As a teaser for a forthcoming paper by Morgan Churchill and myself, I thought I’d introduce a (short) new series of posts (fewer than the last series, I promise). As the publication is not out quite yet, I thought I could at least give an introduction to the extinct “killer” walrus from the Sharktooth Hill Bonebed.
The Sharktooth Hill bonebed in Kern County, California is a widespread horizon within the Round Mountain Silt member of the Temblor Formation. It’s exposed near Bakersfield, California, and is middle Miocene in age. It’s approximately 10-50 cm thick, generally lacks calcareous invertebrate fossils, but is extraordinarily rich in teeth and bones of sharks, bony fish, birds, sea turtles, pinnipeds, dolphins, sperm whales, baleen whales, and occasionally sea cows, desmostylians, and terrestrial mammals. I visited Sharktooth Hill several times as a high school student, trying to find “local” vertebrate fossils – digging well through the night in the trenches with tiki torches and a headlamp. At many localities frequented by amateur fossil collectors, the bonebed is exposed on a hillside and a large linear scar follows the position of the bonebed, dug out by collectors removing overburden to get to the fossil layer. Amateur fossil collectors have done so much digging that a trench reminiscent of World War 1 battlefields encircles many hills in the region where the bonebed is exposed. Although some collectors will spend days at a time digging through overburden - admittedly backbreaking work - some decide to risk it and tunnel into the trench to get at the bonebed. Some collectors have paid for this tactic with their lives: on my first visit in 2002, a cross was placed at one of the localities where a collector had tunneled in about ten feet and was killed when the hillside slumped down onto him; it took the authorities several days to dig out his body. The rest of the Round Mountain Silt is mostly barren with respect to vertebrate fossils, not only explaining the attention given by collectors to the bonebed itself – but also suggesting a “unique” environment temporarily persisted in order to concentrate vertebrate remains. A number of strange biologic explanations have been offered, including red tides, extensive shark predation, and even a marine mammal calving ground. Several authors have quite rightly scrutinized these biologic explanations, and have suggested sedimentologic processes as a cause (Mitchell, 1966; Prothero et al., 2008; Pyenson et al., 2009). These studies have specifically suggested that a depositional hiatus (slowdown in the accumulation rate of sediment) permitted marine vertebrate remains to be concentrated on the seafloor. I have some minor taphonomic reservations, but those are best discussed another day.
One of the Sharktooth Hill localities, wife for scale.
According to Barnes (1976), the Sharktooth Hill bonebed is the most extensively studied and richest marine mammal locality in the eastern North Pacific; a faunal list compiled by amateur collectors can be viewed here, and it includes roughly 140 vertebrate taxa. Some of the species on the list are not yet described or published (“Neotherium ernsti”, for example) and other taxa are based on old identifications and may not be borne out in the long run (aff. Herpetocetus). Regardless of issues pertaining to the taxonomic identity of some fossil vertebrates, the ballpark number is probably accurate. It’s also fairly spectacular: I recently tallied up fossil vertebrates from the Purisima Formation, and there are roughly 70 taxa present – still impressive as hell, but not quite as gargantuan as Sharktooth Hill. Depending upon whose publication you look at, there are anywhere from seven (Barnes, 1972; Barnes and Hirota, 1995) to four pinnipeds present (Deméré et al., 2003). Papers by L.G. Barnes and colleagues list several desmatophocids, including Allodesmus gracilis, Allodesmus kelloggi, Allodesmus kernensis, Desmatophocine B, and Desmatophocine C in addition to the imagotariine walruses Neotherium mirum and Pelagiarctos thomasi. According to Deméré et al. (2003), only four taxa are present – Allodesmus kernensis (with A. kelloggi and A. gracilis subsumed as junior synonyms), an indeterminate desmatophocid (Desmatophocine B), and the two walruses. While it’s nowhere near as diverse as the cetacean assemblage from the same locality, it’s fairly comparable with other fossil pinniped assemblages from the eastern North Pacific.
The skeleton of Allodesmus kelloggi as exposed in the field. From Mitchell (1966).
In 1980, future chief preparator of the Los Angeles County Museum of Natural History (LACM) discovered a curious chunk of bone with teeth at Sharktooth Hill. Several years later, he brought it in to LACM and showed it to Dr. L. G. Barnes (colloquially known as ‘Larry’ within the field), and insisted that it was the piece of a snout of some extinct mammal – it even had two small holes which look like nostrils to the uninitiated. Barnes kindly pointed out that those were mental foramina on the “chin” end of a very large jawbone of a pinniped. Larry and Howell enthusiastically recalled this whole story for Morgan Churchill and I when we sat at the very same table last January, thirty or so years later (Larry Barnes has an incredible, near photographic and certainly encyclopedic memory of marine mammal fossil specimens). Howell Thomas donated the fossil for study, and within a few years was hired as the Chief Preparator, and Barnes began to study the specimen. At the time, the marine mammal assemblage was already enormous, and the pinniped assemblage well documented by hundreds of specimens. Most of the fossils could be assigned to the large seal-like Allodesmus, although a single jaw described by Barnes (1972) as “Desmatophocine B” didn’t appear to be referable. “Desmatophocine B” was probably similar to Allodesmus, which has a long narrow skull, enormous eye sockets, single-rooted teeth, and a relatively large body. Furthermore, we know Allodesmus retained the ability to rotate its hindflippers forward for sea-lion like terrestrial locomotion, and it was probably a sea-lion like underwater “flyer”. Numerous small pinniped elements appeared to be similar to a handful of elements described by Remington Kellogg (1931) as Neotherium mirum.
Skulls of Allodesmus (left) and Neotherium (right) roughly to scale. From
Barnes and Hirota (1995) and Kohno et al. (1995).
Neotherium was an enigma for over 60 years, and it wasn’t until more complete remains of the early walrus Imagotaria downsi were recovered from the Santa Margarita Sandstone near Santa Cruz, California, that Neotherium began to make sense. Imagotaria was a sea lion-like walrus that lived about 9-12 million years ago – a bit younger than the 15-16 Ma Sharktooth Hill Bonebed – and by the close of the 1970’s was known by a number of well preserved skulls and partial skeletons from Santa Cruz County. Fossils of Neotherium, although never as common as Allodesmus, continued to trickle in from the bonebed and were referred to Neotherium piecemeal, one or two bones at a time by Mitchell (1961), Mitchell and Tedford (1972) and Repenning and Tedford (1977). By the 1980’s, Barnes had amassed a collection of nearly every skeletal element of Neotherium, identifiable as miniature and slightly more primitive versions of that found in Imagotaria – including partial skulls and several mandibles (eventually a complete skull was published by Kohno et al. 1995). Barnes has been for many years working on a monograph on Neotherium – I’m looking forward to seeing it published.
The holotype of Pelagiarctos thomasi. From Barnes (1988).
Howell Thomas’ mystery jawbone appeared more similar to Neotherium relative to Allodesmus, with the exception of its comparably gigantic size as well as having a fused intermandibular joint (mandibular symphysis) and deep grooves on the sides of the canines. Eventually, several isolated teeth that were similar to Neotherium, but several times larger in size – were discovered from the bonebed. Some of these teeth even fit right in to the tooth sockets in the mandible fragment. Barnes published the fossils in 1988 and described them as Pelagiarctos thomasi, the species name honoring Howell Thomas. The genus name Pelagiarctos refers to the primitive dental anatomy, as ‘arctos’ refers to bears, the traditional sister taxon of pinnipeds (the root arctos is frequently used in pinniped genus names – Arctocephalus, Phocarctos, Hydrarctos, Pteronarctos, etc.), as well as the inferred pelagic ecology of the animal.
The isolated teeth referred to Pelagiarctos by Barnes (1988).
Several aspects of the anatomy of Pelagiarctos, although based on scant material, suggested a different approach to feeding in this fossil walrus relative to other Sharktooth Hill Pinnipeds. The teeth of Pelagiarctos were huge – very robust canines, and postcanine teeth with multiple large cusps and sharp crests. He likened the premolars and molars to those of modern hyenas and extinct borophagine dogs, two groups which (by observation or inference) crack and ingest bones, suggesting that Pelagiarctos had dental adaptations for large bite forces related to feeding on large prey items. Furthermore, the robust mandible and fused symphysis further suggested high bite forces. Barnes (1988) additionally noted that Pelagiarctos is very large and numerically rare in the Sharktooth Hill Bonebed – only known by five teeth and a mandible fragment at the time of his study, as opposed to the hundreds of specimens known of other pinnipeds such as Allodesmus and Neotherium. This suggested to Barnes that Pelagiarctos was rare in California waters during the middle Miocene, further supporting his hypothesis that it was an apex predator (apex predators at the top of the food chain can never be very abundant because they rely on a constant stock of abundant prey items). Barnes further postulated that the type specimen was a male, as it had proportionally large canines; modern and fossil pinnipeds are sexually dimorphic, including early walruses like Neotherium, Imagotaria, and Proneotherium. One of the canines in the holotype is broken and polished down, suggesting the tooth had been broken and worn down after continued use in life – damage which Barnes attributed to male combat, which occasionally results in such damage in modern pinnipeds. Furthermore, Barnes identified some of the fossil teeth as males because they fit right into tooth sockets on the type specimen, and those that didn't were of similar size.
As a result of these hypotheses, numerous fanciful reconstructions of Pelagiarctos have been produced by paleoartists (fanciful depictions can be seen here, here, and here). and Pelagiarctos has achieved the nickname "killer" walrus by some enthusiasts. But what do we really know about Pelagiarctos? Stay tuned...
Barnes LG (1972) Miocene Desmatophocinae (Mammalia: Carnivora) from California.
University of California Publications in Geological Sciences 89: 1-69.
Barnes L.G., 1976, Outline of eastern Northeast Pacific fossil cetacean assemblages:
Systematic Zoology, v. 25, p. 321–343,
Barnes LG (1988) A new fossil pinniped (Mammalia: Otariidae) from the middle Miocene Sharktooth Hill Bonebed, California. Contributions in Science, Natural History Museum of Los Angeles County 396: 1-11.
Barnes LG, Hirota K (1994) Miocene pinnipeds of the otariid subfamily Allodesminae in the North Pacific Ocean: Systematics and Relationships. The Island Arc 3: 329-360.
Deméré TA, Berta A, Adams P (2003) Pinnipedimorph evolutionary biogeography. Bulletin of the American Museum of Natural History 13: 32-76.
R. Kellogg. 1931. Pelagic mammals of the Temblor Formation of the Kern River region, California. Proceedings of the California Academy of Science 19(12):217-397
Kohno N, Barnes LG, Hirota K (1995) Miocene fossil pinnipeds of the genera Prototaria and Neotherium (Carnivora; Otariidae; Imagotarinae) in the North Pacific Ocean: evolution, relationships, and distribution. The Island Arc 3: 285-308.
E. D. Mitchell. 1961. A new walrus from the imperial Pliocene of Southern California: with notes on odobenid and otariid humeri. Los Angeles County Museum Contributions in Science 44:1-28
Mitchell ED (1966) The Miocene pinniped Allodesmus. University of California Publications in Geological Sciences 61: 1-46.
Mitchell ED, Tedford RH (1972) The Enaliarctinae: a new group of extinct aquatic Carnivora and a consideration of the origin of the Otariidae. Bulletin of the American Museum of Natural History 151: 203-284.
Repenning CA, Tedford RH (1977) Otarioid seals of the Neogene. Geological Survey
Professional Paper 992: 1-87.
D. R. Prothero, M. R. Liter, L. G. Barnes, X. Wang, E. Mitchell, S. McLeod, D. P. Whistler, R. H. Tedford, and C. Ray. 2008. Land mammals from the middle Miocene Sharktooth Hill Bonebed, Kern County, California. New Mexico Museum of Natural History and Science Bulletin 44:299-314
Pyenson ND, Irmis RB, Lipps JH, Barnes LG, Mitchell ED, et al. (2009) The origin of a
widespread marine bonebed deposited during the Middle Miocene Climatic Optimum.
Geology 37: 519-522.