Saturday, March 27, 2010

Pinniped ancestry: the "Oligocene Seal" from South Carolina

One of the most fascinating aspects of the study of marine mammals are the origins of the land to sea transition in various groups. Often marine mammals are so adapted for marine existence, that it is difficult to identify what group (fossil or modern) they descended from. Such has been the case with the origin of whales, for example. While I'm certainly interested in cetaceans, there has been a ton written about them, and I'm currently more intrigued by pinnipeds.

While modern cetaceans don't necessarily look like their artiodactyl relatives, pinnipeds generally retain enough features that allows them to be readily identified as carnivoran mammals, straight down to the dog-like bark of California sea lions. Certainly, this should make identifying the terrestrial progenitors of pinnipeds an easier task. Or does it?

The prevailing opinion throughout much of the 20th century was that the Pinnipedia are diphyletic; the sea lions and walruses formed the "Otarioidea", who descended from the Ursidae, and the true seals (the Phocidae) descended from mustelid (or musteloid) ancestors. In 1973, Ed Mitchell and Richard Tedford described Enaliarctos mealsi from the Early Miocene Pyramid Hill member of the Jewett Sand in Kern County, California. Enaliarctos still bore an ursid-like shearing dentition, which is absent in all modern pinnipeds (which have generally homodont postcanine teeth). Enaliarctos was purported to be the common ancestor of the Otarioidea.

Later work (which at some point will be detailed on here) by Andre Wyss (UC Santa Barbara) and Annalisa Berta (San Diego State University), proposed a drastically different phylogeny of the pinnipeds, which suggested that pinnipeds were instead monophyletic, and that walruses (Odobenidae) were more closely related to the true seals (Phocidae), forming a new group, the Phocomorpha. Dozens of molecular phlyogenetic analyses have unequivocally supported pinniped monophyly, chucking the old diphyletic view out the window. The major tenets of the new studies are the following: 1) all pinnipeds descended from a common ancestor; 2) Enaliarctos is basal to the Otariidae, Odobenidae, and Phocoidea; 3) the Otarioidea is paraphyletic, as odobenids and desmatophocids are more closely related to the phocids; and 4) pinnipeds were derived from an ursid (bear) like ancestor sometime during the Late Oligocene, probably in the Northeast Pacific, where the oldest pinniped fossils are known.

Putative seal femora from the Late Oligocene of South Carolina, from
Koretsky and Sanders (2002).

In 2002, Irina Koretsky and Al Sanders reported on some partial femora from the Late Oligocene of South Carolina. According to Koretsky and Sanders, these femora are most similar to those of extant phocids, and list several features they share in common. However, they argued that the Late Oligocene age of these, as well as the occurrence of these in deposits of the Atlantic Coastal Plain, indicate that true seals were already present in the Atlantic Ocean when the very primitive Enaliarctos was just appearing in the Pacific Ocean. Thus, the monophyletic origin of pinnipeds was not supported by this fossil, and it appeared that true seals evolved in the Atlantic separately from the Otarioidea, which are (during the Oligocene and the Miocene) restricted to the North Pacific.

There are, of course, some issues with this study and its inherent implications, irrespective of the incomplete nature of the fossils. For starters, the fossil record is notoriously crappy; in fact, the Early Oligocene marine record is the poorest of all with regards to marine mammal fossils. Very few cetaceans are known from the Early Oligocene worldwide, for example, primarily due to the scarcity of marine rocks for this time, due to low sea levels caused by the Eocene-Oligocene climate crash. Pinniped fossils just "show up" after sea levels rise and deposit more marine sediments during the late Oligocene, in both the Pacific (Enaliarctos) and Atlantic (Phocid femora). So - it is entirely possible for even more primitive pinnipeds to be found in earlier sediments (or, as "lazarus taxa" in Late Oligocene rocks).

The completeness of these fossils requires additional scrutiny. Some of Koretsky's other work focuses on the major elements of the fore- and hind-limb of true seals (and lower jaws) but generally placing low importance on cranial material (Koretsky and Ray, 2008). Koretsky and Sanders (2002), however, did not compare these specimens with femora of Enaliarctos, or basal odobenids such as Proneotherium and Neotherium (or an unnamed basal odobenid described by Naoki Kohno early on, ~1990). Additionally, now that the putative stem-pinniped Puijila darwini has been described, it's femora should be compared with these specimens as well. These could very well turn out to belong to something more like Enaliarctos or Puijila that we don't yet have a record of in the Atlantic. However these fossils are interpreted, their Late Oligocene occurrence (if the provenance is accurate) is intriguing, and further field investigation of Late Oligocene sedimentary rocks of the Atlantic Coastal Plain (and elsewhere!) should be considered.


Koretsky, I.A. and A.E. Sanders, 2002. Paleontology of the Late Oligocene Ashley and Chandler Bridge Formations of South Carolina, 1: Paleogene pinniped remains; the oldest known Seal. Smithsonian Contributions to Paleobiology 93: 179-183.

Koretsky, I.A., and Ray, C.E. 2008. Phocidae of the Pliocene of Eastern USA. In: C.E. Ray, D. Bohaska, I.A. Koretsky, L.W. Ward, and L.G. Barnes (eds.), Geology and Paleontology of the Lee Creek Mine, North Carolina, IV. Virginia Museum of Natural History Special Publication 14: 81-140.


Morgan said...

well, the biggest problem with the with pinniped nonmonophyly is that you have to absolutely ignore every single genetic analysis EVER DONE for it to work. That's a little hard to swallow.

IIRC, Koretsky also didn't really compare the partial bones to a wide variety of carnivorans, fossil and modern. It would not surprise me at all if this fossil had nothing to do with pinnipeds.

Robert Boessenecker said...

Yes, that is definitely the single worst problem with pinniped diphyly or paraphyly. Which, if you read some more recent work by Koretsky (i.e. Koretsky and Barnes 2006), ignoring molecular analyses is effectively what they've done.

Yet another thing I forgot to elaborate on: I wouldn't be surprised either if those femora were from some completely different carnivoran group.

J. Velez-Juarbe said...

I am not as well acquainted with fossil pinnipeds as with other marine mammals (though maybe I should), but it seems that Enaliarctos had a fairly derived morphology for a Late Oligocene critter. What I want to say is that finding Late Oligocene Enaliarctos and the phocid femur on opposite sides of North America, doesn't necessarily imply different origins (diphyly). It can also mean a much earlier divergence between the two groups and rapid dispersal (Caribbean seaway?!). There is definitely a need to get better samples from rocks of the right age (Early Oligocene).
There are indeed some discrepancies as to the provenance of the South Carolina fossil. Also there were other morphological features of those femora that could have been discussed in the paper.

Morgan said...

Indeed...It wouldn't surprise me if Enaliarctos is more closely related to otaroids than to phocids. My guess is that the true ancestral pinniped are going to be Early Oligocene in age (and possibly arctic in distribution).

Brian Lee Beatty said...

I agree with Jorge, this is not a decisive fossil, and I also agree with Morgan and Bobby that this might not be a phocid at all. I remember when Irina was showing me this specimen and I was hard pressed to get a clear discrete character from her that she could say made it a phocid and not another carnivoran - but, either way, there is much, much more that needs to be found. Speaking of which, I hope to be tackling some Late Oligocene of the NE very soon... heheheheh...

Robert Boessenecker said...

It'll be difficult to find more Arctic critters aside from Puijila, I'll wager, but it's a fascinating possibility - an early Oligocene Arctic origin could explain (as Rybcynsky et al discussed) could explain the early phocid/everything-else split.

Brian: Late Olig. of the Pacific Northwest?

Jorge: Perhaps there are more Early Oligocene rocks in New Zealand (not helpful if pinnipeds didn't reach the S Pac. until the Miocene, though) and Japan; some rapidly subsiding basin from that time must have a good record of it.

Robert Boessenecker said...

I believe the Alsea Fm. in Oregon/Washington (type area of Simocetus, and Mark Uhen's new edentulous mysticete from the Emlong collection that's more basal than Eomysticetus) is Early Oligocene; there are plenty of late Eocene rock units (which have yielded potential archaeocete teeth) and more than enough Late Oligocene units; I bet there's something Early Ol. up there.

J. Velez-Juarbe said...

There are some neat Early Oligocene formation in Puerto Rico, those and the Late Oligocene ones are actually where I collect most vertebrate fossils. The Early Olig. was colder even in the Caribbean; so who knows, it might be a possibility.

As for the Pacific NW, there sure must be some Early Oligocene there (other than the Alsea), but it seems so hard to look for fossils there. I was really disappointed last October when I was in Oregon. Emlong must have been really persistent to be able to find all the material he collected. Fortunately, I think that some of his unprepared material at the USNM will finally see the light of day in the coming years.

Erich Fitzgerald said...

Koretsky's 'Late Oligocene phocid': hmmm. Pretty ambiguous to me given that there is a 50/50 chance it is derived from late Neogene/Quaternary strata that unconformably overlie the Paleogene sequence in Sth Carolina.

As for SW Pacific options for early pinnipeds: no firm pre-Late Miocene record. I currently have a student writing up the Mio-Pliocene 'monachine' material from Australia (reported in Fordyce & Flannery, 1983), and I am working on a recently collected Late Miocene skull from this part of the world.

Robert Boessenecker said...


Yes, the stratigraphic provenance of those specimens is another issue, at least Jorge has mentioned the same to me. Another issue is this: those units are very thin, and during my thesis fieldwork, I've documented marine bioturbators reworking bones and other clasts up to two meters below their original horizon, and it very well could be much more than that. So, "backwards" reworking is possible as well, especially in cases of thin stratigraphic units.

Glad to hear that someone's working on southern pinnipeds again! Remind me about this and I'll email you pictures - there's a collector in New Zealand who's collected some pinniped material (including a complete phocid braincase) from Taranaki - I'm not sure how much material is already in museum collections, but F.Y.I.