus does seem a little different (it has a different tooth count). To my knowledge there is nothing like S. whitmorei in Europe.

  3. Alton Dooley says:

    Incidentally, I think Cope’s Charles County specimens are demonstrably different in small ways from S. calvertensis, and so far I haven’t seen anything similar figured from Europe.

  4. Doug says:

    Perhaps the same applies to all those dinosaurs named from just teeth back in the mid 19th century. I know that differences in teeth are sufficient to declare new species of lane mammals (well, most of them), but it seems you just demonstrated that the same doesn’t apply to marine mammals. I always thought the skull was the only diagnostic part, but i guess it helps to have the actual skull. Oddly, Palagiarctos thomassi is a marine mammal known from just a small collection of teeth and part of the lower jaw.

  5. Alton Dooley says:

    In general I’d agree that it’s not a good idea to erect new species based only on teeth, with the exception of certain groups of mammals, although in many groups teeth are useful to some higher level of classification (family, for example). I’m not as sure about dinosaurs, although I’ve been told that Allosaurus and Torvosaurus teeth (both theropods that are found in the Morrison) are quite distinct from each other.

    I can’t really fault the early workers for naming all those taxa, however. Individual variation and its relationship to species was just starting to be understood (and still isn’t fully understood). Moreover, often teeth were the only things that were available at the time. My attitude is that you should not ignore the material just because it’s not very good; if we did that there would be no vertebrate paleontology at all.

  6. Doug says:

    Yeah, maybe i was a bit harsh. They didn’t know any better. But connecting remains to isolated teeth has been troublesome for dinosaurs. Like Astrodon. A lot of material has been referred to it, and the whole taxonomy has been a mess. Pleurocoelus remains from Texas, a dinosaur thought to maybe be synonymous with Astrodon, has been christened Paluxysaurus.

    Perhaps theropod teeth hold more promise than those of herbivores, seeing as Phil Curry can identify a theropod based on it’s teeth.

  7. Brian Beatty says:

    Yes, I think identifying taxa from teeth, or even toothrow lengths, like some people still do to a small degree, is worrisome. What is more of a problem is simply our lack of understanding of intraspecfic variation, even among modern species, that is necessary to recognize what can be clearly diagnosed as a taxon, and what is polymorphism. If more people did population aggregation analyses of taxa, including fossil taxa, for which we had decent samples, I think we would be closer to having real estimates of past diversity.
    But I must agree, we cannot blame those in the past for their naming… one should read the first 100 pages of Osborn’s titanothere monograph to gain a better understanding of how many of these guys came up with their taxonomic rationale, it is very telling.
    And in reality, we know far fewer species of animals than their must have been in the vastness of time.

  8. Alton Dooley says:

    Brian said:

    “If more people did population aggregation analyses of taxa, including fossil taxa, for which we had decent samples…”

    I had to laugh when I read that (as I’m sure you were doing also, Brian), because of course vertebrate fossils rarely have large sample sizes. Of course, this is all too rarely done even with modern vertebrates for which we do have potential large samples (or, for that matter, fossil invertebrates).

    That said, over the next few weeks I’ll be making some posts about vertebrates for which we do have pretty good sample sizes (just a teaser).

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