The Hidden Necks of Seals

Harbor Seal at Roger Williams Park Zoo

I’d bet that most casual observers of Harbor Seals (Phoca vitulina) are under the impression that the little pinnipeds are almost neckless. Of course, it’s not immediately obvious what the skeletal proportions are underneath that blubber, but judging by the forelimb origin, it would seem the neck is less than half the length of the head. Harbor Seals are hiding more neck than what they let on, as their cervical vertebrae are actually similar in length to the skull:

Harbor Seal skeleton from Museum of Comparative Zoology

Clearly the live animal above was not holding its neck in the same posture as this skeleton, and (not having X-Ray specs) I’d assume it was similar to the deep “S” curve in this Harp Seal. As for why seals would shorten their apparent necks so dramatically, King (1983) suggested phocids (“true” seals) require a spindle-like shape when swimming since their propulsion comes from oscillating the posterior end of their body. But why bother having a neck at all? A strongly retracted neck gives seals “slingshot potential” to capture prey (Rommel & Reynolds 2002), as alarmingly demonstrated by this Leopard Seal. It’s surprisingly difficult to find photographs of Harbor Seals showing off their full neck — presumably they only do it rarely and briefly — but blogger Kitty Kono has an amazing snapshot:

From Kitty Kono’s blog Back in the U.S.A. Used with permission.

I of course covered all this in my Weddell’s Long-Necked Seal article, but now, I’ll take things further by determining just how long seal necks are. The tails and sacral vertebrae varied much more than what I expected (plus one source often lumped the two), so rather than measure necks relative to the length of the entire spine, I did so relative to only the thoracic and lumbar vertebrae (T–L from here on). And yes, there is going to be an appendix to this blog post.

Black = Skull (gray for no data); Red = Cervical, light blue background equal to shortest neck, pink equal to longest neck; Blue = Thoracic; Green = Lumbar; Yellow = Sacral; Orange = Caudal.
Species are (top to bottom): Dog, Bearded Seal (Erignathus), Weddell Seal (Leptonychotes), Southern Elephant Seal (Mirounga leonina), Leopard Seal (Hydrurga), Hooded Seal (Cystophora), Harbor Seal (Phoca vitulina), Ringed Seal (Pusa hispida), Harp Seal (Pagophilus).

Phocid necks range from 21% T–L in Bearded Seals to 35% T–L in Harp Seals*, which falls short of dogs with necks 40.5% of their T–L. Weddell seals were initially described as “long-necked”, so it’s amazing to see they’re at the low end of the spectrum with necks only 23% T–L. Contemporary sources almost always describe Leopard Seals as “long necked”, however their necks are only 29% T–L, so perhaps “thin-necked” would be a more apt description. The extinct seal Acrophoca is also typically described as “long-necked”, although judging from this skeletal illustration and mounted specimen, the neck is around 30% T–L. The closest true seals have to long necks come from members of the “tribe” Phocini (Phoca, Pusa, Pagophilus), although I have no idea why that would be the case.

* Piérard (1971) cites a source giving 35% for Harbor Seal T–L, so my figure above is probably a shorter-necked young individual.

Northern Fur Seals (Callorhinus ursinus) at Mystic Aquarium.

As for otariids (“eared seals”) — sometimes described as having “snakelike” necks (e.g. Riedman 1990) — do they truly have long necks, or is this a deception from thin necks held out straight? Otariids require a considerable amount of mass before and after their foreflippers (their source of propulsion) for stability, and so they keep their necks out fairly straight (King 1983). Phocid and otariid necks have been described as “similar” in length (Rommel & Reynolds 2002), although just to be on the safe side, I’ll quantify this as well:

Same deal as before, but now with: Walrus (Odobenus rosmarus), Australian Sealion (Neophoca cinerea), Australian Fur Seal (Arctocephalus pusillus doriferus), California Sealion (Zalophus californianus), Steller’s Sealion (Eumetopias jubatus), Northern Fur Seal (Callorhinus ursinus).

Otariid necks vary little, from 34.5% T–L in Australian Sealions (and most other being only slightly higher) to 41% T–L in Northern Fur Seals. So there is certainly quite a pronounced difference in neck length between Bearded Seals and Northern Fur Seals, it does not seem that the average phocid, otariid or walrus is not truly that different in neck length. It should be cautioned that proportional neck length can change considerably during maturity, so there is certainly some error in these figures. The Bearded Seal was reportedly an old female (Hayden 1880) so if anything, it may have had a longer neck than average.

One thing I haven’t mentioned so far is that pinnipeds have far longer tails than most observers would expect, ranging from 13.5% T–L in Northern Fur Seals to 41% T–L in Ringed Seals. I really have no idea why this would be the case.

References:

Allen, J. (1880) History of North American Pinnipeds. Available.

King, J. (1983) Seals of the World.

Piérard, J. (1971) Osteology and Myology of the Weddell Seal Leptonychotes weddelli (Lesson, 1826). Available. IN: Burt, W. (editor) Antarctic Pinnipedia.

Riedman, M. (1990) The Pinnipeds: Seals, Sea Lions, and Walruses.

Rommel, S. & Reynolds, J. (2002) Skeletal Anatomy IN: Perrin, W. et al. (eds.) Encyclopedia of Marine Mammals. Relevant Passage.

	Skull	Cerv.	Thor.	Lumb.	Sac.	Caud.
Dog (King)	?	17.00%	24.00%	18.00%	4.00%	37.00%
Bearded (Hayden)	230	250	800	390	175	350
Weddell (Piérard)	0	274	820	385	150	385
S. Elephant (Hayden)	480	570	1690	670	250	680
Leopard (King)	?	17.00%	37.00%	22.00%	3.00%	21.00%
Hooded (Hayden)	265	275	630	320	190	290
Harbor (Hayden)	220	210	445	216	120	230
Ringed (Hayden)	163	200	410	190	100	245
Harp (King)	?	19.00%	37.00%	17.00%	7.00%	21.00%
Walrus	390	400	1170	380	?	550
N. Fur (Hayden)	275	430	770	270	160	140
N. Fur (Hayden)	245	360	680	245	145	145
N. Fur (Hayden)	200	200	520	185	105	160
N. Fur (Hayden)	185	172	470	173	95	120
Aus. Sealion (King)	?	20.00%	43.00%	15.00%	7.00%	15.00%
Aus. Fur (King)	?	21.00%	44.00%	15.00%	6.00%	13.00%
Cal. Sealion (Hayden)	236	320	640	230	?	280
Steller’s (Hayden)	374	500	1050	340	?	440
Steller’s (Hayden)	385	540	1090	400	?	520

Data from King (1983) were in percentage of the entire spine with no data on skull length. Some Data from Hayden (1880) is in millimeters, and for instances where sacral and caudal vertebrae were lumped, I signified this with “?” in the area for sacral measurement.

Cutting off the Trunk to Spite the Whale

A new look for Makaracetus bidens

In Trunks, Proboscides & Makaracetus, I argued the extinct whale Makaracetus bidens probably didn’t have a “trunk” or “proboscis”, contra Gingerich et al. (2005). For one thing, an overhaul of terminology resulted in ‘trunk’ no longer referring to a nasal appendage, ‘proboscis’ being restricted to tubular nasal/lip structures capable of grasping food, and all other elongate nasal structures being termed ‘prorhiscis’ (Milewski & Dierenfeld 2013). Makaracetus certainly had over-developed facial features compared to its cetaceous relatives, but it didn’t appear any more likely to have a “trunk” than, say, a camel.

Boar skull, courtesy of Markus Bühler.

One feature of Makaracetus I didn’t discuss in the last round were its maxillary fossae, which Gingerich et al. noted were also present in pigs. This feature is plainly visible on the boar skull above — it’s the virtual trench running from the orbits down the snout. These fossae provide attachment for the maxillolabialis superior and inferior, which are responsible for moving the protruding snout (Gregory 1920).

Pig snout musculature. From Gregory (1920)

What’s interesting is that pigs almost fit the definition of having a proboscis. Pigs can apparently grasp food items with both their nose and lip musculature, but fall short of the definition since only the disc is mobile (Milewski & Dierenfeld 2013). Does this mean Makaracetus may have had some sort of proboscis after all? Probably not. It’s notable that one trait shared by proboscides owners is directional olfaction (Milewski & Dierenfeld 2013), which seems like an unlikely trait for a cetacean to evolve, considering they lack the ability to smell underwater. Furthermore, pigs don’t have exclusive rights to maxillary fossae.

Top: Makaracetus from Gingerich et al. (2005)
Middle: Potamochoerus sp. from Gregory (1920)
Bottom: Macropus sp. from Gregory (1920)

Kangaroo maxillary fossae bear a surprising resemblance to those of Makaracetus — they’re similar in extent, only present anterior of the antorbital foramen, and are on the lower half of the rostrum. Unlike pigs, the maxillary fossae of kangaroos provide an origin for part of the buccinator muscle (Gregory 1920). The fossae are also present in baboons, where I suspect they have something to do with this display. Maxillary fossae also appear to be present in llamas but apparently not camels, for some reason. Considering the apparently variable functions of maxillary fossae and the additional premaxillary fossae of even more mysterious purpose, I’d agree with Gingerich et al.’s interpretation of “hypertrophied facial muscles” for Makaracetus and hesitate to speculate much further.

So if the nose of Makaracetus didn’t bear a proboscis or prorhiscis, what did it do? Perhaps the expanded nasal vestibule was merely a side effect of the snout warping into a down-turned shape. Another parallel could be found in Gray Seals, which have nasals retracted to a similar degree as Makaracetus and enlarged noses which function as displays (Miller & Boness 1979); however more specimens will be required to conclude that the trait was sexually dimorphic. Milewski & Dierenfeld (2013) hypothesized that the enlarged nose of Moose may act as a buoy to counterbalance to forelimbs when foraging underwater, a trait which could be beneficial to the apparently benthic-feeding Makaracetus. However, Makaracetus appears to have a much smaller nasal chamber than Moose, so the potential benefits could have been minimal.

I reconstructed Makaracetus as a Hippo/Manatee/Walrus-thing up top, but I suspect my endeavors were far too conservative. Makaracetus was a beast without obvious parallels and its true appearance can probably never be known. Still, there’s no reason to go slapping a “trunk” on strange-looking skulls, and I hope Milewski & Dierenfeld (2013) will get the attention it deserves and open minds to the myriad other possibilities of big, weird noses.

References:

Gingerich, P. et al. (2005) Makaracetus bidens, a New Protocetid Archaeocete (Mammalia, Cetacea) from the Early Middle Eocene of Balochistan (Pakistan). Contributions from the Museum of Paleontology 31(9) 197—210. Available

Gregory, W. (1920) Studies in comparative myology and osteology, No. V. — On the anatomy of the pre-orbital fossæ of Equidæ and other ungulates. Bulletin of the American Museum of Natural History 42 265—283. Available

Milewski, A. & Dierenfeld, E. (2013) A structural and functional comparison of the proboscis between tapirs and other extant and extinct vertebrates. Integrative Zoology doi: 10.1111/j.1749-4877.2012.00315.x

Miller, E. & Boness, D. (1979) Remarks on display functions of the snout of the grey seal, Halichoerus grypus (Fab.), with comparative notes. Canadian Journal of Zoology 57 140—148. Available

Trunks, Proboscides & Makaracetus

The terms ‘trunk’ and ‘proboscis’ have been applied to a heterogeneous array of bits danging off vertebrate faces such as the bulbous nose of male Proboscis Monkeys, the pendulous resonating chamber of male Elephant Seals, the hydrostatic facial tentacle of proper elephants, and even the chins of Elephantnose Fish. Milewski & Dierenfeld (2013) argue the terminology has become “dysfunctionally vague” and in need of a massive overhaul. The authors trash the term ‘trunk’ as it is vague and redundant (being a synonym for ‘torso’) and lacks precedence over ‘proboscis’, which was originally applied to elephants and roughly translates as ‘forager’. Due to this etymology and the extreme morphology exhibited by elephants, the term ‘proboscis’ was redefined to be a tubular extension of nasal and lip musculature capable of grasping food (Milewski & Dierenfeld 2013). The only extant non-Proboscideans to fit this definition are tapirs, and while numerous extinct mammals have been interpreted with proboscides (Palorchestes, Cadurcodon, Brachycrus, Macrauchenia, Pyrotherium) only Astrapotherium has a good case for having a proboscis under this new definition, since it doesn’t appear to have been capable of feeding itself otherwise.

This Mountain Tapir demonstrates that if narial tubes are wrapping around food, it’s a proboscis. From Wikipedia Commons.

As for other nasal structures that are incapable of grasping food, those have now been termed ‘prorhiscis’. These, uh, ‘prorhiscides’ (?) have disparate functions ranging from a directional sense of smell in Elephant-Shrews, dust filtration in Saiga, thermoregulation and water conservation in Dik-Diks, amplification of roars in Elephant Seals, and possibly acting as a buoy for diving Moose (Milewski & Dierenfeld 2012). There are other weird structures formerly labeled ‘proboscis’ in fish, but that’s really a story for a different time.

The snout of a Saiga rivals that of Tapirs in length, but does not incorporate the lips or grasp food, and is thus a prorhiscis and not proboscis. From Wikipedia Commons.

Gingerich et al. (2005) interpreted the extinct whale Makaracetus as having a “trunk” or “short, muscular proboscis” but confusingly referenced tapirs, manatees and walruses as “imperfect models”. Were they proposing an elephantine proboscis, prehensile lips, or something in between? This is a great example of how the now-archaic terminology was “dysfunctionally vague” and in light of Milewski and Dierenfeld’s efforts, it’s time for a reassessment.

Dorsal and lateral views of: Makaracetus (top) from Gingerich et al. (2005); Artiocetus (bottom) from Gingerich et al. (2001)

Makaracetus is classified as a protocetid and appears fairly similar to species such as Artiocetus (above), although with four pronounced differences: a nasal vestibule extending to the end of the snout, a downward-deflected rostrum with two rather than three pairs of incisors, “extraordinary” fossae (labeled LFM, LFP above) suggesting massive facial muscles, and enlarged antorbital canals (labeled AF above) indicating increased blood supply to the end of the snout (Gingerich et al. 2005). Clearly something odd was growing on the face of Makaracetus, but without living protocetids, it’s a bit hard to tell just what. I’m not even certain if a Tapir-style proboscis can be distinguished from a prorhiscis without live specimens, so, unfortunately, it appears there’s just no way of knowing things for certain. That won’t stop me from rampantly speculating.

Top: Makaracetus
Second: Brazilian Tapir from Witmer et al. (1999)
Third: West Indian Manatee from Husar (1994)
Last: Walrus from Wikipedia Commons

Imperfect models indeed. Gingerich et al. (2005) compared Makaracetus and Tapirs on the basis of expanded nasal vestibules, however, the nasals of Makaracetus are only around half as retracted. Makaracetus and Manatees were compared on the basis of being aquatic and having down-turned rostrums, which Gingerich et al. suggested to be an indicator of benthic feeding. Walruses were suggested as an ecological model, namely as a species that uses facial muscles to prey on bivalves, despite having some very different anatomy. The proposed ecology seems reasonable — although it requires better-known teeth to confirm — but as for what was happening with the nose of Makaracetus, there are some more interesting models.

Dromedary from Harvard Museum of Natural History.

Superficially, Makaracetus looks kinda camel-y. The rostra of the two species, while differing considerably in depth, appear to be deflected downward at a similar relative place and to a similar degree. Camels have one less pair of incisors than Makaracetus and, oddly, the placements of the remaining incisor pair, canine, and first premolar are comparable. The antorbital foramen (“AF”) of the camel appears to be larger than that of Makaracetus. There are of course several pronounced differences between the two, namely, the fossae in Makaracetus (LFM, LFP) do not appear to have equivalents in camels and, perhaps as a result of these structures, Makaracetus has a rostrum that looks spoon-like when viewed above and camels don’t. Makaracetus and camels undoubtedly had enlarged noses for very different reasons, but this comparison makes a proboscis or even prorhiscis seem unlikely in the whale.

But I’m not quite done with Makaracetus yet, I haven’t even talked about pigs and moose! That will have to wait for a followup to this increasingly out-of-control article.

References:

Gingerich, P. et al. (2005) Makaracetus bidens, a New Protocetid Archaeocete (Mammalia, Cetacea) from the Early Middle Eocene of Balochistan (Pakistan). Contributions from the Museum of Paleontology 31(9) 197—210. Available

Gingerich, P. et al. (2001) Origin of Whales from Early Artiodactyls: Hands and Feet of Eocene Protocetidae from Pakistan. Science 293 2239—2242. Available

Husar, S. (1978) Trichechus manatus. Mammalian Species 93 1-5. Available

Milewski, A. & Dierenfeld, E. (2013) A structural and functional comparison of the proboscis between tapirs and other extant and extinct vertebrates. Integrative Zoology doi: 10.1111/j.1749-4877.2012.00315.x

Witmer, L. (1999) The proboscis of tapirs (Mammalia: Perissodactyla): a case study in novel narial anatomy. Journal of Zoology 249 249—267. Available