Thursday, January 12, 2017

2016 in review: Advances in marine mammal paleontology


Welcome to the fifth annual post chronicling the year's publications on the marine mammal fossil record. It's only a couple weeks into January, so I'm doing pretty good this year. Note that I include anything that was published first in 2016 - some of these already have 2017 publication dates as they were published "Online Early" in 2016 but assigned final volume/issue/page numbers since Jan 1. Because I do this, I do NOT transfer summaries from articles published online in 2015 but published in print in 2016 from last year's post to this one; I don't have the time, and you can just relax.

The usual disclaimer applies - I may have forgotten something. If so, leave a comment, send me an email, or facebook message - and I'll be sure to correct it. Note that Blogger has a nasty habit of deleting spaces and de-italicizing anything within hyperlinks, and I've tried re-adding spaces but won't bother with italicizing anything in the hyperlinks. I do this in my spare time, so don't you dare complain to me about it.



The fossil record of marine mammals is heavily biased toward the northern hemisphere - making localities like New Zealand (where I did my Ph.D.), Australia, Antarctica, and especially South America ripe targets for paleontological exploration. Recent efforts have yielded the first archaeocetes from South America, one of the only Aquitanian (earliest Miocene) marine mammal assemblages anywhere on earth, and the embarrassingly productive regions of the Pisco basin (Peru) and the Bahia Inglesa area (Chile). Marine mammals are today linked to cold water temperatures and high primary productivity - making good marine mammal watching (typically) latitudinally constrained. Similar biases affect the fossil record of marine mammals - marine mammals are geologically young (Ischyrorhynchus vanbenedeni, a longirostrine relative of the modern Amazon river dolphin Inia). Recent analyses suggest Ischyrorhynchus may actually be a platanistid - and the new rostra referred to this taxon is really reminiscent of Platanista in some regards. A scrappy iniid skull is also described, as well as a pair of vertebrae referred to Zarhachis flagellator, a species known only from the Chesapeake Group of the eastern USA. Other remains include a couple of partial baleen whale skulls. The authors suggest that the river dolphin Saurocetes may be a synonym of Ischyrorhynchus, and highlight the possibility of a south American record of Zarhachis - which awaits further field study.




 
Multiple groups of marine mammals have made the plunge (excuse the pun) and returned to life in the sea. At least two of these - the sea cows and cetaceans - have become so modified for aquatic life that they cannot support their weight on land. Others, like pinnipeds and sea otters, return to land to mate or for rest. While modern animals can simply be observed by someone with a notebook and binoculars, fossilized marine mammals have been dead for a long time and don't exactly move around a whole lot - so paleontologists interested in the evolution of secondarily aquatic tetrapods have to examine anatomical (chiefly, skeletal, owing to the nature of the fossil record) evidence. Much of the research on aquatic habits have focused on fore/hindlimb functional morphology and flexibility of the vertebral column; other important studies have examined the histology of postcranial bones as increases in bone density often happen early in semiaquatic lineages. However, few - if any - studies have examined changes in ribs (aside from chopping them up for histology, of course!). This study alleviates this problem by compiling a dataset of rib thickness measurements, intended to serve as a rough proxy of bending strength - the idea being that fully aquatic species which never return to land would have ribs that would break under the animal's body weight if it laid down on a hard surface out of water (as often happens with stranded whales and dolphins). They found that, unsurprisingly, all modern cetaceans have relatively weak ribs which cannot support their weight on land, and that pinnipeds and otters do have strong ribs (which matches modern observations). Surprisingly, the early semiaquatic archaeocete whale Ambulocetus may not have been able to support its weight on land without breaking its ribs - surprising, and suggestive that Ambulocetus was obligately aquatic. Further surprising are their findings of desmostylians - Desmostylus could support its weight on land, but Paleoparadoxia and Neoparadoxia could not. This study needs to be tested using CT data, but as simple as the current study is it is provocative and certain to generate some interesting future discussion.




Sirenians are relatively common in the fossil record as far as marine mammals are concerned and include two major families: the Trichechidae (manatees) and Dugongidae (dugongs and sea cows); the latter is quite diverse, and includes a host of species which managed to disperse worldwide including the modern Dugong and the recently extinct Hydrodamalis gigas (Steller's sea cow). Dugongids diversified early in the North Atlantic and Tethys, later dispersing into the Pacific in the early Miocene. This study reports a new species of Prototherium, Prototherium ausetanum, from the middle Eocene of Spain. The species is characterized by a rather elongate and narrow cranium and a rostrum that is not deflected much. Cladistic analysis links this species with Eotheroides aegyptiacum, and suggests that either Prototherium may not be monophyletic or perhaps that the cladistic matrix needs work; similarly, Halitherium and Eotheroides (other early dugongids) were also found to be problematic genera.


 
Remingtonocetids are among the smallest of the archaeocetes, previously known from the middle Eocene of Indo-Pakistan. Remingtonocetids are unique in that they are one of the only groups of archaocetes that likely forms an actual clade - pakicetids, ambulocetids, protocetids, and basilosaurids (in order of increasing aquatic adaptations and closeness on the tree to modern cetaceans) commonly plot out as paraphyletic 'families'. Remingtonocetids are relatively small and otter-like in body form with relatively shorter legs than the decidedly more terrestrial long-legged pakicetids and powerful tails, yet possess a greatly elongated rostrum, an enormous saggittal crest, and dorsally placed orbits, not unlike a crocodilian in general form. Indeed, remingtonocetids have been inferred as ambush predators similar to crocodilians, and I colloquially refer to them as "furry crocs" or "otter-crocs". This new study by Ryan Bebej and others reports a new remintonocetid from the middle Eocene of Egypt - extending the range of this clade from the north central Tethys Sea (Indo Pakistan) a bit further to the southwestern Tethys (north Africa). The new species Rayanistes afer lacks skull parts, and so this was only possible owing to earlier work describing more complete skeletons of Remingtonocetus - naturally building upon some of Bebej's Ph.D. work. The holotype consists of a pelvis and sacral vertebrae, a caudal vertebra, and a well-preserved femur - one of the only remingtonocetid femora known. The pelvis exhibits a greatly enlarged ilium and the femur is more robust than Remingtonocetus, indicating greater musculature involved in the hindlimb power stroke; the hip socket appears to have permitted greater range of movement. Additionally the vertebral column exhibits features indicating it was more 'passively flexible' than Remingtonocetus, suggesting more efficiant swimming and paddling than in Remingtonocetus. The discovery of a remingtonocetid in Egypt 1) indicates that remingtonocetids had a greater capacity for dispersal than previously assumed (protocetids were assumed to be the first cetaceans to leave Indo-Pakistan) and 2) raises the possibility that other early semiaquatic archaeocetes (e.g. ambulocetids) may be found in Egypt as well, or similarly that remingtonocetids may be found further yet a field - which would prompt reevaluation of early whale evolution.


 
Baleen whales are of course one of my favorite research subjects - I was a bit sick of them after my Ph.D. on eomysticetids, but that's finally beginning to wear off. Quite a lot has been published in the last decade: the description of Janjucetus and redescription of Mammalodon, the discovery that aetiocetids may have had teeth and baleen, the discovery and naming of numerous eomysticetids from New Zealand and Japan, discovery of aberrant early chaeomysticetes (Whakakai, Horopeta, and the possible eomysticetid Sitsqwayk - see below), the discovery that eomysticetids may have retained teeth, major advances in mysticete phylogeny including the provocative hypothesis that the pygmy right whale is related to cetotheriids, the identification of genetic processes leading to tooth loss, clarification of mysticete hearing and olfaction, and the "cetothere revolution". This review article by Annalisa Berta and others covers all recent advances regarding the dramatic evolution of the feeding apparatus of baleen whales, and is a fine starting point for new researchers to become acquainted with baleen whale evolution (heaven forbid baleen whales attract any additional research attention...). Unfortunately this paper came out just too late to incorporate information from the Marx et al. study (see below) on evidence of suction feeding in aetiocetids.


 
A majority of the beaked whale fossil record consists of partial crania dredged from seafloor deposits without age information. The Pisco Formation of Peru is one of the few continental localities where well-preserved partial skeletons of ziphiids are preserved in abundance. This new study reports two new genera and species of ziphiids from the Pisco Formation: Chavinziphius maxillocristatus and Chimuziphius coloradensis. Chavinziphius has a facial region of the skull convergent with giant beaked whales Berardius spp., yet retains a functional homodont dentition like most odontocetes (most extant ziphiids only retain tusks). Chimuziphius does not resemble modern species but exhibits a mesorostral canal roofed over by the premaxillae, and triangular nasal bones. Two additional ziphiids from the Pisco indicate the presence of five species from the mid-late Miocene – they do not represent either genus named in this paper, nor are they referable to Messapicetus gregarius; this assemblage represents an unusually diverse ziphiid fauna. This study includes a new cladistic analysis with a new phylogenetic hypothesis – many extinct and quite bizarre ziphiids form a clade with the longirostrine and polydont ziphiid Messapicetus, called the “Messapicetus clade” (completely extinct). This new phylogenetic framework permits a reevaluation of character evolution in beaked whales. Both crown Ziphiidae and the Messapicetus clade independently trend towards reduction of dentition, increasing bone density of the facial region, elevation of the vertex, and increasing body size.



Many modern whales and dolphins occur in both the southern and northern hemispheres but rarely cross the equator - resulting in two semi-disparate populations that are somewhat or completely reproductively isolated. Warmer temperatures and lower productivity at the equator make it difficult for individual whales and dolphins to cross because of overheating and lack of food - though some larger baleen whales evidently do cross some of the time (e.g. blue whales) and there are cold-water areas of upwelling that permit intermittent cooling (Humboldt current, eastern equatorial Pacific). This pattern is called antitropicality, and it represents a classic example of allopatric speciation (speciation occurring in different areas); it's resulted in restricting gene flow between two populations of the same species (e.g. fin whales, blue whales, killer whales), and splitting of a single genus into two species (e.g. right whales, bottlenose whales, giant beaked whales). We have evidence of antitropical distributions in the fossil record of marine mammals such as the porpoise Piscolithax, known from the late Miocene of California and Peru. This study reports new specimens of the longirostrine beaked whale Messapicetus longirostris from Italy, confirming species-level separation between this and Messapicetus gregarius from Peru. This relationship suggests that these ziphiids were an antitropical species pair, inhabiting the Mediterranean and eastern South Pacific; no remains of Messapicetus have yet been recorded from the eastern North Pacific, where beaked whale fossils are relatively rare.



Balaenopterids (aka rorquals) are the baleen whales with throat pleats, such as minke, fin, humpback, and blue whales; these are perhaps my favorite group of cetaceans. Balaenopterids are quite common in late Miocene and Pliocene marine sediments worldwide, but until the last decade or so have remained very poorly known; the fossil record of balaenopterids has been plagued by poor taxonomic practices dating from the 19th century to the 1970s. P.J. Van Beneden described a number of Pliocene balaenopterids from Belgium, but designated no holotypes and furthermore assembled chimera skeletons composed of isolated, non-associated bones based on preconceived notions of what each species should look like. Using modern diversity as a guide, he named different species (all probably nomina dubia) for their sizes supposedly mirroring modern species: Balaenoptera musculus ("B. sibbaldina", "B. musculoides"), Megaptera novaeangliae ("Megapteropsis robusta"), and B. acutorostrata ("B. rostratella") and further re-named some of these species. In Italy, individual specimens have been designated too many names, and all Pliocene balaenopterids were once rolled into modern Balaenoptera acutorstrata (none actually belong to minke whale-like species). This taxonomic paralysis has plagued the study of fossil balaenopterids, and crania associated with mandibles and earbones are needed to evaluate old names (or, throw them out entirely). This new study reports a fragmentary but informative skull from the Pliocene of Belgium that represents a new genus, named Fragilicetus velponi. It appears to be different than all of Van Beneden's figured material (taxonomic problems aside), and anatomically is relatively primitive in comparison to other balaenopterids. Most critically, this taxon exhibits a suite of features that are intermediate between balaenopterids and gray whales (Eschrichtiidae). Gray whales and rorquals are known to be closely related, but whether gray whales are sister to or included within the balaenopteridae is a contentious issue as the modern species are very divergent in morphology. Fossils like Fragilicetus suggest that the earliest crown Balaenopteroidea may have exhibited features of both families, with later diverging members of each clade evolving gray whale and rorqual features more and more until the present.


 
Modern river dolphins were formerly grouped into a single group, the Platanistoidea, based on shared primitive features (e.g. long, narrow rostra). Later morphological and eventually molecular studies revealed that these modern dolphins (Inia, Amazon River dolphin; Lipotes, Yangtze River dolphin; Pontoporia, Franciscana; and Platanista, Indus/Ganges River dolphins) belong to different parts of the odontocete tree and thus constitute a paraphyletic grouping. A second hypothesis, first promoted by Christian de Muizon, proposed a new Platanistoidea that included extant Platanista and fossil relatives – which have variably included extinct platanistids, the squalodelphinids, allodelphinids, and in some studies, waipatiids and squalodontids. This study came on the heels of a monographic reevaluation of the Allodelphinidae – an Oligo-Miocene clade of longirostrine dolphins from the North Pacific (see Kimura and Barnes, 2016, below) – including Allodelphis, Goedertius, Ninjadelphis, and Zarhinocetus. This new paper reports an additional genus and species of allodelphinid, Arktocara yakataga, from the “middle” Oligocene of southern Alaska. This new small-bodied odontocete is known only by a braincase, and amongst the Allodelphinidae it is most similar to Allodelphis from the Jewett Sand. This study critically includes the first cladistic analysis including any allodelphinids, which confirmed the monophyly of the group and position within the Platanistoidea sister to a squalodelphinid-platanistid clade. Arktocara represents one of the earliest known crown odontocetes (if platanistoid monophyly is accepted; larger cladistic analyses indicate this group may be paraphyletic), though part of this relates to the rather poor age control for this species (29-25 Ma, late Rupelian to late Chattian).


 
The entire world's fossil record of true sea otters (excluding Holocene age specimens) could probably fit inside of a briefcase. Scrappy fossils of Enhydra have been reported from various localities of supposed Pliocene and Pleistocene age along the California coast, Arctic coast of Alaska, and England. Other giant otters include Enhydriodon from Europe, Africa, and Asia, and Enhydritherium from the Pliocene of California and Florida. In the 1960s-1970s, supposed Pliocene age of some specimens of Enhydra from Los Angeles, coupled with the occurrence of more primitive giant otter specimens (Enhydritherium), led early researchers to propose that sea otters evolved in place in the eastern North Pacific. Many discussions have revolved around the age of Enhydra reevei from England – which is an isolated tooth nearly identical to modern sea otters, and approximately 2 Ma in age. Because few anatomically informative Enhydra specimens have been found, sea otter evolution and biogeography have hinged upon the identification and age of individual occurrences. A new specimen of Enhydra collected by a buddy of mine consists of a nearly complete femur from the middle Pleistocene Merced Formation near San Francisco – a rock unit that is dated exceedingly well. Available dates indicate an age of approximately 620-670 Ka – a fifty thousand year age control, which is staggeringly narrow. This specimen prompted a reevaluation of the ages of various sea otter occurrences in the Pacific. As it happens, all supposed Pliocene and early Pleistocene specimens of Enhydra from the Pacific are actually much younger, and either middle or late Pleistocene in age, and none older than 700 Ka – with this femur being the oldest accurately dated specimen in the Pacific. Specimens from Alaska and England are early Pleistocene (~1-2 Ma) and predate the Pacific record – negating prior arguments which sought to invalidate the role of Enhydra reevei. The available record of Enhydra fossils thus support a North Atlantic or even Arctic origin of true sea otters, followed by dispersal through the Arctic into the Pacific – indicating that sea otters are recent invaders.


 
The fossil record of pinnipeds is heavily biased toward the northern hemisphere – and the north Pacific record is particularly densely sampled. One of the most famous north Pacific pinnipeds is the smelly, gigantic, bizarre, violent, and utterly charismatic elephant seal (Mirounga angustirostris). It's one of only two temperate latitude true seals in the eastern North Pacific, aside from the harbor seal. Whereas the closest relatives of harbor seals all live in the North Pacific, North Atlantic, and Arctic, the closest relatives of northern elephant seals (southern elephant seal, Mirounga leonina, and other lobodontines) are all in the southern ocean – making their evolutionary history a bit of a puzzle. Modern biogeography suggests that the modern northern elephant seal originated in the southern hemisphere with other lobodontines (antarctic seals). Unfortunately, the fossil record of elephant seals is a bit crap; fragmentary specimens indicate the presence of modern species in the mid-late Pleistocene of California and Chile – indicating that the north-south split between the species is at least a few hundred thousand years old, but not giving us a better idea of which direction. Morphological studies initially proposed that the northern elephant seal is more primitive, indicating a North Pacific origin; however, phocid seals are unknown in pre-Pleistocene rocks in the North Pacific, and the sample is densely sampled enough in my opinion that we can confidently say that true seals are recent invaders. A “new” fossil seal – a fragmentary skull collected in the 1920s and informally known as the “Waipunga seal” was studied by Dr. J.A. Berry, a zoologist in NZ but never formally published; other specimens of his were eventually published by the late Dr. Judith King, one of the foremost pinnipedologists of the 20th century. Morgan Churchill and I studied this specimen together when he visited U. Otago for an EAPSI grant. This fragmentary skull is about the size of a modern harbor seal, but evidently had a prenarial shelf and likely a small proboscis like an elephant seal, but primitively retained double rooted teeth with cusps. The bulbous crowns are shared uniquely with modern Mirounga, but the more primitive features and incomplete preservation precluded us from identifying it to the genus level, though identification to the Miroungini (a group within the lobodontines). The age of this specimen is late Pliocene, predating all fossils of Mirounga. This specimen thus represents some of the first fossil evidence for a southern hemisphere origin of elephant seals, sometime in the Pliocene – followed by a Pleistocene dispersal to the northern Pacific. The specimen further suggests that Mirounga evolved from a fairly small-bodied ancestor, perhaps only the size of a gray seal – and evolved gigantism within the last 2 million years.
 

 

This study is the last major chapter from my Ph.D. thesis and reports the new genus and species Matapanui waihao, the geochronologically oldest eomysticetid baleen whale from New Zealand. Eomysticetids are the earliest obligate filter-feeding baleen whales, were mostly toothless, but likely had a few non-functional teeth in the front of the jaws. Matapanui is known by several partial skulls with well-preserved earbones, and is somewhat older than other NZ eomysticetids like Tokarahia, Tohoraata, and Waharoa. Matapanui has some features reminiscent of northern hemisphere eomysticetids, but cladistic analysis indicates it is sister to the Tokarahia-Tohoraata-Waharoa clade – indicating that most New Zealand eomysticetids form a clade. Whereas Matapanui is not the most completely known eomysticetid, but it is known from numerous referred specimens that reveal aspects of ontogenetic and intraspecific variation. One of the specimens, a partial braincase, is from a large adult and exhibits some features that are a bit of a departure from other eomysticetid braincase anatomy including thickened nuchal crests, a greatly inflated exoccipital, and a convex (rather than triangular) supraoccipital shield. At approximately 27 Ma, Matapanui represents one of the oldest described chaeomysticetes. In addition to description and cladistic analysis, I decided to include within this paper a taxonomic review of the Eomysticetidae, essentially summarizing all published literature on the clade. This includes reporting updated geochronologic ranges for all named eomysticetids, and formally recognizing Micromysticetus spp. as eomysticetids, and suggesting that Cetotheriopsis lintianus – the type species of the “Cetotheriopsidae” - is too incomplete to be the foundation of a family-level clade, but has some features consistent with eomysticetids. Lastly, I'll add that we originally published this as Matapa waihao; Matapa means “flat face” in Maori. With prior names I found no evidence of occupied names, given the unique structure of Polynesian words. However, Matapa is a bit shorter and more simple than Tokarahia or Tohoraata, and unbeknownst to me, Matapa had already been used as the genus name for a butterfly in Sri Lanka – which I was informed of within a day or so of the paper being published. So, we went ahead and proposed the new genus name, Matapanui – meaning “big flat face” in Maori.



Basilosaurid whales are the first truly pelagic whales to have evolved - protocetids, though capable of crossing ocean basins, were evidently still tied to land for reproduction similar to seals. Protocetids are essentially unknown from the Pacific basin, though they escaped the closing Tethys sea into the North Atlantic; a possible protocetid from Peru has affinities with Basilotritus, a basilosaurid formerly thought to be a protocetid. Archaeocetes in general are quite rare from the southern hemisphere - their record includes some well-preserved skeletons from Peru, a partial skull and various teeth from New Zealand, and some scraps from Antarctica. A number of kekenodontids are known as well from NZ - late-surviving Oligocene archaeocetes. New material from the mid-late Eocene La Meseta Formation of Seymoura Island, Antarctica, reported in this study by Monica Buono and others significantly expands the record of southern archaeocetes, and includes some nice mandibles as well as isolated teeth and a nice pelvis. The fossils all represent basilosaurids, though none of the material is complete enough to identify to the genus level. This material supplements the scrappier remains reported by earlier researchers, and critically none of the material appears to represent Neoceti - the geochronologically earliest (but not necessarily most primitive) mysticete, Llanocetus denticrenatus, originates from about the Eocene-Oligocene boundary about 100 meters up section in the La Meseta from some of these specimens. New (albeit fragmentary) specimens of Llanocetus have been discovered by this same team.





Fossils of sea cows (Sirenia) are rather common owing to their dense bones that are resistant to fragmentation and abrasion. Like other marine vertebrates, sirenians are typically discovered during weathering of fossiliferous rocks exposed at the earth's surface. It's generally rare to find anything during digging if no bones are exposed at the surface - a common mistake in movies and TV shows. On occasion the improbable does happen, and this study reports an isolated rib of a dugongid sea cow from the Floresta Calcarenites Formation of southern Italy, recovered during sawing and polishing of decorative tile slabs. Ribs are not identifiable to the genus or species level, but it does extend the range of dugongids in southern Italy back to the earliest middle Miocene or latest early Miocene. Additionally, the specimen is absolutely beautiful. Similar finds include the protocetid Aegyptocetus tarfa, recovered in Eocene Egyptian limestone, and a sea cow (Prototherium, recently reported at the 2016 SVP meeting) in limestone paving slabs in a sidewalk in Spain.



Fur seals and sea lions – also known as eared seals – are numerically common in the North Pacific but arguably most diverse in the southern hemisphere. Three of the five species of modern sea lions and all but two species of fur seals live in the south. However, the fossil record of this family (Otariidae) is almost entirely restricted to California and Japan, with fossils in California extending back to the early Miocene (e.g. Eotaria). One of the few anatomically informative fossils of an otariid from the southern hemisphere is a nearly complete skull collected from Ohope Beach on the North Island of New Zealand – studied first by JA Berry and subsequently named by Australian pinnipedologist Judith King as Neophoca palatina. Neophoca is the modern genus of Australian sea lion, and the New Zealand sea lion, Phocarctos hookeri – does not have a pre-Holocene fossil record. This suggests that Neophoca had a formerly wider distribution than the current species does, inhabiting both Australia and New Zealand – or alternatively, inhabiting New Zealand prior to dispersing to Australia. However, no pinniped specialists had even examined the holotype skull since King named the species in 1983. Was the generic allocation correct? We took the holotype skull on loan in 2013 while Morgan visited New Zealand, and reconstructed it – it had shattered in the mail on return from Australia, and we painstakingly glued many of the surviving bone fragments together so that we could reassemble the skull for the first time in 30 years. Owing to cranial incompleteness of the holotype, earlier cladistic analysis failed to resolve the position of Neophoca palatina within the otariid tree. We instead used a principal components analysis to examine relationships and found that cranial measurements successfully segregated all southern otariids by genus. Neophoca palatina was confirmed using this quantitative approach to be allied with extant Neophoca cinerea. This suggests that Neophoca formerly had a wider range in temperature tolerance and that New Zealand geography may have played an important role in the evolution and dispersal of southern hemisphere otariids.





Xenorophid dolphins are a bizarre, highly derived and short-lived monophyletic group of early odontocetes (toothed whales). Xenorophids are known only from the Oligocene of South and North Carolina – principally from the lower Oligocene Ashley Formation (SC) and the upper Oligocene Chandler Bridge (SC) and Belgrade (NC) formations. In 2014, the new genus and species Cotylocara macei was described, a xenorophid which possessed a series of deeply excavated cranial sinuses and cranial asymmetry indicating that it had the ability to produce high frequency sounds like modern echolocating dolphins. But sound production is just one half of the equation – were xenorophids capable of hearing high frequency sounds? Such an ability is required for echolocation. In this study, Morgan Churchill and colleagues report a new genus and species, Echovenator sandersi, which is a somewhat smaller species of xenorophid. Echovenator lacks the derived sinuses of Cotylocara, but the well-preserved earbones of Echovenator permitted CT scanning of the cochleae – the organ of hearing embedded deep within the periotic bone. Numerous features of the cochlea, including looser cochlear coiling (tighter coiling = low frequency hearing, which characterizes mysticetes), indicates that Echovenator – and likely all other xenorophids – were capable of high-frequency hearing, and thus echolocation. Because xenorophids are the earliest diverging odontocetes, this finding indicates that echolocation evolved very early and likely represents a key innovation driving the diversification of odontocetes. Lastly, the analysis found that archaeocete whales possess adaptation for higher frequency sounds than their ancestors, indicating that mysticetes and odontocetes diverged towards high and very low frequency hearing at the Neoceti split, rather than mysticetes inheriting low-frequency hearing from archaeocete ancestors with odontocetes splitting off.


 


Barnacles are one of my favorite invertebrates, and along the rocky shores of California, they are practically everywhere; it's tough to find them here in South Carolina because they generally require hard substrate. Most barnacles - the weirdest of crustaceans - are adapted to attaching to rocky substrate and occasionally to wood, making them pests for boating/shipping - some are specialized towards "fouling" invertebrate shells, and no doubt if you go to the lobster tank at the local grocery store you'll probably find a barnacle or two on one of the crabs in there. Barnacles either attach to substrate via attaching their carbonate skeleton directly to the substrate by cement or protein (e.g. acorn barnacles), or by a soft tissue stalk (e.g. gooseneck barnacles). Certain barnacles, however, have adapted towards living on the shell of sea turtles and within the skin of whales (chiefly baleen whales). Sea turtle shell provides less of a problem - it is hard like rock, but because shell is coated in sheets of horn - keratin - which exfoliate slowly during the turtle's life, turtle barnacles have to continually pierce through layers in order to not be shed off. Whale barnacles often inhabit much softer tissue, and many have evolved elongate "dagger" like plates that pierce through the skin to maintain a hold as skin around the barnacle periodically sloughs away. Owing to their unique and rare substrate, both turtle and whale barnacles are as rare as hen's teeth in the rock record. This paper reports a balaenid right whale skeleton from the Pliocene of Italy with a number of turtle barnacles found preserved in association. These authors suggest that rather than the skin, these turtle barnacles (Chelonibia sp.) were attached to the callosities - large cornified white pads on the snout and chin of right whales which house sensory hairs and host a wealth of parasites (whale lice, and typically whale barnacles). Right whales are relatively slow swimming with this unique hard tissue - permitting the jump from turtle shell to a whale. More interestingly, Chelonibia is related to whale barnacles and is an early diverging lineage within the group Coronuloidea. This raises the possibility that marine turtles, which predate pelagic whales by about 100 million years, may have served as an evolutionary stepping stone - both intermediate in time and substrate quality - to whales.



The land to sea transition in the early evolution of whales has been revealed by a series of well-preserved skeletons of pakiccetid, ambulocetid, remingtonocetid, and protocetid archaeocetes largely from the Tethys region and North Atlantic. These fossils show that early whales evolved from terrestrial running ancestors, spent a fair amount of time in the water, at first resembling wolves and eventually bizarre giant otter-like critters with crocodile-like snouts (I lovingly refer to the remingtonocetids as the “crocodile-otters”). The protocetids chronicle the move to marine settings and the transformation of the hindlimb – later diverging protocetids lose fusion between the sacral vertebrae, fusion between the innominate and the sacral vertebrae, and ultimately the number of sacral vertebrae as well – prefacing the decoupling of the hindlimb from the vertebral column and reduction in hindlimbs in the basilosaurid archaeocetes. These aspects of the land to sea transition were established entirely based upon the functional anatomy of the skeleton as inferred from gross skeletal anatomy. However, in the last 15 years or so, a number of studies reporting on bone microanatomy and histology as well as stable isotopes have yielded further insights. Until this study, these sorts of data have been considered in isolation but not together. This study reports new microanatomical data (from CT and histology) and stable isotopes from pakicetid and remingtonocetid whales as well as the curious deer-like cetacean relative Indohyus. Stable isotopes indicate that pakicetids and Indohyus likely inhabited freshwater settings (consistent with their preservation in terrestrial deposits) and that remingtonocetids inhabited marine settings (consistent with their marine sedimentological context, though terrestrial mammals get swept out to sea and preserved in marine sediments on occasion). Bone microanatomy suggested more variation, with some pakicetids being more aquatic (Pakicetus – somewhat thicker cortex) and others more terrestrial (Ichthyolestes – somewhat thinner cortex like a land mammal), though cortical thickness in remingtonocetids was consistent with aquatic habits. This study highlights the utility of both types of data in inferring aquatic habits in cetaceans and other artiodactyls (e.g. anthracotheres).



Beaked whales (ziphiidae) are some of the more peculiar modern cetaceans - and cetaceans are already quite strange as far as mammals go. Most ziphiids lack dentition other than mandibular tusks, have enormous hyoid bones, and rather dense rostra. One extant beaked whale, Blainville's beaked whale (Mesoplodon densirostris), has the densest bone of any mammal (if not vertebrate) in its rostrum - and apparently serves no function in diving or combat. The bottlenose whale, Hyperoodon, bears paired fan-shaped crests formed from much spongier bone with a deep furrow between for the melon. Fossil ziphiids expand this diversity and include ziphiids with an inflated ridge down the top of the rostrum (Aporotus, Caviziphius, Tusciziphius, Ziphirostrum), paired crests on the maxilla (Africanacetus, Imerocetus) and perhaps the strangest of all - Globicetus hiberus, with an enormous hemispherical tuberosity on the base of the rostrum. Earlier histological studies showed that ziphiids independently grew all sorts of weird cranial structures and used different growth pathways towards forming them. How was the bizarre bony "soccer ball" formed on the rostrum of Globicetus? These authors report the results of histological study of a referred skull of Globicetus dredged off the continental shelf near Spain. Histology shows that the bone was formed by continued periosteal accretion of the premaxillae long after the whale reached physical and sexual maturity. The bone is extremely dense (osteosclerotic) and laminar and lacks any evidence of remodeling. Curiously, the orientation of bone growth seems to have rotated during growth of the tuberosity. These authors conclude that the "internal antlers" hypothesis proposed by Pavel Gol'din a few years ago (that weird sexually dimorphic cranial structures are visible via echolocation and permit easy recognition of species and sex to whales which spend much of their time at depth) is the most likely explanation for rostral features in ziphiidae like the rostral tuberosity in Globicetus.


Paleocetologists are crazy about whale and dolphin earbones. They are loosely attached to the skull, and typically fall out of the skull early on during decomposition. Externally, they are weird shaped bones that are highly distinctive in shape and easily identifiable, simultaneously permitting diagnosing new extinct species when associated with more complete skeletal remains and permitting identification of a particular species at a new locality when found in isolation. Internally, the periotic, also known as the petrosal, houses the bony labyrinth – which consists of the spiral-shaped cochlea and the semicircular canals. These internal structures can only be seen by the naked eye if the specimen is broken – and older studies destructively sampled these by serial grinding: literally grinding away the bone micron by micron and taking hundreds of photographs or camera lucida drawings. CT scanning is now routinely used in paleontology and numerous studies have published on 3D models made from micro CT data of the bony labyrinth of modern and extinct whales. I remember when I was an undergraduate student, nearly everyone seemed to be doing this, but few publications arose. Finally, many of these are being published, mostly by this author (Eric Ekdale) and Rachel Racicot (below). This study presents new data and analyses of modern and extinct baleen whale bony labyrinths. At a gross level, this paper demonstrates that morphometric analysis of cochlear landmarks is insufficient to clearly separate baleen whales and echolocating dolphins – though when the semicircular canals are included, resolution becomes much higher. This study supports an earlier paper by Ekdale and Racicot (2015) which hypothesized that, based on cochlear morphology, basilosaurid archaeocetes (which all Neoceti are thought to have evolved from) appeared to have adaptations for low frequency hearing like modern mysticetes (but see Churchill et al., above); the current study finds that Zygorhiza and the archaic toothed “archaeomysticete” (from Charleston!) plot within the “morphospace” of baleen whales. Other interesting findings: some cetotheriid baleen whales including Herpetocetus and Metopocetus have extreme cochlear coiling, and may have had some unknown super-low frequency hearing specialization. Curiously, the middle Miocene allodelphinid dolphin Zarhinocetus errabundus plots within baleen whale morphospace, near Metopocetus – suggestive of low frequency hearing, or perhaps that certain extinct odontocetes had a much wider range of hearing frequencies.
 

Fitzgerald, E.M.G. 2016. A late Oligocene waipatiid dolphin (Odontoceti: Waipatiidae) from Victoria, Australia. Memoirs of Museum Victoria 74:117-136.

In 1994 my former Ph.D. Adviser Dr. R.E. Fordyce found and collected a particularly complete dolphin skull from a cliff exposure of the upper Oligocene Otekaike Limestone near Duntroon, NZ. The specimen included a cranium missing only the tip of the rostrum, both mandibles, earbones, and the atlas vertebra. Oligocene dolphins are rare worldwide, and at the time, the taxonomy of odontocetes was hopelessly confused; most heterodont cetaceans of Oligocene age were assumed to be squalodontids, known from better presrved Miocene deposits. Incomplete specimens and lazy taxonomy plagued the study of dolphin evolution – so discovery of such a specimen offered a unique opportunity to clarify the early evolution of dolphins. This new dolphin was named Waipatia maerewhenua, and represented the Waipatiidae - one of now several recognized groups of heterodont dolphins from the Oligocene. Other waipatiids now include Otekaikea (NZ), Awamokoa (NZ), and possibly Papahu (NZ) and Sulakocetus (Caucasus). A fragmentary skeleton collected from the upper Oligocene Jan Juc Marl of Australia is described in this new paper by Erich Fitzgerald and includes a partial dentition and associated postcrania (the Jan Juc Marl also yielded the type specimens of the mammalodontids Mammalodon colliveri and Janjucetus hunderi). The teeth are quite small but clearly are separable into two general categories: anterior conical teeth and posterior subtriangular cheek teeth with accessory cusps. Some of the anterior teeth were even procumbent tusk-like teeth, as seen in Waipatia maerewhenua and Otekaikea huata (named for its spear-like tusks). The specimen preserves a nice forelimb – one of the only known examples for an Oligocene odontocete. The combination of forelimb features and the dentition identify this specimen as the first waipatiid from Australia. More complete remains will eventually be found and further refine the identification. Furthermore, some specimens from the Chandler Bridge Formation of South Carolina (including CCNHM 107, our dolphin nicknamed “whacky”, and a bunch of skulls I prepared this summer) likely represent waipatiids from the North Atlantic. I expect future discoveries will illuminate a worldwide geographic range for this group – but specimens from nearby Australia are a critical first step.




Mammalodontid whales are the earliest recognized toothed mysticetes, and include Mammalodon colliveri and Janjucetus hunderi from the Jan Juc Marl of Australia – the same unit as the waipatiid reported by Fitzgerald (see above). Mammalodontids are small (~3 m long) cetaceans with a blunt rostrum, enormous eyes, and denticulate leaf-shaped cheek teeth; they are the most primitive known mysticetes, and reflect early ecological specializations within Mysticeti. Janjucetus is a seal-like apex predator, and Mammalodon was likely a benthic suction feeder owing to extreme tooth wear. Similar to the Fitzgerald article reporting the first waipatiid from Australia, this paper reports the first mammalodontid from New Zealand – Mammalodon hakataramea – based on a fragmentary skull including the top of the braincase, a well-preserved tympanic bulla, and teeth from the late Oligocene Kokoamu Greensand. The teeth are worn down to the gumline and only roots remain; the bulla and skull are similar to M. colliveri and differ only in a few minor details. The type specimen is, however, the only published example of a NZ mammalodontid – and though a couple other specimens needing reidentification exist (if you look hard enough in the literature), I'm surprised by the rarity of mammalodontids in NZ. Mysticete skeletons are not uncommon – eomysticetids and Mauicetus-like specimens dominate Otago collections, and I've seen many more in the field – but only a few precious scraps of mammalodontids. Then again, toothless mysticetes and kekenodontids are not yet reported from the Jan Juc Marl (though the purported kekenodontid “Squalodongambierensis is known from the early Oligocene Gambier Limestone of Australia). This discovery extends the range of mammalodontids to New Zealand – perhaps unsurprisingly, the Oligocene whale assemblages are beginning to resemble one another more and more.
 
 

Baleen is the organ which baleen whales use during filter feeding to strain tiny prey items from the water column – and contrary to popular belief, relatively little of it consists of microscopic plankton – most prey of baleen whales are amphipods, krill, and even schooling fish. Baleen is not part of the skeleton or dentition – though in the mouth, it is a structure unrelated to teeth, and is instead made out of keratin – the same protein that skin, hair, claws/hooves, and horns are composed of within most vertebrates. Like horns, keratin grows continuously as it is abraded and worn down by the tongue. Baleen is lightly mineralized, but because it is a soft tissue, it is rarely found in the fossil record. It has been found at a few localities in the eastern Pacific: Purisima Formation (California), Monterey Formation (California), Empire Formation (Oregon), and most importantly, in numerous skulls and skeletons from the Pisco Formation of Peru. Because soft tissue preservation is unexpected relative to skeletal preservation, what processes permit baleen preservation? This new study reports the first study of baleen preservation using a combination of microscopy and X-ray diffraction. This analysis indicates that early diagenetic formation of dolomite – a type of carbonate mineral similar to limestone – rapidly formed around the carcass and preserved a natural cast around the baleen plates. Preservation is good enough to show individual baleen tubules. Some microstructure of the baleen was presrved by phosphatization, either during or after precipitation of dolomitic cement around the baleen and skeleton. Furthermore, this study indicates that baleen preservation was mediated by sinking of the carcasses into soft, soupy sediment as hypothesized for exceptional preservation of Jurassic ichthyosaurs in the Posidonia Shale of Germany – directly contradicting diatomite deposition rates published for Pisco whales by young earth creationists, founded with intent to disprove radiometric dating and prove a young, biblical age of the earth.



Sometime in the 1840s a curious skull was dug out of a bank near Middleton Place plantation right here in Charleston, SC, only a few miles west of my apartment in West Ashley. The skull was studied initially by various naturalists including Louis Agassiz, eventually named Agorophius pygmaeus (though other names have been applied to the same skull, e.g. Phocodon holmesii) but disappeared sometime in the late 19th century. Later researchers determined that, based on the beautifully illustrated published plates of the specimen, that it represents an echolocating toothed whale (Odontoceti), albeit interpreted as an archaeocete by earlier authorities; based on the location and the morphology, the specimen likely originated from the lower Oligocene Ashley Formation. The whereabouts of the skull are unknown, and Dave Bohaska thinks it may be rediscovered if the historical trash heaps at the estate Agassiz was staying were ever excavated by archaeologists. In 1980, my former Ph.D. adviser R.E. Fordyce was working on his postdoc at the Smithsonian, and on a trip to Harvard he found an isolated, mislabeled tooth that matched the single tooth of the holotype (again, thanks to the highly detailed illustrations). This, unfortunately, is the only surviving part of the holotype. New specimens reported in this study - certainly one of the publications I was happiest to see this entire year - alleviate this problem. Godfrey et al. refer two new skulls to Agorophius pygmaeus, both including well-preserved braincases and one with a nice periotic (earbone). The morphology more or less conforms to the published figures, but gives us crucial new reference specimens to base the genus and species on. A similar problematic example is Zygorhiza kochii - the holotype is garbage, but we colloquially use the most complete skeleton and cranium (USNM 11962) as the de facto reference specimen rather than the type. we can now do this for Agorophius, the concept of which has been based on nothing more than illustrations and a tooth for over a century. The new material permits inclusion of Agorophius pygmaeus within a cladistic analysis, and in the analysis reported by Godfrey et al., Agorophius plots out low on the odontocete tree, between the Xenorophidae and more derived odontocetes like the "squalodontids" and Waipatia. We have more material of Agorophius pygmaeus here in CCNHM collections... so stay tuned!
 

The last decade has witnessed a revolution in the study of "cetotheres" - primitive baleen whales within the crown group that evidently do not belong to any modern family (but see Marx and Fordyce, below). In the 19th and 20th century, a number of extinct chaeomysticetes were named and lumped into the wastebasket "Cetotheriidae" - including species now known to be eomysticetids (e.g. Tokarahia lophocephalus, Micromysticetus), and by the 1990s it was widely acknowledged that "cetotheriids" were either paraphyletic or polyphyletic, but few researchers had anything constructive to say about it. In 2006, a new study on the peculiar dwarf mysticete Piscobalaena emanating from Virginie Bouetel's Ph.D. thesis included a cladistic analysis of mysticetes, and found that a number of "cetotheres" formed a clade with Cetotherium rathkii - including Herpetocetus, Nannocetus, Piscobalaena, Metopocetus, and Mixocetus. Recognition of this clade revitalized the study of "cetotheres", and given the inclusion of the type species of the horribly overbloated Cetotherium, the family Cetotheriidae was redefined. A number of additional studies by Ukrainian cetologist Pavel Gol'din and Russian paleontologist Konstantin Tarasenko have reported new genera and species as well as reevaluated old 19th and early 20th century taxa. This new study by Gol'din and Startsev clarifies the taxonomy and provides new descriptions for many problematic and poorly described late Miocene taxa from the northeastern margin of the Black Sea (Crimean peninsula and Caucasus), including the new genus Mithridatocetus, which includes the new species M. eichwaldi and Mithridatocetus adygeicus (originally placed in Kurdalogonus). "Cetotherium" mayeri is recombined as Mithridatocetus sp., and Kurdalogonus is left with only the type species, K. michedlizei. Additional species are designated as nomina dubia (Cetotherium priscum and Cetotherium incertum) and others simply require additional material and study to be reevaluated ("Cetotherium" klinderi, "Cetotherium" pusillum, "Cetotherium" maicopicum, Eucetotherium helmersenii). A fascinating result of this research is that most of these taxa - true Cetotherium, Kurdalogonus, Mithridatocetus, Eucetotherium, and Brandtocetus - form a clade restricted to Paratethys (a redefined Cetotheriinae), suggesting a short-lived endemic radiation of cetotheriids within the Paratethys. These authors speculate that the endemic mysticete fauna likely went extinct owing to the warming associated with the Messinian Salinity Crisis (for the uninitiated, a period of time where the Mediterranean almost completely evaporated).


Few studies of marine mammals document assemblages from a faunal perspective – many palentologists tend to focus on a subset of taxa, which has an unfortunate side effect – marine mammal assemblages are rarely published in their entirety. I discussed this a bit in my 2013 Purisima Fm. marine mammal monograph. The world's best known marine mammal fossil assemblage – the Pungo River Limestone and Yorktown Formation of the Lee Creek Mine, North Carolina – has been extensively published on, and followup papers to a series of monographic works in the Lee Creek Mine edited volumes are supplementing published material with newly discovered specimens and revising identifications. Another similar assemblage of equivalent early Pliocene age like the Yorktown Fm. - and, also from a strip mine – is the assemblage from Langebaanweg in South Africa. This assemblage is well known for fossils of the lobodontine seal Homiphoca capensis – however, few, if any, cetaceans have been reported from the assemblage up until now. This study reports baleen whale earbones from the mine, mostly identifiable as whales similar to the late Miocene Italian rorqual Plesiobalaenoptera quarantellii. Interestingly, perhaps three species of archaic rorqual may be present. The earbones possess some features similar to gray whales, including a fenestra rotunda that is teardrop shaped and open dorsally. More complete material is obviously needed in order to name any species.

 

Kekenodontids are poorly known archaic cetaceans from the Oligocene – founded upon the fragmentary basilosaurid-like Kekenodon onamata from the upper Oligocene Kokoamu Greensand of the South Island of New Zealand. Other possible kekenodontids include Phococetus vasconum from the Oligo-Miocene of France and “Squalodongambierensis from the Oligocene Gambier Limestone of Australia. These latter taxa are known only from isolated teeth – though a recent SVP presentation by my former labmate Josh Corrie reports new material of “S.” gambierensis from New Zealand which firms up a link with Kekenodon. Most of the research on kekenodontids is unpublished (Corrie, in prep) but indicates these critters are basilosaurid-like with some features of Neoceti, and likely to be late surviving archaeocetes rather than early archaeocete-like toothed mysticetes. This new study reports a very archaic periotic from the upper Oligocene El Cien Formation of Baja California – equivalent in age with the Yaquina and Pysht formations of the Pacific Northwest, and already known to yield aetiocetids and eomysticetid-like baleen whales. This new periotic has a rather high superior process – a ridge that articulates with the squamosal, characteristic of basilosaurid archaeocetes. Several features of the pars cochlearis are shared only with Kekenodon onamata – which is known from a pretty weird-looking periotic (in addition to a bulla, frontal, quite a lot of the dentition, and an atlas). I'm not necessarily convinced just yet – and more complete material from Baja is needed. Furthermore, Kekenodon onamata, and other new kekenodontids, await description from New Zealand (get on it Josh Corrie!).



Here is yet another paper on fossil beaked whales this year (see also Bianucci et al., Dumont et al., and Mijan et al.). Dredged ziphiids have been extensively reported by now – and one of the first major descriptions of a dredged assemblage was published by Bianucci et al. (2007) on a diverse assemblage of Miocene or Pliocene ziphiid crania from the continental shelf off of South Africa. One ziphiid they named, Africanacetus ceratopsis, has a relatively elongate, narrow rostrum somewhat similar to Mesoplodon but has a pair of conical “horn”-like tuberosities on the maxilla. Africanacetus is most closely related to the bottlenose whale Hyperoodon, which similarly sports paired maxillary crests. This study reports a new species of Africanacetus collected by an ROV from a slope of the Sao Paulo Ridge off the shore of Brazil. It's represented by a well-preserved skull coated in manganese oxide – a common mode of preservation for deep marine cetacean specimens (in contrast, the South African specimens were coated in phosphate nodules, much more similar to preservational modes in continental deposits of shallow marine sediments. Additionally, the skull was found in a field littered with manganese nodules. Isotopic analysis of the manganese indicates the skull sat exposed on the seafloor for at least the last five million years, with manganese oxide deposition commencing about 5 Ma. A maximum age of middle Miocene – the period of time when uplift of the Sao Paulo Ridge terminated. This new species, Africanacetus gracilis, differs in minor ways from Africanacetus ceratopsis as well as being more gracile in skull proportions (hence the species name). Like many other extinct and nearly all modern ziphiids, Africanacetus lacks an upper dentition and was likely a suction feeder specializing on squid. Discovery of Africanacetus this close to the equator, and relatively close to Antarctica (Gol'din and Vishnyakova, 2013) suggests that this taxon had a relatively widespread distribution during the Miocene.




In the 1930s the cetacean assemblage from the lowermost Miocene Jewett Sand near Bakersfield was reported, and one of the better specimens consisted of a partial braincase with earbones and postcrania named Allodelphis pratti. In the past 20 years, a number of new specimens have been collected, which have allowed naming of the new species Allodelphis woodburnei, and reassignment of the problematic taxon “Squalodonerrabundus from the Sharktooth Hill Bonebed to the new genus Zarhinocetus. Allodelphinids are smallish, long-snouted dolphins with a number of features allying them with platanistids (Ganges river dolphin and extinct relatives). This new study critically expands the fossil record of allodelphinids, and reports many new specimens and taxa. More detailed descriptions of Allodelphis spp. and Zarhinocetus errabundus are provided, and the new species Zarhinocetus donnamatsonae is described from the lower Miocene Astoria Formation of western Washington. A beautifully preserved skull with articulated mandibles from the lower Miocene Nye Mudstone of Oregon is named as Goedertius oregonensis after “amateur” paleontologists Jim and Gail Goedert who collected the specimen. A new allodelphinid, Ninjadelphis ujiharai, is also named from a partial skull with earbones and vertebrae the lower Miocene Awa Group of Japan. Allodelphinids share similarly long rostra, bizarrely curved zygomatic processes of the squamosal, elongated and enlarged cervical vertebrae, and a secondarily elongate humerus. Allodelphinids represent an Oligocene-middle Miocene diversification of longirostrine platanistoid odontocetes that never left the North Pacific and went extinct sometime after the middle Miocene.



This summary is going to be brief owing to the fact that the original article is entirely in Japanese (which I cannot read). This study reports a rare example of an articulated odontocete forelimb, and includes the distal radius, ulna, carpals, metacarpals, and phalanges. Details of carpal articulations are described in detail – and carpals rarely fossilize, likely owing to disarticulation of the flipper during prolonged floating of carcasses. Unfortunately, functional evaluation within a phylogenetic context is not possible since the skull is missing.



This problematic attempts to resurrect the largely discarded idea of pinniped diphyly – the hypothesis that modern pinnipeds evolved from two completely different groups of arctoid carnivores: eared seals and walruses (Otarioidea) from bears, and earless seals (Phocidae) from mustelids (weasels and otters). The idea was first posited in the 1950s and adopted by most pinniped paleontologists in the 1960s, and remained more or less unchallenged until a series of papers were published by Andre Wyss and Annalisa Berta on pinniped osteology and paleontology. These papers included the first computer-run cladistic analyses of morphological character data, and proposed abundant evidence for pinniped monophyly. Shortly thereafter, molecular analyses unequivocally supported pinniped monophyly as well – and though this new paper cites a few oddball examples of  early karyological and molecular studies supposedly not supporting pinniped monophyly, literally dozens upon dozens of molecular analyses published since the early 1990s on pinniped and carnivore relationships have supported pinniped monophyly. This study provides a reinterpretation of several “important characters”, some of the anatomical interpretations being quite problematic. For example, many cranial characters supporting monophyly are argued ad nauseum with emhasis on how different each pinniped family is from one another; shortening of the humerus and femur is argued as being uninformative because it also characterizes other marine tetrapod groups. Shortening and anteroposterior flattening of the femur is argued away  because this is claimed to characterizes penguins – however, all of the penguin femora I've seen are perhaps shortened, but have a cylindrical shaft. According to these authors, many features cannot be used in cladistic analysis because they are convergent in phocids and “otarioids” - however, how are we supposed to know? These authors poke holes, some of which are valid and suggest some refining of character definitions, but ultimately we're supposed to just take their word for it. If the character evidence is truly convergent, a cladistic analysis incorporating as much data as prior analyses supporting monophyly should in theory produce a diphyletic result. However, the cladistic analysis they report includes a paltry seven taxa coded for 12 characters – a small fraction of the character data published 23 years ago by Berta and Wyss (1994). With such a small matrix, it begs two questions: why were the authors unable to find more character evidence supporting diphyly? And is the matrix small because the authors cherry picked available characters in order to produce the result they wanted? Neither option is desirable.




Beaked whales are perhaps one of the most surprisingly diverse groups of modern cetaceans. Many new species have been discovered, proposed, and named within my lifetime despite never being seen alive – rare and unnamed species are often discovered entirely based on strandings. Earlier this year, a dwarf species (as yet unnamed) of the giant beaked whale genus Berardius was reported from Alaska based on stranded specimens. The fossil record of baked whales is now extensive, yet most of the record consists of reworked specimens dredged from seafloor deposits of unknown age, and few records of extant genera in the fossil record are known. Molecular divergence dating offers a method to estimate the timing of phylogenetic divergences – yet changes in the rate of molecular evolution mean that parts of the tree need to be calibrated with fossil occurrences with good dates. Because few fossils nested within the part of the tree that includes modern species exist, few calibrations exist for Ziphiidae. This study reports a new species of the modern genus Mesoplodon from the lower Pliocene, Mesoplodon posti, named after Dutch paleontologist Klaas Post. A series of skulls are referred to this new species, all of which were collected by unknown collectors in the late 19th century. Fortunately, associated matrix indicates that most of the specimens were collected from the lower Pliocene Kattendijk Formation (3.9-4.9 Ma). This is the first time good age control is available for fossil Mesoplodon, indicating that diversification of Mesoplodon had begun by at least the early Pliocene.



Modern sperm whales (Physeteroidea) are highly specialized deep divers that feed almost exclusively on squid. Some extinct physeteroids exhibited similar feeding adaptations, but an increasingly well-documented group of early sperm whales sport enormous teeth with enamel (lost in extant physeteroids) and robust rostra and mandibles. Zygophyseter, Brygmophyseter, and Acrophyseter were discovered, and eventually the gigantic Livyatan melvillei - similar in size to modern Physeter (giant sperm whale), but with upper and lower teeth the size of 2 liter beverage bottles and a robust skull the size of a small car - was discovered in the Peruvian desert. Even more recently, Albicetus oxymycterus from the middle Miocene of was redescribed, just a bit smaller than Livyatan. This study is a follow-up to the original paper on Livyatan in Nature, and includes a more detailed anatomical description, as well as a similar description of Acrophyseter deinodon - also from the Pisco Formation. Earbones of Acrophyseter are described for the first time. The authors also report a new species, Acrophyseter robustus, a species somewhat older and with a more robust skull than Acrophyseter deinodon. A third possible species is recognized but not named (Acrophyseter sp.). All of these large-toothed sperm whales are killer whale-like and were the apex predators in the Miocene - with competition for prey only from the giant shark Carcharocles megalodon. Cladistic analysis interestingly supports recognition of a clade including Acrophyseter spp., Brygmophyseter, and Zygophyseter, as one of the earliest diverging physeteroid clades; Livyatan is positioned one node crownward, as sister to the Physeteridae + Kogiidae clade. Livyatan and Acrophyseter likely coexisted, suggesting some degree of niche partitioning - for example, that each preyed upon marine vertebrates appropriate for their size.



Modern walruses are one of the largest pinnipeds and are also simultaneously the most charismatic owing to their bizarre feeding apparatus, including a formidable pair of tusks formed from enlarged canines. The modern walrus is an effective suction feeder and has a voracious appetite for bivalve mollusks (clams) - and does not even use its teeth; it simply sucks the flesh right from the clam shell. Other extinct walruses (Valenictus, Ontocetus, Aivukus, Gomphotaria) are similarly interpreted as being molluskivirous, however this is a recent event in walrus evolution - most walruses were remarkably sea lion-like for the majority of their evolution (~5-17 Ma). One middle Miocene walrus in particular, Pelagiarctos thomasi, was initially interpreted to be a "killer walrus" that fed on marine mammals and perhaps seabirds. In a 2013 paper, Morgan Churchill and I found that on morphological and relative abundance evidence, Pelagiarctos was unlikely to be anything but a generalist fish eater like most modern seals and sea lions and found no anatomical adaptations for feeding on large prey. While still in New Zealand, labmate Carolina Loch approached me about sampling Pelagiarctos for an SEM study of enamel ultrastructure as another way to test the "killer walrus" hypothesis. We sampled the enamel of a new tooth of P. thomasi collected by colleague J.P. Cavigelli, as well as the New Zealand sea lion and fur seal for comparison - both extant fish-eating generalists. Hunter-Schreger bands are zigzag-like convoluted bands in enamel that prevent cracks from propagating; mammals that bite hard prey tend to have this developed more strongly. Instead, there was not really much of a difference between any of the three pinnipeds - leading us to propose that Pelagiarctos lacks obvious ultrastructural adaptations in its enamel for feeding on large fish, and is no more specialized than extant otariids.






Marx, F.G. and N. Kohno. 2016. A new Miocene baleen whale from the Peruvian desert. Royal Society Open Science 3:160542.

The fossil record of balaenopterids has been plagued by a problematic history, including specimens which have been assigned too damn many names in the literature (e.g. Pliocene balaenopterids from Italy) and taxa based on too damn many specimens, often chimaeric in nature (e.g. Pliocene balaenopterids from Belgium). Many are poorly figured and present in museums on different continents - meaning that paleontologists with local focus, or young paleontologists, are unable to really publish anything meaningful on new balaenopterid material if 1) unable to visit these collections and 2) unable to use the available literature. As a result, there is an embarrasingly large volume of undescribed balaenopterid skulls and skeletons across the globe, chiefly including California, Oregon, Florida, North Carolina, Japan, Peru, New Zealand, Australia, Italy, and the North Sea region. A number of recent studies have begun to erode away at this problem, both by fixing ancient taxonomic problems and redescribing old material and by describing new, beautifully preserved specimens. This new study reports the new genus and species of Incakujira anillodefuego - the genus name referring to the Peruvian origin of the specimens and their current disposition in Japan (Kujira is "whale" in Japanese) and the species name referring to the the Pacific ring of fire, which both Japan and Peru sit upon. Incakujira is based upon two absolutely gorgeous specimens from the upper Miocene Pisco Formation of Peru, and is characterized by having a relatively narrow rostrum, slender ascending maxillae, elongate nasals, and lacking a squamosal crease (a synapomorphy of Balaenoptera). Many aspects of the skull are similar to the modern humpback whale, and indeed the cladistic analysis supports placement of this species as sister to extant Megaptera. This analysis used a molecular partition, which places Megaptera within Balaenoptera, making the latter paraphyletic. Incakujira is inferred to be a lunge feeder like modern balaenopterids, though it possesses a twisted postglenoid process which may suggest an ability to skim feed. The holotype of Incakujira preserves baleen, and spacing of baleen is similar to Balaenoptera acutorostrata (minke whales).


This new book is a welcome text on the evolution and fossil record of whales and dolphins. An enormous volume of research has been published in recent years, with perhaps half of the studies on fossil cetaceans published within my lifetime – and the science is progressing at such a rapid pace that new review articles are being published almost once a year (whether we're asking for them or not). While I've been privately critical of the need for such review articles, this new text is not indulgently long (just over 300 pages) and includes introductory chapters, a pertinent chapter on modern cetaceans, methods used in cetacean paleontology, chapters on functional morphology, phylogenetics, paleoecology, biogeography, and cetacean macroevolution. Summaries of each cetacean family and other clades are succinct, comprehensive, highly informative, and supplemented with dozens of new illustrations all of consistent style (which greatly pleases my anal retentive tendencies). The book also includes a number of gorgeous color illustrations by Carl Buell. I've not read the book in full, but am seriously impressed with the depth of knowledge, comprehension, and level of detail. This book is a must-have for paleocetologists new and established.



Baleen whale fossils are some of the more commonly preserved marine mammals, and their large size ensures that they are some of the first fossils to be found in an area with fossiliferous Cenozoic outcrops. Indeed, legendary paleontologist Edward Drinker Cope published extensively on many early baleen whale discoveries from the Miocene Chesapeake Group of Maryland and Virginia. One of these is a peculiar skull which he named Metopocetus durinasus – which was found ex situ on a riverbank. Metopocetus has nasals fused at the midline, and a plug-like posterior process of the periotic like Herpetocetus. Metopocetus is now known to be a cetotheriid whale closely allied with Herpetocetus, Piscobalaena and Cetotherium – but its age is completely unknown. It possibly originated from the mid Miocene Calvert Formation, but similar mysticetes of equivalent evolutionary grade occur within the younger upper Miocene St Marys Formation. This study reports a new species of Metopocetus, M. hunteri, from the upper Miocene of the Netherlands. It differs in a few minor features of the periotic, and also includes a tympanic bulla and more of the basicranium than M. durinasus.Both species are characterized by a deep fossa on the paroccipital process in the basicranium, which they interpret as a fossa for the stylohyal. Lastly, the age of this new species is well known – 8.8-7.6 Ma, or Tortonian equivalent. This suggests that Metopocetus durinasus likely originated from the younger St. Mary's Formation.




Approximately one year ago, Marx et al. named the new toothed mysticete (Aetiocetidae) Fucaia buelli, and proposed that large gums preceded baleen. Other toothed mysticetes in the family Aetiocetidae have a series of minute palatal foramina which are homologous with the larger and more extensive foramina present in modern mysticetes. This was originally published by Tom Demere and others in 2008 about the remarkably well-preserved holotype skull of Aetiocetus weltoni from the Oligocene Yaquina Formation of Oregon – showing that some toothed mysticetes exhibited teeth and baleen. This new study reports a nice cranium (collected by Jim Goedert) of an as-yet unnamed aetiocetid that is morphologically intermediate between Fucaia and Aetiocetus from the Pysht Formation of Washington. This specimen has a peculiar style of tooth wear – the lowermost part of the enamel crowns are pristine and unworn, whereas the tips of the teeth are highly worn – indicating that the base of the crowns were embedded in thickened gums (normally, the base of the enamel crown marks the gumline). Furthermore, the wear is manifested on the mesial and distal edges of the teeth – suggestive of sediment moving between the teeth. These authors rightly interpret this to be evidence of benthic suction feeding. We made a similar case for benthic suction feeding in the bizarre porpoise Semirostrum ceruttii (Racicot et al., 2014). They further suggest that rather than possessing baleen, all aetiocetids possessed thickened gums instead as an adaptation for suction feeding – and that palatal vascularization originated first to supply blood to enlarged gums, and the gum tissue was later co-opted for filter-feeding baleen-like structures (modern baleen is derived from gingival tissue after all). This would indicate that eomysticetids – the group I studied for my Ph.D. – where the first cetaceans to be obligate filter feeders. They point out that having teeth and baleen simultaneously presents problems as teeth would damage/interfere with baleen – though I’m not really convinced. They make a good case for suction-based feeding with enlarged gums based on behavior in this one unnamed taxon. However, it is still possible that baleen was present in Aetiocetus proper, and given that analogous tooth wear is not present in any Aetiocetus spp., a different method to look for evidence of enlarged gums would be necessary. 



It's been a pretty damn good year for beaked whale paleontology (see Bianucci et al., Dumont et al., Ichishima et al., Rammasamy) - following several years of some rather spectacular beaked whale fossil discoveries. One of the most unexpected sources of beaked whale fossils have been offshore dredging, which have recovered all sorts of bizarre critters (e.g. Globicetus; Dumont et al., above). Not all discoveries reveal spectacular new skeletons representing unknown genera, however – but some seemingly minor discoveries often serendipitously help solve century-old problems and questions. This study reports a new species of the archaic ziphiid Beneziphius. The new species, B. cetariensis, was dredged from offshore Spain – the same area as Globicetus, Choneziphius leidyi, and Tusciziphius atlanticus. Beneziphius has a generalized Ziphius-like skull in many respects yet belongs to the extinct “Messapicetus clade” (Bianucci et al., above). Like other dredged specimens, the age of the holotype of B. cetariensis is not precisely known. The genus Beneziphius was originally named based on a skull from Belgium – but because it was collected without locality information in the 19th century, its age is unknown as well. The desire for clarified age control for this genus led the authors to sample matrix hidden inside the cranial cavity of the 19th century holotype of Beneziphius brevirostris. Matrix yielded Serravallian-aged (late middle Miocene) dinoflagellate cysts and acritarchs (microfossils useful for biostratigraphy). This study further remarks upon the faunal similarity of the Iberian dredged assemblage and the North Sea – both include Beneziphius, Caviziphius, Choneziphius, and Ziphirostrum; however, the Iberian assemblage includes Tuscizphius, formerly reported from Italy and South Carolina – but not the North Sea.


Murakami, M. 2016. A new extinct inioid (Cetacea, Odontoceti) from the upper Miocene Senhata Formation, Chiba, central Japan: the first record of Inioidea from the North Pacific Ocean. Paleontological Research 20:3:207-225.
Modern river dolphins include four different species, formerly grouped into the Platanistoidea – and subsequently regarded as being a paraphyletic or polyphyletic assemblage. The Platanistoidea is now redefined as Platanista (Ganges river dolphin) and extinct relatives. The two river dolphins unique to South America – Inia, the Amazon river dolphin, and Pontoporia, the Franciscana or La Plata river dolphin – likely are closely related and frequently grouped into the Inioidea. True inioids are generally unique to the Americas but some far-flung (but fragmentary) examples have been reported from western Europe and the Mediterranean. This study reports a fragmentary inoid dolphin skull from the Senhata Formation of Japan – named Awadelphis hirayamai. This species is characterized by a laterally overhanging premaxillary eminence, unique amongst inioids, a short zygomatic process, a posteriorly enlarged nuchal crest, and other features of the facial region. This new taxon is phenetically similar to pontoporiids yet Inia plots out within the Pontoporiidae in the cladistic analysis – and thus Murakami refers the new genus only to the Inioidea rather than a more exclusive clade within the group. Awadelphis extends the range of inioids to the western North Pacific – though the lipotid dolphin Parapontoporia has been preliminarily reported, formerly known only from California and Baja California. Curiously, despite this other record, no comparisons were made with the lipotid Parapontoporia – which was originally confused for a pontoporiid dolphin owing to some convergent features.


 Nance, J.R., M. Kricun, and S.J. Godfrey. 2016. Ankylosis and oosteonecrosis in the pectoral limb of a baleen whale (Cetacea, Mysticeti) from the Miocene Calvert Formation of Calvert Cliffs, Maryland, U.S.A. Marine Mammal Science. Online early.
Obligate marine mammals like whales and dolphins and sea cows do not have to support their weight on land (see Ando and Fujiwara, above). Because of this, certain injuries may be more prevalent in marine mammals – not because they are more susceptible to injury, but because they likely have a higher chance of surviving an injury. Skeletons of modern marine mammals frequently have broken bones – including vertebrae that are fused together. This new study reports a fossil baleen whale forelimb from the middle Miocene Calvert Formation of Maryland. This specimen includes a humerus, radius, and ulna – all fused together at the elbow. The radius is quite enormous, and when examined via CT scanning, is osteonecrotic on the interior along the entire length; smaller osteonecrotic zones are present in the ulna and humerus. Unfortunately, the forelimb was found isolated, so the extent of the injuries in the skeleton is unknown. What could have caused osteonecrosis (bone death) and fusion of the elbow in this extinct baleen whale? The authors interpret fusion of the elbow joint as a result of immobilization of the elbow caused by osteonecrosis of the forelimb bones. Osteonecrosis itself was likely specifically dysbaric osteonecrosis – caused by decompression syndrome, otherwise known as “the bends”, as already documented in other middle Miocene baleen whales.


Park, T., E.M.G. Fitzgerald, and A.R. Evans. 2016. Ultrasonic hearing and echolocation in the earliest toothed whales. Biology Letters 12:20160060

http://rsbl.royalsocietypublishing.org/content/12/4/20160060

Modern baleen whales (Mysticeti) and toothed whales (Odontoceti) differ from one another in many regards, and aside from teeth v. baleen the most important features differentiating the two are related to hearing. As described above (Churchill et al.), echolocating odontocetes are adapted to producing and hearing high-frequency sounds, whereas baleen whales are adapted toward hearing low-frequency sounds and are incapable of echolocating. Some studies using CT data have found evidence of high frequency hearing in Miocene and Pliocene odontocetes - but how early did it evolve? This new study reports CT data from the cochlea of an isolated periotic of a xenorophid dolphin - xenorophids are some of the most common odontocetes in the Oligocene deposits here in the Charleston embayment, and are the earliest diverging clade of odontocetes. The bizarre dolphin Cotylocara is a xenorophid and exhibits a series of sinuses relating to echolocation, showing that xenorophids could produce high frequency sounds. This periotic reported by Travis Park et al. from the upper Oligocene Belgrade Formation of North Carolina has a loosely coiled cochlea and other features indicative of high frequency hearing - leading these authors to propose that echolocation was an adaptation present in the earliest odontocetes. See Churchill et al., above, which builds significantly on this study.


Peredo, C.M. and M.D. Uhen. 206. A new basal chaeomysticete (Mammalia: Cetacea) from the late Oligocene Pysht Formation of Washington, USA. Papers in Palaeontology 2:4:533-554.

http://onlinelibrary.wiley.com/doi/10.1002/spp2.1051/abstract

As outlined above, eomysticetids are the earliest diverging functionally toothless baleen whales. Eomysticetids are typically mid-late Oligocene in age, have elongate mostly toothless rostra, large jaw muscle attachments, earbones intermediate between archaeocetes and Miocene mysticetes, and archaic basilosaurid-like postcranial skeletons. They've been reported from South Carolina, Austria, Japan, and New Zealand; an occurrence in Japan suggests we should be finding them in the eastern North Pacific. This study reports an eomysticetid-like baleen whale, Sitsqwayk cornishorum, from the upper Oligocene Pysht Formation of the Olympic Peninsula (formerly known for the small aetiocetid Fucaia goedertorum). Sitsqwayk has a few key differences from published eomysticetids, including a wider intertemporal region and the lack of a sagittal crest. However, Sitsqwayk does possess at least two eomysticetid synapomorphies – a zygomatic process that lacks a supramastoid crest and is longitudinally rotated so that the lateral surface faces dorsolaterally. Sitsqwayk showed up on a cladogram one node below Eomysticetidae. I imagine that with discovery of better earbones and a skull with complete nasals, eomysticetid relationships may become more firm. Regardless, Sitsqwayk represents the first named Oligocene chaeomysticete from the eastern North Pacific and I heartily welcome additional finds to be published! Similar undescribed skulls exist in collections from the Oligocene of New Zealand, and perhaps species like Whakakai waipata may be somewhat related.



Until recently the fossil record of marine mammals was generally quite localized and consisted of published occurrences that were either 1) outdated and improperly identified or 2) spotty and too incomplete to reveal any patterns of biogeography. Recent advances have published assemblages in their entirety – not just focusing on the best preserved material, and instead giving a more honest view of the fauna – refined old identifications – and reported new specimens and taxonomic occurrences. In the last decade, a number of studies have taken advantage ot the rapidly improving published record and begun broad-scale analyses of marine mammal distribution, biogeography, dispersal, evolution, as well as basic sampling problems. A classic biogeographic hypothesis in marine mammal paleontology is the use of the central American seaway – open until 3 million years ago prior to closure by uplift of Panama – whether walruses, baleen whales, seals, sea cows, or various dolphins. This new study uses the paleobiology database to statistically analyze patterns of faunal similarity between different ocean basins. Notable findings include 1) broad Eocene similarity between the North Atlantic and Mediterranean/Tethys, 2) faunal differentiation between basins driven by the Messinian Salinity Crisis at the end of the Miocene, and 3) faunal similarity between the Pacific and Atlantic pretty much only during the middle Miocene. Perhaps most significantly, this study suggests that the Central American Seaway was a much less used corridor for dispersal – and many marine mammals which inhabit the North Pacific and North Atlantic instead likely dispersed through the Arctic instead, which opened up around the time the CAS closed. This is not the first time this has been proposed, as it was originally proposed by Kohno et al. (1995) for tusked walruses and later by Lambert (2008) for phocoenid porpoises; I discussed this idea at some length in my 2013 Geodiversitas monograph, in addition to Pliocene faunal provinciality (e.g. dissimilarity). The current study is the first to provide a semi-quantitative statistical demonstration of these ideas, which feels quite vindicating!



 Baleen whales like the blue whale (Balaenoptera musculus) are the largest vertebrates of all time (sorry sauropod researchers). Living in water over terrestrial settings has its benefits – energetically more efficient locomotion thanks to buoyancy, and generally more abundant food. The lower constraints on on biomechanics have permitted gigantism – which in turn allows for greater capacity for long distance migration and dispersal; it's no coincidence that marine mammals with the largest body sizes have some of the largest ranges as well (in stark contrast to tiny seabirds like terns, which have similar migration distances yet weigh only a few ounces). In marine mammals, body size likely reflects niche as well as food availability- the latter of which is controlled in a “bottom up” fashion by primary productivity (the amount of chemical energy as biomass that primary producers – e.g. plankton – produce through photosynthesis or similar means). This study tracks maximum body size in in two different groups: filter feeding marine mammals (Mysticeti) and herbivorous marine mammals (Sirenia, Desmostylia) throughout the Eocene-Recent. Curiously, baleen whales do show much smaller sizes throughout most of their fossil record, attaining a large size only in the Plio-Pleistocene interval; similarly, minimum body size for baleen whales was substantially lower during the Oligocene and Miocene. This is demonstrated for the second time – a pattern already discovered and acknowledged by Olivier Lambert and colleagues (2010: Nature, Livyatan melvillei paper). This suggests that mysticete gigantism corresponds to Pleistocene-age increase in primary productivity driven by intensified chemical weathering. On the other hand, most sirenians and desmostylians show a comparably stable trend in body size, with the hydrodamalines (the western sea cow Dusisiren, and the extinct giant sea cows Hydrodamalis) – coinciding with the late Miocene-Recent diversification of kelps and proliferation of kelp forests. For more analytical examinations of body size trends in other marine mammals, be sure to check out Morgan Churchill et al.'s paper on pinniped body size and Cope's Rule.
 


These short papers are a comment and reply on Crerar et al. (2014), a paper proposing  that a second population of Steller's sea cow (Hydrodamalis gigas) inhabited the shoreline of Saint Lawrence Island, located in the northern Bering Sea over a thousand miles to the northeast of where the first sea cows were discovered (Commander/Komandorskiye Islands) and subsequently hunted to extinction in the mid 18th century. The original paper apparently used a number of specimens of “mermaid ivory” - the bones of the Steller's sea cow are dense enough for carving and polishing by native artists like regular ivory (thick teeth/tusks). These suspected Hydrodamalis bone carvings were acquired by the authors from ivory carvers and artists in Alaska, sampled for molecular analyses, and radiocarbon dated. However, the collection to which the physical bone samples are simply listed as “Crerar” - in other words, the private collection of the lead author. Pyenson et al. point out that the specimens should have been preserved as voucher specimens in an established museum collection – and since this was not done, the morphological identification of the specimens is essentially untestable. They further point out that serious ethical concerns are associated with the purchase and study of specimens from fossil dealers – outlined in more detail by the Society of Vertebrate Paleontology Ethics Statement. Crerar et al. respond by highlighting the significance of their discoveries, as well as indicating that the samples had been acquired as part of an investigation into illegal trade of other marine mammal remains – but because Hydrodamalis is extinct, it is exempt from restrictions imposed by CITES, US Endangered Species Act, or the Marine Mammal Protection Act. I agree that all publications involving fossils absolutely must rely on specimens in museum collections, but there are plenty of papers out there reporting purchased specimens that ended up in museum collections (e.g. Marx and Kohno, above). There is a bit of grey area which in all likelihood is never going to go away.



It's been a good year for the study of cetacean earbones and hearing using CT data – see for example Ekdale, Churchill et al., and Park et al. (above). These analyses have effectively made the destructive and time-consuming practice of serial grinding completely obsolete. The lead author, Dr. Racicot, CT scanned the pterygoid sinuses of porpoises as part of her master's thesis at San Diego State (now published), plunging her into the study of porpoise anatomy and evolution and including the description of the bizarre porpoise Semirostrum ceruttii. This study originated during her Ph.D. and benefited from data collection in Japan. This study presents new scan data and an analysis of the bony labyrinths of extinct and extant true porpoises (Phocoenidae). Unsurprisingly, all extinct phocoenids exhibit features consistent with high-frequency hearing. This analysis identifies characteristics permitting inference of an extinct odontocete's sensitivity to head movements – with modern pelagic phocoenids having a high sensitivity. The extinct, presumed benthic feeding porpoise Semirostrum had low sensitivity to head movements – consistent with life on the continental shelf. This study furthermore proposes a method to infer body size from measurements of the labyrinth. However, body length estimates of 7 meters for two Japanese fossil phocoenids – Haborophocoena and Numataphocoena – seems unbelieveably long as the skulls are not quite that large. Perhaps this is a result of highly derived bony labyrinths and some unknown specialization in these species. Regardless, the promise for estimating body size from an isolated earbone is tantalizing.


Rahmat, S.J. and I.A. Koretsky. 2016. First record of postcranial bones in  Devinophoca emryi. Vestnik Zoologii 50:1:71-84.

The true seals (Phocidae) are one of the more challenging groups of pinnipeds to study in the fossil record. The majority of the fossil record of phocid seals consists of isolated postcrania - aside from scattered mandibles and fragmentary skulls and a few skeletons from South America (Acrophoca, Hadrokirus, Kawas, Piscophoca). The fossil record of phocids from Europe and western Asia is particularly difficult, being almost completely based on postcrania with confusing taxonomic history. One of the few early phocid taxa represented by well-preserved crania is Devinophoca - two species within this genus have been reported from the same quarry in Slovakia. Both species (Devinophoca claytoni, Devinophoca emryi) are known from nearly complete skulls differing in only a few minor details - some of which appear to be incorrectly diagnosed (for example, the infraorbital foramen of D. claytoni is claimed to be smaller than D. emryi, yet the converse appears to be true). Devinophoca emryi was named in 2015, and I'm skeptical that it represents a second species. Regardless of how many species are recognized, the Devinophoca material from the "Bonanza" quarry clearly represents an early phocid. This study refers various postcranial elements to Devinophoca emryi, including scapula, humerus, radius, ulna, innominate, femur, tibia, fibula, calcaneum, and astragalus. Some of the postcranial features are clearly phocid-like (e.g. laterally everted innominate, posterior extension of astragalus limiting motion of ankle) yet other bones are remarkably similar to the early pinnipedimorphs Enaliarctos and Pinnarctidion (e.g. humerus). The isolated bones are referred to D. emryi over D. claytoni owing to the "Ecomorphotype" method of Koretsky (2001); modern species occupying different niches have specific morphologies of the cranium, mandible, and postcrania corresponding to their niche. This is taken a step further, and isolated postcrania are referred to taxa with craniomandibular material based on shared "ecomorphology" by analogy with extant taxa. Unfortunately, this method is perhaps circular reasoning at best and untestable at worst. For example - these authors explain that, based on the different ecomorphotype groups outlined by Koretsky (2001), the mandible and postcrania of D. emryi clearly represent a single taxon - but they never explain why the postcrania do not belong to D. claytoni, which is not known by a mandible. Because the ecomorphotype method focuses on mandibles and postcrania and not skulls, the question of whether or not this material represents D. claytoni is simply left unaddressed. 



As mentioned several times earlier, this has been a bit of a ridiculous year for advances in beaked whale paleontology – with the description of Chavinziphius and Chimuziphius (Bianucci et al.), new specimens of Messapicetus (Bianucci et al.), histology of the bizarre rostrum of Globicetus (Dumont et al.), and a new species of Africanacetus dredged from the Atlantic off Brazil (Ichishima et al.). This new study reports a new genus and species of beaked whale, Dagonodum mojnum, from the upper Miocene Gram Formation of Denmark. The Gram Formation has previously yielded the holotype skulls of the “kelloggithere” mysticetes Uranocetus gramensis and Tranatocetus argillarius, and some poorly preserved pontoporiid dolphin skulls. Dagonodum is a reference to the Lovecraftian sea god Dagon (which is hilarious, given a conversation with colleagues Brian Beatty and Morgan Churchill about how awesome it would be to name an extinct whale after Dagon). Dagonocetus is similar to Messapicetus and has a pair of enlarged mandibular tusks and a polydont, homodont dentition set in an elongate narrow rostrum. Enlarged tusks further identify the holotype as a male. The type also includes a decent pair of earbones. Dagonodum was likely not as specialized for suction feeding as modern mostly-toothless ziphiids, and furthermore possessed somewhat more elongate cervical vertebrae indicating a higher degree of neck flexibility than modern ziphiids.

 Image by Gabe Santos


Desmostylians are a bizarre group of hippo-like marine mammals that are closely related to elephants (Proboscidea) and sea cows (Sirenia) and are within the Afrotheria (African mammals). The earliest desmostylians are Oligocene in age (see Beatty and Cockburn, above) and fossils of Desmostylus are some of the more common occurrences in the Miocene of California, Oregon, and Japan. Desmostylians have highly derived teeth – tusk like canines and incisors, and premolars and molars with cusps arranged into finger-like tubes of enamel that superficially resemble six-packs of beer cans. Much attention has been afforded to the taxonomy and functional anatomy of desmostylians, but little research on the ontogeny and tooth replacement of these strange mammals has been undertaken. This study reports an enormous mandible of Desmostylus hesperus from the mid-upper Miocene Sespe Formation of Orange County, CA. This mandible is quite a bit larger than all previously published specimens of Desmostylus, and uniquely has a downturned lower tusk, resembling the strange giant elephant Deinotherium. Of greater importance is the lack of teeth – though a large bony dental capsule is present as a large knob on the inside of the mandible. This suggests that Desmostylus continued to thrive after loss of its last pair of molars – and perhaps fed on kelp in old age. Kelp may be consumed without teeth, as the extinct Steller's sea cow Hydrodamalis gigas fed solely on kelp without having any teeth. The extreme size of this specimen, along with the absence of cheek teeth and infilling of alveoli – indicates that Desmostylus had a dental eruption pattern most similar to other afrotheres, where the last molar is not erupted until the mandible attains full size. A recent study suggests that desmostylians may be early diverging perissodactyls (odd toed ungulates); either the the dental eruption pattern means that it is due to shared ancestry, or that this dental eruption pattern evolved twice – once within Afrotheria and once within Perissodactyla. As a side note, I saw this spectacular specimen in 2013 when I visited the Cooper Center and because of its gigantic size, I thought it was some weird elephant like a gomphothere – to which Gabe Santos was delighted to inform me that it was a giant Desmostylus the size of a small elephant.




These two new studies report newly discovered specimens of the porpoise Numataphocoena from the Pliocene of Japan. Porpoises are currently distributed worldwide but now known to have been restricted to the Pacific for most of their evolution, with diverse Mio-Pliocene assemblages now known from Japan, California, and south America. Numataphocoena was originally described from a partial skeleton (including a fragmentary skull with well-preserved earbones and a nice postcranial skeleton) from the early Pliocene of Hokkaido. Despite cranial differences, Numataphocoena already had the hallmark "spatulate" teeth of modern porpoises - most modern odontocetes have conical teeth. A new periotic reported in the first of the new papers suggests closer affinities with another Pliocene porpoise from Hokkaido, Haborophocoena. The referred periotic further illuminates some subtle ontogenetic trends in odontocete earbones - an understudied topic in my opinion. The second new paper (Tanaka and Ichishima) report a new cranium of Numataphocoena, including a nearly complete braincase but incomplete rostrum. Similar to the prior paper, this specimen is ontogenetically younger and reveals some information about ontogeny in extinct porpoises, and more critically expands the known morphology of the species. Cladistic analysis supports a novel clade of phocoenids including practically every porpoise from Hokkaido from the latest Miocene and early Pliocene: Haborophocoena spp., Numataphocoena, Archaeophocaena, and Miophocaena. I've long suspected that Japanese phocoenids are oversplit, and this grouping may suggest that the number of species and genera ought to be trimmed.
 



The earliest Miocene (Aquitanian) is a period in time with abundant marine rocks but rare fossil marine mammals. Some of the only areas with earliest Miocene marine mammals include California and Oregon (e.g. Jewett Sand, Skooner Gulch Formation, Nye Mudstone), Italy (Belluno Sandstone), and New Zealand (Milburn Limestone, Mt. Harris Formation). Quite a few fossil odontocetes including complete skulls from the Belluno Sandstone of Italy have been reported – however, dating of this unit is poor and the unit possibly Burdigalian. The Belluno Sandstone assemblage is dominated by longirostrine odontocetes of various families (Dalpiazinidae, Eoplatanistidae, Eurhinodelphinidae, Squalodelphinidae), sperm whales, and squalodontids – and notably lacks baleen whales. New Zealand has offered a few earliest Miocene cetaceans with good dates – including the early “kelloggithere” Mauicetus parki, the squalodontid “Prosqualodon” hamiltoni, and the recently named waipatiid-like dolphin Papahu taitapu. The holotype of Papahu was collected from Nelson (northwestern part of the South Island, NZ), and includes a nearly complete skull with partial periotic, and several vertebrae including the axis. Originally suspected to be a kentriodontid dolphin, cladistic analysis suggested Papahu was further down on the odontocete tree, and well outside crown Odontoceti. This new paper reports new material of Papahu – a specimen identified as cf. Papahu sp., including a fragmentary skull with a more completely preserved periotic and a complete tympanic bulla. Inclusion of the new specimen pulls Papahu into two different phylogenetic positions: either as an early diverging member of Platanistoidea (see Boersma and Pyenson, above) or as the earliest member of Synrhina (Delphinida + Ziphiidae). Regardless – this new specimen and analysis strongly suggests that Papahu is indeed a crown odontocete, and highlights the utility of fragmentary specimens.
 


This paper marks one of the last entries in Yoshi Tanaka's Ph.D. thesis on Oligocene odontocetes from New Zealand. Earlier papers from his thesis reevaluated "Prosqualodon" marplesi and reassigned it to the new genus Otekaikea, named the new species Otekaikea huata, and redescribed "Microcetus" hectori and transferred it to Waipatia. Waipatiid dolphins are generalized heterodont odontocetes from the Oligocene and likely early Miocene, likely filling in a bottlenose dolphin-like niche - though waipatiids notably differ from later groups of odontocetes in having moderately enlarged tusk-like incisors. This new study reports the new genus and species Awamokoa tokarahi, a new waipatiid sister to Waipatia spp. This new species differs from Waipatia spp. in a few features of the basicranium and periotic. Awamokoa is represented by a partial skull, well-preserved earbones, a partial mandible, various teeth, vertebrae, ribs, partial scapulae, ulna, and phalanges. Awamokoa represents the fifth confirmed Oligocene waipatiid from New Zealand (Papahu may represent a sixth) and one of the best examples of waipatiids preserved with postcrania; other waipatiids are likely to be reported from elsewhere (e.g. see Fitzgerald, above). Awamokoa has a relatively shallow and elongate temporal fossa, suggesting that the feeding apparatus was adapted for slower but more powerful bites, and that later diverging platanistoids (squalodelphinids, Platanista) with anteroposteriorly shorter and deeper temporal fossa housed muscles adapted for rapid snapping of the lower jaws. This transition similarly parallels the development of homodonty within the Platanistoidea. Lastly, Awamokoa is from the early late Oligocene Kokoamu Greensand, and approximately 28-25.2 Ma in age - making Awamokoa one of the earliest occurring platanistoids. Arktocara yakataga, an allodelphinid (see Boersma and Pyenson, above) may be older but has poorer age resolution. Furthermore, waipatiids may not be within the odontocete crown and instead may not be related to true platanistoids (e.g. Squalodelphinidae + Platanistoidea) - see various phylogenies published by Jonathan Geisler and colleagues.
 


Not all whale specimens are spectacular, with completely preserved skulls, mandibles, and postcranial skeletons. In fact, owing to the large size of baleen whales in particular, most are rather incomplete; it's difficult to catch a large skull eroding out and nab the whole thing. Earbones are often fairly informative, and in the case of a new species reported in this study, Whakakai waipata, well-preserved earbones clearly demonstrate that this baleen whale is a new genus and species; the earbones are distinctive and well-preserved enough that they serve as the major features that diagnose this new species. Whakakai is from the upper Oligocene Kokoamu Greensand of New Zealand, collected nearby the type localities of the odontocete Waipatia maerewhenua and the eomysticetid Tohoraata raekohao, though from somewhat stratigraphically lower than each. The preserved portion of the skull indicates that this mysticete lacked a sagittal crest – immediately distinguishing it from eomysticetids, but resembling such mysticetes further up the tree as Horopeta and Mauicetus, both from the overlying Otekaike Limestone. The earbones on the other hand are quite strange. The periotic is enormous and inflated, with proportions somewhat resembling other archaic mysticetes – yet has a relatively small pars cochlearis; the posterior process is unfused, like other archaic mysticetes. The tympanic bulla is eomysticetid-like with well-defined medial and lateral lobes, but has a partially closed elliptical foramen – which is completely closed in later diverging mysticetes (Balaenomorpha). Cladistic analysis demonstrates that Whakakai is more derived than eomysticetids and plots further up the tree – close to Horopeta and Mauicetus, and in one of two analyses, as sister to Horopeta (also named in a paper resulting from C-H Tsai's thesis at U. Otago. These species evidently lived alongside the more primitive eomysticetids, yet illustrate a suite of morphologies broadly intermediate between eomysticetids and Miocene “kelloggitheres”.



Perhaps one of the strangest and more obscure extinct marine mammals is Kolponomos - an ursid-like semiaquatic carnivore with enlarged, sea otter-like teeth, binocular vision, and enlarged procumbent canines. Originally discovered from the lowermost Miocene Clallam Formation of Washington in the 1960s and thought to be a giant marine raccoon, additional well-preserved fossils including crania were reported in the 1990s and revealed ursid affinities, and a possible sister taxon relationship with pinnipeds (I suspect that Kolponomos along with Potamotherium and Puijila represents an aquatic diversification of arctoid carnivores related to pinnipeds). The dentition of Kolponomos was inferred to be adapted for crushing mollusks, and enlarged mastoid processes were interpreted as anchoring strong neck muscles. This new study investigates the feeding morphology of Kolponomos using finite element modeling (FEM). This study notably looks beyond arctoid carnivores for analogies, and found that the deep "chin" of the mandible shares certain biomechanical similarities with the sabertooth cat Smilodon. In carnivorous mammals that chew on hard food, there is a tradeoff between "stiffness" and mechanical efficiency: a stiff mandible does not store much energy, and a more flexible mandible stores energy during a bite. Based on analogy with Smilodon, Kolponomos was hypothesized to possess high mandibular stiffness - which was borne out by the analysis. Surprisingly, though similar in feeding morphology, the sea otter possessed the most flexible mandible. Kolponomos is inferred to have utilized an "anchor bite" stage prior to chewing, where the mandible and lower dentition helped pry mollusks from rocks, followed by otter-like crushing of the mollusk with the large cheek teeth. Enlarged neck muscles gave Kolponomos the ability to apply enormous torque onto attached mollusks (e.g. mussels). This study indicates that Kolponomos and the sea otter Enhydra crush hard shelled prey while independently relying on mandibular stiffness and high bite force (respectively), using somewhat different pathways which have nonetheless resulted in craniodental convergence. Because higher stiffness is required for the anchor bite, I imagine that the lower stiffness in Enhydra is permitted by the fact that Enhydra removes mollusks and crustaceans from the substrate by using its forelimbs rather than its feeding apparatus.



If you've read everything above you're already familiar with the utility of isolated cetacean earbones – they can help us diagnose species, inform us about the hearing ability of extinct cetaceans, intraspecific variation, and the diversity of extinct cetacean faunas. The Yorktown Formation at the Lee Creek Mine in North Carolina has yielded one of the largest Pliocene marine mammal assemblages in the world, yet the cetacean assemblage is dominated by isolated (yet diagnostic) earbones. Particularly common earbones include delphinid dolphins and small physeteroid sperm whales. In 2008, the Lee Creek IV volume was finally published after being in press and in preparation for decades, and reported the presence of large periotics of some sort of unknown kogiid (pygmy sperm whale). These earbones are much larger than extant Kogia – and this study reports the first records of periotics of true Kogia, as well as additional specimens of the larger species. These same large kogiid periotics are also reported from the Bone Valley Formation of Florida, the marine mammal assemblage of which is mostly undescribed. A large kogiid, Aprixokogia kelloggi, was named in the Lee Creek IV volume yet the skull did not include earbones. It is possible that these large periotics are attributable to Aprixokogia, but skulls with associated periotics are needed to be sure. What is troubling is that a recent donation to our museum (CCNHM) from the mine includes two fragmentary kogiid skulls – one of which appears to be Aprixokogia, and the other is a second non-Kogia kogiid whale – indicating the likely presence of three kogiids, though only two appear to be represented by periotics.