Narwhals are certainly among the strangest and most immediately recognizable of all marine mammals, owing to their fantastic tusk. Narwhals grow to about 20 feet (~6-7 meters), are white and brown and frequently spotted, and inhabit the Atlantic and Russian parts of the Arctic, especially the west and east coasts of Greenland. They are a member of the 'white whale' family - the Monodontidae - though nearly the entire fossil record of monodontids have been tuskless beluga-like forms. Narwhals have always been a bit mysterious, with their tusks being sold in Medieval times as 'unicorn horns' - traded south to temperate Europe by Vikings. What do we know about the evolution of the narwhal, and its fossil record?
Biology and Ecology of the Narwhal
The narwhal is not a terribly large whale - it attains a length of 3-5 meters, and unsurprisingly for a species known with secondary sex characteristics like tusks in males - the males are slightly larger than females (maximum size of 5 v. 4 meters). It's also got unusual coloring - it's frequently spotted with a white background and brownish spots concentrated on the back for some countershading, generally resembling harbor seals; spots are also seen in the spotted dolphins (Stenella attenuata, Stenella frontalis) and some bottlenose dolphins, especially in the Gulf of Mexico (Tursiops truncatus). Their caudal flukes have a distinctive shape, being nearly a diamond with a posterior notch rather than the more triangular flukes of nearly all other whales; in this they resemble belugas, and to a lesser extent, the sperm whale. Narwhals live chiefly in the Arctic between latitudes of 65 and 85 N, and mostly between 70 and 80N, around both coasts of Greenland in Baffin Bay and the Greenland Sea, east of Svalbard towards the Laptev Sea, and formerly south towards Nova Zemlya and east towards the Chukchi Sea. For comparison, belugas range throughout the Arctic, south of the Bering Strait and to the Aleutians, northern Hudson Bay and throughout the Canadian Archipelago, to the mouth of the St. Lawrence River in eastern Canada, and the northern coast of Norway and the entire northern coast of Russia.
A group of male narwhals photographed by a drone. Photo credit: Brian Skerry.
Male narwhals become sexually mature at about 11-13 years and a length of about 4 meters (~13 feet), and females at 5-8 years after attaining a length of 3.5 meters (~11 feet); females maturing faster and at smaller body sizes than males is a hallmark feature of mammals that have sexually competitive males that mate with multiple females - otherwise known as polygyny. Polygynous marine mammals are usually also sexually dimorphic, like narwhals, and have secondary sex characteristics - frequently something as subtle as having proportionately more robust skulls and teeth in something like a sea lion, or having enormous bizarre tusks like a narwhal (or walrus, for that matter). Much of the reproductive behavior of narwhals is unknown (see below) but they are presumed to by polygynous based on their rather extremely dimorphic tusks and body size dimorphism. Gestation lasts about 15 months, with births in late summer most common. Calves are born at a length of about 1.5 meters (five feet) and nurse for about a year and a half. Analysis of amino acids in the eyes of narwhals suggest that males can live for over 80 years and females for 115 years - and females furthermore undergo menopause and live for decades later, helping younger females rear calves. Inuit peoples have hunted narwhal since the arrival of the Thule culture about 1,000 years ago or so, and during the early medieval period, Vikings began hunting narwhal and selling the tusks south to Europe.
Molecular phylogenetic analyses consistently recover a sister taxon relationship between the porpoises (Phocoenidae) and the white whales (Monodontidae), leading Geisler et al. (2011) to name this clade the Monodontoidea. Molecular clock analysis by McGowen et al. (2009) found that the Monodontoidea originated (and split) about 15.5 Ma in the early middle Miocene, at least 6-7 million years prior to the earliest phocoenid fossils (Salumiphocoena stocktoni from the Monterey Formation was used as a fossil calibration for these clades). The narwhal-beluga split was dated to about 6.3 Ma, in the latest Miocene - post-dating some fossil monodontids (see below).
The spectacular Narwhal tusk
Most tusks within the Mammalia are curved somehow, and frequently have non-symmetrical cross-sections. Walrus, hippo, elephant, etc. Modern elephant tusks are circular in cross-section, but curve about their long axis, and mammoth tusks curved in at least two. Narwhal tusks are considerably unique in the fact that the tusk is ramrod-straight - which is accomplished by the tooth twisting about its longitudinal axis (Kingsley and Ramsay, 1988); you can follow individual ridges in the dentine like the stripes on a candy cane. This is critical as it makes the tusk rather distinctive and easy to identify relative to other commonly encountered mammal tusks in Northern hemisphere marine deposits (e.g. walrus, elephant). There are at least a couple of fossil elephants with relatively straight tusks that spiral - the Plio-Pleistocene gomphothere Cuvieronius, for example - but it's only somewhat spiraled.
Cross-section of a narwhal rostrum showing the location of each tusk alveolus and implantation of each tusk in the maxilla, identifying the tusk as a canine. From Nweeia et al. (2012).
Male narwhal skull with the tusks removed, clearly showing the alveolus completely within the maxilla. From Nweeia et al. (2012).
There has been a bit of controversy about the homology of the tusk. Tusks after all are enlarged teeth - they have the same layers present in a tooth - chiefly dentine and cementum, but some mammal tusks have a layer of enamel that frequently gets worn away (some extinct gomphothere elephants have a layer of enamel on one side of the tusk). One problem with determining the likely tooth position, however, is that narwhals generally only have a single tooth that erupts from the skull - the tusk! They are more or less functionally toothless: they lack teeth used for grasping prey, and only retain highly modified tusks - not unlike the "toothless" walrus Valenictus, and the sheer majority of beaked whales (Ziphiidae). If tusks are teeth, than the question remains - what tooth position? In elephants (=proboscideans) the tusk is generally an enlarged incisor. Perhaps owing to this determination, for a long while the narwhal tusk was assumed to be an incisor as well. However, a 2012 study by Nweeia et al. found that the tusk is embedded in the maxilla, rather than the premaxilla. This is critical, as the premaxilla only houses incisors, whereas the canines, premolars, and molars are embedded in the maxilla in all mammals that have them. Because smaller vestigial teeth were found lateral to the tusk in dissections and CT imaging of narwhal skulls, and the tusk is embedded in the medial part of the maxilla, this study identified the tusk as actually being a canine tooth. The remaining vestigial teeth are quite variable in their morphology, but are typically 'procumbent' - forward facing - parallel with the tusk.
Males typically are the only individuals with erupted tusks, though it's critical to note that ALL narwhals have two tusks. Males typically erupt the left tusk (measuring 1.5-3 meters in length), whereas the right is much smaller and rarely erupts from the skull. In some individuals, the right tusk erupts as well (in ~1% of males), rarely becoming as long as the left tusk*. Females also occasionally possess an erupted tusk, though generally shorter and narrower than males. This variation indicates that the tusk is very sexually dimorphic.
*I've seen more photographs of double-tusked narwhals on display in museums than single-tusked, however. This is an interesting, but very understandable bias.
It's nice to know that double-tusked narwhals exist in a state other than dead, skull mounted in a museum somewhere! Drone image from Greenland by Jens Ascanius, from Garde and Heide-Jorgensen (2022).
But what is this incredible tusk for, how is it used, and how did it evolve? Tusks (defined here as an enlarged tooth that permanently sticks out of the mouth), after all, are frequently used for sparring in elephants and walruses, a bit like deer antlers - and in these other species, tusks are also sexually dimorphic, but far less extremely so: female walruses and elephants typically also have tusks that happen to be shorter or more gracile than males. Narwhal tusk size dimorphism is admittedly far more extreme, with 1.5% of females presenting an erupted tusk. Whenever such dimorphism is clear, the structure is most often used in either some form of sexual display or combat between males. Think about deer: antlers are almost always only present in reproductively mature males, who lock their antlers together and spar in order to compete for females. Intuitively, it only makes sense that combat is involved somehow. However, in the past decade or so, a new narrative has emerged - that the tusk is *not really* used for combat, and that it evolved for other reasons.
Statistical relationship between the mass and length of narwhal tusks v. testes mass. From Kelley et al. (2014).
What have we observed? Tusk-crossing behavior has long been observed in narwhals, but hardly the 'sword-fighting' that is frequently depicted in cartoons - this behavior, observed at the sea surface, usually involves some rubbing of the tusks together. Hardly aggressive behavior. However, it's very likely that combat does occur - either below the waves and out of view, or in regions where humans are not observing the behavior (the Arctic is a very big place - and this is what prevented the observation of walrus tusk use for quite a long while). A landmark study by Silverman and Dunbar (1980) found that there are many scars on the heads of male narwhals with erupted tusks, and that there's a strong correlation between body length and the number of head scars. Additionally, these authors report from the same sample that the tips of tusks are broken in about 60% of the sample in adult males, whereas only 10% of juveniles have broken tusks*. In one case, these authors report a 9 cm long tusk tip embedded in the rostrum of another male. Unfortunately, this spectacular specimen was not figured by them. A later study by Gerson and Hickie (1985) found that the number of head scars correlated strongly with body width, tusk width, and tusk weight, but not to tusk length or body length. Male narwhals with heavy, wide tusks also tended to be quite robust in their mass and 'girth' - but typically no longer in body length. This suggests that after achieving maximum body length, male narwhals continue to grow to be more robust in their proportions - developing greater muscle mass and tusk mass. These authors also measured the mass of testes, finding that testis mass was also strongly correlated with heavier tusks and body mass (but again, not body length, which seems to plateau. All of this suggests that male narwhals with the strongest tusks are the most likely to be the victors in as-yet-unseen narwhal combat, and the strong correlation with testes mass further indicates that the tusk is very likely to be a signal of male reproductive viability. A more recent study by Kelley et al. (2014) confirmed the strong link between testes mass and tusk mass. Given the strong sexual dimorphism, almost assuredly a result of sexual selection - females must be very, very choosy - driving the evolutionary trajectory of this bizarre, incredible structure.
*One caveat to this is that because tusks do not re-grow, the probability of tusk damage from any behavior (so long as tusk breakage is possible) increases with age and does not a priori indicate male combat. In other words, if narwhals are using their tusks to flip rocks on the seafloor or poke holes in ice, juveniles will have a lower incidence of broken tusks as well.
Mechanical studies of the tusks of narwhals indicate that they have low stiffness but high toughness - meaning that the tusk is not very resistant to bending or loading along its longitudinal axis (e.g. front to back) but is quite resistant to sudden impacts (Brear et al., 1990, 1993) - think tusks crossing or even the comical sword fighting frequently. Curiously, the mechanical properties of the tusk are similar to bone and compare quite well with reindeer antler (Brear et al., 1990).
However, the story doesn't end there. Some studies of the microanatomy (histology) of the narwhal tusk indicate that the dentine is absolutely riddled with tubules - like many other mammals - but more critically, these tubules extend through the cementum and are open to the marine environment (Nweeia et al., 2009). These tubules typically hold fluid that communicate with nerves in the pulp cavity; in human teeth, tubules are very dense near the pulp, and radiate outward and terminate at the enamel-dentine junction. These tubules, when covered by enamel - help sense pressure, air v. fluid, salinity, sugar, and of course temperature - complimenting the sensory capabilities of the tongue and gingiva. When tooth decay begins, and the enamel is compromised - a much-dreaded cavity forms at the junction, with dentine eroding - and the tubules are exposed to the fluid in the 'oral environment' (saliva). This is precisely what causes a toothache. What this finding implies is that narwhals - with these tubules exposed permanently - have very, very sensitive tusks. But, sensitive to what? Preliminary field experiments detailed by Nweeia et al. (2009) suggested that the tusk was sensitive to changes in salinity.
A followup study by Nweeia et al. (2014) reported a myriad of data on narwhal tusks, including the finding that the pulp of the tusk is highly innervated. Perhaps the most interesting and remarkable finding are the experiments performed with temporarily captured narwhals; these individuals were captured in nets using the same methods for tagging wild narwhals, and then a special "jacket" placed over the tusk; this jacket was filled with freshwater and saltwater, while a series of electrodes monitored heart activity - literally, and EKG (electrocardiogram). When the salinity was changed, the heart rate changed in all six briefly captured whales (Nweeia et al., 2014).
In 2017, footage emerged showing narwhals stunning fish by clubbing them with their tusks - suggesting that perhaps they serve some purpose in foraging. While this does not seem to be published yet, it could mean that the tusk has evolved for use in catching fish. Clubbing has been proposed to explain many longirostrine dolphins in the Miocene (though I don't really buy it, to be honest).
So, what did the tusk evolve for? Sex, detecting changes in water chemistry, or feeding? Any argument that the sensitivity of the tusk drove the evolution of the structure requires vastly different behavior between males and females - such a biological sensor would no doubt be useful to females as well as males. Nweeia et al. (2014) suggest that perhaps males might be the navigators and lead females and juveniles to waters with ideal salinity for feeding - or perhaps might guide males to waters with ideal salinity to find mates.
The surprising fossil record of tuskless Monodontidae
Before we get to the fossil record of narwhals, I need to at least briefly summarize the fossil record of other white whales. While there has been some debate over whether fossil white whales were adapted to cold or warm waters (see below), the truth is that most fossil white whales are found at latitudes far, far lower than the modern two species - which in and of itself illustrates a separate problem (see the last section). The lower latitude occurrence of fossil white whales has been obvious for quite a while, with the discovery of the skull of Denebola brachycephala in 1965 from Baja California - though not named until 1984, it was recognized as a beluga as early as Larry Barnes' PhD thesis in 1972. Denebola brachycephala is one of several odontocetes named in Barnes' (1984) paper on odontocetes from the Almejas Formation of Cedros Island. Denebola is small - the skull is only 40 cm long and about 30 cm wide, roughly the size of a bottlenose dolphin skull - but with a relatively short, blunt rostrum that is quite similar to certain 'blackfish' whales (Globicephalinae) like pilot whales (Globicephala), Risso's dolphin (Grampus), and the Irrawaddy/snubfin dolphins (Orcaella). Barnes specifically compared the rostral proportions with pilot whales, but stopped short of hypothesizing convergent evolution. Unlike narwhals, Denebola has a full compliment of homodont teeth and nothing resembling a tusk - there are 14-15 tooth sockets, though they are oriented anteriorly, like modern belugas - and there are no teeth in the premaxilla, like both modern monodontids. Denebola is one of the only monodontids with an associated periotic - inner ear bone - and it is much more similar to Delphinapterus than to Monodon. The periotic has a very large and anteriorly thrusted 'pars cochlearis', the bulbous part housing the cochlea. The marine vertebrate fauna from the Almejas Formation is broadly similar to other assemblages further north in California including the Purisima, Capistrano, and San Mateo formations, and in my opinion, doesn't seem to have much in the way of obvious warm water species except for a possible pufferfish (Diodontidae indet.), a butterfly ray (Gymnura sp.), and a devil ray (Mobula sp.) suggestive of warm temperate waters; instead, there are a number of temperate to cold temperate sharks and fish present (salmon shark, cf. Lamna; California sheepshead, Semicossyphus sp.; wolf eel, Anarchichadae indet.; Barnes, 2008: table 1). In general, the non-marine mammal fauna seems quite similar to the modern fauna off the coast of northern Baja and southern California today. The Almejas Formation is approximately 6-8 million years in age, making Denebola one of the oldest named white whales.
An unnamed white whale was reported, but not named, on the basis of a fragmentary skull from the Pliocene Kattendijk Formation of Belgium by Lambert and Gigase (2007). This skull is also somewhat smaller than modern belugas, about the size of a bottlenose dolphin. Though most of the rostrum is missing, it clearly does not have a large tusk like Monodon. The configuration of the facial bones is similar to both belugas and narwhals, and does not necessarily seem more closely related to either extant species. The Kattendijk Formation was deposited under cold temperate conditions. Lambert and Gigase (2007) also reported several isolated periotics (earbones) from late Neogene strata in Belgium - all have a relatively large pars cochlearis like Delphinapterus, and differing from the relatively small condition of this feature in periotics of Monodon.
One of the earliest discovered fossil monodontid skulls indicating that belugas had been inhabiting temperate latitudes for a long while was a Beluga-like skull, identified for years as Delphinapterus, and named Bohaskaia monodontoides in 2012 by Jorge Velez-Juarbe and Nick Pyenson. Superficially this skull looks quite a lot like Delphinapterus leucas: it has a broadly triangular but blunt rostrum, and a polydont, homodont dentition lacking an obvious tusk - and like Delphinapterus, the teeth are slanted forwards somewhat (judging from their sockets). Tooth sockets are also not present in the premaxilla, like both modern white whales, and differing from the extinct Baja beluga Denebola brachycephala, which has one socket per premaxilla (recall that three is primitive for mammals and most cetaceans). The skull is somewhat larger than Denebola, being slightly wider (~35 cm) but having a much longer rostrum (~55-60 cm total length). However, the skull has a mix of features that are primitive, more similar to Delphinapterus, or alternatively more similar to Monodon. Amongst these latter features is a skull vertex that is quite narwhal-like; features of the rostrum are more similar to Delphinapterus. As it happens, these rostral features are likely symplesiomorphic, as the rostrum of Monodon is quite modified for possession of a big-ass tusk; indeed, these vertex features evidently led phylogenetic analysis to support a Bohaskaia + Monodon clade in one of the only phylogenetic analyses of the Monodontidae - undertaken later by Bianucci et al. (2019, see below). Isolated periotics were reported earlier from the Yorktown Formation by Whitmore and Kaltenbach (2008) in the Lee Creek IV volume, identified as Delphinapterus sp. - being nearly identical in morphology according to those authors. Additionally, an expansive analysis of the many, many isolated humeri from the Yorktown Formation by Kazar and Bohaska (2008; also in the Lee Creek IV volume) revealed that among the sixteen humerus morphotypes, at least two, and possibly even three monodontids were present. Morphotype 11 was identified as Delphinapterus sp., but one unusually long specimen is similar to Monodon and identified tentatively as cf. Monodon. A second suite of similarly shaped, but generally smaller humeri were labeled Morphotype 12, and identified as cf. Delphinapterus. This makes the Pliocene Yorktown Formation of North Carolina the only multispecies assemblage of white whales known from the fossil record. Sadly, no tusk fragments are yet known from the Lee Creek Mine - though it is possible to envision a tuskless Monodon ancestor.
The juvenile, very short snouted holotype specimen of the Pliocene-aged Haborodelphis japonicus, the first formally named monodontid from Japan. From Ichishima et al. (2018).
In 2018 the new dwarf beluga Haborodelphis japonicus was named by Hiroto Ichishima et al. The specimen was collected from the Embetsu Formation near Haboro on the island of Hokkaido - perhaps the island with the richest marine mammal fossil record in Japan. The Embetsu Formation is early Pliocene in age, so Haborodelphis is about 4-5 million years old. Haborodelphis is downright tiny - the skull is about 34 cm long and 21 cm wide - smaller than a bottlenose dolphin (about 2/3 the size) and just a little larger than a harbor porpoise; Ichishima et al. note that it was likely about two meters long when alive - but that it was also likely a juvenile or subadult, so perhaps reaching a size similar to Denebola as an adult. It has a very short, cute little rostrum that makes up less than half of the length of the skull, about 40% - and like all other named fossil white whales, a 'normal' dentition with numerous teeth and no sign of a tusk (a tooth count was not noted by Ichishima et al., because the specimen is so young that there is a groove but no individualized sockets). Haborodelphis preserves very nice earbones - and, once again, the periotic looks much, much more like Delphinapterus than Monodon. Ichishima et al. rightfully highlight this discrepancy and that all published fossil periotics of the family have an enormous pars cochlearis, differing from the condition in Monodon and more closely resembling Delphinapterus. They entertain two possibilities: one is that we haven't really found any "monodontine" periotics yet in the fossil record, and the second is that the smaller pars cochlearis in Monodon is a recent evolutionary feature, and that narwhals evolved from a more generalized ancestor. At face value, this sounds good - it would explain the fact that all fossil periotics of white whales look kind of similar, and among these, maybe some other features of Monodon might be able to be picked out - though from past conversations, I already suspect Hiroto might be a bit skeptical as he is generally less positive than I am about identifying isolated odontocete periotics.
The holotype skull of Casatia thermophila, named by Bianucci et al. (2019) from Pliocene rocks in Italy.
The most recently named fossil monodontid was named Casatia thermophila by Bianucci et al. (2019) based on a partial skull in a concretion with a well-preserved facial region, discovered in lower Pliocene (5-4.5 Ma) strata of Tuscany, Italy. An additional specimen was reported later by Merella et al. (2022), consisting of a superficially more complete specimen but it is a bit more crushed and fractured. Most of the facial structures are shared with Delphinapterus, though the specimen does have raised premaxillary eminences just in front of the bony nares, like Monodon - and is unique in having a depression just in front of these eminences and anteriorly converging rather than parallel sutures between the maxilla and premaxilla. The referred specimen also preserves some features of the ventral side of the skull, including a couple more monodontid synapomorphies of the palate. Otherwise, the remaining noteworthy findings by Bianucci et al. (2019) and Merella et al. (2022) are concerned with phylogenetics and climate adaptations of extinct monodontids, covered below.
Phylogeny of monodontids - the only analysis yet attempted - from Bianucci et al. (2019). Based on only 21 characters and ten taxa.
Surprisingly, there has only been a single phylogenetic analysis of fossil monodontids ever attempted, though some previous analyses of broader relationships amongst fossil and modern odontocetes included Denebola brachycephala. It wasn't until Bianucci et al. (2019) that the phylogenetic position of a fossil monodontid was analyzed by the describers. In their analysis, Bianucci et al. used a very small matrix composed of 21 characters coded for ten taxa, seven of which are monodontids: Monodon, Delphinapterus, Denebola, Casatia, Bohaskaia, Haborodelphis, and the unnamed specimen from the Kattendijk Formation, IRSNB M 1922. Haborodelphis and Denebola were unsurprisingly found outside the Monodon + Delphinapterus crown clade, perhaps suggesting a Pacific center of origin for the clade (further suggested by the old age of these). Bohaskaia was recovered as the sister taxon to Monodon, and Casatia and IRSNB M 1922 as a clade sister to Delphinapterus.
Southern Hemisphere white whales?
In a short review paper on the stratigraphy and vertebrate paleontology of the Pisco Formation, Muizon and DeVries (1985) listed an indeterminate monodontid, perhaps a delphinapterine, as being derived from the Sacaco horizon of the Pisco Formation (now considered to be latest Miocene). The specimen was finally figured by Ochoa et al. (2021) - well-preserved periotic, similar to Denebola and Delphinapterus. Curiously, this specimen overlaps in time with Odobenocetops - long suspected to be a monodontid relative - though is very, very clearly not even remotely confusable with the periotics of Odobenocetops. This tantalizing little earbone is, at present, the only solid evidence of the family being found south of the equator. The Pisco Formation marine mammal assemblage shares quite a bit in common with the Chagres Formation of Panama (Benites-Palomino et al., 2023) and the Almejas and Capistrano formations of California/Baja California, and it's most likely that a bit of periodic equatorial cooling permitted some Denebola-like early monodontid across the equator.
How exactly does Odobenocetops fit into this picture?
Speaking of the Pisco Formation, we finally get to the 'elephant' in the room - what the hell is Odobenocetops? One of the most bizarre and confusing mammal fossils ever discovered, excavated by none other than Pete Larson of the Black Hills Institute on one of the last field expeditions to Peru prior to the laws being enacted to protect vertebrate fossils. Most of the important fossils from these expeditions ended up at the Smithsonian, including the Odobenocetops peruvianus holotype, which Larson donated for study. Odobenocetops was, according to Larry Barnes, mistaken for an actual walrus until - and I repeat, according to Larry - pointed out that the skull had premaxillary sac fossae and pterygoid fossae, pretty clearly indicating that it was in fact some kind of really, really effed-up odontocete. Apparently there's an abstract from the late 1980s or 1990s, allegedly in JVP, where the specimen is presented as a walrus until Larry set 'em straight - though I've not been able to find it.*
*My recollection from my conversation with Larry, which took place in his office in 2013, was that it was JVP but it just as easily could have been a Society for Marine Mammalogy or NAPC conference.
Odobenocetops is quite strange because it has reversed much of its cranial telescoping and rather than having a typical tongue-shaped or triangular rostrum like virtually all other odontocetes, it's got this strange pair of "alveolar processes", the right side of which possesses a robust tusk (opposite from Monodon monoceros; recall that in narwhals, the left tusk is the one that is typically erupted); in more complete specimens of the later-named Odobenocetops leptodon, this tusk is nearly four times as long as the skull. However, unlike Monodon, the tusk points backwards and ventrally - like the downwards pointing tusks of the modern walrus. A vaulted palate and lack of non-tusk teeth parallels both the walrus and Monodon, leading to a benthic suction feeding hypothesis for the strange whale. More recently, it's been recorded in similarly aged deposits near Caldera, Chile, but remains undescribed.
Is Odobenocetops a relative of the monodontids? That big-ass tusk seems to suggest so - though it's curious that no fossil white whales that are obviously so much closer in skull anatomy have one. Muizon (1993) identified a few features in a hand-drawn phylogeny (an "intuitive cladogram" was what my Ph.D. adviser, R.E. Fordyce, called these), including a lateral lamina of the palatine bone that floors the optic groove and contacts the frontal bone, a thickening of the alisphenoid bone between the zygomatic process and the "foramen ovale" in the basicranium, and an anteriorly placed eye socket. The first two seem to be good synapomorphies to me at first glance; the latter could easily be convergent given the enlargement of the eye sockets and relative weirdness of the skull of Odobenocetops. Muizon and Domning (2002) reported some additional possible features linking the Odobenocetopsidae with the Monodontidae, including a loss of the stapedial foramen within the stapes - the stirrup in the three middle ear ossicles. This hole is the part of a stirrup where your foot goes in - the stapes is essentially just a rod, though not figured by these authors. Other features regard the flexibility of the neck, but this could very well be convergent as well - an obvious adaptation for benthic feeding in Odobenocetops, and a flexible neck is widely regarded as being plesiomorphic (primitive) in modern white whales.
The periotic (inner ear bone) of Odobenocetops is, well, baffling. It barely has any resemblance to any modern delphinoid. The overall shape of the periotic is much more reminiscent of an Oligocene dolphin than a delphinoid - it's sort of rectangular in medial view, and has a long, blade-like anterior process seen in Xenorophidae, Agorophius, and one big old dolphin near and dear to my heart - Ankylorhiza. The periotic is also gigantic: nearly 6 cm long in O. peruvianus, and nearly 8 cm long in O. leptodon: these are the size of baleen whale periotics. They possess one major feature that links them with the Delphinoidea (the clade formed by the Delphinidae, Monodontidae, and Phocoenidae - the most derived group of modern odonceotes) - the loss of an articular facet for the tympanic bulla on the anterior process. However, this independently happened in mysticetes, and could be convergent. The shape of the pars cochlearis, in my opinion, is also very reminiscent of the condition in ziphiids (beaked whales) and physeteroids (sperm whales). The bulla is oddly shaped as well, lacking a transverse groove seen in most non-delphinoids, though in my experience this feature is variable and absent in many Oligocene odontocetes. Superficially, the preserved part of the bulla resembles some features seen in sperm whales. Further, Muizon and Domning (2002) note that the rather bizarre and surprisingly enormous malleus (the largest of the three middle ear ossicles) has more in common with archaic odontocetes like Squalodon than with delphinoids. Curiously, Odobenocetops has only been coded into one phylogenetic analysis with broad sampling of all crown odontocete clades - Mizuki Murakami's (2014) paper on Eodelphinus kabatensis. In this paper, he finds it to be the sister taxon to the Monodontidae.
In sum, I think Odobenocetops may have some sort of relationship with monodontids and the delphinoids, but a lot does not add up and we need some earlier fossils from lower levels of the Pisco Formation.
The disappointing fossil record of Narwhals (Monodon)
There are surprisingly few clear fossils of narwhals despite a century and a half of dedicated research and searching of marine mammal fossils. The few reported occurrences mostly consist of slap-you-in-the-face-obvious examples - fossil tusks. The earliest such reported example is a tusk fragment from the "Cromer Forest Bed", a peaty shallow marine/coastal deposit with many middle Pleistocene land mammals dating to about 500,000-800,000 years old, and exposed in Norfolk on the east coast of the UK. This specimen was reported by Newton (1882), and there is sadly not a pdf available online with a good image of the specimen illustration. I myself received a copy on loan during my time at U. Otago in the rare books library, and had to use an archival pillow and gloves to read it. It was in no shape to be scanned. However, as noted by Newton, the tusk fragment clearly had the spiral structure unique to Monodon monoceros. Another specimen that, to my knowledge was not figured by Newton, consists of a rostrum fragment with partial tusk from the same locality.
A much more interesting specimen was reported and figured by Post and Bosselaers (2017), though it is quite fragmentary as well. The specimen consists of a rostrum fragment - the left maxilla to be precise - with an elongate, straight tusk embedded within. This specimen was dredged off the seafloor near the mouth of the Schelde Estuary in the Netherlands in 1961 during an expedition by G. Kortenbout and C.J. Overweel, curators of the National Museum of Geology and Mineralogy in Leiden, Netherlands, targeting Pleistocene land mammals; marine mammal material was generally considered to be fossil "bycatch" and just put in boxes. Post and Bosselaers (2017) sorted through this large collection and identified a number of interesting specimens of middle Miocene through early Pleistocene age, including this The specimen is a bit smaller than what you'd see in an adult narwhal, however, it is "modern-sized" in the sense that it absolutely dwarfs most of the fossil belugas mentioned above, corresponding to a rostrum that was likely about 24-25 cm wide and likely 35-40 cm long. The tusk is small, and at least 30 cm long at death, suggesting that this is either a female, juvenile male, or perhaps that this early narwhal had a shorter, smaller tusk. These authors identify the specimen as Monodon sp., and consider the specimen to probably be derived from a late Pliocene or early Pleistocene unit, owing to similarity in preservation with the walrus Ontocetus emmonsi from the same collection - maybe from a unit like the Lillo Formation or Westkapelle Ground Formation. This age makes this specimen quite interesting, as it is quite old - at least 2 Ma (there are no middle Pleistocene marine rocks in the Netherlands, and the specimen is mineralized, unlike poorly preserved late Pleistocene specimens; Post and Bosselaers, 2017) - and also quite far south (southern North Sea, approx. 51N).
And, lastly, there is the single skinny monodontid humerus from the Yorktown Formation reported by Kazar and Bohaska (2008), but not figured. And, we're at the end of the narwhal fossil record.
Were fossil white whales adapted to warm or cold water?
Ever since the first fossil monodontid, Denebola brachycephala, was named from the warm temperate Almejas Formation of Baja California (latitude 28N), there has been widespread recognition that monodontids have, in the past, not been tied to polar climates. Specimens from the North Sea are preserved in strata like the Kattendijk Formation deposited under relatively cool sea surface temperatures, around 7C (Lambert and Gigase, 2007), slightly cooler than today, and much cooler than the environment off the Carolinas during deposition of the contemporaneous Yorktown Formation, 13-24C. Still, cold temperate waters of the North Sea being slightly cooler than today is hardly polar.
Ichishima et al. (2018) presented some interesting arguments turning much of this on its head. They pointed out that while Isla Cedros is far south, it is located on the cool water California Current, and was likely not subtropical in terms of its climate - reinforced by contemporary microfossils from elsewhere in Baja California. Likewise, something not highlighted by Ichishima et al. (2018), are the fossil fish listed by Barnes (2008) suggesting more temperate conditions typical of the coastline of northern Baja and southern California - further suggesting against a strictly subtropical climate. They did, however, note that alcids (puffins and auks) suggest a cooler climate (think central California - alcids are quite rare here in southern California; for example, there are more inaturalist records for boobies in San Diego County than there are for alcids); they also highlighted the presence of Hydrodamalis cuestae in the Almejas Formation, which fed on kelp - suggesting nearby kelp forests requiring cool water. These authors further pointed out that during the Pliocene, there are short climatic oscillations between warm and cool climates, perhaps explaining why other interpretations of the Kattendijk Formation range between warm and cool temperate: different studies may be sampling different stratigraphic horizons. The Pliocene is, after all, the last phase of earth's history without a northern hemisphere ice cap, and as soon as it begins to form towards the end of this period, climatic fluctuations become more pronounced. Are southerly-located fossil belugas truly associated with other fossils of warm water species, or are they preserved in strata corresponding to short-term pulses of cold conditions? This very important question was raised by Ichishima et al. as most prior studies only considered formation-level climatic assignments. The Embetsu Formation which produced their new species, Haborodelphis japonicus, was also found in rocks the same age as occurrences of Hydrodamalis elsewhere in Hokkaido (Takikawa Formation) and crabs from the Chippubetsu Formation similarly suggest a cold temperate climate. They acknowledge that Bohaskaia monodontoides likely inhabited warmer than present waters, however.
The discovery of Casatia thermophila was a direct response to this hypothesis by Bianucci et al. (2019). Casatia is quite clearly preserved in an environment characterized by quite warm waters. These waters were tropical to subtropical, owing to the presence of bull sharks (Carcharhinus leucas), tiger sharks (Galeocerdo aduncas), seagrass-feeding dugongids (Metaxytherium subappeninum), and the warm water clam Pelecyora gigas (Bianucci et al., 2019). Because the Mediterranean is low latitude and nearly a closed basin, arguments cannot really be made for Casatia to be migratory, either. If their phylogenetic analysis is correct, the warm water habit of Casatia thermophila and its sister taxon relationship with Delphinapterus suggests that high arctic polar water adaptation of white whales may have occurred separately in belugas and narwhals sometime during the Pleistocene, rather than earlier. However, there may be more to the story (see below).
The "Narluga" - a model for a tuskless Monodon ancestor?
In the late 1980s, an Inuit hunter killed a trio of strange narwhals off the west coast of Greenland where he lived. These narwhals were not spotted but evenly gray in color. Most of the whales taken in the summer (for subsistence) One of these skulls was acquired by Jens Larsen; it was so unusual that Larsen stuck it to the roof of his tool shed as a trophy. When visiting Larsen's village of Kitsissuarsuit in March 1990, marine mammalogist Mads P. Heide-Jorgensen noted the skull and asked Larsen all about it, eventually negotiating loan and study to the Zoological Museum of Copenhagen. The skull is strange - it is large, nearly 10 cm longer than average narwhal, has a wide rostrum apex like a narwhal with a blunt, rectangular end, differing from what you see in the more triangular rostrum of a beluga. It has raised premaxillary eminences and a vertex more similar to a narwhal, but it has multiple teeth. However, these teeth are fewer than in a beluga - four teeth per mandible, and five per maxilla for a total of 18 (v. 10 per quadrant in Delphinapterus; note that Monodon typically never has any lower teeth), and they are elongated, procumbent, and there are greatly enlarged canines - considered by Heide-Jorgensen and Reeves (1993) to be analogous to the unerupted tusks of adult narwhals. However, these tusks are positioned further forward - they are erupted from the skull, and the posterior tip is anterior to the eye socket, rather than extending to overlap with the eye socket. This specimen was proposed to be a narwhal-beluga hybrid by Heide-Jorgensen and Reeves (1993), and more recently confirmed using genetics. The two species are sympatric, broadly similar in external body shape, closely related, and - and this is key - share the same breeding season and general location. The interaction of narwhal and beluga genetics is also pretty fascinating: there's some gene in the narwhal that has not only produced tusks but made all of the teeth, including the mandibular teeth, longer and larger than seen in a beluga. Additionally, at least one pair of the mandibular teeth are spiraled like a corkscrew (it is, however, unclear if the upper canines of this specimen are spiraled).
What did the ancestral narwhal look like, and did it resemble the 'narluga'? The intermediate nature of the hybrid makes for an attractive option for a potential common ancestor, or at least, ancestral tuskless or short-tusked narwhal. Narwhals almost certainly evolved from an ancestor that resembled a beluga in possessing a high tooth count and lacked premaxillary teeth, since that's common to both. The earliest ancestral narwhals must have had some sort of incipiently enlarged canine - large enough to erupt from the gums - and at some point, this canine must have been large and deeply embedded enough to be used for some sort of fighting. It's important to note that Heide-Jorgensen and Reeves (1993) considered the canine tusks to not even be erupted from the gums - an unlikely situation for an ancestral narwhal. Instead, I think some sort of jousting with the teeth in a beluga-like ancestor likely drove elongation of the roots of these teeth and eventual enlargement of them. If competition consisted of biting, as is present in many other species of odontocetes, more ritualized 'jousting' could be a selection pathway towards enlarging these anteriormost teeth. This hypothetical short-tusked early form is broadly analogous to beaked whales, which have short, stubby mandibular tusks used for fighting and have lost the rest of the dentition, with the exception of the Sheperd's beaked whale (Tasmacetus sheperdi), which has a homodont dentition and a pair of mandibular tusks. Many ancestral ziphiids similarly had tusks and a decidedly dolphin-like dentition. If ziphiids are a good analogy, and I think they are, as soon as a short tusk emerged even more than a few centimeters from the soft tissues, it could have been used for male-male combat with proto-narwhals sparring and leading towards runaway sexual selection as soon as some males started fighting with even incipiently enlarged teeth - leading to the objectively ridiculous tooth seen in modern narwhals. Whether that tooth elongation gene applied to the other tooth positions as it does in the narluga is unknown.
Is it possible that the tusk evolved for sensory perception? It's possible, and while Nweeia et al. (2014) highlighted the sexually different diet - it's important to remember that the additional fish species consumed by males are deeper living fish, and the larger body size of males confers a greater diving depth. Again, I think that having each sex with wildly different sensory capabilities doesn't make much sense as an evolutionary driver. So, I'm not sold on that. Though limited, the observations of sparring between male narwhals, and the much greater body of evidence for tusk breakage, scarring, and the very well supported correlation between gonad mass and tusk mass - instead supports sexual selection for male "weaponry" seems to be the most likely driving cause for the evolution of the tusk in narwhals. I strongly suspect that the sensory adaptations may have evolved later, either as an adaptation - or perhaps entirely by accident or exaptation. What about visual display? Displaying teeth to females is a possibility as well - all it would take are a few females to have shown mate choice after some male proto-narwhals with ancestral beluga-like teeth flashed their pearly whites, and females choosing those with larger teeth, and bam - within a few generations you're on your way to short tusks.
One other suggestion on the topic of sensory perception. Recently - well, jesus, a decade ago now - Pavel Gol'din (2014) published a pretty thought provoking study where he hypothesized that the extremely diverging morphologies of beaked whale (Ziphiidae) rostra, both among modern species and among some of the absolutely whacked out fossils like Globicetus and Tusciziphius, functioned as internal display structures that were obvious during echolocation. In other words, Pavel pointed out something nobody else had really considered: because of the way sound travels through soft tissues and bounces off bone rather than skin and muscle, odontocetes effectively have a form of "X-ray vision". This is the "antlers inside" hypothesis: echolocating beaked whales might be able to identify members of their own species by pinging them with sonar and seeing the unique skull structures that distinguish their own species - ironically, in a way similar to what paleontologists treat them. Could the initial enlargement of the narwhal tusk have arisen to distinguish between the "ur-narwhal" and other more beluga-like white whales? I actually think this is unlikely, because the unerupted tusk is embedded deeply within the rostrum, so
How to test between combat and visual display? I think we need an early "proto-narwhal" fossil, caught in the proposed short-tusked stage; if they're kind of skinny and delicate - too delicate for combat - you've got your answer. If, however, they're like early walrus tusks - short and stout - that would support the beaked whale-informed combat hypothesis. Sadly, these fossils don't exist yet, and it's possible they never will, which brings us to the last section...
Why are true narwhal fossils so rare?
There's a pretty simple answer here, and it's devastatingly sad. This blog post has more questions than answers, but I think I have an answer for why there are so many questions. It's possible that narwhals have a longer fossil record, but have been cold water taxa during the entirety of their evolution. There are perilously few Pre-Pleistocene marine mammal fossil localities (and specifically, Pliocene age) further north than 40 degrees north in the western North Atlantic, 55N in the eastern North Atlantic/North Sea, 38N in the eastern North Pacific, and ~45N in the western North Atlantic. Why is this the case? The very thing that the modern narwhal is adapted to - the Arctic - is also the site of extreme glaciations over the past two and a half million years. In much of the world north of 50N, glaciers have scraped the soft Neogene 'scum' off of the harder basement rocks, grinding away sediment and fossils alike and dumping the sludge into large piles of debris (moraines), many of which are either completely submerged or are extant piles that have turned into large sand bars (e.g. Long Island, Cape Cod). Otherwise, ice literally covers much of the ground at these high latitudes, so if there are good fossil bearing rocks they might be completely covered. That ice also makes the erosion and weathering process different and difficult to adapt to as a field scientist - and, most critically, an unrelenting pain in the ass to explore. The Pacific coast is obviously ice free but highly uplifted north of San Francisco, and very few fossiliferous basins of the right age for pre-glacial marine mammals exist. There is only one fossil site that I know of within the Arctic Circle that has marine mammal fossils older than late Pleistocene - the Gubik Formation of Alaska - and it is difficult to get to, remote, cold, hazardous, infested with mosquitos, and generally unpleasant. I imagine that Scandinavia and Greenland likely formerly had some interesting deposits, but those have all since been scraped away; the Tjornes Formation of Iceland, where we reported a Pliocene right whale from a few years ago (Field et al., 2017), is a possible place to target. It's possible that continued exploration of the Aleutians, Sakhalin, Kamchatka, and other coasts in the Russian Far East (e.g. Chukotka) may yield more fossiliferous deposits. But - if I had to wager where I think the next informative discoveries of fossil narwhals will be made? I think the Netherlands, after the specimen reported by Post and Bosselaers (2017). This single specimen demonstrates that maybe not *all* early narwhals strictly inhabited high latitudes. Perhaps further exploration of Hokkaido might produce some more interesting monodontids.
References Cited
Barnes, L. G. (1984). Fossil odontocetes (Mammalia: Cetacea) from the Almejas Formation, Isla Cedros, Mexico. PaleoBios, 42, 1-46.
Barnes, L. G. (2008). Miocene and Pliocene Albireonidae (Cetacea, Odontoceti), rare and unusual fossil dolphins from the eastern North Pacific Ocean. 41, 99-152.
Benites-Palomino, A., Velez-Juarbe, J., De Gracia, C., & Jaramillo, C. (2023). Bridging two oceans: small toothed cetaceans (Odontoceti) from the Late Miocene Chagres Formation, eastern Caribbean (Colon, Panama). Biology Letters, 19, 20230124.
Best, R. C. (1981). The tusk of the narwhal (Monodon monoceros L.): interpretation of its function (Mammalia: Cetacea). Canadian Journal of Zoology, 59, 2386-2393.
Bianucci, G., Pesci, F., Collareta, A., & Tinelli, C. (2019). A new Monodontidae (Cetacea, Delphinoidea) from the lower Pliocene of Italy supports a warm-water origin for narwhals and white whales. Journal of Vertebrate Paleontology, 39, e1645148.
Brear, K., Currey, J. D., Pond, C. M., & Ramsay, M. A. (1990). The mechanical properties of the dentine and cement of the tusk of the narwhal Monodon monoceros compared with those of other mineralized tissues. Archives of Oral Biology, 35, 615-621.
Field, D. J., Boessenecker, R. W.,
Racicot, R. A., Asbjornsdottir, L., Jonasson, K., Hsiang, A. Y., . . . Vinther,
J. (2017). The oldest marine vertebrate from the volcanic island of Iceland: a
partial right whale skll from the high latitude Pliocene Tjornes Formation. Palaeontology, 60, 141-148.
Geisler, J. H., McGowen, M. R., Yang, G., & Gatesy, J. (2011). A supermatrix analysis of genomic, morphological, and paleontological data from crown Cetacea. BMC Evolutionary Biology, 11(112), 1-33.
Gerson, H. B., & Hickie, J. P. (1985). Head scarring on male narwhals (Monodon monoceros): evidence for aggressive tusk use. Canadian Journal of Zoology, 63, 2083-2087.
Heide-Jorgensen, M. P., & Reeves, R. R. (1993). Description of an anomalous monodontid skull from west Greenland: a possible hybrid? Marine Mammal Science, 9, 258-268.
Ichishima, H., Furusawa, H., Tachibana, M., & Kimura, M. (2019). First monodontid cetacean (Odonoceti, Delphinoidea) from the early Pliocene of the north-western Pacific Ocean. Papers in Palaeontology, 5, 323-342.
Kazár, E., & Bohaska, D. J. (2008). Toothed whale (Mammalia: Cetacea: Odontoceti) limb bones of the Lee Creek Mine, North Carolina. Virginia Museum of Natural History Special Publication, 14, 271-324.
Lambert, O., & Gigase, P. (2007). A monodontid cetacean from the early Pliocene of the North Sea. Bulletin de l'Institut Royal des Sciences Naturelles de Belgique Sciences de la Terre, 77, 197-210.
McGowen, M. R., Spaulding, M., & Gatesy, J. (2009). Divergence date estimation and a comprehensive molecular tree of extant cetaceans. Molecular Phylogenetics and Evolution, 53, 891-906.
Merella, M., Collareta, A., Granata, V., Casati, S., & Bianucci, G. (2022). New remains of Casatia thermophila (Cetacea, Monodontidae) from the lower Plicoene marine vertebrate-bearing locality of Arcille (Tuscany, Italy). Rivista Italiana di Paleontologia e Stratigrafia, 128, 229-240.
Nweeia, M. T., Eichmiller, F. C., Hauschka, P. V., Donahue, G. A., Orr, J. R., Ferguson, S. H., . . . Kuo, W. P. (2014). Sensory ability in the narwhal tooth organ system. The Anatomical Record, 297, 599-617.
Nweeia, M. T., Eichmiller, F. C., Hauschka, P. V., Tyler, E., Mead, J. G., Potter, C. W., . . . Black, S. R. (2012). Vestigial tooth anatomy and tusk nomenclature for Monodon monoceros. The Anatomical Record, 295, 1006-1016.
Nweeia, M. T., Nutarak, C., Eichmiller, F. C., Eidelman, N., Giuseppetti, A. A., Quinn, J., . . . Angnatsiak, D. P. (2009). Considerations of anatomy, morphology, evolution, and function for narwhal dentition. In I. Krupnik, M. A. Lang, & S. E. Miller (Eds.), Smithsonian at the Poles: Contributions to International Polar Year Science (pp. 223-240). Washington, D.C.: Smithsonian Institution Scholarly Press.
Ochoa, D., Salas-Gismondi, R., DeVries, T. J., Baby, P., Muizon, C. d., Altamirano, A., . . . Carré, M. (2021). Late Neogene evolution of the Peruvian margin and its ecosystems: a synthesis from the Sacaco record. International Journal of Earth Sciences, 110, 995-1025.
Post, K., & Bosselaers, M. (2017). Cetacean fossils from a 1961 expedition at the Schelde estuary, province of Zeeland, the Netherlands. Cainozoic Research, 17, 11-21.
Silverman, H. B., & Dunbar, M. J. (1980). Aggressive tusk use by the narwhal (Monodon monoceros L.). Nature, 284, 57-58.
Velez-Juarbe, J., &
Pyenson, N. D. (2012). Bohaskaia
monodontoides, a new monodontid (Cetacea, Odontoceti, Delphinoidea) from
the Pliocene of the western North Atlantic Ocean. Journal of Vertebrate Paleontology, 32, 476-484.
No comments:
Post a Comment