New Zealand Eomysticetidae: baleen whale origins

Waharoa fossil baleen whale and RW Boessenecker, Geology Museum, University of Otago

Holotype skull of the extinct eomysticetid baleen whale Waharoa ruwhenua (OU 22044) with Dr. Boessenecker for scale, Geology Museum, University of Otago. Photo by R.E. Fordyce.

Modern baleen whales are the largest animals to have ever existed (sorry, dinosaurs). Baleen whales (Mysticeti) and echolocating toothed whales (Odontoceti) share a common ancestor which likely lived sometime during the late Eocene epoch (~35 mya). Baleen whales differ from toothed whales in the lack of teeth as adults, lacking the ability to echolocate, and most notably by possessing baleen, a unique keratinous tissue (e.g. same general tissue type as mammalian hair, claws/nails/hooves, horns, etc.) which “hangs” from the roof of the mouth and serves to strain tiny prey (fish, krill, amphipods) from the water. The earliest baleen whales possessed functional teeth and appear to have lacked baleen altogether – likely feeding in a manner similar to archaeocete ancestors. The Aetiocetidae (Oligocene, North Pacific) were small (bottlenose dolphin-sized) whales with long palates and possessed both teeth and baleen, given a series of tiny foramina (holes) in the palate which provide blood supply to baleen (the same holes are present in modern mysticetes). The somewhat more advanced Eomysticetidae were some of the largest cetaceans of the Oligocene, similar in size to modern minke whales – but at a maximum of 7-8 meters in length, still much smaller than “medium” sized baleen whales today (e.g. humpback whales, Megaptera, ~15 meters). The first described member of this family, Eomysticetus whitmorei, was reported in 2002 from right here in Charleston, South Carolina (upper Oligocene Chandler Bridge Formation). Eomysticetus was noteworthy for its apparent lack of any teeth, a narrow, elongate skull, and an archaeocete-like postcranial skeleton. The lack of teeth meant that Eomysticetus was one of the earliest obligate filter feeders – since aetiocetids had both teeth and baleen, they could use either (facultative filter feeders). A much larger collection of eomysticetid skeletons had been made since the 1980s from the Oligocene Kokoamu Greensand and Otekaike Limestone of the South Island of New Zealand by Ewan Fordyce and colleagues, and in 2012 I began working on this amazing collection for my Ph.D. thesis research at the University of Otago in Dunedin, New Zealand.

Skull, excavation, and life restoration of the eomysticetid whale Tokarahia kauaeroa. Artwork by Chris Gaskin. From Boessenecker and Fordyce, 2015B.

Skull, excavation, and life restoration of the eomysticetid whale Tokarahia kauaeroa. Artwork by Chris Gaskin. From Boessenecker and Fordyce, 2015C.

When I started my Ph.D., I spent quite a bit of time trying to ask questions about the evolution of early baleen whales that could be answered with this large collection. The first question was “how many species are there? Some are obviously new – which specimens belong to what species?” Without this sort of a basic working framework for a project like this it might mean having to rewrite a substantial part of the research. Some individuals were clearly immature – which allowed some questions about baleen whale growth to be addressed. Certain species had multiple specimens from different stratigraphic levels represented, and other similar species appeared to be stratigraphically separated- permitting observations about relatively short-term changes in diversity to be made. Several eomysticetid specimens had well-preserved mandibles and palates, which provided some insight into the feeding adaptations of the earliest dedicated filter feeding mysticetes. Where did eomysticetids fit in to the tree of baleen whale relationships? Were eomysticetids really toothless? What did their postcranial skeleton look like?

Figure X - ML skull Marples 1956

The holotype skull (left, now lost) and surviving tympanic bulla of Tokarahia lophocephalus (from Boessenecker and Fordyce, 2015C)

Eomysticetid Diversity in the Southwest Pacific

Prior to my thesis and eventual published parts thereof, only Eomysticetus whitmorei had been published. A “small” mysticete from the underlying Ashley Formation (early Oligocene) of Charleston, South Carolina, Micromysticetus rothauseni, known only by a braincase, had been placed into a completely different family (“Cetotheriopsidae”) for some reason that was unclear to me; it really struck me as being eomysticetid-like. Within a few months of starting my Ph.D., Yamatocetus canaliculatus was named by Japanese paleontologist Yoshihiko Okazaki from the Oligocene Ashiya Group of Japan – the specimen included a virtually complete skull, hyoid apparatus (bones in the throat which anchor the tongue), both mandibles, much of the anterior vertebral column, ribs, and forelimbs including the scapulae, humeri, radii, and ulnae. It was an excellent point of comparison for the New Zealand material, though it lacked well-preserved earbones. of Tokarahia lophocephalus (from Boessenecker and Fordyce, 2015C)

Not all of the fossils I studied were recently collected. One of them was originally named Mauicetus lophocephalus, discovered from the same rocks in the 1940s and named in 1956 by former Otago Zoology chair Brian J. Marples. He placed it in the already-existing genus Mauicetus, the type species of which (Mauicetus parki) was based on an unassuming skull fragment from Milburn Quarry, just south of Dunedin. Other fossils that Ewan Fordyce had collected from the Otekaike Limestone turned out to be more complete specimens of Mauicetus parki, which showed that this species is certainly not an eomysticetid but rather more similar to Horopeta umarere recently described by my Otago labmate Cheng-Hsiu Tsai and Ewan (also from the Otekaike Limestone), as well as whales like Diorocetus hiatus from the middle Miocene Calvert Cliffs of Maryland. All this pointed towards Mauicetus lophocephalus needing a new genus name. Original illustrations of Mauicetus lophocephalus showed that it looked quite similar to Eomysticetus – but unfortunately, within 10 years of being named, the skull disappeared, likely thrown out when the Zoology Collections moved buildings in the 1960s. The remainder of the holotype including the earbones, mandible, and vertebrae were all spared, and these elements strikingly differed from Mauicetus parki and most fossil mysticetes except Eomysticetus. A few specimens shared quite a bit in common with M. lophocephalus, and in 2015 we named the most spectacular of these Tokarahia kauaeroa – the genus name after the Maori name of the type locality, Tokarahi, known as “Island Cliff” to English-speaking New Zealanders. The species name means “long jaw” in English, referring to the nearly 2 meter long gracefully narrow mandibles. We referred the species lophocephalus to the new genus, with the new combination Tokarahia lophocephalus – and referred an isolated earbone to each species, as well as a partial skull and skeleton to T. lophocephalus.

A separate specimen with similarites to another of Marple’s species, Mauicetus waitakiensis, was named Tohoraata raekohao, meaning “dawn whale with hole in jaw”; this specimen had particulary weird looking earbones, and was one of the largest (albeit incomplete) eomysticetids from New Zealand. We named this Tohoraata raekohao, and referred “M.” waitakiensis to Tohoraata, as the new combination Tohoraata waitakiensis. We named two other genera and species – Waharoa ruwhenua, based on a series of specimens with gorgeous skulls (showing the first growth series for an archaic mysticete) and Matapanui waihao, the oldest eomysticetid from New Zealand. Some of these were known only from the ~27-26 ma Kokoamu Greensand (Tohoraata waitakiensis, Matapanui waihao) or the somewhat younger (~26-24 Ma) Otekaike Limestone (Tohoraata raekohao, Waharoa ruwhenua). The two species of Tohoraata are stratigraphically separated, perhaps indicating an ancestor-descendant species pair, whereas both species of Tokarahia are known from the Kokoamu Greensand and the Otekaike Limestone, indicating that both species were at least somewhat sympatric (i.e. lived in the same region side-by-side, or perhaps during different seasons). Several closely related, anatomically similar mysticetes inhabiting the same region parallels modern balaenopterid whales (e.g. minke, blue, fin, humpback whales) which are commonly sympatric and diverse in today’s oceans.

Eomysticetid phylogeny

Cladogram showing eomysticetid relationships amongst baleen whales (from Boessenecker and Fordyce, 2016A)

Eomysticetid Relationships

One of the biggest questions when I started my thesis was “is the Eomysticetidae a ‘real’ biological group?” In modern biology, we use cladistics to study evolutionary relationships (phylogeny). In order to be biologically “real”, a group has to include an ancestor and all of its descendants. Many historically named groups of organisms have turned out to be “grades” – nonavian (e.g. non-bird) dinosaurs are a classic example of an evolutionary grade, given that birds are within Dinosauria. Discussing the name “Dinosauria” only makes sense in modern biology if birds are included within. Many older groups of extinct whales and dolphins are similarly grades – what we call paraphyletic. Examples include the “Crenatoceti” (toothed mysticetes), “Archaeoceti” (ancient whales), “cetotheres” (a para- or polyphyletic group of baleen whales, which has recently been re-defined around type species Cetotherium rathkii), “Squalodontidae” (shark toothed dolphins), and various antiquated concepts of the Platanistoidea (river dolphins). In order to examine eomysticetid relationships, I assembled the largest cladistic dataset ever attempted for baleen whales – anatomical characters such as “presence of baleen” or “adult dentition” are coded into a matrix as 0 or 1 (or in some multistate characters, 2, 3, 4, etc.) for nearly 80 different species and analyzed in the computer program TNT. The end result is the best available phylogeny for baleen whales – if you follow the mantra that “more data is better”. These results indicate that 1) Eomysticetidae is indeed a clade – a natural grouping of species that briefly flourished during the Oligocene and went extinct around the Oligo-Miocene boundary (~23 Mya) and 2) placed as the “sister taxon” to all extant baleen whales. Neither result is particularly surprising, but in both cases no prior study had adequately demonstrated either case to be likely based on robust data. We further discovered that Micromysticetus rothauseni is indeed an eomysticetid as I originally suspected prior to starting the thesis.

Tohoraata 3 copy

Life restoration of Tohoraata raekohao, showing the elongate and narrow snout of eomysticetids. Note that I did this reconstruction prior to the discovery of teeth in Tokarahia. Artwork by R.W. Boessenecker.

Eomysticetid Feeding Adaptations

The elongate narrow snout of eomysticetids is one of the hallmark features of the group. The snout is proportionally more narrow than any other mysticete – indicating some sort of specialization. Holes on the palate (foramina) which transmit arteries to the baleen in modern whales are present along nearly the entire palate; in Waharoa ruwhenua (genus name ‘long mouth’, species name ‘shaking land’, referring to the type locality known as “The Earthquakes”) these foramina are restricted to the posterior half of the palate – suggesting that baleen either may have been absent from the front of the mouth or at least reduced. The jaw joint of Tokarahia and Waharoa has a distinct “socket” for the mandible, which in most mammals indicates that a regular synovial joint capsule was present. Lunge-feeding rorquals (e.g. minke, blue, fin whales) swim at top speed and open their jaws, rapidly inflating a large throat pouch around prey items; the water is slowly expelled through the baleen. These whales have a much more flexible fibrocartilaginous joint and lack such a ball-and-socket construction – indicating that eomysticetids probably were not able to feed in the dramatic fashion of humpback and fin whales. The posterior mandible of eomysticetids is very delicate, and constructed as a hollow shell with bone thinner than cardboard – further suggesting an inability to lunge feed. Enormous jaw-closing muscles likely stabilized the elongate mandibles, and we inferred that the elongate snout acted to increase the cross-sectional area of the filter feeding apparatus (baleen) in a manner similar to right and bowhead whales. These whales are skim feeders – slowly cruising through the water column, a bit like a vacuum cleaner. These whales also have extraordinarily long palates which are also vertically expanded. Eomysticetids have flat snouts, but were well on their way towards feeding like right whales. Interestingly, in some phylogenies eomysticetids and balaenids appear as successively positioned branches on the mysticete tree – perhaps indicating that skim feeding is the ancestral feeding behavior for all mysticetes.

Life restoration of Waharoa ruwhenua showing dentition; coloration inspired by modern bowhead whale. Artwork by R.W. Boessenecker.

Life restoration of Waharoa ruwhenua showing dentition; coloration inspired by modern bowhead whale. Artwork by R.W. Boessenecker.

Tooth Loss in Baleen Whales: were Eomysticetids Toothless?

Conventional wisdom held that Eomysticetus was toothless, and that it was one of the first obligate filter feeders – distinguishing it from earlier diverging baleen whales like Aetiocetus which likely had both teeth and baleen. However, when Eomysticetus was first described, the assessment of its dentition was based on the type specimen which had an incompletely preserved rostrum (snout) and a damaged pair of mandibles. The more completely preserved Yamatocetus actually had a completely preserved and well-prepared margin of the thin palate and complete mandibles, and stunningly exhibited a series of poorly defined alveoli (tooth sockets). However, there were no teeth, and the alveoli were restricted to the anterior 1/3 of the mouth – still quite different from Aetiocetus. Okazaki interpreted these as housing adult teeth – and I was quite skeptical, at least until we found one. During preparation of a referred skull of Tokarahia lophocephalus in my last year (actually, after our manuscript on the specimen had already been submitted for review!), Ewan found a strange tooth near the edge of the palate; the tooth lacked a crown, but the root had clear cementum and a core, clearly not a shark or fish tooth and its flattened cross-section distinguished it from Oligocene odontocetes. Rather, it matched the shape of maxillary alveoli in Yamatocetus; furthermore, specimens of Waharoa all have three alveoli in the premaxillae and anterior mandible – suggesting adult incisors at minimum. The tooth would have been peg-like and likely non-functional; the alveoli are shallow and small, and wide diastema (gaps) are present between each alveolus. We interpreted this dentition as being non-functional as far as feeding is concerned, but perhaps there may have been some social function such as the tusks which beaked whales use to joust with one another. The presence of such teeth in the front of the oral cavity suggests that the posterior teeth were resorbed prior to baleen formation; in modern baleen whales, teeth are developed and then completely resorbed prior to birth. Eomysticetid dentitions could predict that the teeth are resorbed from back to front in modern baleen whales – but more dissections of fetuses would be necessary to find out.

Skull growth in the eomysticetid whale Waharoa ruwhenua (from Boessenecker and Fordyce, 2015B)

Skull growth in the eomysticetid whale Waharoa ruwhenua (from Boessenecker and Fordyce, 2015A)

Growth in early baleen whales elucidated by Waharoa ruwhenua

When I began my Ph.D., nearly nothing was known about skeletal growth in fossil cetaceans. Unfortunately, the same was true for modern whales and dolphins; most of our “good” data on modern whale skeletal anatomy hails from the days of whaling, and whalers tended to only kill adult individuals. Occasional studies of fetuses were conducted when pregnant females were taken. But as far as juvenile whales – newborns to 2/3 full size – nearly nothing is known and few specimens exist in museum collections. Though this constitutes less than 10% of published specimens and specimens altogether in museums, perhaps half of the baleen whale fossil record hails from this part of the whale’s lifespan. This poses serious problems for interpreting juvenile fossils: are the anatomical differences caused by growth changes, or are they entirely different species? How do we recognize different growth stages in the fossil record? These basic questions have generated substantial controversy particularly within dinosaur paleontology – different species of horned dinosaurs (Triceratops, Torosaurus), dome-headed dinosaurs (Pachycephalosaurus, Stygimoloch, Dracorex) and tyrannosaurs (Tyrannosaurus, Nanotyrannus) have all been identified as ontogenetic synonyms – and violently debated by researchers. Growth in mammals is generally not a problem because most of mammalian taxonomy is based upon teeth – and since mammals only get two sets of teeth (milk, adult) the taxonomy is quite stable. However, cetaceans generally either 1) lack teeth altogether (e.g. most baleen whales) or have teeth that are identical (most toothed whales) – differing wildly from the ease at which noncetacean mammals can be interpreted. To make matters worse, cetaceans grow to enormous sizes and may not be recognizable as different growth stages of the same species. Occasionally enough fossils of a particular species will turn up with features that clearly unite them to the exclusion of other relatives – such was the case with the species we eventually named Waharoa ruwhenua. Three skulls of an adult, a large juvenile, and a small juvenile (all from the Otekaike Limestone) had completely different proportions of the mandibles and rostrum as well as the atlas vertebra – but all three preserved the anatomically informative and diagnostic earbones. Some very slight differences in the earbones were obvious, but generally showed less growth-related change than modern baleen whales – instead being somewhat more similar to the growth changes in modern odontocetes. The tympanic bulla (outer earbone) increased in length by about 1cm from small juvenile to adult, potentially affecting identification of isolated tympanic bullae which were previously assumed to be less variant in size during growth. The most significant change was the elongation of the rostrum, which proportionally doubled during postnatal (e.g. post-fetal) ontogeny, strongly reinforcing the interpretation of a long rostrum in eomysticetids as being a specialization for skim feeding. Perhaps a short rostrum was important for suckling milk as a calf – shorter snouts are better for suction feeding.

Further Reading

Boessenecker, R. W., and R. E. Fordyce. 2015A. Anatomy, feeding ecology, and ontogeny of a transitional baleen whale: a new genus and species of Eomysticetidae (Mammalia: Cetacea) from the Oligocene of New Zealand. PeerJ 3:e1129.

Boessenecker, R. W. and R.E. Fordyce. 2015B. A New eomysticetid (Mammalia: Cetacea) from the Late Oligocene of New Zealand and a reevaluation of “Mauicetuswaitakiensis“. Papers in Palaeontology 1:107-140.

Boessenecker, R. W., and R. E. Fordyce. 2015C. A new genus and species of eomysticetid (Cetacea: Mysticeti) and a reinterpretation of “Mauicetuslophocephalus Marples, 1956: transitional baleen whales from the upper Oligocene of New Zealand. Zoological Journal of the Linnean Society 175:607-660.

Boessenecker, R. W. and R. E. Fordyce. 2016A. Matapanui, a replacement name for Matapa Boessenecker & Fordyce, 2016. Journal of Systematic Palaeontology. Online Early DOI:10.1080/14772019.2016.1210070

Boessenecker, R. W. and R. E. Fordyce. 2016B. A new eomysticetid from the Oligocene Kokoamu Greensand of New Zealand and a review of the Eomysticetidae (Mammalia, Cetacea). Journal of Systematic Palaeontology. Online Early DOI: 10.1080/14772019.2016.1191045